When you use a laptop, you might assume that it will always perform the same, whether you use it while plugged in or unplugged. After all, these systems are designed to be used on the go. But for a long time, that hasn't been the case.

More recently, some silicon vendors have been boasting about closer parity in plugged and unplugged performance, offering you similar power whether you're working at your desk or anywhere else. To see the current state of this, we've taken samples of the latest laptop from each major CPU manufacturer: Intel, AMD, Qualcomm, and Apple, and run a few tests to see how they compare.

For the Windows systems, we used the balanced power profile, which was already the default on the majority of these laptops, while the MacBook Pro has an "automatic" power management profile out of the box.

Where we could, we went for top-end processors, including the Intel Core Ultra X9 388H, Qualcomm Snapdragon X2 Elite Extreme, and Apple's M5 Max. In AMD's case, we had two Gorgon Point processors, a Ryzen AI 9 465 and a Ryzen AI 7 445. To see some generational gains, we also included an Intel Core Ultra 7 258V "Lunar Lake" chip.

We tested three ways: in Cinebench 2026 single-core and multi-core workloads, as well as using Handbrake, an open source video conversion tool, to transcode a 4K video to 1080p.

Note that we're not comparing the scores of the systems to each other. We're comparing the systems to themselves, to see how performance changes on the battery.

With those tests run, we can calculate percentage differences to see percentage changes. Notably, a number of systems actually increased performance ever so slightly on battery. This occurred the most on multi-threaded Cinebench, where the Zenbook A16 with Snapdragon X2 Elite Extreme increased by 13.06%, while the ThinkPad X9 bumped up by 4.98%.

Qualcomm's chip was by far the most variable. While it dramatically increased in performance on battery for Cinebench nT, it dropped a similar amount for Handbrake.

But it was Apple and Intel that led the pack. While the Core Ultra X9 388H had the most consistent score on average, it did show 3% swings in multi-threaded Cinebench and on Handbrake. Apple's M5 Max results were the most consistent throughout, never swaying even a full percentage point on battery.

The least consistent system on battery power was the Acer Swift Go 16 AI, running on an AMD Ryzen AI 7 445. That system lost a ton of performance when unplugged from the wall, dropping as much as 56.25% performance off the wall. On average, benchmark scores were 50% lower on that system. On the HP OmniBook X Flip 14, Ryzen AI 9 465, it still dropped, but far less, losing an average of 5.45% performance on power based on these tests.

On the bright side, things are going in the right direction. Intel and Apple's most advanced systems give you very similar experiences on battery and power — at least in these tests. Snapdragon can, but sometimes it swings a bit. It's really AMD that needs to catch up in this regard. And we know that can be done. The Steam Deck, which was released in 2022, was specifically designed to have similar performance on and off the charger, and it uses a semi-custom AMD SoC. Its AMD-based opponents, however, still get better performance on a charge. You not only have to have your chip designed around battery and performance. You also need to make sure that the system around it is optimized for that.

The other side of this coin is efficiency. If you're able to perform heavy tasks on battery without losing performance, you're likely going to do it. That's especially the case if you do field work in creative or industrial fields. That means ensuring that your system can handle your workload long enough to get it done. But hey, there's always an external battery, right?

Batteries that can last decades between recharges or renewals are the stuff of tech enthusiast dreams. However, DARPA’s ‘Rads to Watts’ program could bring forth unrivaled portable power, lasting decades, via the development of a new class of radiovoltaic devices. Morgan State University recently won a $3.37M DARPA contract to advance this tech, and a report by Defense One suggests upcoming nuclear waste-powered batteries could last as long as 30 years and power next-gen drones.

Morgan State now has a considerable sum to keep its researchers employed on a project dubbed SYMPHONEE (Strontium-Yttrium Multi-junction PIN-based High-Density Output Nano-system for Extreme Environments). Radioisotope power sources have been used for quite some time in low-power draw applications such as smoke detectors and space systems for decades. However, the new initiative, led by Morgan State and with team members such as Northrop Grumman, PNNL, Project Omega, ARA, and Widetronix, aims to take the tech into the next generation. More specifically, under the auspices of (and with funds from) the 'Rads to Watts' program, they are aiming to build next‑generation radiovoltaic micro‑power systems for extreme environments.

The proposed next-gen power cells make use of radioisotopes extracted from nuclear waste, especially Strontium‑90. Research partner Project Omega has already made some devices and will build the DARPA prototype. We noticed that it has published a promotional video that highlights sea drones with a 10-year battery life, for example. Coordinated by Morgan State researchers, PNNL labs will handle nuclear materials/testing, Northrop Grumman and ARA will provide AI-driven modeling, with Widetronix designing nuclear micro‑generator radiovoltaic converter architectures.

“Our team is pushing the boundaries of radiovoltaic technology, developing high-power, long-life systems that were not previously achievable,” said Professor Michael Spencer, the technical lead for the project. “By integrating advanced materials, device engineering, and nuclear science, we are laying the foundation for a new generation of persistent power systems for extreme environments.”

With this research, development, and fabrication powerhouse assembled, it is hoped a step change will be achieved to boost radiovoltaic micro‑power system power density. The goal is to be able to power persistent autonomous operations in remote or difficult-to-access environments like space, underwater, and contested regions – for decades. A Northrop Grumman rep emphasized the potential of a ‘persistent power source’ becoming available for “next-generation defense systems.” Project Omega is also interested in developing “persistent underwater security systems” – sea drones.

It is also somewhat positive to imagine that substances long considered to be highly undesirable waste products could end up being strategic assets.

A Finnish startup’s claims to have created a ‘miracle’ solid-state battery appear to have collapsed under independent scrutiny. Ryan Inis Hughes, host of the Ziroth YouTube channel, worked with “over 20 independent battery experts” to investigate the product from Donut Lab. They concluded that the battery was, in fact, a repacked lithium-ion chemistry one. Probably most worryingly, Donut Lab has collected $25M in investments from thousands of small investors. Authorities are now investigating, reports Electrek, what is already being described as the Theranos of the battery industry.

The video begins with Donut Lab CEO Marko Lehtimäki announcing at CES 2026 that his solid-state battery tech is “going to transform the industry.” Donut Lab’s valuation subsequently rocketed to $1.25B off the back of claims its 400 Wh/kg, 100,000-cycle, 5-minute-charging solid-state battery was shipping "in production vehicles."

This startup, coming out of nowhere, boasting of an achievement that the likes of Toyota and Samsung are still working towards, made some suspicious and others eager to invest. But even before any independent technical analysis of a Donut Lab Solid State battery pack could be completed, cracks started to appear in its claims. For example, a whistleblower from a partner company, the ex-COO of Nordic Nano Group, came forward to allege that the CES 2026 outlined battery specs weren’t accurate.

Results of the investigations

Even before any battery characteristic tests or chemical analyses were done, much of the backstory of the ‘miracle’ solid-state battery was debunked. Adding to the whistleblower’s revelation, a Fraunhofer Research Institute scientist who talked to Donut Lab technology provider CT Coatings was agog that “these people have no idea how a battery actually works.” Then there was the revelation that a motorcycle firm partnership was merely at the development test phase. The Donut Labs batteries were not in the “first production motorcycle,” as claimed at CES and on the firm’s website.

Moving on to the scientific analysis of the Donut Lab Solid State battery, things get even worse for the Finnish startup. Scientists and experts from the aforementioned Fraunhofer, plus a number of universities, “confirmed the tested cell is lithium-ion,” reports Electrek.

Two key pieces of evidence point to high-nickel lithium-ion cells (NCM chemistry) powering the ‘miracle’ solid-state battery. Firstly, the voltage curves precisely match a lithium-ion battery. Secondly, the ‘kink’ in the battery’s physical expansion curve around 50-70% state of charge is exactly as expected for a lithium-ion battery. It was also observed by testers that the advertised 400 Wh/kg density was way off, with measurements indicating the Donut Lab battery packed 298 Wh/kg.

Tech enthusiasts might feel disappointed by the apparent hollowness of the battery startup’s claims. However, one must pity the 1,300+ small investors who have put their savings into Donut Lab under what may have been false pretenses. Finnish financial authorities and criminal authorities are reportedly investigating this case.

The Dell XPS 14, which features an LCD panel with a variable refresh rate (VRR) of 1Hz to 120Hz, is designed to increase laptop battery life endurance. According to tests conducted by Hardware Canucks, the Dell laptop achieved 43 hours and 3 minutes running completely on battery while doing just light Chrome Web Browsing. By comparison, the 15-inch M5 MacBook Air achieved 14 hours and 30 minutes on the same test — a massive difference of 28 hours and 33 minutes.

The XPS also outlasted the MacBook Air in 4K YouTube playback, with the Windows laptop achieving 20 hours and 21 minutes before running on empty, versus the MacBook’s 14 hours and 2 minutes. The only time Apple outshone Dell was in gaming, where the MacBook Air lasted for 4 hours and 10 minutes (versus the XPS’s 2 hours and 38 minutes).

To ensure that the settings between the two laptops were comparable, both devices were set at 150 nits brightness with HDR off. While it does help that the XPS has a slightly larger 70Wh battery compared to the MacBook Air’s 66.5Wh, that still does not explain the massive difference in endurance between the two laptops.

It’s apparent that the LG display found on the XPS 14 drives the power savings, which allowed the Dell laptop to shift down to a 1 Hz refresh rate when a higher rate is unneeded.

We’ve seen impressive results ourselves in our review of the Dell XPS 14 (2026). While we did not achieve the nearly two-day battery life, we were still able to hit 20 hours and 41 minutes in our test. This is the best result we’ve seen in an x86 laptop and beats the M5 MacBook Pro’s 18 hours and 14 minutes endurance. Our tests are a bit different, though, as we combined web browsing, video streaming, and OpenGL tests for our battery run with the screen set at 150 nits of brightness. This is quite a bit more intense than the light web browsing tests conducted by Hardware Canucks.

The only downside to the Dell XPS 14 (2026), which had the amazing battery life, was that it had an LCD panel. While it does give you excellent battery endurance and a higher maximum brightness, it compromises on resolution and color accuracy. The OLED XPS 14 offers 2880 x 1800 pixels and 89.7% coverage of the DCI-P3 color space and 126.6% of the sRGB gamut. By comparison, the LCD model only had a 1920 x 1200 resolution, while its DCI-P3 coverage only sat at 71.4%. Still, that does not mean that you’ll have to pick between battery life and color accuracy in the future, as LG said that it plans to launch an OLED version of this VRR display by 2027.

The Nintendo Switch 2 will reportedly get a removable battery specifically for the European market. According to Nikkei [machine translated], the company is making the hardware revision due to Regulation (EU) 2023/1542, which says that “Any natural or legal person that places on the market products incorporating portable batteries shall ensure that those batteries are readily removable and replaceable by the end-user at any time during the lifetime of the product.” Although the law was passed in 2023, the removable battery clause will not take effect until February 18, 2027.

This change will also affect the Joy-cons, making it easier for users to replace the batteries on the console and its controllers without needing special tools, solvents, or other materials — at least in the EU. Nevertheless, there is hope for the rest of the world, as the company is reportedly planning to make the same change if the U.S. and Japan adopt similar right-to-repair rules.

It’s unclear why Nintendo didn’t build the Switch 2 with a removable battery from the get-go, especially as Europe is its second-largest market after the United States. Still, this is good news for European fans as it would allow them to easily replace the battery on future Switch 2 consoles, letting them keep their gaming handheld for longer.

This is another challenge that Nintendo has faced in recent months. Despite the Switch 2’s record-breaking opening weekend sales, the company is facing weaker game sales. One report blamed the AI-driven chip shortage for making storage more expensive, although others point to pricier titles and a lack of compelling options as the reason gamers are purchasing fewer games. It has even sued the U.S. government over tariffs, demanding a $200 billion refund with interest.

Note that the EU regulation isn’t limited to the Nintendo Switch 2 or gaming handhelds. In fact, it’s expected to affect nearly every portable battery-powered device, including smartphones, laptops, earbuds, headphones, and more. Other companies have already started responding to this change, with Sony making it easier to replace the batteries on its DualSense controllers and the new MacBook Neo being much easier to repair than older MacBook models.

An Australian research team has developed a quantum battery that could be the stepping stone towards a revolution in battery technology. This prototype is only a proof of concept, with a capacity of only a few billion electron volts, which lead researcher Dr. James Quach of the Australian national science agency CSIRO describes as "very small and not enough to power anything useful." However, the work towards a fully functioning prototype, showcased in the Light: Science and Applications journal (h/t The Guardian), is a promising sign of the future.

First, the battery took femtoseconds (or quadrillionths of a second) to completely charge, and stored its energy for nanoseconds. The six orders of magnitude difference that was recorded is, Quach believes, similar to a battery that took a minute to charge but would ultimately stay charged for "a couple of years."

A fully functional quantum battery would, in theory, be able to charge with incredible speed and, using lasers, be remotely charged. Quach mentions its potential application in drone technology, where "you could charge it while it's in flight." Quach also believes it would revolutionize the electric car industry, suggesting that "once the technology matures... you would no longer need to stop your car at a petrol station to charge it up; you could charge it on the go."

It all boils down to how a battery is designed to work. With conventional batteries, the bigger the battery, the longer it takes to charge. Quantum batteries flip that on its head, with what Quach describes as a "really peculiar property" called collective effects, which allows quantum battery cells to charge faster when there are more of them grouped together. Quantum batteries don't rely on the chemical nature of conventional counterparts but instead rely on quantum mechanics, where the strange properties of subatomic particles entangle all of the cells together.

Professor Andrew White, a lead researcher in quantum technology at the University of Queensland, who wasn't involved in this research, is certainly impressed. The "really nice piece of work," White believes, demonstrates that the tech isn't just a theory but "a working prototype." While White believes that the tech won't be turning up in any new electric cars "any time soon," he does instead believe that "probably the first place that it'll have an impact is actually for quantum computers." White goes on to say that quantum batteries would power these machines "coherently... with the minimum energy cost to these computers."

EVs, quantum computers, drone technology — the real-world applications of a battery like this could be revolutionary. CSIRO, the Australian agency behind the research, is already looking for "interest from potential development partners" to help it take it to the next level. With only nanoseconds of energy storage right now, there's some way to go before this experimental technology is ready for prime time, but the researchers are already looking to "overcome that hurdle" as quickly as possible.

Creator and enthusiast ScuffedBits successfully ran a desktop PC for about five minutes on AA batteries alone and was even able to complete one round of Minesweeper on it. However, it seemed that that wasn’t long enough, so he decided to redo the project, intending to address the problems of the first build.

The biggest issue they fixed first was the high resistance of the tiny wires used to connect the batteries to the motherboard. Instead of replacing them, ScuffedBits modified the wiring of the battery packs to get around 25 volts, meaning that the PC would now require lower current, preventing instability because of the thin wiring. Of course, 25 volts isn’t good for the motherboard, so they also added three voltage regulators wired in parallel to get a consistent 12 volts until the batteries died. To further ensure that the system works, they also eliminated the alligator clips and soldered the wires directly to the capacitors that came from the original project.

With everything in place, ScuffedBits inserted the last three AA batteries, completing the circuit and bringing the PC to life. Unlike the last time, which needed an external power source to “jumpstart” the computer, it was able to boot straight up using just the AA cells this time. With the voltage holding steady, they were able to log into Windows and show us the specifications of the PC — an Intel Core i3-530 paired with 8GB of RAM and a WD SATA SSD.

The first thing they did was play one round of A Short Hike, which they finished some three and a half minutes after turning on the computer purely on AA batteries. But what’s more interesting was that they were able to run Cinebench on the system, pushing the rather old CPU to its limits. Even though this processor is rather old, TechPowerUp reports that it still has a power draw of 73 watts. Nevertheless, the system completed the test almost eight minutes after booting up, with no signs of slowing down.

So, after the benchmark, ScuffedBits decided to install Minecraft and play several rounds of Party Games. After that, he tried to install Portal 2, but the system finally died after 33 minutes and 19 seconds — quite a remarkable feat for a home-made AA-battery power solution for a desktop PC. This isn’t groundbreaking technology, especially as we have better battery solutions available to us nowadays. Besides, who would want to purchase 64 AA batteries at a gas station multiple times a day just to keep their laptops fully charged?

The creator also ran a test using a 12-volt car battery on the system, showing that it will work. But they also said that it was going to be quite boring, as the car battery should have enough juice to run the desktop computer for three to four hours without any problems. Nevertheless, the entire thing is still a fun project, tinkering with batteries and PCs, and we hope to see more similarly crazy experiments from ScuffedBits in the future.

A TechTuber who previously hit the headlines after making a home powerwall from a concoction of thrown-away vape components has shared details of his vape battery-powered car. Chris Doel actually repurposed the 500-vape battery powerwall device, put it in an aluminum case, and plugged it into a quarter-century-old electric car. Conclusion: It was a great success, but it is well worth tuning in to enjoy the full journey by watching the full video below.

In the intro, Doel complains about how the throwing away of single-use vapes is still a ‘normalized’ activity. He reminds viewers that as many as a million of these electronic devices are thrown away every day by UK vapers. However, Doel is on a mission to reuse these cast-aside components, specifically the Lithium batteries, in various projects. He started small with powerbanks, and progressed to a mighty powerwall in a previous video, which managed to take his workshop ‘off grid,” and now he’s going mobile.

Doel picked the old Reva G.Wiz electric car from the early 2000s, which he found to be a surprisingly good fit for his 500 vape battery array of power. This compact car weighs 400kg, and had a top speed of about 50 mph, factory fresh. Originally, it used a 48V battery pack using lead-acid technology. So it is a good target for the 50V power supply from the prior vape battery powerwall project.

Contrast that to a modern Tesla electric car. According to Doel, a Tesla would need about 12,000 vape cells and complex battery management to get off the driveway. Moreover, Musk’s autos have peak draw power measured in hundreds of kW.

So, it sounded like the old G.Wiz and the vape cells behind the previous powerwall project were a good match. And indeed it was, as the first power-up test we see in the video was a success. With the battery still on the driveway, Doel was extremely cautious rolling the car forward and back.

At this point, we also saw Doel plug in a charger to top up the vape cells. It was a USB-C charger, and it also worked the first time. So, as well as the world’s first vape-powered car, this little G.Wiz also became the first USB-C charged car.

An original G.Wiz can pull 350A if you put the pedal to the metal, which made Doel concerned that “the fuses will probably start popping,” as the vape battery pack is only good for ~120A, according to his calculations. Luckily, the inverter could be programmed, and the car profiles can be reworked to limit the power to safer levels.

Test drive shows vehicle has an 18-mile range

With the surprisingly quick development and installation of the new vape-powered battery into the G.Wiz, it was time for a test drive. The TechTuber started off slowly, naturally. A smartphone monitor constantly updated on the condition of the vape battery cells. However, the car was paused, and temperatures were thoroughly checked for safety after a short run.

On the road trip, we saw the little car speed up to a maximum of 35 mph, and it was impressive to see that regenerative braking (10A) also worked as intended. With building confidence, the TechTuber drove to a nearby McDonalds drive through, parked up to eat, then went to a tool store. Next up, the car faced its first major uphill section, which saw a power draw of 100A continuous, so the cautious drivers slowed and caused a little traffic jam.

The weather turned worse, and the car’s wipers and headlights were required, eating away at the battery some more. However, the biggest challenge yet was a hill that saw 150A pulled at peak. The car made it. Overall, the vape-powered G.Wiz managed 18 miles before it lost power as battery cell bank 12 gave up. The observed max temperature of the cells was 19 degrees Celsius (it is still winter in the UK).

Doel mused whether the 500 vape smokers ever thought their trash would one day power a house, or a car. The video ended with a punchline, showing the car had a ‘vape exhaust’ installed.

Gaming laptops come with power-dense Lithium-based batteries to give you on-battery gaming for a few hours — but what would happen if you ditched these specialized power cells and instead chose garden-variety AA batteries to run your desktop gaming PC?

This is exactly what YouTube creator ScuffedBits decided to find out when he pulled out the old 450-watt CX430 power supply and replaced it with an ATX power supply plug from AliExpress that lets you connect its 12V cables to practically anywhere you like. Unfortunately, the creator did not share the specifications of their PC, but it seemed to have an entry-level Intel CPU chip, two RAM sticks, and a 2.5-inch SATA SSD running on Windows 10. They also said that they’ll try running a GPU on the system later.

Now, AA batteries are just 1.5 volts, so you need to connect 8 cells in series to hit the required 12 volts. However, we seriously doubt you can run an x86 system with that small a number of batteries. The motherboard’s status did LCD light up when he attached the batteries to the power cable, but when he measured the power draw from the AAs, he measured a measly 0.06 amps. He powered up the system, and the fan spun for a second before nothing happened.

In an attempt to achieve their goal, ScuffedBits added two more 8-battery packs in parallel, but the computer just refused to boot.

In an attempt to solve the problem, they switched from carbon to alkaline batteries, which can output a higher current and last longer. However, ScuffedBits still ran into the same problem: the system would turn on for only a couple of seconds before dying again.

Wiring tweaks, mains booting, and more

They figured out that the issue was with the thin wires they were using to connect the batteries, which cannot handle the voltage spikes that the PC demands when turned on. To address that issue, they added two big capacitors for some power buffers, used multiple wires to ensure that multiple cables carried power for the entire system, and are now running the entire thing off 56 alkaline batteries. They also kind of cheated by using an external power supply to boot the PC, and then would only switch over to the AA system once the game was running.

With the setup ready, ScuffedBits switched over from the power supply to the AA batteries, and the PC actually ran on battery power alone. However, it seems that launching Steam requires too much power, causing the PC to shut off after just 52 seconds. So, they decided to have A Short Hike running on their next try, which is a rather modest game when it comes to its system requirements. Still, the computer only lasted for five seconds before flat-out dying. But because they were adamant on completing at least one round running purely on batteries, they decided to lower the bar even further and just tried to defeat Minesweeper on easy.

ScuffedBits was eventually able to successfully do this, finishing one round on easy in 4 minutes and 35 seconds, with the PC’s batteries running out at the same time — but that wasn’t enough. They wanted to completely beat the game on AA power alone, so they replaced their LED monitor’s power supply with eight rechargeable AA batteries. They didn’t have to replace their wireless mouse and keyboard, though, as they were already running on AAs. This time, the PC only lasted for 2 minutes and 14 seconds, probably because they didn’t use fresh batteries.

They seemed happy with the results of their experiment, but, for one last attempt, they decided to install a GPU and run the game at full screen. They switched over to battery power alone, and, to their surprise, the “gaming” desktop PC actually ran for around nine seconds.

This fun and wacky experiment just showed how far battery technology has moved forward, allowing us to get relatively good gaming performance even on small handhelds. A few commenters suggested that the YouTuber try again, but to use laptop components designed for low power draws this time around. I would actually be interested to see how many batteries you need to run one of the highly-efficient Snapdragon laptops or an Apple silicon device.

When U.S. defense planners talk about supply chain risk, semiconductors usually dominate the conversation. But a quieter dependency is drawing growing attention inside the Pentagon and across the tech industry: batteries. From AI training clusters and hyperscale data centers to drones, satellites, and electric military vehicles, modern computing and warfare now rely on lithium-ion batteries and the materials that enable them.

According to a New York Times special report, the U.S. government is concerned that China sits at the center of the ecosystem, controlling a large share of battery manufacturing and processing.

Three high-level meetings on the battery supply chain are understood to have been held at the White House in recent weeks, and the National Energy Dominance Council, established by President Trump, has been meeting with battery companies. The Energy Department also recently announced up to $500 million for battery materials and recycling projects.

Structural dominance

U.S. dependence on Chinese batteries did not emerge overnight; it is the product of decades of industrial policy, aggressive investment, and vertical integration that have allowed Chinese firms to dominate nearly every stage of the battery supply chain. While the U.S. and allies have focused on downstream innovation and system-level design, China has built scale in mining, refining, cell manufacturing, and component processing. That imbalance is now colliding with surging demand from AI, electrification, and modern defense systems.

Chinese companies control the majority of global processing capacity for lithium, cobalt, nickel, and graphite, materials essential for lithium-ion batteries. Battery-grade graphite processing, for example, is overwhelmingly concentrated in China, and lithium iron phosphate cell manufacturing is similarly dominated by Chinese firms. Even when raw materials are mined elsewhere, they are often shipped to China for refining before entering the global supply chain.

This is significant because processing and component manufacturing are the hardest stages to replicate quickly. Mines can take a decade or more to permit and develop, but refineries, cathode plants, anode facilities, and separator production lines also require years of capital investment and technical expertise. China built these capabilities early, often with state support, and scaled them to a level that made competing facilities elsewhere economically unattractive. Western manufacturers increasingly opted to source components from China rather than to build redundant capacity domestically.

As demand for batteries has increased, driven first by consumer electronics and electric vehicles and now by grid storage and AI infrastructure, it’s apparent that this decision has become a significant liability. Battery packs for data centers, backup power systems, and industrial AI deployments are large, expensive, and complex to substitute. As a result, for many U.S. operators, Chinese suppliers remain the only realistic option at scale.

Military systems tied to the same supply chain

The Pentagon faces a much more profound version of this problem. Modern military platforms depend on batteries to power sensors, communications, guidance systems, and increasingly autonomous operations. Drones, loitering munitions, electric ground vehicles, and directed-energy weapons all require high-density energy storage. Even legacy platforms rely on battery-backed electronics that cannot function without a supply of critical materials.

Defense supply chain assessments have found that a significant share of U.S. weapons systems contain components derived from Chinese-sourced materials, particularly rare earths and battery inputs, says the New York Times. This is not always because the final part was made in China, but because upstream processing took place there. Once those materials are embedded in magnets, cathodes, or battery cells, tracing and replacing them becomes extremely difficult.

This is compounded by the fact that the proliferation of drones and electronic warfare systems has pushed energy storage from a supporting role into a central one. This creates a planning challenge for the Pentagon, as stockpiling batteries is not straightforward due to degradation over time and rapid changes in chemistry and form factor. Relying on just-in-time global supply chains for critical military energy storage is also untenable.

Congress has begun to respond by tightening sourcing rules for defense procurement, including future restrictions on batteries that rely on Chinese materials, but policy changes do not create factories. Even with aggressive use of the Defense Production Act, building domestic or allied capacity for battery materials and processing will take years.

A new layer of pressure

The rapid expansion of AI computing has added another dimension to the battery problem. Training and inference at scale require enormous amounts of power, and data centers increasingly rely on large battery installations for backup and load balancing. These systems are not small UPS units but rather industrial-scale energy-storage deployments measured in megawatt-hours.

As utilities struggle to deliver new generation capacity fast enough, battery-backed power systems have become a critical enabler of AI growth. In practice, many of these batteries are produced by Chinese manufacturers because they are available at scale and at prices that competitors have struggled to match. Cloud providers and AI firms typically face two options: Chinese batteries or no batteries at all.

This dependence sits uneasily alongside U.S. export controls and broader technology competition with China. The same companies racing to build sovereign AI capacity are relying on Chinese energy storage to keep their data centers running. That contradiction is not lost on policymakers, but alternatives remain limited in the near term.

The situation also complicates efforts to localize AI infrastructure. Even if advanced chips are fabricated in Taiwan, packaged elsewhere, and assembled into servers in the United States, the power systems that support those servers may still trace their supply chains to China. That weakens the overall system from a resilience perspective.

Unwinding the dependency

Washington naturally wants to respond and unwind the dependency that has built over several decades, but the timeline for doing so is unforgiving. The Inflation Reduction Act and related legislation have triggered a wave of announced battery and EV investments in the United States, often led by foreign firms seeking to qualify for incentives. These projects will expand domestic cell manufacturing capacity, but many still rely on imported materials and components.

Upstream efforts to build lithium refining, graphite processing, and cathode production in North America are underway, supported by federal loans and grants. However, these projects face opposition and cost pressures that slow deployment. Even optimistic projections suggest it will take most of the decade to meaningfully reduce reliance on Chinese processing for key battery materials — and we’re about to enter 2026.

In the meantime, both the Pentagon and the AI industry are being forced to manage risk rather than eliminate it. Defense planners are exploring alternative chemistries, recycling, and limited stockpiling, while tech firms are diversifying suppliers where possible and investing in efficiency to reduce battery demand per unit of compute. None of these measures fully solves the underlying issue.

The only factor that has meaningfully changed is how we view batteries; they are no longer viewed as a peripheral component, but as a technology on par with semiconductors. Energy storage sits at the intersection of computing, transportation, and military power, so control over its supply chain carries significant weight.

China recognized this years ago and treated batteries as a national priority, accordingly building the industrial base to match. The U.S. is only now beginning to respond seriously. Whether its response will be fast enough to keep pace with AI and evolving military needs remains to be seen.

Scientists reckon that the McDermitt Caldera might hold the largest deposit of lithium ever found. Situated on the Nevada – Oregon border, this huge crater was formed when a magma chamber collapsed around 16 million years ago. Estimates point to there being between 20 and 40 million metric tons of lithium-rich clay at this site. The deposit could be worth $1.5 trillion at current prices, says Earth.com, and be ample to supply the lithium battery industry (gadgets, cars, Powerwalls) for decades. Moreover, it is undoubtedly a resource of strategic significance.

Investigations into the size and nature of this deposit were led by Thomas R. Benson, PhD, working on behalf of Lithium Americas Corporation (LAC). The research suggests that several million years ago, a supervolcano situated about 28 miles north to south and 22 miles east to west along the Nevada-Oregon line formed a large crater, after its magma chamber collapsed in the wake of a violent eruption.

Subsequently, thick sheets of hot ash cooled on the caldera floor, and a lake was established for eons, collecting volcanic ash and mud sediments. Volcanic activity continued, though, with mineral-rich hydrothermal water being carried upwards into the lake sediments.

Millions of years later, the lake long-gone, we are left with a variety of materials, with high concentrations of lithium within. For example, at the Thacker Pass, there are potassium-rich and lithium-infused clay bands that are 100 feet thick. A significant amount of high-grade lithium-holding clay is also close enough to the surface to allow for open-pit mining.

Strategic and industry importance

Of course, it is good for nations to have key industrial resources on their home turf. The U.S. wasn’t lithium-barren before this discovery, but it has been largely reliant (strategically speaking) on the Silver Peak Mine in Nevada since the 1960s. Last year, we reported on significant deposits of 5.1 to 19 million tons of lithium being found in Southern Arkansas, though.

Some analysts expect lithium global demand to expand to 8x the world’s 2022 output by 2040. So, commercial mining of newly discovered deposits in places like the McDermitt Caldera and Southern Arkansas could be extremely important for U.S. interests in the coming decades. Moreover, lessons learned during this supervolcano find could help unlock many more deposits.

Lithium's importance could decline, though. Battery tech is being continuously improved, with refinements to both lithium-based and alternative chemistries being frequently flagged by researchers. Some of the most promising lithium-ion alternatives still have a way to go, though. Sodium-ion batteries show promise, based on this far more abundant element, but are currently second-class in terms of energy density (and thus cell size/weight) and cycle life. So it looks like lithium will continue to be the best choice in the near term.

In a recent DIY project, YouTuber Chris Doel salvaged 500 rechargeable batteries from disposable vapes to build a powerful power bank that runs his workshop and even his entire house. The process appears quite tedious and requires a fair amount of technical expertise, but the result takes a solid swing at addressing the growing e-waste problem.

The UK has been facing an e-waste crisis caused by disposable vapes for the past few years. In fact, a report published by Material Focus in December 2024 estimates that around 8.2 million vapes, including single-use, pod, and big-puff models, are discarded or improperly recycled each week. Doel took this as an opportunity and has been working on repurposing batteries from discarded vapes since last year. His previous projects include a fast-charging power bank and an e-bike powered entirely by recycled batteries.

His latest experiment involved collecting thousands of discarded empty vapes returned by customers to vape sellers. He then popped them open, removed the internals, and desoldered the circuitry from the battery. This extraction process roughly takes him about 3 minutes. Notably, he mentions that a battery is more or less useless if its charge drops below 3V, which meant that he had to separate the bad ones from the good ones. To simplify the sorting process, he then used a small pump to ‘puff’ air into each vape to check whether the battery still had some charge left.

The next step involved testing each recycled cell to determine its exact capacity using cell testers, which was probably the most labor-intensive part of the process. Since all the cells had to be connected in parallel, they needed to be charged to the exact voltage. After some DIY tinkering and making flexible 3D-printed contact pads to fit smaller cells, Doel managed to charge and sort cells of similar capacity into groups. He also found that focusing on the larger cells made more sense, as using the smaller-capacity cells would have required nearly 2,000 units.

After extracting, desoldering, and testing 500 large cells, he moved on to connecting them all by sourcing custom 3D-printed modules and aluminum rails. Each 3D-printed module would hold nine cells, which were then connected in parallel by soldering copper wires. Four modules were then connected to form a larger bank of cells with a 3.7V output. Once complete, a total of seven large banks are stacked together in two groups and connected in series to deliver around 50V. For protection, a fuse was added between the two groups, along with a battery management system (BMS) module, to ensure that all cells connected in parallel are working as intended.

This massive battery was then connected to an inverter, which converted the 50V of DC output into 240V AC. The setup was capable of powering Doel’s entire workshop, including lights, an oscilloscope, and other testing equipment. Since the inverter was installed between Doel’s workshop and his house, it not only powered the workshop but also sent electricity back to his home, allowing him to run appliances and his editing rig. Doel claims that the entire setup can produce enough power to run his workshop for up to three days or his home for about eight hours, depending on the devices in use.

Doel's project is a great example of how recycling can be turned into something useful and showcases the potential of repurposing e-waste for sustainable energy solutions. Having said that, it’s important to note that this project was carried out with advanced technical knowledge and safety precautions. It is highly advised not to replicate any of the steps at home, as working with lithium-ion cells can be very dangerous if not done properly.

Chinese PC gaming peripherals specialist Angry Miao (AM) is vigorously promoting its new Infinity Mouse on social media platforms. Its latest innovation is an ultra-light gaming mouse hewn from metal but weighing under 50g, with premium features and a headlining “infinite battery.” The tri-mode rodent is currently around halfway through a very successful Kickstarter campaign. However, a previously teased infinite battery explainer has missed its advertised Feb 26 reveal, and we will have to wait until March 10 for this important info. Is something amiss? We’ve had a nose around and think a better name for the AM Infinity Mouse would probably be the ‘AM Instant Charging Mouse.’

The AM Infinity Mouse’s extraordinary battery claim is right there in the name of the product. Thus, it is extremely important that it should live up to the expectations such a moniker brings.

AM’s Kickstarter project must surely credit at least part of its rip-roaring success (raised over $71,000 – many multiples of the initial goal) on this ever-enduring battery implication. However, the product summary near the top of the crowdfunder page includes an asterisk beside the “infinite battery life*” claim. Later, you can read that the asterisk warns, “*We are currently fine-tuning some critical features and will continue to share more final details as soon as they become available.” Fine-tuning might be understating the work required, as we are quite sure the finished product won’t come packing an “infinite battery,” unless you accept a very loose definition of such a concept.

Later on the crowdfunder page, a section about the battery in the AM Infinity Mouse reveals that the designers of this rodent were inspired by the current situation where users “dial down specs for longer battery life.” AM isn’t happy with that compromise, but we get the first hint here that the battery tech will eliminate “the constant worry of recharging.” The firm doesn’t repeat any infinite battery life blurb in this copy.

As mentioned in the intro, the Feb 26 reveal of the battery tech was a no-show, and now we’re looking at March 10. For those on tenterhooks, we might have some relief, though. In a Tweet on Friday, an animated GIF seems to give the (battery technology) game away. This new tease appears to indicate that the AM Infinity Mouse should more correctly be named the AM Instant Charging mouse. The tease says that the Infinity Mouse can be charged in just 1.2 seconds.

We might be a little disappointed with AM’s infinite battery marketing shenanigans, but instant charging sounds pretty cool, too, but with the caveat that is an equally questionable claim. However, the thought, attention to detail, and the rest of the mouse specs seem attractive. The mouse design is Lotus sports car inspired, features premium switches, and the PAW3950 sensor offers 8K polling. This is wrapped inside an ergonomically carved magnesium alloy design. Check out the top video for (English subtitled) details on the iterative design work behind the AM Infinity Mouse to find out what you should get for your $150 early bird pledge.

Remember that crowdfunding a project is not a guarantee of receiving a finished product. Backing a crowdfunded project is akin to an investment; you believe in the project and want it to succeed. You are not purchasing a retail product. Given the details (or lack thereof) surrounding this project, you should also remember that the end products can also end up not living up to the billing.

Researchers from Fudan University in China have developed a method to refresh and extend the lifespan of lithium batteries, a breakthrough that could improve battery sustainability and reduce electronic waste. The findings, recently published at Nature, are based on a technique that rejuvenates degraded lithium-ion cells, potentially increasing their usability for long periods.

As we know, lithium-ion batteries power everything from smartphones and laptops to electric vehicles, but their efficiency declines over time due to electrode degradation and chemical side reactions. The research led by Peng Huisheng and Gao Yue from Fudan University, along with other collaborating Chinese research institutions, began by identifying and synthesizing a lithium carrier molecule - lithium trifluoromethanesulfinate (LiSO₂CF₃). The researchers described the white powdered functional salt as a "precise-treatment medicine for lithium-ion batteries."

Over time, lithium-ion batteries lose efficiency as lithium becomes trapped in inactive compounds and electrode materials degrade. The new method involves introducing an electrolyte solution containing dissolved LiSO₂CF₃ into a degraded battery. When a voltage is applied, the compound breaks down, releasing lithium ions that reintegrate into the electrode material, effectively restoring lost capacity.

Simultaneously, the breakdown process generates gaseous byproducts, including sulfur dioxide (SO₂), trifluoromethane (HCF₃), and hexafluoroethane (C₂F₆), which naturally escape from the electrolyte if the battery is designed to allow venting.

Extending lifespan from 1,500 to 12,000 charge cycles

Once the lithium restoration is complete, the battery is sealed and tested to confirm its performance. Early experiments have shown that lithium-iron phosphate batteries treated with this method can regain nearly all of their lost capacity, extending their lifespan up to 12,000 charge cycles. For comparison, a typical electric car battery has a lifespan of around 1,500 charge cycles.

According to Gao Yue, “The battery only lost 4% of its performance after 11,818 cycles. For an EV that charges twice a day, this means the battery could last up to 18 years. In comparison, current EV batteries usually lose 30% of their performance in just 2.7 years with the same charging routine.”

However, practical implementation requires that batteries be designed with electrolyte replenishment and gas venting in mind, which may limit the new technology's feasibility for widely used consumer electronics. While still in the experimental stage, this process presents a promising approach to reducing battery waste and increasing the longevity of rechargeable lithium-ion cells, particularly in large-scale energy storage systems.

The breakthrough not only has the potential to extend battery lifespans and reduce e-waste, but it can also lower the environmental impact of lithium mining. More testing and regulatory approvals will be needed before this method can be implemented widely, but it seems to be a promising step toward a more sustainable battery industry.

A welcome redesign of the Windows 11 battery indicator is on the way with more pixels and colors. Windows sleuth PhantomOfEarth has tested this UI tweak and shared some images of the new look on Twitter/X. The new icons can be enabled in Insider build 26120.2510 which was released to the dev channel recently.

Microsoft Windows 11's battery icon is shown in the System Tray area, as well as the Quick Settings and Settings app – on appropriate systems like laptops, tablets, and other portables. However, the current design leaves a lot to be desired, with its stunted size and lack of color. Your smartphone or tablet probably has a battery icon that delivers more info at a glance.

The screen real estate available on typical portable Windows devices vastly outclasses mobiles, so it doesn't seem right that the Windows battery indicator is so short on pixels. In Insider build 26120.2510 testers can beef it up and enjoy a clearer at-a-glance indicator of battery life remaining. It hasn't been scaled so much that it looks out of place, but at such a small scale even a couple of extra pixels helps.

If you enable this UI update, you will benefit from a sprinkle of color, too. It isn't a massive departure, with the icon remaining monochrome unless the energy saver is enabled (goes orange), or the battery is charging (green).

Windows users who are used to seeing a plug-over-battery icon, signifying charging, may notice there is a lightning-bolt-over-battery icon in this latest update. Moreover, the lightning bolt is located over the center of the better icon, rather than to the left, in this UI update.

Last but not least, there are also updates to the battery usage graph's labels.

Enabling the new battery icons

Even if you are using the aforementioned Insider build 26120.2510, you won't see these new battery icons unless you enable them. PhantomOfEarth explains that to update to the new icons users will have to download the ViVeTool app from GitHub and input a few command line strings then reboot. We would only do this kind of tinkering on a test system, not a daily driver PC.

Do you like the new battery indicator? According to PhantomOfEarth's small-ish survey, about 75% of respondents approve.

An advance in battery chemistry promises to deliver a "smaller, lighter, and cheaper battery without sacrificing end-of-life battery performance." On September 13, researchers from around the United States released a paper in collaboration with the Department of Energy's Argonne National Laboratory titled "Solvent-mediated oxide hydrogenation in layered cathodes", with additional comments and a summary in an official press release on the Argonne National Laboratory website. In short, the researchers used the cutting-edge X-ray technology available at Argonne National Laboratory and its Advanced Photon Source, which pairs X-ray studies and electrochemistry to allow researchers to examine their batteries at a molecular level.

But why go through all this trouble? You may or may not know this, but the rechargeable batteries available in your phones, electric vehicles and such are most likely lithium-ion batteries. Lithium-ion batteries degrade over time, getting steadily worse and worse at holding charge until they simply no longer function or can't hold a charge for a meaningful amount of time. Just because we've known how to make lithium-ion batteries for so long doesn't mean we've perfected them — advancements into "perfection" require deeper study than just being able to make a functioning item. 

According to Argonne Senior Chemist Zonghai Chen, "Self-discharge is a phenomenon experienced by all rechargeable electrochemical devices. The process slowly consumes precious functional battery materials and deposits undesired side products on the surface of the battery components. This leads to continuous degradation of battery performance."

In this case, using high-end equipment allowed researchers to bridge a key missing link in existing research by getting an otherwise impossible molecular view of a lithium-ion battery in action. This allowed researchers to identify a specific process: cathode hydrogenation, which is the process of "dynamically transferring the protons and electrons from the electrolyte solvent into highly charged layered oxides in the cathode". The chemical reaction here leads to battery degradation, and while lithium-ion batteries have been known to have degradation issues for a long time, this greatly narrows down the cause.

So, what does it mean in the long term? As Zonghai Chen says, "By mitigating self-discharge, we can design a smaller, lighter, and cheaper battery without sacrificing end-of-life battery performance." This should be a pretty big step toward a future where rechargeable batteries are not awful. Here's hoping!  

For a college student, there is no more important tool than a laptop. It's a catch-all for note-taking, studying, homework, collaboration, procrastination (don't tell your professors), and extracurricular activities. When you pick one, you're likely choosing a tool for at least four years of education, if not further studies or just life after college.

Picking a college laptop can feel like homework in itself. There are a ton of laptops on the market, and you may get some help by checking out recommendations from your school or department. In our opinion, most students should prioritize portability, battery life and then performance, unless your school recommends something specific.

This semester, the latest parts in laptops include Intel's Core Ultra Series 2 CPUs, AMD's Ryzen AI 300 processors, Apple's M4 chips, and Nvidia's RTX 50-series GPUs. Looking ahead, there's rumors of a budget MacBook that could be a serious player on college campuses.

If you're taking classes remotely, be sure to take the webcam into consideration. More and more laptops are finally getting 1080p webcams or better, which often include higher-quality lenses for better images. (Many, but not all, also have privacy shutters for when you're not making calls.)

Also, don't forget whatever student discounts are available. Many vendors may give you some percentage off, a gift card, or a free extra when you can prove that you're a student, either with a .edu email address or a college ID card.

Best College Laptops You Can Buy Today

Best College Laptop Overall

Apple's MacBook Air is reliably one of the most popular notebooks on college campuses. It's thin, fanless, and the latest sky blue color will be sure to attract some attention.

Whether you're taking a virtual class or just calling home, Apple has bumped up the webcam to 12MP, which will make you look sharper than ever. The M4 chip also supports two external displays while you have the lid open, which is a bump up from M3.

Perhaps most important, the MacBook Air is $100 cheaper than the previous model, starting at $999 for the 13-inch laptop and $1,199 for the 15-inch notebook. That being said, upgrade pricing for more RAM or storage is outrageous.

In our testing, we found the 15-inch MacBook Air ran over 15 hours on a charge, so you shouldn't have to worry about charging it too much between classes. And the 15-inch notebook has powerful speakers alongside its large display, making it a great multimedia machine.

Read: MacBook Air (M4) review

Best College Laptop for Notetaking

The Microsoft Surface Pro is a sleek, portable system that's easy to take with you, and is great for notetaking with a pen. Add in long battery life, and you have a study tool that can go with you from class to class.

The Qualcomm Snapdragon X Elite chip has enough power to keep up with x86. Thanks to an increasing number of Arm-compatible apps, including Google Chrome, you won't be missing out on a lot from Intel and AMD machines. That being said, you'll want to make sure any dedicated software you use for class will work.

Qualcomm's NPU, with 45 TOPS, is great for Windows Studio effects, but mostly these effects are cool tricks for now. Don't buy this just for AI. But if you want a great note taking device that's extra slim, it's worth considering. Just don't forget to add a keyboard and stylus to the base price.

Read: Microsoft Surface Pro review

Best 2-in-1 College Laptop

If you want a laptop that can flip into a tablet, the Lenovo Yoga 9i 2-in-1 is our top choice. It combines a gorgeous OLED touchscreen, great build quality, and long battery life into one package.

You might question having a 2880 x 1800 resolution on a 14-inch screen, but it obliterated competing notebooks when measuring for color volume.

The system looks nice, too, with a "cosmic" blue color that will stand out in a lecture hall full of black and gray laptops. The laptop is built like a tank (even if there are some quirky design elements blending rounded edges and flat corners) and has a clicky keyboard. You also get a 5MP webcam, great for presentations, virtual classes, and calling your family.

If you want to make DIY upgrades, this machine is tough to repair, but that probably won't bother the majority of students.

Read: Lenovo Yoga 9i 2-in-1 Aura Edition review

Best College Laptop with a GPU

Some majors, like engineers, graphics designers, or some programmers, may need a bit of extra graphics power. That's where having a discrete GPU comes in. We tested the Lenovo Yoga Pro 9i with an RTX 4050, along with a beautiful 16-inch, 3200 x 2000 display.

The Yoga Pro 9i also has an excellent 5MP webcam for when classes are remote, and deep key travel for taking notes. The display could be a bit brother, and there's a bit of bloatware you may want to remove. But the Yoga came in cheaper than some competing laptops, like the Dell XPS 16, with a discrete graphics card.

Those who like to tinker will appreciate the extra SSD slot, which you can use to bulk up storage.

Read: Lenovo Yoga Pro 9i review

Best College Laptop for Business Majors

For business majors who want to get a feel for the type of laptop their first job will give them, there's the Lenovo ThinkPad X1 Carbon (Gen 13). The 13th Gen "Aura Edition" laptop is lighter than ever at just 2.17 pounds.

We loved the keyboard on the TrackPoint, which is snappy and tactile. And the TrackPoint in the middle of the notebook means that you never have to remove your hands from the home row if you don't want to.

The X1 Carbon also has plenty of ports, including two Thunderbolt 4/USB Type-C ports, a pair of USB Type-A ports, HDMI to connect to a monitor, and a headphone jack.

In our testing, the X1 Carbon ran for 11 hours and 28 minutes on a charge. That's not bad, but competitors lasted hours longer. You'll want to top this one off every night.

Read: Lenovo ThinkPad X1 Carbon (Gen 13) review

Quick College Laptop Shopping Tips

Finding Discounts on the Best College Laptops

Whether you're shopping for one of the best gaming laptops or another model that didn't quite make our list, you may find some savings by checking out our lists of Dell coupon codes, HP coupon codes, Lenovo coupon codes, Razer promo codes or Newegg promo codes.

Tom's Hardware Gaming Laptop Coverage

Best Premium Laptops | Best Gaming Laptops Under $1,500 | Laptop Reveiws | Apple MacBook Pro (M5) Review | Apple MacBook air (M4) Review | Lenovo Yoga 9i 2-in-1 Review | Dell 16 Premium Review | Lenovo ThinkPad T14s Review | How to Upgrade a Laptop SSD | How to Upgrade Laptop RAM

Microsoft and Qualcomm just launched Copilot+ PCs highlighting so-called game-changing AI features. However, a Bloomberg report says that most buyers have purchased these laptops for their battery life, instead of their on-device AI capabilities. Microsoft introduced three AI-powered features exclusive to Copilot+ PCs — Cocreator, Windows Studio Effects, and Live Captions with Translation. However, many users find that these aren’t essentials in day-to-day use. Microsoft also delayed the release of the interesting Recall feature due to privacy concerns.

Snapdragon-powered laptops have still precipitated some good sales numbers, with Copilot+ AI PCs raking in 20% of global PC sales during launch week. However, industry analyst Avi Greengart said that most users purchased these AI laptops for their better battery life, rather than their AI capabilities.

Although AI capabilities have been gaining popularity in recent months, many software companies still run user-facing AI features on their data centers via the Cloud, instead of running them on-device. The giant software corporations — Adobe, Salesforce, and SentinelOne — declined Microsoft’s request for them to tweak their apps to use the NPUs onboard the Snapdragon chips in time for the launch of Copilot+ PCs.

While tweaking apps to run AI processes locally will save companies processing power, doing so costs resources and manpower. Gregor Stewart, SentinelOne’s VP for AI said, “SentinelOne is looking at optimizing its products for AI PCs in future development, but it will likely take years for these devices to reach a ‘sufficient proportion of deployed machines.’” Salesforce is also developing on-device generative AI tech, but it didn’t give information about when it will deliver this feature to its users.

With only 3% of PCs sold this year expected to have AI capabilities, it does make sense for major developers to hold-off building AI features into their apps until they have more widespread adoption. Estimates suggest that it will take at least a couple of years before AI PCs are in the hands of one in five new PC buyers. Even by 2028, only 40% of new computers shipped to users will be AI-capable, according to the latest analysis.

Consumers are more likely to purchase AI PCs, especially those who want the longer battery life that Arm-based chips afford laptops (like Apple-silicon MacBooks), but still want to stick with the familiar Windows OS. However, many corporate clients might hold off switching to Windows on Arm, especially those that need legacy apps that run into issues with or aren’t even supported by Microsoft’s Prism emulation software.

Marketing millions

But Qualcomm is still hopeful that the AI boom will help drive sales of its chips and it is taking a page directly from Intel by investing heavily in marketing. Team Blue’s Intel Inside branding was a massive win for the CPU company, with the average PC buyer looking for that sticker when buying a new computer in the ‘90s and 2000s. 

Microsoft’s Copilot+ PCs are currently synonymous with Qualcomm’s Snapdragon X chips, but it won’t be for long as AMD’s Ryzen AI 300 PCs will get the same treatment later this year. So, the chip maker is investing massive amounts directly in marketing, with the company paying more than $75 million annually, sponsoring Manchester United. It’s also reportedly paying more to computer manufacturers to aid their marketing efforts.

The AI PC is currently just a marketing gimmick, especially with the limited on-device AI capabilities you get with Microsoft’s Copilot+ PCs. We expect on-device AI features will eventually receive widespread adoption, but it will take several years for that to happen. 

However, the simple fact that we have another CPU option over Intel and AMD is already a big win among consumers, helping break the duopoly that Team Blue and Team Red have built over the years. Furthermore, the improved battery life with comparable performance that Snapdragon laptops bring will force x86 chip manufacturers to innovate on power efficiency or be left in the dust.

A Swedish startup could revolutionize building lighter laptops and other portable devices through its work with multipurpose materials. Specifically, its work with carbon fiber could lead to “massless” batteries where the material serves as energy storage and as part of the device’s outside case.

Tech startup Sinonus sprang from research at Chalmers University of Technology in Gothenburg, Sweden. A research team there has been working for years on developing carbon fiber that offers both the stiffness required for structural integrity and the electrochemical performance needed to store energy.

Chalmers University of Technology and venture capital firm Chalmers Ventures spun the project into the startup, and Sinonus is now working to commercialize carbon fibers that double as electrodes. Sinonus has already developed a proof of concept, replacing AAA batteries in the lab with its carbon-electrode batteries.

Of course, AAA batteries are a far cry from the ones you find in your laptop or smartphone. The challenge lies in engineering the carbon fiber crystals to have the energy density required to be effective. While Sinonus hasn’t published its current energy-density metrics, the 2021 prototype at Chalmers University only came in at 24 Watt-hours per kilogram (Wh/kg). By comparison, existing lithium-ion batteries in laptops typically have an energy density between 80 and 120 Wh/kg.

The recently appointed CEO at Sinonus, Markus Zetterström, acknowledged this challenge. However, he also pointed out that the material doubling as a structural integrity component helps justify the loss of electrical storage efficiency. “Storing electrical energy in carbon fiber may perhaps not become as efficient as traditional batteries, but since our carbon fiber solution also has a structural load-bearing capability, very large gains can be made at a system level,” he told New Atlas.

As Sinonus describes it, using “the mass that is already there” to store energy could serve to reduce the weight and volume of the device while still improving overall system performance. After all, if your laptop is smaller and lighter while still giving the same battery life, it hardly matters that the material storing the energy isn’t as efficient as a LiON battery pack.

Chinese company Betavolt has announced an atomic energy battery for consumers with a touted 50-year lifespan. The Betavolt BV100 will be the first product to launch using the firm’s new atomic battery technology, constructed using a nickel-63 isotope and diamond semiconductor material. Betavolt says its nuclear battery will target aerospace, AI devices, medical, MEMS systems, intelligent sensors, small drones, and robots – and may eventually mean manufacturers can sell smartphones that never need charging.

Buying an electronics product that can go without charging for 50 years would be amazing. But the BV100, which is in the pilot stage ahead of mass production, doesn’t offer a lot of power. This 15 x 15 x 5mm battery delivers 100 microwatts at 3 volts. The company says multiple BV100 batteries can be used together in series or parallel depending on device requirements, and Betavolt has plans to launch a 1-watt version of its atomic battery in 2025.

The new BV100 is claimed to be a disruptive product on two counts. Firstly, a safe miniature atomic battery with 50 years of maintenance-free stamina is a breakthrough. Secondly, Betavolt claims it is the only company in the world with the technology to dope large-size diamond semiconductor materials, as used by the BV100. The company is using its 4th gen diamond semiconductor material here.

In its press release, Betavolt says its atomic battery is very different from similarly described power cells developed by the US and USSR in the 1960s. It says that the old nuclear batteries were large, dangerous, hot, and expensive products. For example, some old-tech atomic batteries used Plutonium as the radioactive power source. Meanwhile, the Betavolt BV100 is claimed to be safe for consumers and won’t leak radiation even if subjected to gunshots or puncture.

The new, improved levels of safety stem from the choice of materials. Betavolt’s battery uses a nickel-63 isotope as the energy source, which decays to a stable isotope of copper. This, plus the diamond semiconductor material, helps the BV100 operate stably in environments ranging from -60 to 120 degrees Celsius, according to the firm. Betavolt boasts that this battery technology is “way ahead” of European and American academic and commercial institutions.

How does Betavolt make this battery? We’ve already mentioned the essential materials, but the above diagram provides an excellent overview. The BV100 is made by “placing a 2-micron-thick nickel-63 sheet between two diamond semiconductor converters.” This construction relies on Betavolt’s “unique single-crystal diamond semiconductor that is just 10 microns thick.”

So, the current BV100 or its 1-watt successor scheduled for next year may not sound so impressive regarding peak power output. Betavolt will be well aware of devices with a greater thirst for power and teases that it is investigating isotopes such as strontium-90, promethium-147, and deuterium to develop atomic energy batteries with higher power levels and even longer service lives – up to 230 years. 

Swapping lithium-ion battery binding material can deliver a 10% increase in battery life and a reduction of 70% in self-discharge, according to a Canadian research team.

The specific finding of the researchers was that swapping from polyethylene terephthalate (PET) tape to polypropylene (PP) tape battery cell binding appears to be highly worthwhile, delivering the noticeable benefits outlined above. If the industry heed this research, it could mean laptops, tablets, and smartphones with improved battery life. Importantly, PP tape is roughly the same price as PET tape, so there should be little to no impact on costs/prices.

Ph.D. student Anu Adamson led the research team at Dalhousie University in Halifax, Canada. Nature published her paper, “Improving lithium-ion cells by replacing polyethylene terephthalate jellyroll tape,” earlier this month.

The reason for the improvement in battery performance when changing tape material is due to a "largely overlooked" interaction between PET tape and Li-ion battery chemistry. Adamson and team noticed PET tape can depolymerize into its monomer dimethyl terephthalate, inducing substantial self-discharge in a lithium-ion cell.

"Nobody, I think, thought that it could be the tape," said Adamson in an interview with CBC Radio news. As people see battery self-discharge as inevitable, mused Adamson, no one thought of checking if there was a simple fix. "This is really the first time we've seen that something that is generally regarded as very inactive in batteries, is essentially anything but," Adamson explained.

Jeff Dahn, professor emeritus in the university’s physics, atmospheric science and chemistry departments, characterized Adamson’s findings as “incredibly nice chemical detective work,” praising the Ph.D. student for her talent and recognizing the extent of connections in lithium-ion battery chemistry.

Dahn was not involved with the research project but agreed switching to polypropylene (PP tape) would make a huge difference to battery cells. The professor reckoned “most manufacturers will make the change” going forward, as it is a simple switch that makes a noticeable difference. 

One of the most extraordinary things about today’s gaming laptops is that it’s hard to find an absolute dud. OEMs have stepped up their game in recent years, and we’re seeing laptops that deliver higher frame rates, last longer per charge, and have better screens than we could have imagined just a few years ago. And more importantly, gamers don’t have to spend an arm and a leg to get decent performance these days.

That is the case with the Lenovo Legion Slim 5, a 16-inch gaming laptop sporting a QHD+ display, an AMD Ryzen 7 7840HS processor, 16GB of RAM and an Nvidia GeForce RTX 4060 GPU. The laptop has an MSRP of $1,529.99, but it sells for as low as $1,149.99 on sale, making it an excellent bargain among the best gaming laptops. 

Lenovo Legion Slim 5 Specifications

Design of the Lenovo Legion Slim 5

The design of the Legion Slim 5 is similar to others in Lenovo's gaming line, like the Legion Pro 5i. In other words, you’ll find a plastic chassis finished in onyx grey and an understated aesthetic. It is a gaming machine, but it wouldn’t look out of place sitting on a desk in an office.

Thin bezels surround the display, although a slightly larger protrusion at the center of the top bezel houses the 1080p webcam. The webcam doesn’t have a physical privacy shutter like some other gaming laptops, but a physical switch on the right side of the chassis turns off the webcam when it is not needed. The only other port residing on the right side of the chassis is the full-size memory card reader. Switching to the left side, there are two USB-C ports and a 3.5 mm headphone jack. The back panel features two USB 3.2 (Type-A) ports, gigabit Ethernet, and a proprietary jack for the power adapter.

The Legion Slim 5 measures 14.16 x 10.25 x 0.99 inches, which is similar in footprint to the Legion Pro 5i, although the former is slightly lighter (5.29 versus 5.6 pounds). The Acer Predator Helios Neo 16 is also within spitting distance for size and weight (5.79 pounds), while the Razer Blade 14 has a much smaller footprint due to its 14-inch display and is over a pound lighter than the other three laptops (4.05 pounds).

Lenovo Legion Slim 5 Gaming Performance

The Legion Slim 5 uses an AMD Ryzen 9 7940HS, an 8-core/16-thread processor. The base clock in this laptop is 3.8 GHz, while the maximum boost clock is 5.1 GHz. The processor is paired with an Nvidia GeForce RTX 4060 and 16GB of DDR5-5600 memory (2 x 8GB).

The competitors I have lined up for the Legion Slim 5 match closely in specs, apart from the Razer Blade 14 (Ryzen 9 7940HS, RTX 4070, 1600p), which has a smaller display and more powerful GPU. The other two competitors are a more even match, including the Lenovo Legion Pro 5i (i7-13700HX, RTX 4060, 1600p) and the Acer Predator Helios Neo 16 (i7-13700HX, RTX 4060, 1600p).

I played Cyberpunk 2077 on high settings with ray tracing enabled, DLSS Balanced, and DLSS frame generation enabled, averaging 94 frames per second (fps) at 1200p. Increasing the resolution to 1600p saw the performance dip to around 62 fps.

Kicking off our gaming benchmark suite, I turned my attention to Shadow of the Tomb Raider (highest settings). The Legion Slim 5 hit 96 frames per second (fps) at 1080p and 55 fps at 1600p. This performance put it just slightly behind the Predator Helios Neo 16 by 1-2 fps but well behind the similarly-equipped Legion Pro 5i.

Grand Theft Auto V (very high settings) testing saw the Legion Slim 5 slide into second place with 91 fps at 1080p and 52 fps at 1600p. However, the rankings were very tight across the board at 1080p and 1600p for the competitors. Only the Legion Pro 5i ran away from the competition at 1080p resolution at 97 fps.

Far Cry 6 (ultra settings) also saw a tight grouping of the laptops. While the Legion Slim 5 technically came in last place at 1080p with 83 fps, we wouldn’t exactly call that a poor showing versus the 85 fps for the Blade 14 and 88 fps for the Predator Helio Neo 16. The Legion Slim 5 put up a respectable 62 fps at 1600p.

Moving on to Red Dead Redemption 2 (medium settings), the Legion Slim 5 tied the Predator Helio Neo 16 for third place, achieving 65 fps at 1080p and 39 fps at 1600p. However, the Blade 14 led the field with 73 fps at 1080p and 45 fps at 1600p.

Borderlands 3 again showed how evenly matched the Legion Slim 5 and Predator Helio Neo 16 are, with each garnering around 80 fps at 1080p and just over 50 fps at 1600p. The Blade 14 again showed its muscle, delivering 98 fps at 1080p and 64 fps at 1600p.

Rounding testing out with the Metro Exodus Benchmark (RTX settings), the Legion Slim 5 ran through 15 loops at 1600p resolution. The Legion Slim 5 hit 49.96 fps averaged across the 15 runs. During the run, the Ryzen 9 7940HS averaged 3.93 GHz at 70.67 degrees Celsius (159.2 degrees Fahrenheit). The GeForce RTX 4070 GPU ran at an average of 2.12 GHz and measured 70.53 C (158.95 F).

Lenovo Legion Slim 5 Productivity Performance

The Legion Slim 5 features an AMD Ryzen 9 7940HS processor, 16GB of DDR5-5600 memory and a 512GB SSD.

The Geekbench 5 synthetic CPU benchmark saw the Legion Slim 5 running neck and neck with the Raptor Lake-based competitors in single-core performance (1,893) and slightly behind the Blade 14, which uses the same Ryzen 9 7940HS processor. However, both the Legion Slim 5 and the Blade 14 were clobbered in the multi-core benchmark, with the latter delivering a score of 11,046 compared to 14,967 for the first-place Legion Pro 5i.

Moving to the file transfer test, which involves shuttling 25GB of files, the Legion Slim 5 garnered 1,733.97 MBps, putting it just shy of the first-place Legion Pro 5i.

In the Handbrake test, where we transcode a 4K video to 1080p, the Legion Slim 5 completed the task in 4 minutes and 34 seconds, nipping the Blade 14 by 11 seconds. However, neither matched the Legion Pro 5i, which completed the task in just 3 minutes and 39 seconds.

Display on the Lenovo Legion Slim 5

The Legion Slim 5 comes standard with a 16-inch FHD+ (1920 x 1200) IPS display, but our review sample arrived with an upgraded QHD+ (2560 x 1600) IPS panel with a 165 Hz refresh rate.

I spent a lot of time playing Overwatch 2 and Cyberpunk 2077. Given that this is an IPS panel (with an anti-reflective coating applied), the display had excellent viewing angles, and the colors were reasonably vibrant. However, you won’t find OLED levels of color here, and darkened areas in Cyberpunk 2077 looked more like a dark grey as the LED backlight shined through.

For video performance, I turned my attention to Guardians of the Galaxy Vol. 3, which was recently released on Disney Plus. As was the case when I watched the movie in theaters, I fought back tears during the gut-wrenching scenes detailing Rocket’s origins. It also gave me a chance to scope out the generally good color performance of the display, particularly when the Guardians arrive at Orgocorp headquarters in their brightly-colored spacesuits (yellow, red, orange, blue and green). While there was a noticeable drop-off in color compared to the Mini-LED display on my daily driver laptop, the performance here for a laptop retailing below $1,500 was admirable.

In our tests, the Legion Slim 5 fell behind its Legion Pro 5i stablemate across the board. Based on our colorimeter, it hit 78.1 percent of the DCI-P3 and 110 percent of the sRGB color space. The display was also the least bright in this field at 318 nits, compared to 349 nits for the Legion Pro 5i and 499 nits for the Predator Helios Neo 16.

Keyboard and Touchpad on the Lenovo Legion Slim 5

The keyboard on the Legion Slim 5 was instantly recognizable to me, as it is essentially the same unit found in the Legion Pro 5i that I reviewed earlier this year. The keys are all in the same place, the handy number pad remains, and even the centrally-positioned power button (right above the F9 and F10 keys) is unchanged.

As always, I visited keyhero.com to measure my typing speed and managed 85 words per minute with 95 percent accuracy. This essentially mirrored my performance on the Legion Pro 5i. (What a shocker!)

The keyboard features four-zone RGB lighting, which is controlled via the Lenovo Vantage software. You’ll find the settings under Lenovo Spectrum, where you’ll find several effects to choose from, along with options for changing the speed and brightness of the RGBs. However, you won’t find per-key controls on the Legion Slim 5.

The plastic touchpad, while not as large as some of the behemoth control surfaces in the laptop world, was easy to use and had a nice “click” sensation. With that said, I spent most of my time with the laptop using a Razer Atheris mouse wirelessly connected via Bluetooth.

Audio on the Lenovo Legion Slim 5

The Legion Slim 5 uses a pair of 2-watt speakers with Nahimic audio. Nahimic audio software allows you to tweak settings to dial in sounds to your tastes. However, after playing around the presets, I didn’t notice enough of a difference to change my mind about the audio quality.

Voices in Cyberpunk 2077 sounded a bit muffled, which was a bit of a disappointment given the dialog-heavy nature of the game. Gunfire, explosions and vehicle engines sounded slightly better, but I’d stick with headphones while gaming anyway.

When listening to music, I turned to “Stranger” by Hooverphonic, from their The President of the LSD Golf Club album. Geike Amaert’s ghoulish vocals landed flat with the Legion Slim 5’s speakers, as did the bass guitar. The keyboard was the only part of the track that survived unscathed, leaving me underwhelmed.

Upgradeability on the Lenovo Legion Slim 5

Like most Lenovo laptops, gaining access to the internals is simple. Ten Philips head screws stand between you and access to the Legion Slim 5’s upgradeable bits. In this case, you can see that the battery is replaceable, and there are two SO-DIMM slots for future memory upgrades. However, note that both slots are currently occupied in this 16GB DDR5-5600 configuration.

The Wi-Fi/Bluetooth combo card is also readily accessible, as are the two M.2 slots for SSDs. One M.2 slot is pre-populated by the 512GB SK hynix SSD, and another is free for additional storage if you so choose.

Battery Life on the Lenovo Legion Slim 5

The Legion Slim 5 was no slouch in the endurance category. As we saw with the Blade 14, the AMD Ryzen 9 7940HS is a highly efficient chip. In our testing, which involves web browsing, light graphics work and video streaming while connected to Wi-Fi with display brightness set at 150 nits, the Legion Slim 5’s battery lasted for 7 hours and 21 minutes. While that couldn’t top the Blade 14’s performance (8:34), it left the Raptor Lake competition in the dust.

As we saw in the gaming benchmarks, the Legion Pro 5i, with its Core i7-13700HX, led the way in several benchmarks. However, the Legion Slim 5 has a clear advantage regarding battery life.

Heat on the Lenovo Legion Slim 5

While running our Metro Exodus stress test, the dual fans in the system kicked into high gear (as we would expect for a gaming laptop). However, the rush of air was not objectionable while gaming. While it was ever-present in the background, it didn’t intrude on my gaming experience once I cranked up the volume to about 40 percent. It would be a non-issue if you’re using a pair of the best gaming headsets.

We measured degrees 22.22 degrees Celsius (72 degrees Fahrenheit) on the touchpad and 39.88 C (103.8 F) between the G and H keys during the Metro Exodus stress test. The hottest spot on the underside of the chassis (near the far-left corner) was 50.11 C (122.2 F).

Webcam on the Lenovo Legion Slim 5

Like most new laptops introduced today, the Legion Slim 5 uses a 1080p webcam, a notable step up from the 720p units that have long been standard equipment.

Image quality was on par with other laptops in this class, delivering realistic skin tones, good detail and low noise. I’d happily use this camera for my Google Meet work calls or when chatting with family over Skype. As I mentioned earlier, there is no physical shutter to block the webcam. Instead, a little switch on the right side of the chassis (next to the SDXC slot) quickly enables/disables the camera.

Software and Warranty on the Lenovo Legion Slim 5

Four branded apps are installed from the factory on the Legion Slim 5. Legion Arena allows access to your games across Battle.net, Epic Games, Steam, Xbox, and Ubisoft services. Lenovo Hotkeys gives you a rundown of function shortcuts available on the system and Lenovo Voice provides controls for the microphone and text-to-speech functionality.

The most important of them all, however, is Lenovo Vantage. This is Lenovo’s command center for controlling and monitoring most aspects of the Legion Slim 5’s hardware. You can oversee system vitals for the GPU, CPU and SSD, change power profiles, and overclock the GPU (among other things). You can also access RGB settings for the keyboard within this app and perform system updates (drivers, software, BIOS, etc.).

Other installed apps include Nahimic (audio controls), X-Rite Color Assistant (display color profiles), Dolby Vision and a Lenovo-sponsored Dropbox promo. Of course, you’ll also find Microsoft’s usual smattering of junkware shortcuts to apps like Netflix, and TikTok sprinkled through the Start menu.

The Legion Slim 5 comes backed by a one-year manufacturer’s warranty.

Lenovo Legion Slim 5 Configurations

I’ve only found two configurations for the Legion Slim 5, with the base system available direct from Lenovo. It features an AMD Ryzen 5 7640HS processor, 16GB of DDR6-5600, a 512GB SSD, a GeForce RTX 4050 and a 16-inch 1920 x 1200 display for $1,019.99. That same configuration is $1,349.99 at Best Buy.

Our review unit is available as a Best Buy-exclusive and carries an MSRP of $1,529.99. That price includes a Ryzen 9 7940HS processor, 16GB of DDR5-5600, a 512GB SSD, a GeForce RTX 4060 and a 2560 x 1600 display. However, at this time, the laptop is on sale for $1,149.99 at Best Buy, representing a $380 discount. That is a lot of kit for less than $1,200, making the Legion Slim 5 a knockout bargain. However, if the laptop were priced closer to its $1,529.99 MSRP, the Legion Pro 5i would likely be a better choice.

Bottom Line

The Lenovo Legion Slim 5 didn’t dominate our gaming benchmark suite, but it was in the running or at least mid-pack in all the tests. It turned out solid productivity numbers, including high marks for its storage performance. However, the laptop was a real trooper regarding endurance, lasting over seven hours per charge in our tests. That was nearly two and a half hours longer than the Legion Pro 5i.

I would have liked for the screen to have been a bit brighter, but it is by no means a deal-breaker. The speakers are rubbish, but which is more of a concern if you plan on gaming with headphones. If you want a much brighter display and slightly better audio, the Acer Predator Helio Neo 16 is an option worth considering.

However, what’s most impressive about this laptop is the pricing. While its MSRP of $1,529.99 would give me a bit of pause, considering the generally superior performance of the Legion Pro 5i, the sale price of $1,149.99 makes the Legion Slim 5 a true bargain that can’t be ignored.

MORE: How to Buy a Gaming Laptop

MORE: Best Gaming PCs

MORE: Best Ultrabooks and Premium Laptops

In a live stream today, Chinese handheld gaming PC specialist Ayaneo unveiled its first device aimed at Android gaming. The new Ayaneo Pocket AIR looks a lot like the company's previous Windows-powered handhelds, but it manages to be thinner and lighter. Also, thanks to the Arm architecture, and retaining the "Windows handheld grade" battery, it is suggested that users can enjoy games "without worrying about battery life."

Android and Windows handheld markets may not really have a lot of overlap. On Tom's Hardware, we look most closely at PCs and PC gaming. PC AAA games titles are quite different from the library available on Android, except some common casual titles (and streaming services). 

Thus we think many readers won't be interested in this type of device for Android gaming, or games streaming, especially when they have powerful smartphones already. Nevertheless this device is worth a closer look to see Ayaneo's efforts in trying to jump into the Android market and compare the device with its Windows-powered lineage.

Starting with the basics, the Ayaneo Pocket AIR has a 5.5-inch 1080p AMOLED screen. The processor used is a 6nm MediaTek Dimensity 1200, which has 1x Ultra Core (Arm Cortex-A78 up to 3.0 GHz), 3x Super Cores (Arm Cortex-A78 up to 2.6 GHz), and 4x Efficiency Cores (Arm Cortex-A55 up to 2.0 GHz). Inside MediaTek's SoC there is also an Arm Mali-G77 MC9 GPU, support for various codecs, 5G, Wi-Fi, Bluetooth, and more. Despite the flagship smartphone guts, Ayaneo says that the Pocket AIR has an "x86-level powerful active cooling system." 

Similarly, it says this compact new device uses a "Windows handheld grade" battery, at 7,350mAh. Sadly, it hasn't been forthcoming with screen-on-time gaming endurance figures, so we will have to wait and see about extended battery life claims.

We have some physical details of the Ayaneo Pocket AIR to share, too. With its 5.5-inch screen you shouldn't expect a very hefty device, but at 380 g (13.4-ounces) it is about twice as heavy as a common-or-garden smartphone. The device's 'waist' is 17 mm thick. Ayaneo asserts that the Pocket AIR "fits easily in your pocket."

As the Ayaneo Pocket AIR is a pre-release, pre-crowdfunding stage device at the time of writing, we don't seem to have a great depth of specs, so details about memory and storage options seem to be absent right now.

Some other interesting details we have access to though, concern the controls. In brief, buyers of the Ayaneo Pocket AIR should expect a very similar extensive control set. So, here we have lots of buttons and pads including hall sensing joysticks and triggers, and there is the SoundTapMagic audio vibration system, an X-axis linear motor, and six-axis gyroscope included for your gaming delights.

On top of the Android installation, Ayaneo has ported its Windows gaming UI app, AYAHome, as well as the AYASpace management app. Another software nicety is the Ayaneo Cloud Game Database for retro gaming fans.

Running emulators looks like a focus of fun on the Ayaneo Pocket AIR, as there are many slides on the Ayaneo blog and Pocket AIR product pages highlighting this aspect of the device. Even its color scheme with "classic red and white," harks back to the 8- and 16-bit eras.

We don't have pricing or availability for the upcoming Ayaneo Pocket AIR. At the time of writing the firm is asking for sign-ups for news of its Indiegogo campaign launch. Let's hope it is priced competitively, to give this device some impulse purchase appeal. We've seen a lot of pricey handhelds from this firm recently, but it has made some sub-$300 devices in the past.

Ayaneo's CEO has just completed a live stream session about an upcoming 8.4-inch screen-equipped gaming PC handheld. Dubbed the Ayaneo Kun, behind the device's 2.5K screen, you will find the AMD Ryzen 7 7840U 'Phoenix' processor (no surprise). Perhaps afforded by the larger form factor, another attractive specification of the Kun is its 75 Wh (19,500 mAh) battery inside. These specs and boundary-pushing handheld dimensions aren't going to come cheap though, with Ayaneo boasting that the Kun is a "high-end masterpiece."

At the time of writing, with the Ayaneo Kun in pre-launch and pre-Indiegogo campaign stages, we don't have a full set of specifications for the device. However, there are enough specs and features that have already been shared to make this device very interesting.

In its pre-launch materials Ayaneo spends quite some time highlighting the Kun's controls, talking-up the variety and quality of the gaming control array. Being one of the biggest handhelds should also make it a comfortable hold (despite the probable extra heft) for adult hands.

Probably the best thing about offering a larger handheld is the opportunity for adding a beefier battery and more capable cooling system. These may be the biggest weaknesses with existing / rival handhelds, so it is good that Ayaneo has grasped the opportunity to improve in these areas.

Considering battery life first, the new 75 Wh cell sounds impressive. Remember that's nearly double Valve's Steam Deck battery and is approaching the limit of 99 Wh for laptops that are allowed on airplanes. Sadly, though it boasts about this battery, Ayaneo is remaining tight-lipped about expected gaming run times on this device.

Recent headlines about the Asus ROG Ally have been focused on an issue with its cooling system, resulting in fried MicroSD cards. Such bad publicity should focus device developers minds on adequate system-wide cooling, and Ayaneo's Kun hopefully makes use of its extra chassis capacity for bigger / better cooling components. The firm says that the Kun has "perhaps the strongest heat dissipation of Windows handhelds to date." Moreover, we learned that the cooling system uses a large main fan coupled with a trio of copper heat pipes.

As mentioned previously, the Ayaneo Kun is still at a pre-launch stage. Therefore the firm is keeping some powder dry with respect to offering the full specifications. Details like pricing, configurations, and availability are also being held back. Today the Ayaneo CEO said we should expect all these details "at the end of August." We expect the Ayaneo Kun to be quite pricey, but there will be early bird pricing alongside the Indiegogo launch, and a special gift box for those among the first adopters, too.

A fortnight ago Ayaneo was in our headlines for preparing what might be the thinnest and lightest AMD Phoenix handheld yet seen. The Ayaneo Air 1S Ultra contrasts with the Ayaneo Kun by featuring a relatively small 5.5-inch screen, measuring 18 mm (0.7 inches) thick, and weighing 405 g (0.89 pounds).

I’m always intrigued to see moderately-priced entries when looking for the best gaming laptops available on the market. While models loaded-to-the-gills with $2,000 (or higher!) price tags often pack tons of performance, gamers with more meager budgets are now awash with less expensive entries that offer a lot of bang for the buck.

With this in mind, that’s an all-new contender in the field, the Acer Predator Helio Neo 16, which offers an Intel Core i7-13700HX processor, Nvidia GeForce RTX 4060 GPU, 16GB of RAM, and a 16-inch QHD (2560 x 1440) display. It's not a budget system, but at $1,199 to start, $1,549 as tested, it's not unreasonable.

Acer wraps that hardware in a chassis that, while designed for 1440p gaming, wouldn’t look out of place in an office setting thanks to its restrained design.

Acer Predator Helios Neo 16 Specifications

Design of the Acer Predator Helios Neo 16

The Predator Helios Neo 16 is an all-new model for Acer, aimed explicitly at hitting a moderate price point for gamers. That said, I find the overall design very conservative and without much to stand out. The only elements of flash in the design are the WASD, arrow, and Predator Sense keys, which feature translucent blue keycaps topped in black versus the clear topped with black.

The black plastic chassis also has a slight shimmering effect when light hits it at certain angles. It’s subtle enough to add visual interest to the laptop without being in your face.

The 16-inch display has some thick bezels all around. The top bezel is roughly twice as wide as the sides, while the bottom bezel is massive. Acer attempts to break up the large visual expanse with a chromed Predator logo to little avail.

On the right side of the chassis, you find two USB 3.2 Gen 2 (Type-A) ports and a Kensington lock. Switching over to the left side, Acer includes a USB 3.1 Gen 2 (Type-A) port, an Ethernet jack, a MicroSD card reader slot and a 3.5 mm headphone jack. The back panel has an HDMI 2.1 port, two Thunderbolt 4 (Type-C) ports and a proprietary jack for the 330-watt power adapter.

The Predator Helios Neo 16 measures 14.18 x 11.02 x 1.05 inches and weighs 5.73 pounds. The Lenovo Legion Pro 5i and Gigabyte Aorus 15X are similar in dimensions and weight, while the Razer Blade 14, with a smaller display, was naturally sleeker at 12.23 x 8.97 x 0.70 inches and 4.05 pounds.

Gaming on the Acer Predator Helios Neo 16

The Predator Helios Neo 16 comes equipped with an Intel Core i7-13700HX CPU, Nvidia GeForce RTX 4060 GPU and 16GB of DDR5-4800 memory (2 x 8GB). The Core i7-13700HX has eight performance and eight efficiency cores and operates at a maximum turbo boost frequency of 5 GHz.

To kick off gaming testing with the Predator Helios Neo 16, I installed Cyberpunk 2077. I averaged 68 fps using high settings with ray tracing enabled, DLSS Balanced, and DLSS frame generation enabled at 1600p. Stepping down to 1200p, that figure rose to 90 fps.

Moving to our usual suite of gaming benchmarks, I lined the Predator Helios Neo 16 up against the Lenovo Legion Pro 5i (i7-13700HX, RTX 4060, 1600p), Gigabyte Aorus 15X (i9-13900HX, RTX 4070, 1440p) and Razer Blade 14 (Ryzen 9 7940HS, RTX 4070, 1600p). Please note that the RTX 4070-based systems will have an edge in gaming benchmarks, while the RTX 4060 in the Legion Pro 5i has a slightly higher boost clock than the one in the Predator Helio Neo 16.

I started with Shadow of the Tomb Raider (highest settings), which saw the Predator Helios Neo 16 land in last place with 97 frames per second at 1080p and 57 fps at 1600p. The Legion Pro 5i with the same processor and GPU achieved 107 fps and 62 fps, respectively. That discrepancy is explained by the slightly lower vertical resolution of the Legion Pro 5i (1440p versus 1600p) and the higher boost clock (2370 MHz vs 2250 MHz). The Aorus 15X leads all systems with 115 fps at 1080p and 73 fps at 1440p.

Next, I queued up Grand Theft Auto V (very high settings), where the Predator Helios Neo 16 tied for third place with the Blade 14 at 1080p with 90fps. However, its showing at 1600p puts it in last place at 49 fps compared to 52 fps for the Blade 14. The Legion Pro 5i took home the gold with 97 fps at 1080p and hit 53 fps at 1440p.

The Aorus 15X slipped into first place in Far Cry 6 (ultra settings) with 99 fps at 1080p and 80 fps at 1440p. The Legion Pro 5i took second place, while the Predator Helio Neo 16 landed in third place, with 88 fps at 1080p and 61 fps at 1600p.

Red Dead Redemption 2 (medium settings) once again saw the Predator Helios Neo 16 in last place with 65 fps/40 fps (1080p/1440p). However, only 8 fps separated first and last place at 1080p, while the difference at 1600p was just 5 fps.

Moving to Borderlands 3 (badass settings), the Predator Helios Neo 16 could only muster 82 fps at 1080p and 53 fps at 1600p. The Aorus 15X hit 98 fps at 1080p, and managed 67 fps at 1440p.

To round out our benchmarks, we ran the Metro Exodus benchmark (RTX settings) through 15 loops at 1600p resolution. The benchmark ran at an average of 53.08 fps (consistently ranging between 52.76 to 53.19 fps). During the run, the Core i7-13700HX performance cores averaged 2.19GHz, while the efficiency cores averaged 1.82GHz at 65.7 degrees Celsius (150.3 degrees Fahrenheit). The GeForce RTX 4060 GPU ran at an average of 2.29 GHz and measured 73.2 C (163.8 F).

Productivity Performance on the Acer Predator Helios Neo 16

The Predator Helio Neo 16 features an Intel Core i7-13700HX CPU, 16GB of DDR5-4800 memory (2 x 8GB) and a 1TB PCIe 4.0 SSD.

Our Predator Helios Neo 16 review unit achieved scores of 1,880 and 14,625 in single- and multi-core performance, respectively, in Geekbench 5. Not surprisingly, those numbers were remarkably close to the Legion Pro 5i, which uses the same processor. The Aorus 15X, with its stronger Core i9-13900HX, took top honors with a single-core score of 2,007 and a multi-core score of 18,129.

The Predator Helios Neo 16 came up well short against the competition in our 25GB file transfer test. The laptop hit 1,047.37 MBps compared to 1,230.04 for the third-place Blade 14. However, the Legion Pro 5i took first place with 1,792.63 MBps.

Finally, the Handbrake transcoding test saw the Predator Helios Neo 16 tied for second place at three minutes and 55 seconds, compared to three minutes and 39 seconds for the first-place Legion Pro 5i.

Display on the Acer Predator Helios Neo 16

The PredatDisplay on the Acer Predator Helios Neo 16or Helios Neo 16 uses a 16-inch IPS panel with a QHD+ (2560 x 1600) resolution and a 165Hz refresh rate. The panel is driven by an Nvidia GeForce RTX 4060 GPU.

16 inches is becoming a popular size for gaming notebooks. 15.6 inches was once the norm, but we’re seeing more laptops adopt 16-inch panels, specifically with a QHD+ resolution. As far as displays go, the IPS panel in the Predator Helios Neo 16 is highly capable. The anti-glare coating muted reflections in almost every possible situation except direct sunlight.

The Predator Helios Neo 16’s panel also gets incredibly bright, hitting 499 nits according to our light meter, which is right at the factory's claim of 500 nits. Instrumented testing with a colorimeter showed that the Predator Helios Neo 16 closely matched the Legion Pro 5i with 81.6 and 115 percent coverage of DCI-P3 and sRGB, respectively.

I watched the Oppenheimer trailer on YouTube in 1080p Premium HD (Enhanced Bitrate) and was impressed with the gritty look and feel. If you know Christopher Nolan films, you’re familiar with the muted colors and realistic tones compared to the more vibrant palettes you’d see in, say, the typical Marvel film.

Switching gears completely, I played the Barbie trailer which had just enough pink plastered throughout for me to reflexively reach for a bottle of Pepto Bismol. The blue of Barbie’s blazer and light blue skies of Malibu Beach also popped on the screen. Perhaps it’s time for a Barbenheimer double feature?

I also played quite a bit of Cyberpunk 2077, which looks gorgeous with the details cranked to high and with ray tracing enabled. The fine particles of smoke wafting through the air, the mirror-esque reflections in water, and the way that red light diffused inside bars looked terrific. The same could be said for vehicles and how the sunlight glistens on the chrome trim.

Keyboard and Touchpad on the Acer Predator Helios Neo 16

Given the price point, you won’t find a fancy mechanical keyboard on the Predator Helios Neo 16. However, the keyboard has nicely spaced keys and is comfortable to type on. Acer even found enough room to fit in a number pad (for those who like that added functionality).

As usual, I leverage keyhero.com to gauge my typing speed and accuracy. I typed an average of 80 words per minute with an accuracy of 95.55 percent. For comparison, I achieved 90 wpm and 94 percent accuracy with the Razer Blade 14.

Acer includes its PredatorSense app for configuring the RGB lighting on the Predator Helios Neo 16. The laptop doesn’t feature per-key lighting but instead employs a four-zone arrangement. You can select any hue that you want for each of the four zones, and there are eight different lighting effects that you can choose.

There’s also a reasonably-sized touchpad, which measures 4.9 x 3.1 inches. My finger glided smoothly over the plastic surface and it had a satisfying click when pressed.

Audio on the Acer Predator Helios Neo 16

The Predator Helios Neo 16 uses stereo speakers with DTS:X Ultra Audio and Acer TrueHarmony. Acer claims that TrueHarmony, which refers to the foam-coated paper cones in the speaker, helps “maximize clarity” and “lower distortion” for more immersive sound.

In practice, the speakers sounded fine, at best. “Cracker Island” by the Gorillaz sounded lifeless, with no bass and muffled voices. The biggest detractor was increased distortion as I approached around 75 percent volume. Fiddling with settings in the Realtek Audio Console and the DTS:X Ultra apps had little effect on the sound.

While playing Grand Theft Auto V, things were more palatable, with explosions and firefights coming through loud and clear. And love him or hate him, hearing Trevor’s gravelly voice still made me smile a decade after he first appeared.

Upgradeability on the Acer Predator Helios Neo 16

Accessing the internals of the Predator Helios Neo 16 is accomplished by removing 11 Philips head screws. Then the whole bottom panel is easily removed using some well-placed fingernails.

With the bottom panel removed, you can access the battery, two SO-DIMM slots (each occupied with 8GB DDR5-4800 modules in our review unit), and two M.2 slots. One of the M.2 slots holds the 1TB PCIe 4.0 SSD, while one is free for future expansion. There’s another slot for the M.2 Wi-Fi/Bluetooth combo card.

Battery Life on the Acer Predator Helios Neo 16

As I saw in the gaming benchmarks, the Predator Helio Neo 16 is near the back of the pack in overall performance. I hoped that the meager gaming results would result in above-average battery endurance. However, I was wrong. The Predator Helios Neo 16 lasted just shy of three hours (2:59) on our battery test, which browses the web, streams videos, and runs OpenGL graphics tests.

The next closest competitor was the Legion Pro 5i (4:54) and then the Aorus 15X (5:20). The Blade 14 outlasted all the assembled laptops with a runtime of 8:34. The Predator Helios Neo 16 uses a relatively beefy 90 WHr battery, which made the results even more puzzling.

Heat on the Acer Predator Helios Neo 16

The Predator Helios Neo 16 uses a Core i7-13700HX processor (55W) and a GeForce RTX 4060 GPU (140W). Acer employs dual fifth-generation AeroBlade 3D fans in the system and Liquid Metal thermal grease for the CPU.

In testing during our Metro Exodus stress test (15 consecutive runs), the Predator Helios Neo 16 measured 25.8 degrees Celsius (78.5 degrees Fahrenheit) on the touchpad and 35.3 C (95.5 F) between the G and H keys. The underside of the laptop, which was also the hottest, measured 41.4 C (106.5 F) near the center.

Webcam on the Acer Predator Helios Neo 16

The Predator Helios Neo 16’s webcam is situated directly above the display, center stage. Acer uses a 1080p sensor, which has become commonplace in most new gaming notebooks but lacks a privacy shutter or Windows Hello facial recognition technology.

While not as good as some of the best webcams available, the Predator Helios Neo 16’s 1080p webcam had good detail and accurately captured my skin color. It also had no trouble capturing detail from my mustache and stubble on my face.

Software and Warranty on the Acer Predator Helios Neo 16

The Predator Helios Neo 16 is loaded with bloatware from the factory, which many customers will rarely use. You’ll Acer Purified Voice Control, Acer Product Registration, Acer Care Center (system vitals, recovery management, etc.), Acer Planet 9 (a social gaming platform), DTS Surround Unbound, DTS:X Ultra, Express VPN (30-day free trial), Dropbox (100GB free offer when you sign up for Acer customers), the Killer Intelligence Center and the Realtek Audio Console.

In addition to all those preinstalled apps, the usual Windows 11 bloatware is included, like shortcuts for Spotify, TikTok, ESPN, and Instagram.

However, the app that I used the most was Acer PredatorSense. This utility has its own dedicated key on the keyboard for easy access and allows you to set power profiles, adjust system fan speeds, and control the RGB lighting for the keyboard (among other things). It also allows you to monitor system vitals like fan speed, CPU/GPU temperature, Wi-Fi statistics and RAM speed/utilization.

The Predator Helios Neo 16 comes with a one-year warranty.

Acer Predator Helios Neo 16 Configurations

The Predator Helios Neo 16 base configuration has a Core i7-13500HX processor, GeForce RTX 4050, 16GB of RAM, a 512GB SSD and a 16-inch FHD+ display for $1,199. The mid-range configuration swaps out the RTX 4050 for an RTX 4060 and adds $100 to the price.

Our review unit came configured with a Core i7-13700HX processor, GeForce RTX 4060, 16GB of RAM, a 1GB SSD and a 16-inch QHD+ display for $1,549 and is available exclusively from Best Buy.

Bottom Line

The Acer Predator Helios Neo 16’s lower gaming numbers are easy to explain with the higher-performing GeForce RTX 4070 GPUs on two of the competitors. In addition, the Lenovo Legion Pro 5i features the same Core i7-13700HX CPU and RTX 4060 GPU, albeit with a slightly higher boost clock. So it’s not a surprising result.

However, when you add the low storage performance and the disappointing battery, which lasted just under three hours, it becomes harder to recommend the laptop. 

The Predator Helios Neo 16 has some positives, like its bright display and attractive chassis. But it’s hard to overlook its performance shortcomings and, especially, its battery life. The final knock against the laptop is its price; at $1,549, the Legion Pro 5i represents a better option for $100 less.

MORE: How to Buy a Gaming Laptop

MORE: Best Gaming PCs

MORE: Best Ultrabooks and Premium Laptops

Dell's Premium XPS laptops are meant to be its best examples of design and quality. The XPS 13 Plus ($1,149 to start, $1,449 as tested) upends traditional laptop design with its capacitive touch bar, edgeless keyboard and borderless touchpad, not all of which work out perfectly, but Dell gets points for thinking like no one else. 

This is the same chassis as the XPS 13 Plus from last year — in fact, so little has changed that Dell hasn't changed the model number (9320), as it usually does with a refresh.

Unfortunately, the XPS 13 Plus might be a little too style-oriented. The whole laptop has just two Thunderbolt 4 ports and its battery life is well below par in this class.

Design of the Dell XPS 13 Plus (9320, 2023)

The XPS 13 Plus delivers the high style and quality that the XPS brand is known for. Its ultra-chic design exudes modernness, from its nearly bezel-less screen to its edge-to-edge keyboard.

The chassis and lid, made of aluminum and glass,  are exceptionally stiff and resisted my attempts to induce flex. This laptop looks too delicate to throw around, but it shouldn’t need to be coddled given that it’s built so well. Picking it up one-handed, for instance, doesn’t feel like it stresses the laptop at all.

I have mixed feelings on the platinum color of my review unit as I’ll explain later in the keyboard section, though it is a good break for the norm. Dell also offers dark graphite.

The XPS 13 Plus is 11.63 x 7.84 x 0.60 inches (295 x 199 x 15.2 mm) and 2.77 pounds (1.26 kg). The Asus ZenBook S 13 OLED (UX504) and the Apple MacBook Air (M2) are about 0.2 inches thinner and the Asus is a full half-pound lighter, but the Dell’s sub-three-pound weight is within reason for an ultraportable. The HP Spectre x360 13.5 is a hair over the three-pound barrier but offers the convenience of a 2-in-1 screen design.

Port selection is a serious weakness for the XPS 13 Plus: It has just two Thunderbolt 4 (USB-C) ports and nothing else, not even a headphone jack. Making matters worse, the included power adapter will take up one of the Thunderbolt 4 ports.

Dell includes USB-C to USB-A and USB-C to headphone jack adapters, but they’re small consolation; remembering to take them is a pain and they’re easily lost. Unless you go with wireless everything – the XPS 13 Plus has Wi-Fi 6E and Bluetooth 5.2 – budgeting for a docking station is a smart idea.

Dell XPS 13 Plus (9320, 2023) Specifications

Productivity Performance on the Dell XPS 13 Plus (9320, 2023)

The XPS 13 Plus I’m reviewing has an Intel Core i7-1360P processor, 16GB of RAM, a 512GB SSD, and Intel Iris Xe integrated graphics. The 28-watt P-class processor is a step up in performance from the 15-watt U-class chips that seem to grace most ultraportables.

The XPS 13 Plus will be competing against the Apple MacBook Air (M2), the Asus ZenBook S 13 OLED, and the HP Spectre x360 13.5 in our performance comparison charts. The MacBook Air uses Apple’s M2 CPU while the ZenBook S 13 sports an Intel Core i7-1355U and the Spectre x360 13.5 a previous-generation Core i7-1255U. The latter two chips have 10 cores (two Performance and eight Efficient) whereas our Dell’s Core i7-1360P has 12 cores (four Performance and eight Efficient). Apple’s M2 chip is always tough to beat, so we’ll see how it stacks up.

The XPS 13 Plus finished first in the cross-platform Geekbench 5 benchmark multi-core test, its score of 9,938 points just ahead of the next-best MacBook Air’s 8,919 points. The MacBook Air led in single-core, though, with 1,932 points to beat the XPS’s 1,814 points. The Asus scored about the same as the XPS 13 Plus in the single-core test though its U-class CPU held it back in the multi-core scenario.

Moving on, the MacBook Air soundly beat the XPS 13 Plus in our Handbrake transcoding test, finishing in 7 minutes and 52 seconds versus the Dell's 8 minutes and 14 seconds. The Asus and the HP trailed significantly, which wasn’t surprising given their weaker CPUs.

Things picked up for the XPS 13 Plus in our 25GB file transfer test, where it finished first with 1,658.62 MBps. The Asus was right on its heels with 1,559.7 MBps. The MacBook Air was much slower, at 958.85 MBps.

We stress productivity laptops by running the Cinebench R23 CPU stress test 20 times in a row, which usually causes heat soak in the laptop’s cooling system and therefore gives a good baseline for its long-term behavior under load.

The XPS 13 Plus dropped from its initial score of 10,095 points to stabilize in the mid-6,000-point range, with a random spike to 8,304 points in the middle. The initial drop isn’t unusual and the XPS 13 Plus does otherwise manage to keep its performance stable.

During the test, the CPU’s Performance cores ran at an average of 2.0 GHz and the Efficient cores at 1.55GHz. The average processor temperature was 72.34 degrees Celsius.

Display on the Dell XPS 13 Plus (9320, 2023)

I tested with the XPS 13 Plus with Dell's optional 3456 x 2160 OLED screen. It's a lofty $300 upgrade over the base 1920 x 1200 IPS screen, but it only took about a minute of the Dune: Part Two trailer to convince me I'd spend the money. 

The OLED technology's infinite contrast ratio shows exquisite details in shadows and highlights that you aren't likely to see or appreciate on a lesser screen. Explosions and gunfire are especially vivid. OLED is also famous for its color reproduction, of which Dune doesn't exercise much, so I switched over to Spider-Man: No Way Home and wasn't disappointed. Looking at this screen is a real pleasure.

The XPS 13 Plus' small size meant I also used its touch-screen functionality more than I thought I would. I often used it in my lap and gripped the screen with one hand, then scrolled with my thumb. The glossy screen surface is slick but has an anti-reflective treatment, so it doesn't act like a mirror. It's not anti-glare or matte, but I could only make out a dull reflection of myself, and I didn't have issues with overhead lights causing eye strain.

The Dell’s screen covers 80.5% of the DCI-P3 gamut, second only to the HP’s 87.7% among its direct competitors. Its 373.6-nit peak brightness is almost the same as that of the ZenBook S 13 OLED and well south of the MacBook Air’s 489 nits, but it’s far from being a dim screen. I didn’t wish for it to be brighter, and that’s a good sign.

Keyboard and Touchpad on the Dell XPS 13 Plus (9320, 2023)

I admit to having serious reservations about the XPS 13 Plus' keyboard design the moment I took it out of the box. Not only is there no spacing between the keys (Dell calls it "zero lattice"), but the top row (F1 through F12 and associated functions) is a touch bar, not actual keys. I nonetheless managed 111 words per minute with 100% accuracy in the Monkeytype typing test, and I can't remember doing better than that.

The large keys have satisfying cushioned feedback with healthy travel. The power button is the blank key at top right, which doubles as a fingerprint reader.

Layout-wise, the keyboard’s half-sized up and down arrows squeezed between full-size left and right arrows are my main pet peeve. The other issue I had was visibility; white backlighting in the daytime made it hard to differentiate key symbols on the platinum keys, so ironically, it was easier to see the key symbols with the backlighting turned off. Here's the keyboard with its backlighting on:

And then off:

The effect isn’t as bad as the photos make out, but you might find yourself tilting the laptop or looking closer if you’re not a touch typist. I'd probably get the XPS 13 Plus in its optional graphite color for this very reason.

But how about that capacitive touch bar? It's much better than I thought, mostly because it seemed highly responsive to my input - as in, it reacted as fast as a regular key. The only downside is that there's no tactile or haptic feedback of any kind, and that takes getting used to. Fortunately, I had no issues with my touches registering. I did find myself looking down to locate the key I needed on the touch bar, though over time I was able to rely on muscle memory, especially for shortcuts I use a lot, like Ctrl + F4 to close tabs. 

Pressing the Fn key instantly switches the symbols to F1 through F12. You can also enable Fn Lock by pressing Fn + Esc if you want F1 through F12 to show as primary, though doing that means you lose dedicated access to the Home, End, and Insert keys. Page Up and Page Down are only available as secondary functions in the up and down arrow keys.

The XPS 13 Plus' touchpad is also controversial because it doesn't have a defined border. Dell’s assumption is that users naturally touch just below the spacebar, and that's valid for me. Touch sensitivity starts a few millimeters below the spacebar and above the front edge of the palm rest and extends just beyond the edges of the spacebar. You can tap to click, but lightly pressing on the surface rewards with a satisfying tactile bump. I always appreciate it when physical clicking is practically silent as it is on this XPS, too.

Audio on the Dell XPS 13 Plus (9320, 2023)

The XPS 13 Plus pumps out impressive sound for a laptop its size. Listening to Breaking Benjamin’s “Angels Fall”, I noticed clear separation between vocals and instruments, with enough low-end to keep the track from sounding brittle as so often happens with rock music on laptop speakers. I also noted that the speakers maintained the airy acoustic guitar and smooth vocals of “Torn” by Natalie Imbruglia.

The included WavesMaxx Audio Pro app provides EQ and other sound tweaks, including 3D spatial audio if you allow the webcam to track your head movement. The hard rock EQ seemed to improve the separation in “Angels Fall”, but I ended up going back to the No EQ setting, putting the details dial at 25%, the width dial to 25%, and turning the artificial-sounding MaxxBass dial to 0% for an all-around setting. I might not have done that had the app included an automatic EQ setting; I don’t want to fuss with EQ all the time.

Upgradeability of the Dell XPS 13 Plus (9320, 2023)

The only user-changeable parts in the XPS 13 Plus are its storage drive and battery. The bottom panel is held on by six Torx T6 screws, which are all the same size but aren’t retainer style. You’ll then need a plastic pry tool to pop the panel’s clips free; I started at the rear near the heat vents, then went up the right edge.

The battery covers about half the chassis. The SSD is M.2 2280 format and is covered by a heatsink. The memory and even the wireless card are soldered down, which is  disappointing, but not surprising on an ultraportable.

Battery Life on the Dell XPS 13 Plus (9320, 2023)

The XPS 13 Plus’ 55 WHr battery and 28-watt processor aren’t a great combination for long battery life. It managed a mere 6 hours and 20 minutes in our battery test, which consists of continuous web browsing over Wi-Fi, video streaming, and OpenGL workloads with the screen set to 150 nits of brightness. That’s over an hour worse than the 7 hours and 34 minutes we recorded from the original XPS 13 Plus.

The ZenBook S 13 OLED had no trouble broaching the 10-hour mark, with its more powerful 63 WHr battery and power-friendly 15-watt CPU. None of these units touched the MacBook Air’s 14 hours and 6 minutes.

Heat on the Dell XPS 13 Plus (9320, 2023)

We measure the surface temperatures of productivity laptops while we run our 20-loop Cinebench R23 CPU stress test. The keyboard center of the XPS 13 Plus (between the G and H key) reached 95 degrees F (35 degrees Celsius) while the touchpad center (which is just below the spacebar) reached 85 F (29.4 C). The underside of the laptop was slightly warmer, reaching 97 F (36.1 C) in the center and peaking at 108 F (42.2 C) near the heat vents.

Overall, the XPS 13 Plus never felt overly warm to my hands. I had no reservations using it on my lap. Using it in your lap is perfectly fine since its air intakes are towards the edges, not under the bottom panel.

Webcam on the Dell XPS 13 Plus (9320, 2023)

The XPS 13 Plus' webcam is properly positioned over the screen to be as close to eye level as possible, though the laptop's size means you'll still be looking down at most of your subjects - ah, chat-mates. It exposed my face nicely despite the horrible lighting conditions in my local coffee shop.

The webcam's resolution is unfortunately just 720p, which is hard to accept on a premium laptop and would have me shopping for one of our best webcams. A 1080p webcam would have produced a sharper image.

The upside of this camera is that it supports infrared for facial recognition, so you can at least log into Windows using Windows Hello. That said, the webcam lacks a sliding privacy shutter, and I didn't see a shortcut key to disable it.

Software and Warranty on the Dell XPS 13 Plus (9320, 2023)

As a premium laptop, the XPS 13 Plus doesn’t have unwanted software or trial apps installed, though unfortunately it has many of the apps that are typically included with Windows 11, like Disney Plus. The included MyDell app has system settings, such as battery charging preferences, and lets you tweak the fan profile for cool or quiet modes. I didn’t find it necessary to change the latter from the default balanced setting since the fans were well-behaved and the laptop didn’t get too hot in my lap.

The mentioned Waves MaxxAudio Pro app is useful for tweaking sound settings.

Last is Killer Control Center. This seems like more of an app that belongs on a gaming laptop, but it can be used to analyze network traffic and give priority i.e., the most bandwidth to your preferred apps. It could be useful if you have little bandwidth to spare.

The XPS 13 Plus comes standard with a one-year warranty with Dell's premium support tier.

Dell XPS 13 Plus (9320, 2023) Configurations

Dell offers the XPS 13 Plus as a built-to-order model on its site. All models appear to include the Core i7-1360P processor and Iris Xe integrated graphics of my review unit.

The base $1,149 configuration ($250 off at the time of this review) includes 16GB of RAM, a 512GB SSD, and a 1920 x 1200 non-touch display. The RAM is factory configurable to 32GB, a wise option if you’re a heavy multitasker – remember, it can’t be upgraded later. Available storage upgrades are 1TB and 2TB.

Dell offers the 1920 x 1200 screen with touch, or you can step up to the 3456 x 2160 OLED touch screen of our review unit. The top screen is a 3840 x 2400 touch panel that Dell rates for 500 nits, though it’s not OLED.

The XPS 13 Plus can also be had in platinum or graphite colors. Graphite would be my pick if I had to do it again because of the contrast issues I had with the platinum keyboard and white backlighting.

Bottom Line

The key reasons to spend extra for the XPS 13 Plus over a more budget-oriented ultrabook like the HP Pavilion 14 Plus are its premium quality and design. The XPS 13 Plus delivers on those fronts with a gorgeous aluminum and glass chassis that’s exceptionally easy to transport. Few laptops look more chic than this one.

Uniqueness is also a factor when you’re spending a lot, and the XPS 13 Plus delivers there, too. Its most controversial feature is its capacitive touch bar, but I was pleasantly surprised with its responsiveness and didn’t miss a real physical keyboard row. I also had reservations about the edgeless keyboard, but it turned out I could type just as fast on it as I can on any laptop keyboard. The borderless touchpad was my least favorite feature because it’s a little strange not to have a defined border, but it never became a problem since I tend to favor touching the center of the touchpad.

Practicality is where things go askew for the XPS 13 Plus. Just two Thunderbolt 4 ports is limiting to begin with, not to mention that you lose one when you plug in the power adapter. No headphone jack is difficult to forgive – laptop designers need to realize that many of us appreciate the simplicity of wired devices that don’t need batteries.

Battery life is the XPS 13 Plus’ other major downside, lasting far fewer hours than the Asus ZenBook 13 S OLED and the Apple MacBook Air. Those laptops also sport better port selections (especially the Asus) and command similar money. The MacBook Air remains our favorite in this category if you can use MacOS, and the Asus is a good Windows alternative. Also consider the HP Elite Dragonfly G3 if your budget permits.

MORE: How to Buy a Gaming Laptop

MORE: Best Gaming PCs

MORE: Best Ultrabooks and Premium Laptops

Ubisoft and OWO may not want to make you bleed, but the duo is certainly hoping you feel the same punches and stabbings that you make the next protagonist endure. That's because the companies have announced the launch of a limited, Mirage edition of OWO's haptic feedback shirt, a piece of wearable technology that constricts sections of your torso to transmit the feeling of impacts, cuts, and compression forces. The byline? You'd better ensure you're a good caretaker for your in-game character, Basim, or you'll feel exactly how good (or bad) of a player you are.

The haptic feedback Mirage shirt will be part of a limited 100-unit run and builds on OWO's own premium haptic offering, the Founder Limited Edition. That product ships with ten haptic sensation locations (the places where the suit can compress and decompress to simulate contact), 100-ohm high-conductivity electrodes, Bluetooth 5.2 connectivity, and an eight-hour rechargeable USB Type-C battery. Depending on the settings, we can't imagine anyone wanting to be inside a walking machine of pain for the duration of its battery, so maybe OWO could have shaved some weight or pulled some additional constricting power from the suit's battery.

But the Mirage edition, being a limited one, brings some additional bells and whistles: according to OWO's page, the device will come bundled with an Assassin's Creed-style lycra shirt (a particularly useful material for all the sweating from playing contact games will generate); 20 gel pads; a storage pouch, and of course, the actual "beat me up" simulator itself and the world to be beat in: the OWO device and a digital copy of Assassin's Creed Mirage.

Luckily, users can customize their individual experiences, feedback intensity (and pain thresholds) through the OWO device and mobile app. Through it, users can make it so that a 50mm bullet feels like the lightest hand-slap in the world. OWO's sensations library (the range of contact that the vest can simulate) includes experiences ranging from wind, stress, and freefall to knife wounds, severe abdominal wounds, and shots... complete with exit wounds...

Uh.

Before you ask: No, it can't play Crysis (the game isn't supported yet), and yes; I checked. But it is compatible with Halo Infinite, Apex Legends, Half-Life: Alyx (VR), CS:GO, and Valorant, among others. Counting the soon-to-be-included Assassin's Creed: Mirage, that's quite a heavy-weight support list already.

Assassin's Creed Mirage is scheduled for release on October 12, 2023, for PC, PlayStation, Xbox, and the Amazon Luna platforms. The game will also be available to all Ubisoft+ subscribers at release.

Valve's Steam Deck runs Linux, while the Asus ROG Ally is a Windows machine. But both are computers, so you can install what you want on them. Phoronix compared Valve's Steam Deck and the new Asus ROG Ally on Linux explicitly to see which handheld console is better at gaming under similar operating systems. With the OS swapped from Windows 11 to Arch Linux on the ROG Ally, the performance margin between the two consoles didn't change much. The ROG Ally exhibited superior gaming performance to the Steam Deck, but the Steam Deck was able to outperform the ROG Ally in power efficiency and battery life.

Testing the ROG Ally on Linux is definitely an interesting proposition, since many Linux distros arguably feature less bloatware than Microsoft's latest Windows 11 operating system, which could improve performance and battery life. Linux versions also come with different AMD-based drivers that can change the performance behavior (scheduling) of AMD's Ryzen CPUs in Linux compared to Windows, which can potentially improve performance in some cases.

But according to Phoronix's testing, installing Linux on the ROG Ally didn't really change its behavior compared to running games on Windows 11. The handheld was still faster than the Steam Deck in the right performance profiles, but still lost the race in efficiency, with the Steam Deck consuming far less power than the ROG Ally.

Even when the ROG Ally was running in its more power-efficient modes, the handheld could only match the performance of the Steam Deck in most cases and was unable to drastically outperform it.

For instance, in Cyberpunk 2077, the ROG Ally ran an average of 61 FPS with Arch Linux's ACPI performance profile but dropped down to 45 FPS when running in the OS's standard power profile. The Steam Deck (which is still running SteamOS) was able to almost match the ROG Ally's in its non-performance power setting, outputting 44.66 FPS average.

In terms of power consumption, the Steam Deck was massively better, consuming an average of just 10.76 watts in Cyberpunk 2077, while the ROG Ally in its performance preset consumed 30W of power - 3x more power than the Steam Deck. Sadly, the ROG Ally's more power-efficient profile doesn't save it here, and still consumed more power than the Steam Deck, coming in at 14.77 watts average.

Cyberpunk 2077 was one of the worst-case scenarios for the ROG Ally, however, most of Phoronix's testing suit still exhibited similar behavior. The only benchmarks that didn't see this behavior were the CPU tests, which unsurprisingly showed substantially better performance with the ROG Ally in all circumstances compared to the Steam Deck — thanks to the far newer Zen 4 architecture and four additional CPU cores.

Coming back to Linux specifically, Phoronix's results align closely with our own review of the ROG Ally (running Windows 11), in that the Steam Deck was able to greatly outperform the ROG Ally in battery life and power efficiency in the console's performance profile. Sure you could back down to a more power-efficient mode, but you'd be getting the same performance as the Steam Deck, with hardware that is two generations newer.

Phoronix's results proved that the ROG Ally's efficiency differences are most likely related to the hardware and cannot be rectified with an OS swap.

Asus's alluringly powerful handheld, the ROG Ally, has been subjected to one of iFixIt's extensive and detailed teardowns. The video teardown was positive, praising the device's easy-to-service battery, thumbsticks, fans, and SSD. However, the Asus ROG Ally loses some points due to a difficult-to-replace screen and the (lack of) availability of some custom spare parts.

iFixit wants to see under the skin of the Asus ROG Ally to check how it fares against the "undisputed king of PC-based gaming handhelds," Valve's Steam Deck. The repairability of a device is essential to tech enthusiasts, so the assessment of sites like iFixit is crucial to many users. Our extensive reviews of the Asus ROG Ally and Valve Steam Deck also opened units up, which can be read via the inserted links. However, iFixit goes considerably deeper in this respect.

Opening up the ROG Ally was a cinch, with just six Philips screws and a couple of tabs/clips needing to be sprung to get the back off. The battery has a do-not-tamper sticker (and two smaller warning stickers on other components). However, this is more safety advice than legal - as Li-ion batteries can be volatile if mistreated. iFixit described the ROG Ally's easily replaceable battery as its "biggest repair advantage" over the Steam Deck.

Despite the warning, the battery in the ROG Ally is more accessible to replace than that of the Steam Deck, as it is screwed down rather than glued down. Similarly, easy-to-unscrew components included the M.2 2230 form factor SSD (which can be upgraded to 2TB at the time of writing).

Further disassembly highlighted more user-friendly considerations to the design. For example, the thumbstick assembly is on a modular board, which is attached using screws and a detachable data cable. When detached, turning two more screws allows a user to separate the physical thumbsticks - which are prone to wear, depending upon the gaming styles and preferences of the user. The heatsink and fan assembly are modular and easy to remove / service.

After removing the mainboard, removing the modular speakers, power button, fingerprint sensor, LED ring, and other bits aren't tricky.

The screen wasn't very user-friendly to remove. Like many smart devices, this was glued in, and iFixit used a combination of scalpel and guitar picks to cut and lever it out to minimize the danger of breaking the glass (perhaps some hot air would be helpful on the glue).

A final drawback highlighted by the iFixit video concerns the lack of official ROG Ally spare parts. Asus hasn't yet committed to providing these to repairers or the public, but if it follows the trend set by its principal rival, it should do so in due course.

iFixit ends its video by slamming the ROG Ally's UI and its use of Windows. However, the good and bad aspects of Windows are very well known to Tom's Hardware readers, and we shared an extensive discussion of Armoury Crate SE on the Asus handheld in our review.

Gigabyte has creative aims with its Aero 14 OLED ($1,600.11 as tested) has. Inside the 14-inch chassis is a powerful Intel Core H-class processor and an Nvidia RTX 4050 graphics card with special video drivers for creative apps.

And if that’s not enough excitement, this Gigabyte also has fantastic build quality in a stylish chassis and a stunning OLED screen that will make anyone look twice. There aren’t too many laptops with screens that look this good. . 

But the Aero 14 OLED doesn’t follow up its successes in key areas, especially battery life, which is hard to look past for a laptop that’s supposed to be highly portable.

Design of the Gigabyte Aero 14 OLED

Gigabyte seems to have spared no expense designing the Aero 14 OLED. There’s no plastic or lack of strength here, thanks to a CNC aluminum chassis. This laptop feels rock solid.

Gigabyte even uses a special anodization process that gives the a shimmer-like effect. It indeed appears to emit subtle rainbow colors in direct light, which vary depending on the viewing angle.

The lid backing has laser-cut "AERO" lettering, which is passively backlit in white by the display. It’s another signal that this is a high-end laptop.

Overall, the Aero 14 OLED’s bright exterior, especially its white keyboard, help it look less pedestrian than your average ultrabook or workstation. It’s a refreshing look that avoids any  polarizing traits.

The Aero 14 OLED is 12.3 x 8.7 x 0.67 inches (WDH) and 3.28 pounds, making it smaller and lighter than the Lenovo Slim Pro 7 (12.81 x 8.92 x 0.61 inches, 3.5 pounds). The Apple MacBook Pro 14 (12.31 x 8.71 x 0.61 inches, 3.5 pounds) is nearly the same size but also heavier. The Aero 14 OLED is less than a tenth of an inch thicker than either laptop.

Something I dislike about this design is that the display doesn’t tilt back any further than shown in our photos, about 45 degrees past vertical. This angle worked fine sitting at my desk but it wasn’t ideal for laying on a couch and propping the laptop on my knees.

Left-edge ports include USB-C 3.2 Gen 2 (10Gbps), a 3.5 mm headphone/microphone jack, and a MicroSD card reader. The power adapter connects to the USB-C port on this edge.

The right side has two Thunderbolt 4 (USB-C) ports.

There are also ports on the rear, including USB-A 3.2 Gen 1 (5Gbps) and HDMI 2.1 video output. These are conveniently located to keep cables out of the way in desktop setups. That said, I wish the side ports were further back  so that plugged-in devices don’t potentially interfere with using an external mouse.

Gigabyte Aero 14 OLED Specifications

Productivity Performance on the Gigabyte Aero 14 OLED

We’ tested the Gigabyte Aero 14 OLED with an Intel Core i7-13700H processor, an Nvidia GeForce RTX 4050 6GB graphics card, 16GB of RAM, and a 1TB SSD. This laptop is aimed at content creation, such as video and photo editing. The RTX 4050 runs special Nvidia Studio drivers that are meant to  maximize compatibility with creative apps, such as Adobe Photoshop.

Gamers shouldn’t look here, though; the RTX 4050 has just a 45-watt maximum graphics power rating, and even if it had a much higher power rating, it would still have trouble driving the Aero 14 OLED’s 2880 x 1800 screen with playable framerates in modern games.

The laptops we’re using for performance comparisons include the mentioned Apple MacBook Pro 14, the Lenovo Slim Pro 7, and the larger 16-inch Samsung Galaxy Book 3 Ultra. The MacBook’s M2 Pro processor is quite strong, so it should stack up nicely to the Core i7-13700H in our Gigabyte, which is also used in the Samsung. The Lenovo’s eight-core Ryzen 7 7735HS processor and previous-generation RTX 3050 graphics card aren’t likely to catch the Aero 14 OLED, though it’s a much less expensive machine for those who don'to have as much to spend ($1,199 as tested).

The Aero 14 OLED completed the cross-platform Geekbench 5 ahead of the Samsung with an identical CPU, with 12,842 points versus 12,212 points in the multi-core scenario, but it was outclassed by the MacBook Pro (15,120 points). The Lenovo’s Ryzen chip scored just 9,031 points. Admittedly, the HS-class Ryzen chip is a little out of its depth here.

Things took a turn for the worse in our Handbrake transcoding test, where the Aero 14 OLED finished dead last with a time of nine minutes, 34 seconds. The MacBook Pro again finished in the lead and was distantly trailed by the Samsung and the Lenovo, the latter continuing to bring up the rear.

At least the Aero 14 OLED was competitive in our 25GB file transfer test; its internal SSD hit a respectable 2,011.6 MBps. The Samsung was a smidge faster, at 2,075.85 MBps.

We also put productivity laptops through 20 loops of the Cinebench R23 CPU benchmark to see how the laptop’s cooling system stands up to heavy demand. A straighter line indicates more stable long-term performance.

The Aero 14 OLED could be considered a poster child for long-term stable performance, as this is about as straight of a line as you can get. A dip between the first and second runs is normal since the laptop’s cooling system is usually heat-soaked by then; we saw the Gigabyte drop from 14,315 points to 13,381 points on the second run. But it wasn’t a big drop, and everything was smooth sailing from there, with the second through 25th runs having a standard deviation of just 25 points. The average score of all runs was 13,397 points.

During the stress test, the Core i7-13700H’s P-cores averaged 2.91 GHz while the E-cores averaged 2.31 GHz. The average core temperature was 74 degrees Celsius (165.2 degrees Fahrenheit).

Display on the Gigabyte Aero 14 OLED

The Gigabyte Aero 14 OLED uses a 14-inch OLED display with a 2880 x 1800 resolution. Its 90 Hz refresh rate is higher than the usual 60 Hz and makes fluid motion appear smoother.

This display resolution shows more detail than FHD (1920 x 1080) and minimizes vertical scrolling, thanks to its 16:10 aspect ratio, which has greatly increased in popularity this year.. Gigabyte doesn’t offer UHD (3840 x 2400) in this laptop, but the utility of an even higher resolution on a 14-inch screen is debatable.

OLED displays usually have outstanding picture quality and that’s certainly the case here. The Aero 14 OLED’s exceptional brightness, contrast, and color give everything such a vivid look that I found myself continuously flipping through my SLR photo collection. Video looks just as amazing; OLED makes anything bright pop off the screen, which was a real treat during the lightsaber and space battles in Star Wars: The Rise of Skywalker. The deep black levels (OLED displays actually shut off individual pixels that show black) transform darker scenes, too; Blade Runner 2049 and Arrival were two movies that I admittedly watched all the way through, despite only intending to watch for a few minutes.

The Aero 14 OLED’s screen was simply unmatched in color range, its 160.3% of DCI-P3 almost double what any of the other laptops produced. (The Samsung Galaxy Book 3 Ultra was the next most colorful, at 86.2%.) OLED displays tend not to be ultra-bright, but the Aero 14 OLED did well there, too, its 397.2 nits a tad short of the MacBook Pro’s 450 nits.

Keyboard and Touchpad on the Gigabyte Aero 14 OLED

I like the Gigabyte Aero 14 OLED’s keyboard, but it does have notable downsides. Let’s start with the all-important layout. The keyboard spans the width of the chassis, which is good, as nearly all the keys except the top function row are full-size. I also appreciate the dedicated Home, End, Page Up, and Page Down keys down the right column. Forcing full-size arrow keys into the main keyboard area caused truncation of the right Shift key, though, and if you usually tap right Shift on its far-right edge, you may wind up hitting the up arrow. The arrow keys should really be their own cluster.

This keyboard’s tactile feel also has inconsistencies. Lengthy key travel means your fingers know precisely when keys are pressed, though the feedback is on the rubbery side. These keys also seem to have higher actuation pressure than normal, which seemed to have slowed down my typing. I managed 94 words per minute with 97.1% accuracy in the 10fastfingers typing test, which is below the 100 to 110 wpm I do on a desktop keyboard.

The keyboard has three levels of backlighting, toggled via Fn + spacebar. I had no visibility issues with the key symbols despite both the keyboard keys and the backlighting being white.

The Aero 14 OLED’s touchpad is rightly sized for its 14-inch screen. The glass surface is inflexible; tactile clicks (accomplished by pressing the pad) are satisfying though they make a highly audible click when doing so. I found myself tapping to click in quiet public places.

Audio on the Gigabyte Aero 14 OLED

Speaker quality isn’t one of the Aero 14 OLED’s assets. Its stereo speakers, which reside under the palm rest, sounded muffled and generally struggled with everything I tried. They’re far from unusable, but this laptop wouldn’t even be a distant second choice as my preferred listening device. I missed the spacious soundstage in Pat Benatar’s "Love is a Battlefield," and vocal clarity was lacking. Movies seemed better – the virtual surround-sound effects in Star Wars: The Rise of Skywalker were surprisingly convincing – but I always found myself wishing for a fuller, crisper sound experience.

The settings in the included DTS:X Ultra app do help; it’s set to Music by default, not Automatic, which I found odd. Switching to Movie mode helps expand the soundstage in movies but it sounds rather awful for anything else.

This laptop includes a DTS Headphone:X license for spatial audio support, which made my inexpensive Koss Porta Pro headphones sound fuller than usual.

Upgradeability of the Gigabyte Aero 14 OLED

Serviceable components in this Gigabyte include its battery, wireless card, and SSD. Getting to them involves removing seven Torx T6 screws that secure the bottom panel.

With the screws removed, mild prying is required to pop the panel free. I used a credit card and started near the rear air vents. The entire panel came up once I had that edge loose.

The battery spans the upper half of the chassis. Just below it are the M.2 2280 slot for SSDs and the M.2 2230 wireless card slot. It’s good to see the latter, as the wireless card is often soldered in smaller laptops like this. That said, you can’t upgrade the LPDDR5 memory since it’s soldered to the motherboard.

Battery Life on the Gigabyte Aero 14 OLED

It’s hard to call a laptop portable if it doesn’t have good battery life. The Aero 14 OLED managed just five hours and 37 minutes in our battery test, which consists of simulated web browsing, streaming video, and OpenGL simulations over Wi-Fi with 150 nits of screen brightness. The other laptops all cleared the nine-hour mark, even the Samsung which is a much larger laptop. Even Gigabyte only claims 7.5 hours of video playback, which isn’t as demanding as our test.

The Aero 14 OLED’s only potential upside here is that its battery is claimed to fast-charge to 50% in 30 minutes.

Heat on the Gigabyte Aero 14 OLED

For productivity laptops, we measure surface temperatures during our 20-loop Cinebench R23 CPU stress test. The Aero 14 OLED’s keyboard center (between the G and H keys) measured 106.7 degrees Fahrenheit (41.5 degrees Celsius), which happened to be the peak temperature on the top of the laptop. The touchpad was just 86.7 F (30.4 C). Meanwhile, the peak underside temperature was 103.8 F (39.9 C). The Aero 14 OLED never felt too warm to the touch, and the keyboard keys were always sub-100 F (37.8 C). The laptop’s twin cooling fans were audible but didn’t get loud enough for me or others within earshot to take much notice.

Webcam on the Gigabyte Aero 14 OLED

The Aero 14 OLED’s webcam is properly centered over the display and produces reasonably sharp images, thanks to its FHD (1920 x 1080) sensor. Grain is minimal and it handles poor lighting situations well, like my coffee-shop selfie below.

The webcam is flanked by dual microphones and an IR sensor for Windows Hello biometric facial login support. What it’s missing is a sliding privacy shutter, and there doesn’t appear to be a way to disable it in an app, either. Windows Hello wouldn't work with the shutter on, but many productivity laptops now offer the choice to cover the lens (though Apple doesn't have a shutter, either).

Software and Warranty on the Gigabyte Aero 14 OLED

The Aero 14 OLED’s major included app is Gigabyte Control Center. It offers what you’d expect from a system management app, including health checks and software updates.

Most of the settings in the General section can be toggled outside the app. The unique options include the Charge Mode, which is somewhat cryptic, but the slider you get when it’s enabled becomes the maximum battery charge percentage. Setting it to 80%, for instance, will cause the battery not to change past that point. Maintaining a lower battery-life charge, especially if the laptop will be plugged in most of time, can extend its lifespan.

Most of the other settings here launch external apps. I wasn’t entirely clear on the scope of AI Boost, but Gigabyte’s site says it auto-detects the performance your apps need and prioritizes power to either the CPU or the GPU. It’s not enabled by default. On the other hand, the modes shown on the left seem to be a collection of other settings. Creator mode boosts screen brightness to max while Power Saving Silence lowers it and restricts fan noise at the expense of performance.

Fan noise can be controlled independently in another section of the app. Normal is the default, but at one point I forgot I switched it to Eco and was wondering why the performance tests I was running were scoring much lower than expected. Setting the fan noise back to Normal fixed that.

Other sections of the app let you view SSD health, check for system updates, and create a recovery disk.

The other important included app is the DTS:X Ultra appmentioned in the audio section.

Gigabyte sells the  Aero 14 OLED has a one-year warranty.

Gigabyte Aero 14 OLED Configurations

The $1,600 Gigabyte Aero 14 OLED reviewed here is model BMF-72USBB4SH, which has a 14-inch 2880 x 1800 OLED screen, an Intel Core i7-13700H processor, an Nvidia RTX 4050 6GB graphics card, 16GB of RAM, and a 1TB SSD. Earlier in our review period, it was selling for $1,649.

The only other configuration is model 9MF-E2USBB4SH, which is $1,499 and differs only by downgrading to a Core i5-12500H processor.

The memory is soldered and not upgradeable, so it appears you’re stuck with 16GB in all configurations. This is disappointing since many creative apps can take advantage of 32GB or more, especially when editing massive files. Having at least one SODIMM slot would have been a real bonus.

Gigabyte also offers a 16-inch version of this laptop, the Aero 16 OLED, which offers up to a Core i9-13900H and a GeForce RTX 4070 8GB graphics card.

Bottom Line

What the Gigabyte Aero 14 OLED does well, it does very well. Its aluminum chassis is downright handsome and feels very well made, and it’s respectably thin and light for an ultrabook, especially one packing an Intel Core H-class processor and dedicated graphics. We were also wowed by its OLED screen.

Other aspects, however, could go back to the drawing board. The Aero 14 OLED’s battery life is especially disappointing for a laptop this size. Performance was good in our tests, but Gigabyte hamstrings this laptop’s potential with an un-upgradeable 16GB of RAM, which may prove insufficient for demanding creative apps.

Alternatives at this price range seem to be priced above or below the Aero 14 OLED. The Lenovo Slim Pro 7 is a few hundred less ($1,199) but it’s a significant step down in performance and also suffers from soldered-down RAM.

If you’re going to spend this much, we’d recommend saving a little more for the Apple MacBook Pro 14 ($1,999), which is superior in most ways unless you need Windows. It doesn’t offer an OLED screen, but that’s about the Gigabyte’s only real advantage. If you do need Windows, dealing with the Gigabyte’s short battery life may be a worthwhile alternative to spending much more for the Samsung Galaxy Book 3 Ultra. All told, the Aero 14 OLED is a handsome laptop with decent performance, but Gigabyte has some work to do on making its Aero line the best. 

MORE: How to Buy a Gaming Laptop

MORE: Best Gaming PCs

MORE: Best Ultrabooks and Premium Laptops

LG has launched the new LG StanbyMe Go (27LX5), which could be summarily described as a 27-inch display in a suitcase (h/t Liliputing). Other attractive features of this relatively expansive portable display include its built-in adjustable stand, carrying suitcase, speakers, and three hours of battery life. LG’s lifestyle images suggest the primary imagined use-case for the 27LX5 is as a device you might take on a picnic. We can imagine a few other uses as well, that don't involve tents.

From the introductory description and the images above and below, you probably already have a good handle (pun intended) on the potential appeal of the new LG StanbyMe Go (27LX5). Here's a look at some of the technical specifications of the product:

With the installed WebOS 22, the LG StanbyMe Go can do a good job as a large ‘tablet’, so you don’t need to connect another device for playback of your favourite videos or music. WebOS also has a range of apps and enables features like voice control, AirPlay, and screen mirroring. An HDMI input is present so you could connect to something like a laptop without poking around in software.

The 1920x1080 resolution seems relatively low for a 27-inch display, but perhaps that was to help with battery life. If you're not taking the StanbyMe Go outside, you could also use it for presentations in a board room... one with WiFi but without a built-in display, presumably.

Of course the special sauce here is the suitcase form factor, which offers rugged (MIL-STD-810H ) portability and easy setup, with a wide range of display position adjustments possible, including portrait. It might be good for shoving in the car for a picnic, as LG’s media department foresees, but seems a bit too heavy for a business traveler. For thinner, smaller, and lighter portable monitors, please take a look at our Best Portable Monitors 2023: Gaming and Laptop Productivity feature.

This isn’t the first member of the StanbyME series of displays from LG. A few years ago, the electronics firm introduced the StanbyME (27ART10AKPL) which was also a 27-inch FHD portable monitor with three hours of battery life, but this one came with a tall lamp-like pedestal stand.

Hacking off the stand and adding a suitcase several years later seems like very little has been accomplished, but perhaps LG will find a market niche for this device among the outdoorsy type who don't want to leave their large screens at home.

Last week Sony took the wraps off its Project Q handheld. A lot of questions about the device were answered during the showcase event, however, a vital handheld specification wasn't provided — battery life. Now Tom Henderson, a games industry tipster with a long track record, has gone public to assert that the Sony Project Q will only last for three to four hours between charges.

Sony's latest handheld could be described as a 'dumb' device, in that it doesn't have onboard processors for tackling PlayStation gaming content. All the onboard processing power needs to do is handle games streaming and user inputs without detrimental lag. It is possible to enjoy PS Remote Play streaming on gadgets like mobiles, tablets, and so on as the PlayStation game processing is done in the cloud, but Sony thought a dedicated device like the Project Q is the best way to do justice to its games catalog.

With its light processing burden in mind, one might expect it would be relatively easy for the Sony Project Q to offer extended battery life, as the Logitech G Cloud can. However, if Henderson is right, users will get a desperately short unplugged experience here. Sony must have decided this was enough, and/or the decision was steered by the overall weight, size, or cost of the handheld.

To put the claimed Sony Project Q handheld battery life into some perspective, let us compare it with a mix of contemporary handheld gaming and smart devices:

• Logitech G Cloud: 'all day' 12+ hours claimed
• Apple iPad 10th Gen (2022): about 10 hours
• Samsung Galaxy Tab S8 Plus (2023): about 9 hours
• Nintendo Switch OLED: between 4.5 and 9 hours
• Valve Steam Deck: usually 2 to 6 hours observed in our review
• Asus ROG Ally: PC gaming 1.43 hours, web activity 5.51 hours in our review testing

In the above examples, you can see devices like the ROG Ally really eat through their batteries when games are using the local CPU and GPU power (actually the AMD Ryzen A1 Extreme APU). However, it looks like it would be able to stream games for nearly six hours. This kind of local processing vs streaming battery life usage differential is a big reason why the touted Sony Project Q handheld battery life is so disappointing. The source takes a swipe at Sony, by adding that "at least Project Q is on brand with the DualSense having a short battery life." Hopefully, Henderson's information comes from a pre-release testing unit with a smaller battery, or there is another pre-release hardware issue at play stunting the battery life.

Another factor that will be important to the success or failure of the Sony Project Q will be pricing, which was also held back from the unveiling event. Henderson expects the product release to be sometime in November, and we should get pricing nearer that time.

Framework's 13-inch laptop has looked effectively the same since launch. That makes sense, right? To keep its library of components working across generations, that means that you can't change too much. But with the Framework Laptop 13 (the first time the laptop is getting a size-based number in its name, but rest assured it's the same model) with 13th Gen Intel core processors, Framework has made a number of small changes that do add up.

Besides 13th Gen P-series processors, there's also a new matte display, which helps protect against glare (and happens to look great); a larger, 61 watt hour battery that makes the laptop last all day; and louder speakers.

While the laptop starts at a barebones $849 DIY config  without any operating system, memory, or storage, we're testing the top-tier, $2,069 "professional" pre-built model — and that's before you add port modules. But that's more than enough to show off how all of Framework's little changes are adding up.

Design of the Framework Laptop 13

If you've been following Framework at all since its initial laptop launch in 2021, it shouldn't surprise you to know that not much has changed design wise. It's still a silver slab with a mirror-like gear logo on top, featuring aluminum construction all around (the CNC aluminum lid option was added in the second generation).

The new Framework Laptop 13 is the same size as the previous model at 11.68 x 9.01 x 0.62 inches and 2.87 pounds. All of this makes sense: to keep the components in the Framework Laptop compatible with prior generations, being roughly the same size helps. It would be nice to see some design changes. Perhaps new colors? Is that too much to ask? You can buy different-colored bezels (orange, white and an upcoming translucent) and different language keyboards (or even blank or clear options) on the Framework Marketplace.

One new addition shows when you open the display: a matte screen. We'll analyze this a bit more down below in its own section, but it's the same 13.5-inch size and 2256 x 1504 resolution in the 3:2 aspect ratio Framework has been using since the beginning.

Framework has room for four "expansion cards" built into the sides of its laptop. These plug into slots that connect to USB Type-C ports directly on the motherboard, with two carve outs on each side of the case. You can try to plug peripherals or chargers straight into those ports, but they're not that easy to get to. The only port that's included on every system is a 3.5 mm headphone jack on the left side. To remove any of the ports, there are buttons on the bottom of the laptop, though it still requires a bit of muscle to get them out. And as I've written for a few years now, I wish they matched the laptop casing in color.

The company recommends at least one USB Type-C port for charging. You can also choose between USB Type-A, HDMI, DisplayPort, an Ethernet jack, extra storage or a microSD card reader.

The Framework is a bit heavier than competitors. The MacBook Air is 2.7 pounds and the Dell XPS 13 Plus is 2.8 pounds. Of our test group, only the 2-in-1 Yoga 9i weighed more, at 3.09 pounds.

Fans of Framework may note that Framework hasn't done anything to really differentiate the design since it released its first laptop. On the one hand, that's in service of a noble goal, ensuring that you can continue to update your own laptop and avoid creating e-waste. But eventually, it will feel dated.

Framework Laptop 13 (Intel) Specifications

Upgradeability of the Framework Laptop 13 (Intel)

Opening the Framework Laptop is as easy as ever, especially considering Framework includes a tool in the box for any repairs you need. The tool is a screwdriver, with a reversible Torx T5 head and a Phillips head 0 if you flip the bit over. The other end of the driver acts as a spudger.

To open the Framework, you should remove any expansion cards. From there, you loosen — but don't need to remove — five Torx screws on the bottom of the case. Then, you flip the laptop over, open the lid, and pull the keyboard cover off (it's guided magnetically, but I found it easier to get in the bottom-right hand corner with a spudger and pull it off.

The one area where you need to take special care is the ribbon cable for the keyboard and touchpad which is attached to the piece you're taking off. For most repairs, you're best off removing the cable from the mainboard so you can set the whole thing aside.

Once you're in there, it's easy to find the battery, SSD, Wi-Fi card and RAM (which is below some heat shielding). But effectively everything, including the speakers and the mainboard itself, are repairable and replaceable. Framework replacement guides on its website for just about any repair you can think of. For more advanced repairs like the webcam or the display, you can pull the bezel off (also held on with magnets using just your fingers.

Just about every part has some sort of QR code on it. You can scan parts with your phone as you repair them to go directly to maintenance manuals. One QR code of interest specifically tells you to keep the battery connected for most repairs due to the "sensitive" connector it uses. While this goes against the advice most laptop vendors offer, I admit I've done plenty of minor repairs without removing the battery connector.

Many of the repairs or upgrades aren't necessary on our unit yet. It has the latest Intel CPU (though maybe we'd want to change it to try an AMD mainboard that’s coming later this year).

One interesting missing item is the CMOS battery, sometimes referred to as the RTC battery. In a blog post earlier this year, Framework wrote that shipping without the battery "reduces environmental impact." If you unplug the main battery, the system clock will reset, but Framework says it will catch up on the time or date when your OS reconnects to the internet. For those using the mainboard as a computer without the battery (perhaps as a maker project), the company says these batteries will be available in the Framework Marketplace.

Productivity Performance on the Framework Laptop 13 (Intel)

The "professional" version of the Framework Laptop 13, with an Intel Core i7-1370P, a whopping 32GB of RAM and a 1TB M.2 PCIe SSD did well on our benchmarks. I have to imagine that extra RAM was a boon. It's DDR4 RAM, which isn't the newest. Intel's Raptor Lake supports DDR5, but Framework hasn't chosen to go that direction (though it will on its upcoming AMD option). This likely can keep the Intel version cheaper, or perhaps Framework just has a stock of DDR4 RAM it wants to get rid of.

On Geekbench 5, the Framework Laptop 13 earned a single-core score of 1,958 (inching past the MacBook Air) and a multi-core score of 11,317, the highest of the group. The Lenovo Yoga 9i, with a Core i7-1360P, fell behind at 1,827/9,954. The Dell XPS 13 Plus (9320) with its Core i7-1280P has an impressive second-place multi-core showing at 10,621.

The SSD in the Framework Laptop 13 was blazing fast, copying 25GB files of 1,697.07 MBps, coming in just ahead of the Yoga 9i and far ahead of the MacBook Air.

On Handbrake, the Framework transcoded a 4K video to 1080p in 5 minutes and 49 seconds. This was the fastest of the group, beating the MacBook Air (7:52) and the other devices.

To stress test the Framework Laptop 13, we put the system through 20 runs of Cinebench R23, simulating an extremely taxing workload. It started at 12,250 before slowly dropping to the high 10,000's. It did have a boost back up around run 15, but then slowed back down.

The CPU's performance cores ran at an average of 2.7 GHz, while the efficiency cores ran at 2.1 GHz. The CPU averaged 72.39 degrees Celsius. The fans ran loud and constant during the test.

Display on the Framework Laptop 13 (Intel)

One of the few changes Framework has made to its laptop outside of the new mainboard is the display.Specifically, there's a new matte screen (being sold as a separate part for previous Framework Laptops) with the same 13.5-inch size and 2256 x 1504 resolution.

I really like the matte. I'm not sure the average person would care, but even under the bright fluorescent lights of our New York City office, the display was able to reduce glare. On top of that, it's bright and colorful. The trailer for Fast X looked great, with one scene at a street race (remember when this series did those all the time?) showing bright clothing and cars popping against the night sky.

Frameworks' panel covers 72.5% of the DCI-P3 color gamut and 102% of the sRGB color space, coming in just a few points behind the MacBook Air. The Yoga 9i and XPS 13 Plus, with OLED screens, were far more vivid.

But at 465 nits of brightness, the Framework Laptop 13 was the second brightest of the group. Again, it fell just a bit behind the MacBook Air (489 nits).

Keyboard and Touchpad on the Framework Laptop 13 (Intel)

Framework promises a deep 1.5mm of key travel on the backlit keyboard, but I wish it were a bit snappier. It's slightly mushy, even as I know I'm hitting the bottom of the keys. That being said, Framework has a nice layout here, including putting volume keys and media keys next to each other (more laptops should do this!), though it does waste one function key on a dedicated button for Framework's website. I also wish the right and left arrow keys were half size like the up and down keys.

The lack of tactility didn't hurt much; I still reached 117 words per minute on the Monkeytype typing test with 98% accuracy.

The touchpad is tall and wide enough for me to navigate Windows without any issues. When I clicked, however, it felt a bit plasticky and even flexed a bit in the topmost areas. This has been an issue since the initial Framework Laptop, and I'd like to see it reworked.

Audio on the Framework Laptop 13 (Intel)

We were excited to check out the Framework's speakers, because the company added in the louder 80 decibel options it debuted in last year's Framework Chromebook. They're louder than previous versions, but still not amazing.

Green Day's "Jesus of Suburbia" easily filled my apartment with sound, but it was muddled. The vocals were clear, but the bass and drums were lost among heavy guitar riffs. This is fine for video calls, but if you want to hear your music well, you'll want to plug in some headphones.

Battery Life on the Framework Laptop 13 (Intel)

The Framework Laptop's new 61 WHr battery allows you to go most of the workday without a charge. On our battery test, it ran for 11 hours and 38 minutes browsing the web, streaming video and running light graphics tests with the display set to 150 nits of brightness. It's a significant update over the 12th Gen version, which ran for just 7:38.

It was beaten only by the Macbook Air (14:06), but surpassed the Yoga 9i (10:38) and the last-gen XPS 13 Plus (7:34). It's a shame Framework isn't including this on the base model laptop, which is keeping the older 55 WHr battery. I think this bump up in capacity might be worth the upgrade alone.

Heat on the Framework Laptop 13 (Intel)

The fans can be loud. In fact, there were plenty of cases when the laptop was sleeping with the lid open and the fans booted up momentarily to keep the system cool, which was surprisingly loud given the circumstances. So it's no surprise that during our Cinebench R23 stress test, when we took skin temperatures, the fans roared constantly and the laptop ran on the hot side.

At the center of the keyboard, the Framework reached 44.7 degrees Celsius (112.46 degrees Fahrenheit), which is toastier than we like to see ultraportables get.

Meanwhile, the hottest spot on the bottom reached 54.6 C (130.28 F). If you're pushing the CPU to its limits, use this laptop on a table or desk.

Webcam on the Framework Laptop 13 (Intel)

There's a 1080p webcam in the Framework Laptop's top bezel, flanked by a privacy switch to shut it off and another to mute the microphone.

The images from the webcam are solid. Even when I was in front of some brightl- lit windows — a challenging condition for any camera — there was still detail in my skin and hair, and the lens still caught the blue in my eyes. 

I wish Framework would consider adding support for Windows Hello with infrared cameras; right now the only biometric option is the fingerprint reader on the chassis. 

Software and Warranty on the Framework Laptop 13 (Intel)

We tested the Framework Laptop 13 with Windows 11 Pro installed - that's the default on the prebuilt "Professional" tier we tested. The base model and "performance" option both use Windows 11 Home.

On the DIY model, you can install Windows yourself, or choose to pick your preferred flavor of Linux. Framework claims Fedora 38 and Ubuntu 22.04 will work straight out of the box, with Manjaro XFCE 22.0 and Linux Mint 22.1 "also working great[.]"

Framework doesn't include the same bloatware you see a lot of laptop vendors deliver. The Start menu does include pinned links to the Framework Marketplace and Framework Support. That's not as bad as a bunch of software, but I hope it's not a slippery slope. Of course, there's also the links to the Microsoft Store that come with most Windows installs, such as those to download Spotify, Kindle, Whatsapp, TikTok, Instagram, and more.

Framework sells its laptop with a 1-year warranty in the US and Canada, while it offers a 2-year warranty in the European Union.

Configurations of the Framework Laptop 13 (Intel)

We tested the top-tier "Professional" Framework Laptop 13 with an Intel Core i7-1370P, 32GB of DDR4 RAM, 1TB of SSD storage and Windows 11 Pro. It also uses Framework's second-generation hinges, which it says are more stable and a 61 watt hour battery. This is $2,069 before adding expansion ports, which start at $9 for USB Type-C and USB Type-A, $19 for HDMI, DisplayPort or a microSD card slot, $39 for Ethernet, and additional storage cards: $69 for 250GB and $149 for 1TB. These storage cards act as secondary drives and promise up to 1,000 MBps read and write speeds over a USB 3.2 Gen 2 (10 Gbps) connection.

One step down is the "Performance" model, with a Core i7-1360P, 16GB of RAM, 512GB of SSD storage, Windows 11 Home, and the same increased battery. This model is $1,469 before you add ports.

The base model uses a Core i5, 8GB of RAM, 256GB of storage, Windows 11 Home and a smaller 55 WHr battery, which starts at $1,049 before ports.

As always, Framework offers a DIY Edition for those who want to bring their own memory, storage and operating systems (though it will gladly sell you those separately, too). That begins at $849.

Framework sells directly from its website. As of this writing, some 11th Gen Intel options are also available, but the 12th Gen configurations are out of stock. There's also a Chromebook and pre-orders for an upcoming Ryzen 7040 model.

Bottom Line

The Framework Laptop 13 is, to some degree, set in stone. The company is locked in to fitting three generations of hardware so that original owners can upgrade. And yet with this latest entry, Framework proves there's still juice to be squeezed out of this design. 

The new battery is a huge improvement over the one in last year’s model, and it's great to see Framework making a laptop you can actually get a day out of, rivaling the MacBook Air (as long as you don't buy the base model). The matte display, too, is a great quality of life improvement that any owner can buy as an upgrade, but is great for consumers getting their first Framework.

There are still some changes that have haunted Framework. The cooling, limited to a single fan, works overtime on stressful workloads, causing loud whirring and some overheating. And the touchpad still feels a bit cheap when you click it, which should never happen on a $2,000 laptop.

The simple repair is still a wonder. Dell can talk a big game with its concepts and Apple will send massive repair kits to your home, but Framework includes one tool in the box that lets me confidently make any change I want. It may not be the thinnest design around, but it's a trade every laptop manufacturer that makes claims about environmental responsibility should be eyeing.

So if Framework wants to keep this design for a bit longer, I can accept that. Apparently old dogs can keep learning some new tricks.

MORE: How to Buy a Gaming Laptop

MORE: Best Gaming PCs

MORE: Best Ultrabooks and Premium Laptops

Laptop docking and display specialist UPerfect has launched a new portable monitor which utilizes twin 15.6-inch FHD displays in a clamshell form factor (h/t Liliputing). The firm says its UStation Delta solution is great for expanding the desktop working area of laptops and computers, and its built-in adjustable stand and VESA mounting option supports such use cases.

Portable monitors are great for road warriors. While bigger is better, up to a limit, a folding display like the UPerfect UStation Delta is a good compromise between having extra portable screen area and physical portability. The twin 15.6-inch hinged IPS technology displays each offer 1920 x 1080 pixels to add to your setup. The ‘portability price’ is the size 358 × 221 × 22 mm (14.09 × 8.7 × 0.87 inch) and pretty substantial weight of 1.55 kg (3.42 lbs). We guess one of the contributors to this physical presence is the built-in 10,800 mAh battery, which is claimed to provide a run time of 3-4 hours. The battery takes about 3-4 hours to charge, too, according to the manufacturer.

We have some uncertainties about the above specs, which we gathered from the product pages. UPerfect mentions that the monitors are 300 nits and offer 100% sRGB, but Liliputing says these figures are 250 nits and 72% sRGB. Also keep in mind that the UStation Delta isn’t a touch-screen device, and doesn’t have built-in speaker(s) – quite common features of portable monitors.

UPerfect has the UStation Delta up for pre-order at a hefty $629 with worldwide shipping. Orders will be fulfilled in late June, according to the company. Currently there is a $200 discount code shared on the product page, but we would wait for independent third-party reviews before recommending anyone spends their hard earned cash.

If you are on the hunt for a portable monitor, don’t forget to check out our Best Portable Monitors 2023 feature. Perhaps the UStation Delta might win a place in our list, we will have to wait and see.

The Raspberry Pi is a popular first choice for many makers looking to create an emulation platform. Its comfortable size and capable specs make it an optimal choice for amateurs and pros alike. It also helps that the community has vast amounts of support for anyone who wants to game on this classic SBC. Today we’ve got another brilliant Pi-powered gaming handheld to share with you. This one, known as PALPi, was created by a developer named Arnov Sharma.

It wouldn’t be much of a handheld if it wasn’t portable, and this build is no exception. It’s designed with hinges so the screen can be folded up, and there's a built-in battery board to keep it mobile. It’s intended for retro gaming emulation and can run classic console games for systems like the NES. This has been a work in progress for a while and it’s worth noting that this is the sixth iteration of Sharma’s PALPi project.

The latest update for PALPi V6 is the addition of its battery board. This is a custom PCB that sports two lithium cells inside of an SMD battery holder. It can be recharged and fits neatly into the underside of the unit. According to Sharma, the battery lasts up to three hours.

The system is powered by a Raspberry Pi Zero W that connects to another custom PCB created by Sharma. It enables the convenient port connections and buttons necessary for user input. There are also LEDs worked into the system for a professional touch. The screen is a 4-inch PAL display panel which, according to Sharma, was taken from an old car monitor. All of the hardware is housed inside of a 3D-printed shell.

Software-wise, you could use just about anything from Raspberry Pi OS to Lakka or RetroPie. In this case, however, Sharma has opted to use Recalbox OS. This operating system is designed for retro gaming as well, but also comes with a selection of free games pre-installed which is useful if you want to start off gaming without having to load a bunch of ROMs into your system.

If you want to get a closer look at this Raspberry Pi project and maybe even recreate it yourself, check out the full build guide over at Instructables, and be sure to follow Arnov Sharma for more cool creations as well as any future updates on this one.

The Intel Core i3-13100F comes to market with big shoes to fill; its predecessor, the Core i3-12100F, built a reputation as the best budget CPU for gaming. Intel is obviously looking to repeat that success, but perhaps the Core i3-13100F follows in the footsteps of its predecessor too closely — the company repurposed the previous-gen design for the Core i3-13100, so it has the same four p-cores paired with an ever-so-slightly-improved 200 MHz higher boost clock of 4.5 GHz.

The 13100F addresses the sub-$150 budget segment, so pricing is critical. However, while Intel launched the previous-gen graphics-less Core i3-12100F at $104, it later quietly raised pricing. So naturally, that leads to higher prices for the new Core i3 models, too. As such, Intel's recommended pricing has jumped to $144 for the full-fledged Core i3-13100 model and $119 for the graphics-less 13100F model. That's a 12% and 15% gen-on-gen price increase, respectively.

Meanwhile, AMD might have moved forward to its new AM5 platform for its high-end Zen 4 Ryzen 7000 chips, but it also refreshed its Zen 3 Ryzen 5000 lineup to address the lower end of the market with its AM4 motherboards. The new Zen 3 processors initially arrived with slightly higher price tags than expected, but they now represent an exceptional value at their current pricing.

That leaves Intel's quad-core chip battling two six-core processors: The $144 Core i3-13100 faces stiff competition from the $140 Ryzen 5 5600, while the $119 Core i3-13100F squares off with the $99 Ryzen 5 5500. Both AMD chips drop into the inexpensive and plentiful AM4 chipset ecosystem, giving builders plenty of affordable options for budget builds, thus creating a fierce contest for a spot on the list of best CPUs for gaming. Let's see how the Core i3-13100F stands up to the pressure.

Intel Core i3-13100F Pricing and Specifications

Retail pricing for AMD and Intel's chips now diverges widely from suggested pricing, so here we've listed the latest street prices. Intel has used re-badged silicon for its lower-end chips in the past, and continues that practice with its new lineup. So while the new Core i3 models slot into the 13th-Gen Raptor Lake family, Intel re-uses the previous-gen 12th-Gen Alder Lake 'Intel 7' silicon and Golden Cove core architecture. Intel also takes this approach with some of the Core i5 models, like the Core i5-13400.

As such, outside of some microcode tuning, the 13100 is identical to the 12100. That is evident from the L2 cache capacity, which weighs in at 1.25 MB per core for the 13100. In contrast, the truly new Raptor Cove cores come with 2MB per core.

Like its predecessor, the Core i3-13100/F comes with four performance cores and eight threads, but no e-cores for background tasks. Intel sprinkled on 200 MHz of extra boost frequency (+5%), bringing it to 4.5 GHz, and 100 MHz higher base clock (+3%), bringing it to 4.3 GHz. The clock speed adjustments are all that Intel has to justify the step up to 13th-Gen branding and the much higher price tag. 

As before, the 13100 has a 60W / 89W processor base/max turbo power, 16 PCIe 5.0 lanes and four PCIe 4.0 lanes, and support up to DDR4-3200 and DDR5-4800 MT/s (caveats apply to DDR5 support). Intel's non-K models don't allow overclocking the CPU cores, but they do support memory overclocking. Unfortunately, the company's nonsensical decision to keep certain voltages locked restricts DDR4's overclocking headroom, so gains are limited.

The standard Core i3-13100 comes with the UHD Graphics 730 engine and 24 EUs that run at 300/1500 MHz base/boost frequencies. You can save some cash by going without graphics with the F-series model, but that means you will lose Quick Sync capabilities and the iGPU fallback that you can use for troubleshooting. Otherwise, the chips provide the same performance.

The 13100 models drop into either 600- or 700-series motherboards and come with a bundled Laminar RM1 cooler that is sufficient for cooling the chip under normal conditions, albeit at the expense of a higher noise level than you'd get with a better cooler.

The six-core 12-thread Ryzen 5 5600 is a 7nm Vermeer model that debuted at $199, but its new $140 price point makes a lot more sense. This chip is the ‘non-X’ version of the Ryzen 5 5600X, so the two chips are nearly identical besides the 5600's reduced 3.5 / 4.4 GHz base/boost clock. Overall, the 5600 has the same feature set we see with other chiplet-based Zen 3 chips.

AMD's lowest-end Zen 3 chip, the Ryzen 5 5500, debuted at $150 but now retails for a mere $99. For this chip, AMD repurposed its monolithic (single-chip) Cezanne silicon that it typically uses for APUs, but disabled the chip’s integrated Radeon Vega graphics engine. That leaves us with a six-core 12-thread chip that looks and largely acts like a standard Vermeer processor.

The 5500 has the same design as the Ryzen 5 5600G, including support for PCIe 3.0 instead of PCIe 4.0. As a result, this chip will make a great pairing for older, lower-end AM4 motherboards (you definitely don’t want to pay for functionality you don’t need by pairing it with a PCIe 4.0-supporting motherboard). The 5500 wields a 3.6 / 4.2 GHz base/boost clock.

In contrast to the locked Core i3 lineup, the Ryzen chips are fully unlocked for overclocking the CPU cores, memory, and fabric, and like their more full-fledged counterparts, both chips come with a bundled Wraith Stealth cooler that is sufficient for stock operation. They also support up to DDR4-3200 memory. Neither chip has a functional iGPU, so you'll need to plan for a discrete GPU. 

• MORE: AMD Ryzen 7 7800X3D vs Intel Core i9-13900K vs Core i7-13700K
• MORE: Best CPU for gaming
• MORE: CPU Benchmark Hierarchy
• MORE: Intel vs AMD
• MORE: How to Overclock a CPU

Intel Core i3-13100F Power Consumption and Efficiency

In several of these tests, the Ryzen 5 5600 and 5500 processors draw more power than the Core i3-13100F. However, the AMD chips have two more cores than the 13100 and still use the 7nm TSMC process, which isn't as efficient as the newer 5nm that AMD uses for Ryzen 7000 models. The Ryzen processors are faster in multi-threaded workloads, thus justifying the extra power consumption. This is reflected in the first chart, which outlines power efficiency as a renders-per-watt metric. Here the Ryzen processors both offer better efficiency metrics than the Core i3-13100F.

Our highest average power measurement for the Core i3-13100F was 57W, which is relatively easy to tame with lower-end coolers. Aside from a few outliers, the 13100F regularly consumed more power than its predecessor, the Core i3-12100. That's expected given the higher clock rates, but it results in less power efficiency. The Core i3-12100 beats the newer 13100F in the renders-per-day efficiency metrics, indicating that Intel's tuning for a higher clock rate has pushed the silicon further out of the sweet spot on the voltage/frequency curve.

Here we take a slightly different look at power consumption by calculating the cumulative energy required to perform x264 and x265 HandBrake workloads, respectively. We plot this 'task energy' value in Kilojoules on the left side of the chart.

These workloads are comprised of a fixed amount of work, so we can plot the task energy against the time required to finish the job (bottom axis), thus generating a really useful power chart. Bear in mind that faster compute times, and lower task energy requirements, are ideal. That means processors that fall the closest to the bottom left corner of the chart are the best. 

This chart brings the Core i3-13100's relatively slower performance into view — the Ryzen processors are faster while consuming less cumulative power, thus explaining their leading power efficiency. We can also see that the Core i3-13100 consumes more power than the 12100 here, but for a minimal performance improvement. 

Intel Core i3-13100F Test Setup

• Intel Core i3-13100F: Corsair H115i 280mm water cooler, default power limits, DDR4-3200 (Gear 1)
• Intel Core i3-13100F DDR4-3600: Corsair H115i 280mm water cooler, default power limits, DDR4-3600 (Gear 1)
• Ryzen 5 5600: Corsair H115i 280mm water cooler, PBO Disabled, DDR4-3200
• Ryzen 5 5600 PBO: Corsair H115i 280mm water cooler, PBO Enabled, Scalar 10X, DDR4-3800, FCLK 1900 MHz (1:1 Ratio)
• Ryzen 5 5500: Corsair H115i 280mm water cooler, PBO Disabled, DDR4-3200
• Ryzen 5 5500 PBO: Corsair H115i 280mm water cooler, PBO Enabled, Scalar 10X, DDR4-4000, FCLK 2000 MHz (1:1 Ratio)

We tested with Windows 11 and DDR4 memory on a Z790 motherboard to maintain an even playing field with the rest of our test pool. Of course, you wouldn't pair this chip with this class of motherboard, but the 700- and 600-series B- and H-series motherboards also support overclocking memory and removing power limits. Given the 13100's tame power consumption, even lower-end motherboards will provide the chip with enough power to assure full operation.

Notably, we were able to overclock the Core i3-12100's memory to DDR4-3800 in our initial review but found that to be unstable for both the 13100 and 12100 during our retesting, which could be due to newer motherboard firmware. In either case, we dialed back to DDR4-3600. Regardless of the board you use, these locked chips have a relatively low memory overclocking limit due to Intel's decision to keep the SA voltage locked for Alder and Raptor Lake chips, which we'll hopefully see rectified in the next-gen refresh models. Your memory overclocking mileage will vary, but probably not by much.

We test Intel processors with the power limits fully removed for our standard measurements, so those chips run beyond the 'recommended' power settings but remain within warranty. We used DDR4 memory for testing, as DDR5's pricing premium isn't as suitable for this class of chip.

Microsoft has advised gamers to disable several security features to boost gaming performance. As such, we disabled secure boot, virtualization support, and fTPM/PTT on all systems for maximum performance. You can find further hardware details in the table below. 

• MORE: AMD Ryzen 7 7800X3D vs Intel Core i9-13900K vs Core i7-13700K
• MORE: Best CPU for gaming
• MORE: CPU Benchmark Hierarchy
• MORE: Intel vs AMD
• MORE: How to Overclock a CPU

Intel Core i3-13100F Gaming Benchmarks — The TLDR

Here we have the geometric mean of our gaming tests at 1080p and 1440p, with each resolution split into its own chart. We're testing with an Nvidia GeForce RTX 4090 to reduce GPU-imposed bottlenecks as much as possible, and differences between test subjects will shrink with lesser cards, which is especially pertinent in this price range, or higher resolutions and fidelity. You'll find further game-by-game breakdowns below.

The first slides are simplified without the overclocking configs, while the remainder of the slides in the album cover the full roster of tested configurations.

The Core i3-13100 and 13100F offer the same level of gaming performance when paired with a discrete graphics card, but have a $40 price delta. That means we have two different contests in this section.

The $150 Core i3-13100 faces off with the $140 Ryzen 5 5600. The Ryzen 5 5600 is 8% faster than the Core i3-13100 at stock settings, and 6% faster after overclocking. That's an impressive advantage, especially given that the Ryzen 5 5600 regularly retails for $10 less.

The Core i3-12100 remains a viable gaming alternative at $130, especially because it offers basically the same performance as the 13100 at stock and overclocked settings, but for $20 less. However, the Ryzen 5 5600 still holds the performance lead by decent margins, but for a 7% upcharge. Given the Ryzen 5 5600's impressive advantages that we'll see in other areas, like power efficiency and performance in heavily-threaded work, it's the easy winner in the $130 to $150 price class.

Meanwhile, the $110 Core i3-13100F faces the $99 Ryzen 5 5500. The 13100F is 8% faster in 1080p gaming than the 5500 at stock settings, and 5% faster after overclocking. The Ryzen 5 5500 has a PCIe 3.0 connection instead of support for PCIe 4.0 and 5.0 like the 13100F, giving the 13100F the victory over the 5500.

The 13100F's win isn't unchallenged, though: The Core i3-12100F is an even better deal. The Core i3-12100F comes into the conversation at the same pricing as the 5500 and with a 7% advantage in gaming, giving it the win over the Ryzen 5 5500 and the Core i3-13100F. If you can find it in stock, the 12100F is the best value in the $100 price range, with the 13100F being a good alternative if it remains around $110.

Intel's decision to use the same design for the 13100/F and merely increase boost clocks by 200 MHz doesn't do enough to deliver any tangible increase in value — in fact, its higher pricing only reduces the appeal of the Core i3. We see a dead heat between the two Core i3 models in gaming at stock settings and a 2% difference after tuning the memory. That certainly isn't enough to justify the Core i3-13100's higher price tag, especially given that both chips can drop into the same motherboards and have identical features.

We also included the $170 Ryzen 5 5600X in our test pool, but as you can see, it isn't a good value compared to the Ryzen 5 5600. Additionally, the Ryzen 5 5600G, a Cezanne APU, falls into this price category. This chip was a wonderful alternative for budget gaming machines during the GPU shortage, but it's really meant for low-end gaming machines that don't have a discrete graphics card.

The AMD vs Intel contest can vary in different game titles, with some clearly favoring one architecture over the other. As such, you should peruse the individual game benchmark results below for a closer look at the test results. 

Cyberpunk 2077 Benchmarks on Intel Core i3-13100F

Far Cry 6 Benchmarks on Intel Core i3-13100F

F1 2021 Benchmarks on Intel Core i3-13100F

Hitman 3 Benchmarks on Intel Core i3-13100F

Microsoft Flight Simulator 2021 Benchmarks on Intel Core i3-13100F

Red Dead Redemption 2 Benchmarks on Intel Core i3-13100F

Warhammer 3 Benchmarks on Intel Core i3-13100F

Watch Dogs Legion Benchmarks on Intel Core i3-13100F

• MORE: AMD Ryzen 7 7800X3D vs Intel Core i9-13900K vs Core i7-13700K
• MORE: Best CPU for gaming
• MORE: CPU Benchmark Hierarchy
• MORE: Intel vs AMD
• MORE: How to Overclock a CPU

Intel Core i3-13100F Productivity Benchmarks — The TLDR

The first slides are simplified without the overclocking configs, while the remainder of the slides in the album covers the full roster of tested configurations. We boil productivity application performance down into two broad categories: single- and multi-threaded. These slides show the geometric mean of performance in several of our most important tests in each category, but be sure to look at the individual benchmark results below.

Again, due to the $40 price gap between them, the $150 Core i3-13100 and $110 13100F each have a different competitor in our test suite. With six Zen 3 cores and 12 threads, the $140 Ryzen 5 5600 easily leads our threaded benchmark suite with 30% more performance than the Core i3-13100. The six-core, 12-thread $99 Ryzen 5 5600 is also plenty impressive with a 20% lead over the 13100 in threaded workloads. 

The Intel 13100/F takes the lead in single-threaded work, with a 20% lead over the Ryzen 5 5600 and a 25% lead over the Ryzen 5 5500. That will result in snappier responsiveness in lighter fare, but the 13100/F's slow-downs in threaded work will be far more noticeable than its performance advantage in single-threaded work. Overall, the Ryzen 5 5600 and 5500 deliver a better blend of performance at their respective price points. 

The Core i3-13100/F exhibits a more meaningful lead over the 12100/F in the productivity benchmarks than we saw in the gaming benchmarks, delivering 6% more performance in both single- and multi-threaded applications. However, this doesn't really justify a ~15% price hike, again making the 12100 the more reasonable choice if you're specifically looking to buy an Intel chip. 

Notably, none of these chips benefit much from overclocking, with low-single-percentage-point gains. The benchmarks below are pretty predictable, with the Ryzen processors dominating the threaded benchmarks while Intel takes smaller leads in lightly-threaded applications. As such, we'll limit our commentary.

Rendering Benchmarks on Intel Core i3-13100F

Here we have a quad-core attempting to fend off two six-core 12-thread chips in heavily threaded rendering applications. As you would expect, that doesn't go too well, with the Ryzen 5 5600 and 5500 both taking substantial leads in all manner of threaded workloads. The Core i3-13100/F does carve out a few wins in the single-threaded rendering tasks.

Encoding Benchmarks on Intel Core i3-13100F

Most encoders tend to be either heavily threaded or almost exclusively single-threaded — it takes an agile chip to master both disciplines. Handbrake, SVT-HEVC, and SVT-AV1 serve as our threaded encoders, while LAME, FLAC, and WebP are indicative of how the chips handle lightly-threaded engines.

Adobe, Web Browsing, Office, and Productivity on Intel Core i3-13100F

Compilation, Compression, AI Chess Engines, AVX-512 Performance on Intel Core i3-13100F

Aside from the compression/decompression tests, most of these benchmarks aren't as important for this class of processors, but we include them for the sake of completeness. As you can see, AMD easily wins the compression and decompression benchmarks due to both the advantage of having more cores and specific architectural optimizations that benefit this type of work. 

The remainder of the benchmarks run the gamut from the exceedingly branchy code in the LLVM compilation workload to the massively parallel molecular dynamics simulation code in NAMD to encryption and compression/decompression performance. The demanding Y-cruncher benchmark computes Pi with the AVX instruction set and has optimizations for both Intel and AMD’s architectures.  

The Core i3-13100/F pulls off a few wins in single-threaded tests, but the rest of the benchmarks are nearly entirely in AMD's favor. 

• MORE: AMD Ryzen 7 7800X3D vs Intel Core i9-13900K vs Core i7-13700K
• MORE: Best CPU for gaming
• MORE: CPU Benchmark Hierarchy
• MORE: Intel vs AMD
• MORE: How to Overclock a CPU

The Core i3-13100F and Core i3-13100 come to market using the same winning formula that propelled the previous-gen Core i3-12100 to the top of our recommendations for budget systems, but AMD's Ryzen 5 5600 has seen substantial price reductions that change the ~$130 to $150 landscape entirely. Additionally, despite its higher price tag, the 13100's slim improvements in clock speeds mean it only delivers small gains over its predecessor.

Below, we have the geometric mean of our gaming test suite at 1080p and 1440p and a cumulative measure of performance in single- and multi-threaded applications. We conducted our gaming tests with an Nvidia RTX 4090, so performance deltas will shrink with lesser cards and higher resolution and fidelity settings. This is especially important to keep in mind with these lower-end processors as they will be paired with lower-end GPUs. 

The Core i3-13100 and 13100F offer the same level of gaming performance when paired with a discrete graphics card, but a $40 price delta separates the two chips. That means we have two different contests in this section.

If you're looking for a gaming processor for around $100, the $110 Core i3-13100F is 8% faster than the $99 Ryzen 5 5500, giving it the win. The Ryzen 5 5500 is also limited to the PCIe 3.0 interface while the 13100F supports the faster PCIe 5.0/4.0 interfaces. However, while the 13100F takes the lead over the Ryzen 5 5500 in this price bracket, the previous-gen $100 Core i3-12100F remains the chip to beat with basically the same performance as the Core i3-13100F in gaming, but for $10 less. 

If you're looking for a chip with a bit more heft in productivity workloads, or if you can't find a Core i3 F-series model in stock, you'll have to jump up to the $130 to $150 price range. The $140 Ryzen 5 5600 is 8% faster than the Core i3-13100 in 1080p gaming, yet costs $10 less, making it an easy recommendation even before we take its other advantages into account. The $130 Core i3-12100 remains a viable alternative here if you absolutely must have an iGPU (the Ryzen model doesn't have one), but the Ryzen 5 5600 holds the lead in gaming with a discrete GPU and is far more performant in productivity workloads.

Speaking of which, both the Ryzen 5 5600 and 5500 offer substantially more performance in multi-threaded productivity applications (30% and 20%, respectively) than the Core i3-13100 and 12100. Conversely, the Intel chips are faster in single-threaded tasks, but by smaller margins. We think Ryzen's big advantage in threaded workloads will be far more noticeable than Intel's smaller advantage in single-threaded work.  

The Ryzen processors leverage the plentiful and affordable AM4 motherboard ecosystem, and like the Core i3 chips, support DDR4 memory. Meanwhile, the Intel chips are confined to 600- and 700-series chipsets, so AMD has the platform pricing advantage — you can find incredible deals for AM4 motherboards. Be aware that you'll lose support for PCIe 5.0 and DDR5 if you go with the Ryzen 5 5600, but we don't think those interfaces are as important with this class of chip. Both the Intel and AMD chips come with bundled coolers that are sufficient for normal operation. 

The $150 Core i3-13100 is simply overpriced for a quad-core in 2023. Intel's decision to use the same design for the 13100/F and merely increase boost clocks by 200 MHz doesn't do enough to deliver any tangible increase in value. The minor improvements we saw in our testing certainly aren't enough to justify the Core i3-13100's higher price tag over the prior-gen model. 

The less-expensive $110 Core i3-13100F makes more sense, but it faces a stiff challenge from its predecessor. If you're looking for a chip in the $100 price range and you can find it in stock, we still recommend the Core i3-12100F for gaming-focused budget builds. If the Core i3-13100F is all that is available, it is also better in the ~$100 price range than the Ryzen 5 5500. If you're looking to step up a pricing tier and want even more performance in gaming and productivity applications, the Ryzen 5 5600 is the best of the bunch in both gaming and productivity work. 

• MORE: AMD Ryzen 7 7800X3D vs Intel Core i9-13900K vs Core i7-13700K
• MORE: Best CPU for gaming
• MORE: CPU Benchmark Hierarchy
• MORE: Intel vs AMD
• MORE: How to Overclock a CPU

Turtle Beach is gunning for rivals Razer and SteelSeries with its new flagship gaming headset, the Stealth Pro. And, as an aside, companies need to start actually naming new products instead of just tacking “Pro” onto the name of an existing line, but I won’t fault Turtle Beach for doing this.

Announced last month and available starting April 23, the Stealth Pro is a wireless over-ear gaming headset packed with a laundry list of features and specs that match up almost exactly with SteelSeries’ Arctis Nova Pro — which currently sits at the top of our list of best gaming headsets.

The Stealth Pro has all the gaming features you wanted and didn’t know you wanted, including dual-mode wireless connectivity with simultaneous audio playback, active noise cancellation, a swappable battery system, and a detachable boom microphone and built-in beamforming mics. It sports 50mm dynamic drivers that are hand-matched for balance and has a remappable multi-function button and wheel so you can control everything on the hardware. The Stealth Pro comes in both Xbox and PlayStation versions, both of which can be pre-ordered now for $329.99.

Design and Comfort of the Stealth Pro

The Stealth Pro is a wireless over-ear (circumaural) headset with a detachable boom microphone. It’s part of Turtle Beach’s Stealth line, but it’s been redesigned to look less like a gaming headset and more like a pair of headphones you might wear on your commute. To that end, it comes in just one colorway, black, with silver accents, and the company has no current plans to release additional colors. While I understand this choice, it’s a little disappointing — the navy-and-copper version of the Stealth 700 Gen 2 Max might be my favorite headset colorway (ever).

The Stealth Pro is on the heavier side: my review unit weighed 14.74oz (418g) with the boom mic attached (14.11oz/400g without). For comparison, its predecessor the Stealth 700 Gen 2 Max weighs 13.4oz (380g), while the Corsair Virtuoso RGB Wireless XT weighs 13.47oz (382g). Both the SteelSeries Arctis Nova Pro Wireless and the Razer Barracuda Pro Wireless weigh around 12oz (340g). The Stealth Pro isn’t overwhelmingly heavy — it’s much lighter than the Audeze Maxwell, which weighs 17.28oz (490g) — but it is, for the most part, heavier than the headsets it’s competing against.

The Stealth Pro has a flexible metal headband covered in a slightly-rubbery, soft-touch material with “Turtle Beach” debossed across the top. Its earcups, which are connected via metal yokes, swivel both directions (and swivel flat for storage), tilt, and are individually height-adjustable. The earcups’ speaker plates also feature a soft-touch (though not rubbery) finish, which attracts a lot of fingerprints and, smudges and other unattractive things. I know these soft-touch finishes are kind of Turtle Beach’s thing, but covering up metal with plastic makes the headset look a lot cheaper than it is.

As far as comfort goes, the Stealth Pro is... okay. It’s not uncomfortable, but it’s not so comfortable that I forgot I was wearing it. Its clamping force is on the stronger side and the headband is a little too wide (and I don’t have a small head). Again, while it wasn’t uncomfortable, the fit felt a little off. Perhaps it would benefit from a suspension-style headband like what’s on the SteelSeries Arctis line or the Audeze Maxwell.

The headband and earpads are padded with memory foam covered in nice-feeling black leatherette. The leatherette is comfortable and does a reasonably good job of passive noise cancellation even without the headset’s active noise cancellation turned on, but it does hold in heat. The earpads are glasses-friendly — the foam is less dense in the area where your glasses sit, so you can wear this headset over glasses without getting a headache. I tried this headset on with my sunglasses (I no longer wear glasses), and it definitely works — the earpads put almost no pressure on the sides of my sunglasses.

Most of the headset’s controls are located on the right earcup. The face of the earcup features a remappable multifunction button and wheel. By default, the button toggles active noise cancellation and the wheel controls volume. The wheel has a secondary function, which is activated by holding down the multifunction button for three seconds. The wheel’s default secondary function is game / chat mix. The multifunction button and the wheel’s secondary function can be remapped using Turtle Beach’s companion software or mobile app.

The bottom edge of the right earcup also contains three buttons — Bluetooth, power, and “Superhuman Hearing.” The power button is just a power button, but the other two buttons are multi-function. The Bluetooth button lets you pair and control a Bluetooth device, while the “Superhuman Hearing” button toggles Turtle Beach’s “Superhuman Hearing” feature — a game audio setting that pumps up key sounds such as footsteps and gunfire. Pressing and holding the Superhuman Hearing button for two seconds toggles mute for the headset’s built-in microphones, so you can mute yourself even if you don’t have the boom mic attached. These buttons are well-spaced and easy to differentiate (the Superhuman Hearing button is textured), but they’re relatively stiff and can be difficult to press.

The right earcup also has a USB-C port for charging and updating the headset’s firmware.

The left earcup houses the boom microphone jack, which is covered by a small plastic cover that I immediately lost, as well as the headset’s hot-swappable battery. The hot-swappable battery works just like the hot-swappable battery on the SteelSeries Arctis Nova Pro: it sits under a magnetic speaker plate, and so long as you remove and replace it within a few seconds, your connections won’t drop. My only issue with the Stealth Pro’s system is that the speaker plate is trickier to remove when the boom mic is attached. It’s not impossible, but it’s not the seamless hot-swap you get from the Arctis Nova Pro.

The Stealth Pro comes with several accessories in the box: the detachable boom mic, the wireless transmitter/battery charging station, two USB-C to USB-A cables (one 6.5-foot/2m cable for charging, one 3-foot/1m cable for connecting the transmitter to your PC), two batteries, and a carrying pouch. The carrying pouch, while not as robust as the rigid travel case that comes with the Razer Barracuda Pro Wireless, is pretty nice — it’s made from thick, stiff velvet fabric, and it has an inner pocket that’s large enough to hold all of the headset’s accessories (including the wireless transmitter).

Spec

Audio Performance of the Stealth Pro

The Stealth Pro sports 50mm “Nanoclear” dynamic drivers with a frequency response range of 10 Hz - 22,000 Hz — the same as SteelSeries’ Arctis Nova Pro Wireless. This is a little wider than you’ll find on most gaming headsets (which typically stick to 20 Hz - 20,000 Hz), but it’s not the widest — Corsair’s Virtuoso RGB Wireless XT has a frequency range of 20 Hz - 40,000 Hz, while Audeze’s new Maxwell has a frequency range of 10 Hz - 50,000 Hz.

The Stealth Pro’s default sound profile (called “Signature Sound” in the EQ presets) is relatively balanced and fairly neutral — though it’s perhaps a bit too mid-range focused for me. Audio is not as detailed or layered as it is on, say, the Audeze LCD-GX or the Audeze Maxwell, but the Stealth Pro still managed to separate out most of the delicate sounds in Billie Eilish’s bad guy. Extremely heavy bass sounded a little muddy, but Kaskade’s POW POW POW didn’t turn into a swampy mess, so that’s something. My biggest issue (coincidentally, considering the Stealth Pro’s mid-focused default profile) was with mid-range heavy songs, such as K-391’s Lonely World — which sounded compressed in the mids and lower-mids on the Stealth Pro (and a little less dynamic overall).

When it came to gaming, however, the Stealth Pro was much more impressive. The Stealth Pro features Turtle Beach’s “Superhuman Hearing,” which is a game-focused setting that boosts certain sounds — such as footsteps, explosions, nearby artillery, etc — in games for a competitive advantage. It works, and it works well. I was especially impressed with how it managed to provide a significant boost to these higher-mid-range sounds without going overboard — I turned on the “footsteps” EQ preset for Audeze’s Maxwell headset last week, and I’m pretty sure my dog thought I was trying to communicate with her over high-range frequencies. But turning on Superhuman Hearing for the Stealth Pro nicely boosted the sound of footsteps and gunfire in Fortnite without aggravating my Pomeranian.

The Stealth Pro also sounded good in games with more ambient environmental noise. The jungle in Uncharted 4 sounded lush and layered, while cities in The Elder Scrolls Online had plenty of depth and busyness — even though the Stealth Pro has a much narrower soundstage than, say, the Audeze Maxwell. This headset might not be my go-to for listening, but it’s a great gaming headset.

The Stealth Pro has active noise cancellation (ANC), which is still a fairly uncommon feature on the best wireless gaming headsets — though we’ve been seeing it more often on premium headsets designed to cross over into the lifestyle sphere. Both the SteelSeries Arctis Nova Pro Wireless and Razer Barracuda Pro Wireless have it, as does the Epos H3Pro Hybrid. The Stealth Pro’s ANC is adjustable, from full transparency/ambient mode to full noise canceling. And it’s very good — even at 50% it was able to completely cancel out the hum of my 47 PC case fans (an exaggeration, but it’s still no small task). At 100% it was too good, and I found myself getting dizzy from the lack of... well, noise. At 0% (full transparency/ambient mode), it actually amplifies the sounds around you.

Microphone of the Stealth Pro

The Stealth Pro comes with a detachable cardioid boom microphone as well as built-in beam-forming microphones in the earcups. The boom mic plugs into the mic jack on the left earcup and sits at the end of a 4.5-inch (114.3mm) flexible arm. The microphone jack is located on the face of the earcup (rather than the edge), and the boom mic can be rotated up and out of the way.

Rotating the mic up mutes it, which is a fantastic feature — I’ve used a lot of headsets and it’s hard to tell if your mic is hardware-muted with buttons or switches (mostly because it’s hard to remember which button/switch direction is mute/unmute), but there’s no question with this mic — if you can’t see it, it’s muted. But if you need an additional visual cue, the wireless transmitter flashes red when your mic is muted. It’s a solid green when your mic is on, and this also works when the boom mic is unplugged and you’re using the headset’s built-in mics. 

All the hardware features of the mic aside, the first thing I heard from my coworkers when I logged on with the Stealth Pro was “You don’t sound as good as you normally do” and “I feel like there’s more... static?” Naturally, I immediately started mic testing when I logged off, and after playing around with some settings in the software I didn’t think it sounded that bad so I logged on the next day only for everyone to say, “You still sound worse than normal.” At this point I thought it might be the connection type — I was connected via Bluetooth because I was playing a full-screen game on my PC and didn’t want to switch windows, so the next day I logged in on my PC. I got the same comments — “You sound a little staticky? Or maybe that’s on my end?” — but ultimately people agreed that I sounded fine and the microphone clearly works well enough, even if it’s not perfect. 

I’m not sure if my coworkers are spoiled by my normally incredible broadcast-quality recording equipment, but the boom microphone on the Stealth Pro sounds fine — it’s not the best gaming microphone you can find, but no headset boom microphone is or ever will be. My voice sounds relatively full and is clear enough, and the microphone is definitely on par with boom mics on similarly priced headsets. 

But you can judge for yourself — here’s a test recording on the Stealth Pro boom mic:

The Stealth Pro’s built-in microphones are not quite as good as the boom microphone, but they’re not bad — they’re pretty sensitive, so they managed to pick up my voice fairly loudly and clearly (though they also picked up a lot more background noise). While I wouldn’t use the Stealth Pro’s built-in mics to do a guest spot on NPR, they seem like they’ll work fine for most casual situations — including communicating with your teammates while gaming. 

Features and Software of the Stealth Pro

The Stealth Pro can be configured using the Turtle Beach Audio Hub V2 PC software or the Turtle Beach Audio Hub mobile app. I initially had trouble getting the PC software to recognize the headset when it was connected to my PC, but this seemed to be a pre-production bug. I was eventually able to connect, but I had better luck in general with connecting to the mobile app — which has the same functionality (minus the PC-specific hotkey mapping).

You can use the Audio Hub V2 or the Audio Hub app to adjust the Stealth Pro’s various EQs, which include a number of “basic” adjustable presets for both audio and mics, as well as a 10-band “advanced” EQ for those who want more control. The Audio Hub also lets you toggle several settings that you can (for the most part) also access on the headset itself, including ANC, Superhuman Hearing, noise gate, and mic sidetone. 

But the main reason you’ll want to use the software is probably going to be so you can remap the multifunction button and the wheel’s secondary function — which you can also do on the mobile app. The button can be programmed to toggle ANC, noise gate, or cycle through game presets, or it can be set as a standard multi-function button. The wheel’s secondary function can be set to adjust mic sidetone, ANC level, Bluetooth volume, Superhuman Hearing volume, bass boost volume, treble boost level,  noise gate volume, or mic sensitivity.

The Stealth Pro’s software is a little rough around the edges — like most gaming peripheral software, unfortunately. But the good news is that any changes you make using either the software or the mobile app immediately save directly to the headset — so you shouldn’t need to use the app too often once you’ve got the Stealth Pro set up.

Wireless Performance of the Stealth Pro

The Stealth Pro offers two forms of connectivity — both of which are wireless: low-latency 2.4GHz wireless and Bluetooth 5.1. It does not have a 3.5mm jack for a wired analog connection, which is (I suppose) not too surprising. I would have liked an analog option, but I’m probably in the minority there.

It connects to 2.4GHz wireless via the wireless transmitter (in lieu of a dongle), which is not particularly convenient for those who don’t want to deal with the extra heft of the wireless transmitter, but also not terribly surprising — after all, the Arctis Nova Pro Wireless connects via its base station. On the back of the wireless transmitter you’ll see a switch for jumping between console/PC mode, a USB-A charging port, and a USB-C port for connecting to your console or PC. It’s decidedly less useful than the Arctis Nova Pro Wireless’ base station, which not only lets you control a variety of hardware settings but also has two USB-C ports so you can connect to multiple systems at once.

The Stealth Pro lets you connect to both 2.4GHz wireless and Bluetooth 5.1 and can play audio from both connections simultaneously. This might seem like a niche feature, but it’s one of my must-haves, because it means I can play a full-screen game on my PC while also taking a meeting on my phone (this is what I do almost every morning during the Tom’s Hardware morning meeting).

The Stealth Pro doesn’t give you an easy way to adjust the mix between the two audio sources, however. You can adjust the volume of each source separately by assigning “Bluetooth volume” as the secondary wheel function (the primary wheel function controls the main — 2.4GHz wireless — source audio), but this is pretty clumsy. There are headsets that handle this better, such as Logitech’s G735, which has a rocker button that you can use to mix the 2.4GHz wireless and Bluetooth sources.

Battery Life of the Stealth Pro

The Stealth Pro comes with two 12-hour batteries that can be hot-swapped for a total of 24 hours of battery life on a single charge. The wireless transmitter doubles as a battery charging dock, so — in theory — you’ll never run out of juice, so long as you’re gaming at home on your PC or console.

This is exactly the same system as SteelSeries’ Arctis Nova Pro Wireless, right down to the base station-slash-charging dock. And while it’s a great system for at-home use, it’s less convenient when you’re away from your PC (or console). Assuming you don’t want to lug a wireless transmitter around with you, the headset now only has 12 hours of battery life when you’re on the go, which is... not much. Most headsets easily get twice that — the Stealth 700 Gen 2 Max gets up to 40 hours on a single charge. Even the Arctis Nova Pro Wireless’ swappable batteries get 18 - 22 hours each. 

I like the hot-swappable battery system in general, because it means you essentially have unlimited battery life while you’re near the wireless transmitter. But when you’re not near the wireless transmitter, you’re probably going to want a little longer than 12 hours.

Bottom Line

I was excited for the Stealth Pro — I’ve been a big fan of Turtle Beach headsets for years, and the Stealth 700 Gen 2 Max is one of the few headsets I have that I know will just work (hey, it’s hard out here for a peripherals editor). But the Stealth Pro is sort of underwhelming. It’s clearly going for SteelSeries’ Arctis Nova Pro Wireless, but it just doesn’t go far enough. 

Don’t get me wrong, the Stealth Pro has its strengths — its audio performance didn’t blow me away when I was listening to music, but games sounded great thanks to Turtle Beach’s Superhuman Hearing audio presets. The microphone, despite being hated on by my coworkers, is solid for a headset mic, and the built-in mics are better than expected. The ANC is good — too good — and the headset also comes with a nice carrying bag that will at least keep everything together when you’re traveling with it. 

But if Turtle Beach is going to go up against SteelSeries, it’s going to have to try a little harder. It’s nice to see a swappable battery in a headset that’s not SteelSeries, but not one that gives half the battery life. And while I’m glad to see Turtle Beach is stepping outside of its comfort zone with a more lifestyle-oriented design, this headset definitely doesn’t look like it costs $330. 

MORE: Best PC Gaming Headsets

MORE: Best Wireless Gaming Headsets

MORE: Best Gaming Keyboards

MORE: Best Gaming Mouse

For $10 Elecfreaks’ WuKong 2040 is an interesting add on for the Raspberry Pi Pico and Pico W. It brings easy connections for four motors, 12 servos, or a mixture of sensors and general electronics components. But there are two things which make this stand out from the crowd. Firstly, there is a large 18650 battery holder on top of the board. This means that a project can be powered for up to an hour, and that the battery can be recharged using the onboard micro USB port. Secondly, underneath the board is a Lego compatible frame that can be used to integrate the board, and your Pico into a Lego based electronics project.

All of these features are great, but do they work and can I build something with them? To learn that I put it on the bench and took it for a test.

WuKong 2040 Specifications

Setting Up The WuKong 2040

There’s no software library to install for the WuKong 2040, and this means we have direct access to the GPIO via any compatible programming language. Inserting the Raspberry Pi Pico into the header pins, my first instinct was to connect the micro USB cable to the micro USB port of the WuKong 2040. This instinct was incorrect. It seems that this particular port is reserved for charging an 186050 cell (not included). 

Speaking of 18650s, you’ll need to pick up a battery with a flat top, not with a tip for the + polarity. A flat top battery will slide into place with minimal effort, whereas others will produce massive force on the spring steel contacts. If you’ve never used an 18650 before, heed this caution. They can store a lot of power and should never be damaged or shorted. Store them in a secure plastic container when not in use.

To write code to the Raspberry Pi Pico, I had to connect to the Pico’s own micro USB port. I’m not a massive fan of this as the number of plug / unplug cycles will ultimately reduce the life of the Pico. But that is just the way it is.

The GPIO is broken out via a series of SVG (Signal (Yellow), Voltage (Red) and Ground (Black)) pins. This means that there is access to 3V and GND for every connection, handy. 

Where it comes into its own is when a servo is attached. The SVG pinout matches a servo perfectly, and with a little code there can be up to 12 servos in operation. There are four motor terminals, marked M1 to M4. These terminals are connected to GPIO pin breakouts for the Raspberry Pi Pico via four chips. The chips should be some form of H bridge controller, where the polarity of the motor can be switched to provide forward and backward motion. But these 75V15 chips have me stumped. I can’t find a datasheet for them. However , they do work as well as a traditional H bridge motor controller (L298, L3110S, DRV8833 etc). I connected up two Lego compatible DC motors and wrote a few lines of MicroPython to spin a robot around my desk.

The onboard Neopixels are bright and easy to use with MicroPython and CircuitPython. Elecfreaks suggested using Adafruit’s CircuitPython and so that is what I used for a quick test. This is also where I hit a snag. Normally, I would save the code to the Pico as code.py, the NeoPixel library would be in the lib folder and everything would just work. But it did not. After a little head scratching, I found that I had to swap the micro USB port from the Pico to the WuKong 2040 in order to see the NeoPixels illuminate. That’s not a great workflow, and one that could cause issues for newcomers.

Lego Compatible Chassis

Underneath the mainboard is a Lego compatible frame. This frame is designed to work with Technic components, and is also compatible with Lego Spike components. I attached two Lego compatible DC motors, purchased from Aliexpress for less than $10 for the pair, then raided my box of Lego Spike parts. 90% of the parts i connected were not official Lego, but be they official or not, they were all solidly connected. The Lego chassis integration is a key selling point. It means we can make projects without the need for the best 3D printer or a laser cutter.

What Projects Can Be Made With Wikong 2040?

First and foremost, this is a board for Pico robotics. It just works with robots. Throw a Raspberry Pi Pico W on top, an 18650 battery in and some time learning Anvil you have a Wi-Fi controlled robot.

The digital and analog GPIO and specialist protocols also tie in nicely with the onboard battery system. It means that I can make a sensor platform with just a few parts, put it in a nice case and start collecting data. Multiple servo connections make this a fun board to build a walking robot or a sensor triggered art installation which gracefully changes form.

Bottom Line

For $10, I can forgive the micro USB issue. Sure it is annoying but the sheer number of features and easements present on the WuKong 2040 mean I can have lots of fun. The onboard 18650 battery holder and charging system are useful, and while I don’t get a full GPIO breakout, what I get is plenty for my needs.

Printing the GPIO pin reference on the board is handy, I just wish that it included the I2C, SPI, UART pin references too. The Lego compatible base is great fun for Lego fans such as myself. If you don’t need it, no worries. Grab a cross head screwdriver and it comes off in a jiffy. In a classroom, makerspace or homebrew project, the WuKong 2040 is a great all rounder for $10.

MORE: Best RP2040 Boards

MORE: Best Raspberry Pi Projects

MORE: Raspberry Pi: How to Get Started

What makes a lightweight mouse lightweight? (Its weight, presumably.) Or maybe you can just toss “Lightweight” in the name? 

MSI’s new Clutch GM51 Lightweight Wireless gaming mouse weighs 3 ounces (85g), which is kind of heavy by lightweight mouse standards. It features an ergonomic design, grippy sides with RGB-infused “diamond lightgrips,” and it sports an excellent PAW-3395 sensor that puts the mouse in league with the best gaming mice on the market today. 

The mouse also boasts an impressive 140-hour battery life (with the lighting turned off, of course), and its $100 price includes a wireless charging dock (which doubles as a USB extender). This is MSI’s upgrade to its Clutch GM41 Lightweight Wireless, but it’s, well... heavier. And maybe that’s the problem.

Design and Comfort of the Clutch GM51 Lightweight Wireless

The Clutch GM51 Lightweight Wireless is — you guessed it — a lightweight(ish) wireless gaming mouse with five programmable buttons (including scroll wheel click). The GM51 features a plastic body with a smooth, matte black finish and an RGB-illuminated MSI dragon logo on the palm rest.

The surface is allegedly anti-smudge, but I started seeing oil from my fingertips after just an hour or two of use.

Its sides are what MSI calls “diamond lightgrips,” which are lightly-textured, soft, grippy-feeling thermoplastic elastomer. Actually, the GM51’s sides are the same as those of the Clutch GM41 Lightweight Wireless, just with holes that let the side plate’s (transparent — for lighting effect) polycarbonate body show through.

While its predecessor had an ambidextrous design, the GM51 is ergonomically contoured for right-handed users. The mouse is shaped similarly to Razer’s DeathAdder line — specifically, the DeathAdder V2 — with a large, comfortable body that flares at the top and has a relatively pronounced hump. The GM51 measures 4.8 inches (122mm) long by 2.44 inches (62mm) wide at its widest point, and is 1.77 inches (45mm) tall.

The GM51 weighs 3 ounces (85g), which is heavier than the standard 80g threshold for lightweight mice. It’s still relatively lightweight compared to MMO mice like the Razer Naga V2 Pro (4.72oz/134g) or the Logitech G502 X Plus (3.74oz/106g), but it’s not really light enough to be considered a truly lightweight mouse. It still has enough heft — however slight — to it that picking it up with a fingertip grip gets tiring pretty quickly. The GM51 is over 10g heavier than the GM41. And it just can’t compete with truly light mice such as the Asus ROG Harpe Ace Aim Lab Edition (1.9oz/54g).

The GM51 has five programmable buttons — left click, right click, scroll wheel click, and two thumb buttons on the left side. The button setup isn’t particularly exciting, but I found no issues with the buttons during my testing. The scroll wheel is notched, tactile, and relatively quiet; it won’t please those looking for ultra-smooth hyperscrolling, but it’s well-balanced for a notched wheel — not too tight, and tactile without being clunky.

The bottom of the mouse has six PTFE skates: two at the top, three along the bottom curve, and one surrounding the sensor. The skates are slim but efficient, allowing for smooth, effortless gliding across a variety of surfaces (including hard, soft, and hybrid mouse pads). Also on the bottom of the mouse is a power switch for toggling between 2.4GHz wireless and Bluetooth, as well as a DPI switch button.

The Clutch GM51 Lightweight Wireless comes with a 6.5-foot (2m) braided USB-C to USB-A cable, a wireless charging dock, and a 2.4GHz wireless USB-A dongle. The dongle can be stored in the front of the mouse (the USB-C port is tucked inside a USB-A-shaped channel) for travel, which is a nice touch. MSI calls its cable “FrixionFree,” but while it is braided, it’s not nearly as flexible or weightless as other mouse cables I’ve used (such as Razer’s Speedflex cables) — you’ll definitely notice the extra tension if you use the mouse with the cable plugged in.

It’s nice to see a wireless charging dock included in the Clutch GM51 Lightweight Wireless’ $100 price. The dock is pretty unassuming: small and black, with MSI’s dragon logo printed on the back in dark gray. It comes with the mouse’s 2.4GHz wireless dongle plugged in (it doubles as a wireless extender), and it’s easy to use — the mouse clicks in easily and charges quickly. The dock also features cable routing on the bottom, as well as a ring of micro-suction tape for sticking it to your desk.

Specs

Performance of the Clutch GM51 Lightweight Wireless

The Clutch GM51 Lightweight Wireless sports a PixArt PAW-3395 sensor, which has a maximum sensitivity of 26,000 DPI, a maximum speed of 650 IPS, and can handle up to 50 G’s of acceleration. It has the same Omron mechanical switches as its predecessor, which are rated for up to 60 million clicks. Clicking the left and right mouse buttons felt consistently satisfying — light enough for speed, but not so light as to trigger misclicks.

The mouse’s thumb buttons are well-placed (but so are most thumb buttons), but interestingly designed: they’re not flat buttons. Instead, they form a pointy ridge that juts out from the side of the mouse. I found this a little strange at first, but once I started using the mouse I kind of liked that I could press the buttons by nudging upward from the bottom. 

I tested the Clutch GM51 Lightweight Wireless with a variety of games, though I focused on competitive FPS games such as Overwatch 2 and CS: GO. The mouse is good, but it’s not amazing, especially for shooting games — it glides well but it's not as light as the competition, which makes it a little more fatiguing for claw and fingertip grippers especially. (The ergonomic shape perhaps lends itself better to a fuller grip, anyway.)

The Clutch GM51 Lightweight Wireless is responsive and has reliable clicks, but it didn't wow me with its performance or impress me with its feel. The side grips were grippy enough that the mouse didn’t slip from my fingers when I picked it up, but they felt like they might get gummy with wear, and the mouse’s otherwise smooth surface started showing fingerprint smudges in the first hour or two of use. It’s still a solid gaming mouse, especially for less competitive gaming, but if you're looking for a basic lightweight mouse you're better off with something like the Corsair Katar Elite Wireless (2.24oz/69g).

Features and Software of the Clutch GM51 Lightweight Wireless

The Clutch GM51 Lightweight Wireless has three onboard profiles, which can be configured using MSI Center. You can use MSI Center to configure the mouse’s programmable buttons and adjust its polling rate, lift-off distance, and DPI steps. You can also customize the mouse’s RGB lighting with a handful of preset lighting effects.

The Clutch GM51 Lightweight Wireless has five programmable buttons (left/right click, scroll wheel click, and two thumb buttons). MSI Center isn’t the most intuitive or polished software (what peripheral software is?), and programming the mouse is...trickier than it should be. It’s also somewhat limited — you can program the mouse buttons to perform mouse button actions, multimedia actions (play/pause/volume up/volume down/etc), DPI switching, or one of 30 macros (you can record macros in the software). Other programs, such as Razer’s Synapse 3 and Logitech’s G Hub perhaps go a little overboard in allowing you to program mouse buttons to do everything from launching applications to executing system commands, but I’d like to at least see an option for programming keyboard functions.

Customizing the Clutch GM51 Lightweight Wireless’ lighting is, for some reason, a separate feature called “Mystic Light.” MSI provides a handful of lighting presets such as “color cycle” and “wave,” which have limited customization options. You can also manually customize the mouse’s three RGB zones with individual effects or colors, but there’s no way to create your own lighting effects/animations, or even blend multiple colors on the side grips, which is sort of disappointing.

Wireless Experience and Battery Life of the Clutch GM51 Lightweight Wireless

The Clutch GM51 Lightweight Wireless offers three forms of connectivity: 2.4GHz wireless, Bluetooth, and wired via USB-C. Gamers will mostly use the 2.4GHz wireless or wired connections; Bluetooth is too high-latency for competition, but it’s a convenient option for non-gaming activities. That said, some lighter lightweight mice, such as the Razer DeathAdder V3 Pro, forgo Bluetooth altogether — and I can’t help but wonder if this mouse might weigh a little less if MSI had gone that route.

The Clutch GM51 Lightweight Wireless’ 2.4GHz wireless connection was low latency and reliable in my testing, with no signal dropping or noticeable lag. MSI’s “Swiftspeed” 2.4GHz wireless technology has a latency of between 1 - 2 ms, which MSI says is almost 10 times faster than a normal 2.4GHz wireless transceiver. While this is probably technically true if we’re talking about all wireless mice (including non-gaming mice), <1 ms latency is on par with what you’ll find with other wireless gaming mice. 

MSI rates the mouse’s battery life at 150 hours with the RGB lighting turned off, which is excellent for a wireless gaming mouse, especially a lightweight one. The mouse also features quick charging: 15 minutes of charging via cable will get you 27 hours of play time, while 15 minutes of charging via the dock will get you 7 hours.

Bottom Line

The Clutch GM51 Lightweight Wireless is a decent gaming mouse, it’s just not really  lightweight enough to compete with the slew of lightweight gaming mice on the market. If you’re looking for a lightweight mouse, the Razer DeathAdder V3 Pro remains my ride or die — but for a budget-friendly pick, the Corsair Katar Elite Wireless is a good option.

Don’t get me wrong, the Clutch GM51 Lightweight Wireless has its pros, including excellent battery life, attractive, flashy RGB side grips, and an included wireless charging dock. Plus, it’s moderately priced at $100. But there are trade-offs — I liked the mouse’s light-up side grips in the promo pics, but RGB customization is limited (and the battery life rating is with the lighting turned off completely). The Clutch GM51 Lightweight Wireless is okay, but it’s not Logitech G502X Plus-beautiful.

An electronics hobbyist has designed a “laptop form factor device” that is estimated to run for two years between charges. The self-effacingly named PotatoP combines a mix of low power components, plus a 12,000 mAh battery pack and a modest bank of solar cells. Eventually, Andreas Eriksen, the PotatoP designer, wants to develop this laptop project to offer endless battery life, plus a bigger display.

The inspiration for the PotatoP came from the creator’s annoyance with his existing laptop always running out of battery. For the “small programming projects” it was used for, powerful hardware was unnecessary. It would better suit the user if the laptop design was skewed toward very low power use, as long as the screen was sharp and readable and the keyboard workmanlike.

To execute the above vision, the PotatoP design relies on the SparkFun RedBoard Artemis ATP developer board as its 'motherboard.' This foundational component choice features the Ambiq Apollo3 SoC, which is claimed to “set a new standard in energy efficiency for battery powered devices.” The SoC’s Arm Cortex-M4F runs at up to 96 MHz while consuming less than six micro-Amps per MHz. The SBC also has 384KB RAM and 1MB flash memory, as well as an extensive variety of interfaces / connectivity options considering its size.

Another component that was key to the PotatoP as you see it today is the Sharp Memory in Pixel display LS044Q7DH01. This is a monochrome 320 x 240 pixels display with a 4.4-inch diagonal. Eriksen says that he wishes that Sharp made a bigger version of this display, and is probably open to an alternative as long as it's as thrifty with power consumption. The display has no backlight, which will help with power efficiency. However, this 4.4-inch display is claimed to be “surprisingly readable,” in well lit conditions. Lastly, we must say the diminutive display gives this PotatoP device a unique style.

While there are obvious sacrifices made in the display and processing power, it's good to see that the PotatoP creator couldn’t stomach a substandard typing experience. Thus, he sacrificed an old Happy Hacking Keyboard Lite 2 to integrate it into the PotatoP.

For software, the PotatoP is running a Lisp environment (uLisp). PotatoP creator Eriksen says that using Lisp to create necessary application software is another part of the fun of using this homemade computer.

Lastly, the PotatoP, with its already impressive battery life already measured in years, is being tuned so that it may never require a charging cable. The creator has been busy working on system optimizations to reduce power consumption, which you can read about in the development project logs. If software tweaks alone can’t extend the battery life to infinity as things stand (with the help of solar), Eriksen is pondering over adding more solar cells to achieve this goal.

Batteries degrade over time. And unfortunately, the Lithium-ion (Li-ion) batteries that are used in most laptops these days are prone to swelling or bulging after a while. There are several reasons why your laptop battery would swell up. It could be from age, overheating, overcharging or a manufacturing defect. Any of these issues could cause battery expansion (a process also known as off-gassing) to occur. But no matter what the source is, as soon as you notice that the battery in your laptop has bulged, it poses a safety hazard that you need to take care of right away. You can remove the battery yourself as long as you are careful, but if you aren’t comfortable with the idea of opening up your laptop, you can go to a professional for help.

Though the process of removing the battery from your laptop is quite simple, know that there are risks involved in the process that could result in a fire or explosion or even toxic gas inhalation (in case of chemical leaks). The key is not to disturb or puncture the battery itself so you can avoid any of these disasters from happening. 

If you decide to go the DIY route, then it is best to be prepared and apply all safety precautions even though you may end up not needing them.

What You'll Need

• Philips Screwdriver (or a torx screwdriver if your laptop’s screws require one)
• Gloves
• Safety Goggles
• Fire extinguisher (optional)
• Fireproof container (to transport battery to recycling location)

How to Remove a Swollen Laptop Battery Safely

Luckily, a swollen or bulging battery is usually easy to spot. If you notice your laptop’s case coming apart at the seams or if you feel your trackpad or some of the bottom keyboard keys lifting when you type, then there is a good chance that you have a battery problem.

Below are steps you can follow to safely remove your swollen battery. You should also consult the device’s manual for instructions specific to your machine. We have a Dell Latitude 5490 here as an example.

1. Power down your system and unplug. If your laptop feels hot, wait for it to cool down before you begin the process.

2. Set up in a well-ventilated area, preferably outside. Be sure to have gloves and safety goggles handy just in case you need them. And in an abundance of caution, you could also keep a fire extinguisher at the ready in case something catches fire.

3. Lay the laptop bottom side up and remove all the screws from the bottom panel.

4. Lift and detach the bottom panel. You may hear a slight pop indicating that the panel has been released. It should come off easily.

5. Remove any tape and screws that are holding the battery in place, being careful not to puncture the battery.

 6. Carefully take out the battery and flip it over so you can unplug the cable connecting it to the computer.

If the cable is stuck, you can gently wiggle it to loosen the connection.

7. Place the battery in a fire-proof container. Keep it in a dry and cool place until you can safely dispose of it. But you should not wait too long to recycle your battery.

Note: The same general steps would apply if you were attempting to remove an old or used battery from your laptop. To replace the laptop battery yourself, just reverse the steps above.

Recycle Swollen Battery Properly: Where to Dispose Your Batteries

Batteries contain toxic materials that can leak and are harmful to the environment. You need to dispose of used or damaged batteries responsibly. Find the nearest e-waste recycling center near you.

Both Best Buy and Staples provide free In-store Recycling Services where you can drop off your used batteries. You can also register as a Staples Rewards member (if you aren’t already) and get a $10 store discount for bringing in and recycling your battery with them.

Get Technical Support

Alternatively, you can go to any computer repair shop near you and have them take care of your issue. They will happily remove, replace and dispose of your used or broken battery for you — for a fee. The cost of replacement will vary, so call ahead to get a quote.

You can also call your laptop’s manufacturer for assistance. If your machine is still under warranty, they will send you a replacement battery.

Tips on Taking Care of Your Laptop Battery

You can prolong and extend your laptop’s battery life by doing the following:

• Keep track of your battery’s health (Check out our guide on How to Get a Battery Health Report in Windows 10 or 11).
• Turn on Battery Saver Mode in Windows.
• Look out for signs of battery degradation like slow charging or unexpected shutdowns.
• Keep your machine cool. Remember, heat is the enemy.
• Clean laptop fans and make sure vents are clear to help prevent overheating.

Disclaimer: Again, if the thought of opening your laptop yourself makes you feel uneasy, you do not need to change your laptop battery yourself. You can take your system to a local computer repair shop where they can take care of removing and replacing your battery for you.

The 2021 redesign of the MacBook Pro supercharged the lineup, adding Apple Silicon, plenty of ports, a 1080p webcam and even made it thicker to accommodate it all. It was a bold move (and in some cases, a big reversal for Apple). Now, the new 14-inch and 16-inch MacBook Pros maintain everything that was great about that design, but updates them to the M2 Pro and M2 Max chips, adding more power and endurance than before.

These are still some of the best ultrabooks, with gorgeous Mini-LED displays, excellent speakers and functionality that feels worthy of professionals and hobbyist creators alike.

Upgrades to the highest-performance parts, including RAM and storage, will cost you a serious chunk of cash over the base models, but you should seriously consider them, because Apple's design doesn't allow for internal upgrades down the line. If you're ever going to want 96GB of RAM with your M2 Max, consider emptying your bank account at the point of purchase.

With M2 Pro and M2 Max, the MacBook Pro is more powerful than ever, and is an alluring replacement for lots of aging.

MacBook Pro (2023) Design

The new 14 and 16-inch MacBook Pros have the exact same chassis as last year's redesign. They still feel new and fresh, and they still allow for a solid port selection and plenty of power, so I'm OK with this decision. Both options have boxy sides combined with rounded corners to mix pro functionality with Apple sheen.

We tested the 14-inch MacBook Pro in silver and the 16-incher in space gray (which I prefer), but the colors aren't that dissimilar. At least this year, the notebooks come with color-matched MagSafe power cables. Either size or color has Apple's logo in a tone-on-tone mirror finish on the lid, but is otherwise minimalist.

Whether you have the 14.2-inch or 16.2-inch Liquid Retina XDR display, it's going to have a notch in the top of it, similar to the iPhone 14 and 14 Plus (the 14 Pro line switched to a pill-shaped cutout last year). This allows Apple to fit a 1080p camera in there. Early last year, I bought a 14-inch MacBook Pro with M1 Pro, and, unlike with my iPhone, I have never stopped noticing the notch. That's the case for me with this year's Pros, as well as the M2 MacBook Air. The screens are actually taller than the usual 16:10, so you don't lose any usable space versus previous models. But if you're annoyed like I am, you might want to look into software fixes, like TopNotch, which make the menu bar black to help make the notch blend in. And yes, the mouse can still go underneath the notch, which seems like an oversight.

All of the other improvements from last year's redesign, however, still feel great. There are full-sized function keys, Touch ID on the power button and plenty of ports.

Both the 14-inch and 16-inch MacBook Pros have the exact same ports. On the left side, there's a MagSafe 3 charging port, a pair of Thunderbolt 4 ports and a 3.5 mm headphone jack. The right side has an SDXC memory slot, a third Thunderbolt 4 port and the HDMI port. That HDMI port has been improved from the prior models, allowing 4K support at 240 hz or 8K at 60 Hz, as well as variable refresh rate. We didn't have a 4K 240 Hz monitor to test with (those are pretty rare), but we're curious to see how it holds up.

The 14-inch MacBook Pro measures 12.31 x 8.71 x 0.61 inches and weighs 3.5 pounds, while the 16-inch MacBook Pro is 14.01 x 9.77 x 0.66 inches and, with the M2 Max, weighs 4.8 pounds. Comparatively, the popular Windows-based Dell XPS 15 (9520) is 13.56 x 9.06 x 0.73 inches and weighs 4.31 pounds.

MacBook Pro (2023) Specifications

Productivity Performance on the MacBook Pro (2023)

Apple's M2 Pro and M2 Max are both powerhouses. And both MacBook Pro configurations we tested: a 14-inch Pro with an M2 Pro, 16GB of RAM and 1TB SSD and a 16-inch Pro with a whopping 96GB of unified memory and 4TB SSD both performed admirably in our testing.

The M2 Pro in our 14-inch review unit has 12 CPU cores - eight performance cores and four efficiency cores. The M2 Max has an identical CPU core setup to the Pro we reviewed. But the M2 Pro has a 19-core GPU, while the Max has a 38-core GPU. Both also include a 16-core Neural Engine and dedicated media engines for encoding and decoding in H.264 and HEVC or accelerating ProRes.

Both the Pro and the Max also have faster access to memory. Apple claims the updated architecture allows for 200 GB/s of bandwidth on M2 Pro and up to 400 GB/s on M2 Max.

Needless to say, both machines were overkill for my workload, which involves lots of web browsing for research, writing and photo editing, often with music and social media apps in the background and some light gaming (I still leave my intense gaming to PC and consoles.)

On Geekbench 5, the M2 Max-based laptop won with a single-core score of 1,993 and a multi-core score of 15,173, with the M2 Pro just a few points behind. It's a nice jump over the M1 Max we tested back in 2021, as well as the Intel Core i7-12700H in the Dell XPS 15 (9520), though we haven't seen what Intel's 13th Gen Core mobile processors can do just yet.

The 14-inch laptop copied 25GB of files at a rate of 1,917.69 MBps, tied with the previous-gen 16-incher. This year's M2 Max-based 16-inch Pro came in at 1,789.85 MBps, which represents being one second behind. It should be noted that there have been reports that entry-level MacBook Pros are reportedly seeing worse SSD performance, Teardowns suggest Apple has used fewer NAND modules on those devices.

The 16-inch Macbook Pro with M2 Max made its strongest showing on Handbrake, transcoding a 4K video to 1080p in 3 minutes and 59 seconds. That's seven seconds ahead of the M2 Pro-based system (4:06) and well ahead of the M1 Max and the XPS 15.

On the Xcode Benchmark, which we're using for the first time here, we found that the similar CPUs meant similar scores. The 14-inch MarBook Pro with M2 Pro took 85 seconds, while the 16-inch Mac with M2 Max completed the task in 83 seconds.

To stress test the systems, we ran each through 20 runs of Cinebench R23 multi-core. Because the laptops have the same CPU cores, it's not surprising that they both shared scores in the mid to  high 14,700's. Both had some peaks and valleys as the chips cooled off, but they largely operated in a similar fashion. Notably, the 14-inch MacBook Pro's fans started earlier and were louder to my ears than the 16-inch MacBook Pro, which was far quieter until the eighteenth run, where I really noticed it.

During the stress test, I used TG Pro to track CPU core temperatures. (Apple doesn't make clock speeds available to the software. For what its worth, Cinebench R23 lists estimates that the M2 Pro has a 3.2-GHz single-core clock, the M2 Max has a 3.4-GHz single-core clock, and both chips have 3.3 GHz multi-core clock speeds. Take that for what you will.) The M2 Pro's CPU efficiency cores ran at an average of 94.44 degrees Celsius, while its performance cores measured 99.02 C. The M2 Max's efficiency cores ran at 92.98 C and its performance cores averaged 97.62 C.

The M2 Max may have run cooler because it's in the 16-inch chassis, with more room for heat and, likely, a larger cooling system.

Gaming and Graphics on the MacBook Pro (2023)

The full-fat M2 Pro has a 19-core GPU, while the M2 Max gets 38 cores. For many creatives, those cores will be put to work in rendering, motion graphics, illustration or other artistic pursuits.

It won't come as a huge surprise that we focus a lot on gaming as a show of graphical prowess here at Tom's Hardware. In Apple's case, newer games are being built on the Metal 3 API, including MetalFX Upscaling to render scenes more quickly with temporal or spatial algorithms. More and more games, especially those on Apple Arcade, are using Metal 3. The biggest example of a major game coming to Apple's App Store with native support for Apple Silicon is Resident Evil Village.

I was eager to test the M2 Max's 38-core GPU and 96GB of unified memory, so when I played Resident Evil Village, I set the resolution to 3456 x 2160 and set the graphics preset to max. The game typically ran between 50 and 60 frames per second (higher in daylight scenes with less action, less in darker scenes with more detail), though it did achieve a flat 60 fps a few times. With MetalFX upscaling set to "quality mode," it often stayed at a locked 60 fps. Using these settings took up 12.54GB of RAM out of the 70GB or so that the game can reserve.

On the M2 Pro's 19-core GPU and 16GB of RAM, I tried some more reasonable settings, using the "prioritize graphics" preset at 1080p. In a dark, haunting scene early in the game, the MacBook with the M2 Pro ran Village around 80 frames per second, but in the daylight scenes, where more details in the game's namesake village and surrounding forests needed to be rendered, it ran between 60 and 70 fps. With MetalFX scaling, the game often jumped to the high 70s or even 80 fps in the same scenes.

In an attempt to try some more conventional settings, I set the resolution to 2560 x 1440 and chose the "prioritize graphics" preset. I was able to get over 100 fps in game play, and that's before MetalFX upscaling.

There's power on Apple Silicon for gaming, but more developers have to be on board. I would love to see more games on Metal, but Apple will need to do some serious developer relations to get more native development released.

That's not to say the new Macs can't game over the Rosetta instruction set. On the (admittedly old) Rise of the Tomb Raider, the M2 Pro reached 49 frames per second at 1920 x 1200, while the M2 Max reached 92 fps at the same resolution. The M2 Max even achieved a palatable 32 fps at 3456 x 2234.

Display on the MacBook Pro (2023)

The screens on both the 14-inch and 16-inch MacBook Pro are excellent. Apple labels them "Liquid Retina XDR," which effectively translates to high resolution mini-LED panels with HDR support and adaptive refresh rates up to 120 Hz using Apple's ProMotion technology.

The 14.2-inch laptop uses a 3024 x 1964 resolution screen, while the 16-inch notebook has a 3456 x 2234 resolution panel. Either way, the first episode of The Last of Us on HBO Max looked great. In a scene late in the episode in which Joel and Tess explore a darkened building with flashlights, the beams made a horrific infected body pop, and the shadows added a ton of atmosphere while not appearing too dark to actually see.

On our colorimeter, we found the screens to be largely identical for SDR content. They both covered 118% of the sRGB gamut and over 83% of the DCI-P3 gamut, and were around 450 nits of brightness. Those colors are better than the prior-gen 16-inch MacBook Pro, though seemingly at the expense of some brightness, though I certainly didn't have any issues.

Some in the PC space have made the move to OLED, which Apple hasn't done on its Macs yet. The Dell XPS 15's OLED touchscreen covered slightly more of both color gamuts, though it wasn't as bright as Apple's displays (the 14-inch  MacBook Pro reached 450 nits, the 16-inch laptop hit 447 nits).

In HDR content, the Macs could climb higher. At 40% coverage, the 14-inch laptop hit a peak brightness at 1,491 nits, while the 16-inch notebook hit 1,470.

By default, Apple uses a feature called True Tone to make the display adapt to different lighting conditions, which is great for watching movies or browsing the web. If you're working with photos, graphic design or videos, though, you may want to turn it off for the most accurate rendition of your work.

Keyboard and Touchpad on the MacBook Pro (2023)

The scissor switches on the MacBook Pro are solid, and in the exact same layouts on both the 14 and 16-inch MacBook Pro. You get backlit keys in a black well, which looks sleek. The full-sized function keys are a welcome reprieve from the ones on the Intel-based models (or the Touch Bar).

On the monkeytype typing test, I reached 120 words per minute using the 16-inch laptop's keyboard, with a standard 2% error rate.

Macs continue to have best-in-class trackpads. The 14-incher's Force Touch trackpad is 5.1 x 3.2 inches, while the 16-incher has a gargantuan 6.3 x 3.9 inch touchpad. Both use haptic feedback, so you can click anywhere on the touchpad. Both laptops have plenty of room for macOS gestures, even the ones that require three fingers.

Audio on the MacBook Pro (2023)

The 16-inch MacBook Pro continues to have my favorite speakers on any laptop, especially outside of gaming. Both the 14-incher and 16-incher have a six-speaker sound system, but for my money, the 16-incher just sounds slightly more detailed.

Both get loud and can fill a room with sound. Fall Out Boy's "Love From The Other Side" was balanced and detailed on the 16-inch Pro, with well-leveled guitars, vocals, drums and bass. It sounds excellent on the 14-inch model, too, and if they weren't together, I wouldn't notice the slight differences. I found the 14 didn't get quite as loud, and that the bass and drums sounded ever-so-slightly better on the bigger model.

Both laptops support Apple's spatial audio on supported apps, and also allow for Dolby Atmos on supported movies and TV. You can also get spatial audio over AirPods, and Apple claims that the 3.5 mm headphone jack offer pros "advanced support for high-impedance headphones."

Upgradeability of the MacBook Pro (2023)

While all that separates you from the inside of your MacBook Pro are eight pentalobe screws, don't expect to be making any upgrades here.

Firstly, pentalobe screws aren't terribly common. But more importantly, the RAM is part of the M2 Pro and M2 Max system on a chip, while the SSD is soldered to the motherboard. Configure your laptop with the memory and storage you're going to need to make it last.

As of this writing, some manuals for the new MacBook Pros are available through Apple's website and Self Service repair, but the full repair manuals have not been uploaded. Some teardowns have shown that there are some internal changes, including around the cooler and layout of the RAM and SSD on the motherboard.

Battery Life on the MacBook Pro (2023)

Apple likes to tout the efficiency of its systems, and for good reason. Both the 14 and 16-inch MacBook Pros proved to be long-lasting on our battery test, which continuously browses the web, runs OpenGL tests and streams video, all while connected to Wi-Fi with the screen set to 150 nits.

The 16-inch MacBook Pro, with its 100 WHr battery, ran for 18 hours and 56 minutes — more than 3 hours of improvement over the 2021 version. The 14-inch MacBook Pro, with a 70 WHr battery, lasted 14:21. The Dell XPS 15 ran for 9:43, though its OLED panel is likely a factor in its shorter runtime.

Heat on the MacBook Pro (2023)

We ran our heat test on the MacBook Pro while running our Cinebench R23 gauntlet to get a sense of how hot the chassis gets to the touch during a stressful multi-core workload.

On the 14-inch MacBook Pro, the center of the keyboard reached 45.8 degrees Celsius (114.4 degrees Fahrenheit), while the trackpad was a cooler 30.9 C (87.62 F). On the 16-inch laptop, the center of the keyboard hit 43.1 C (109.58 F), while the touchpad hit 29.9 C (85.82 F).  The 16-inch clearly kept these common touchpoints cooler, though neither seemed too uncomfortable to use.

The hottest point on the bottom of the 14-inch Mac measured 41.7 C (107.06 F), while on the 16-inch laptop it reached 43.7 C (110.66  F), reversing the trend from the keyboard and touchpad.

Webcam on the MacBook Pro 2023

Apple's FaceTime HD camera hasn't changed much since Apple redesigned the MacBook Pro in 2021, and that's largely good. The 1080p webcam is color accurate and catches great detail. Even in challenging light, the camera captured my royal blue sweater (and differentiated from the blue in my eyes), and showed detail in my hair and skin. I was near a window, and the lighting didn't blow out the image.

That may be in part due to Apple's Neural Engine on the M2 Max and M2 Pro, which the company says use "computational video" to improve output on the webcam and use machine learning to reduce noise in the image. It's definitely not as extreme as what I see on some of my iPhone photography — it's all pretty natural looking.

But I said it in 2021, and I'll say it again: I wish the notch that the camera is housed in also supported Face ID. I imagine the engineering to get the facial recognition that's already in the iPhone crammed into a thin laptop lid would be challenging, but many of Apple's competitors running Windows use facial recognition with IR cameras. Touch ID is accurate, but that notch really makes it feel like Face ID is something that's missing from the Mac.

If you need an even better webcam, macOS Ventura also lets you connect your iPhone to your laptop with a feature called Continuity Camera (see below), though I don't think that will be necessary for most. That's a feature that feels meant more for the older Intel or Apple M1-based laptops that run Ventura and don't have 1080p webcams. 

macOS Ventura and Software

The new MacBook Pros ship with macOS Ventura (version 13) installed. Ventura launched in the fall of 2022.

New features include some long-needed features in mail, like undo and schedule send; improved image search in Spotlight; Passkeys that never leave your device for more secure sign-in; editable messages in iMessage; Stage Manager, a way to organize windows in an attempt to better stay focused; Continuity Camera, which lets you use your iPhone as a webcam; and redesigned System Settings, which take some getting used to but more resemble what's on iOS.

Unlike with Windows machines, there really isn't any bloat in macOS. The software includes Mail, Messages (great for syncing with your iPhone or iPad), Notes, Maps, FaceTime and Safari. There are also dedicated apps for subscription services like Apple Music, Apple TV, Apple News and Podcasts. Productivity apps like Numbers, Pages and Keynote are also free.

Rosetta 2 is still around to translate apps designed for Intel x86 programs so that they can run on Apple Silicon, but more and more programs are now native for M-series chips or are using universal binaries. A small group of games, like Resident Evil Village, Disco Elysium , RollerCoaster Tycoon 3 and Minecraft all have native Apple Silicon support and run using Apple's Metal API.

MacBook Pro 2023 Configurations and Warranty

We tested two MacBook Pros for this review. A 14-inch MacBook Pro with M2 Pro (12-core CPU and 19-core GPU), 16GB of RAM and a 1TB SSD in a silver chassis; and a 16-inch MacBook Pro with M2 Max (12-core CPU and 38-core GPU), 96GB of RAM and 4TB of SSD in a space gray finish. The 14-incher costs $2,499.00, while the 16-inch version with all of that RAM and storage, is $5,299.00.

The 14-inch MacBook Pro starts with a cut-down M2 Pro with a 10-core CPU and 16-core GPU, along with 16GB of RAM, 512GB SSD and a 67-watt USB-C charger for $1,999. It's cheapest M2 Max option, with a 12-core CPU and 30-core GPU, along with 32GB of RAM and 1TB SSD is $3,099.

Apple's 16-inch MacBook Pro starts with an M2 Pro (!2-core GPU, 19-core GPU), 16GB of RAM and a 512GB SSD for $2,499. The way we reviewed it was effectively maxed out, unless you want to bump to 8TB of SSD storage, making it a whopping $6,499.

The prices to bump up RAM and storage are jaw-dropping. It's $400, for instance, to move a base 14-inch MacBook Pro with M1 Pro from 16GB to 32GB of RAM. With an M1 Max, it's $400 to move from 32GB to 64GB, or $800 to get to 96GB like in our review unit.

To get from 512GB to 1TB is $200, and it's $400 on top of that to reach 2TB. If you're a pro who wants 8TB of on-board storage, that's going to be $2,400 in storage upgrades.

Apple offers a one-year warranty on its MacBook Pros, unless you pay for AppleCare Plus. That starts at $99.99 annually for the 14-inch MacBook Pro, or $279 for three years. Coverage for the 16-incher costs $149.99 annually or $399 for three years. 

Bottom Line

The 14-inch and 16-inch MacBook Pros aren't huge changes over the previous generation, but that's largely not an issue. The design still feels new, the displays, while not OLED, are excellent, the speakers are the best I've heard and, perhaps most importantly, performance and battery life are excellent.

For most, I suspect the M2 Pro is going to be enough, but if you're doing serious graphics work, the M2 Max's extra GPU cores and higher memory bandwidth may be worth it.

The Dell XPS 15, the representative Windows-based sample here, didn't benchmark as well as Apple's latest, though it's using a 12th Gen Intel Core chip, with 13th Gen Core shipping shortly in a wide variety of notebooks. Those laptops still have more configuration options, including OLED, touch screens and in many cases, upgradeable RAM and storage. And Windows supports far more games, if that's a priority for you.

The new MacBook Pros take all of the design's existing strengths and add better chips with updates to Wi-Fi and HDMI. If you're coming from a 2021 model, that may not be much, but these will prove incredibly enticing to anyone coming from an Intel-based MacBook Pro or looking to make the jump from Windows.

MORE: How to Buy a Gaming Laptop

MORE: Best Gaming PCs

MORE: Best Ultrabooks and Premium Laptops

The $221 Intel Core i5-13400 comes armed with ten cores and 16 threads, but you can also buy the chip as a graphics-less Core i5-13400F model that delivers identical performance for $196, giving AMD's Ryzen 7000 lineup stiff competition — particularly in the mid-range tier where AMD's chips suffer the most from high platform pricing. 

The Core i5-13400/F is a solid improvement over the previous-gen 12400 model that has long dominated our best CPUs for gaming list due to its exceptional performance and attractive pricing. This new 13th-Gen Raptor Lake chip delivers even more gaming performance than its predecessor, and Intel's addition of e-cores to its 65W Core i5 processors, a first, brings additional threaded horsepower for productivity and game streaming, too.

Intel's Raptor Lake processors have taken the lead in gaming, productivity, and value over AMD's Zen 4 Ryzen 7000 processors, coming out on top in our CPU benchmark hierarchy. The rivalry remains heated, though: AMD recently released three new 'non-X' processors, including the $229 Ryzen 5 7600 that's the 13400's closest competitor. However, although the non-X chips are a better value than AMD's X-series processors, they still have other pricing problems. 

Despite Ryzen's impressive performance, the chips have been plagued by high B-series motherboard pricing and Ryzen's strict requirement for DDR5 memory, both of which lead to higher platform costs. In contrast, Intel's motherboard ecosystem remains less expensive, and the Raptor Lake chips support either value-centric DDR4 or the pricier DDR5. As our tests show, you should only buy the latter if you intend to splurge on a heavily-overclocked kit — at stock settings, a cheap DDR4 memory kit will give you roughly the same gaming performance as a pricey DDR5 kit. Additionally, DDR5 overclocking gains come with eye-pricing markups that largely aren't worth the small uplift.

Like AMD, Intel also throws in a bundled cooler with its lower-end 65W chips to sweeten the deal, and even though we generally recommend buying a better cooler than the bundled models, the Core i5-13400 cooler is at least serviceable for most users. 

AMD does have three new Ryzen 7000X3D processors coming this month with its disruptive 3D V-Cache tech that delivers explosive gains in gaming performance, but those chips start at $449, so they aren't for the mid-range market. As such, the faceoff between the Core i5-13400 and the Ryzen 5 7600 will define the ~$200 AMD vs Intel segment for the year to come. Let's see who wins the battle. 

Intel Core i5-13400 and 13400F Specifications and Pricing

The $221 Core i5-13400 comes with six performance cores (p-cores) and four efficiency cores (e-cores) etched onto the 'Intel 7' process node, for a total of 10 cores and 16 threads. The 13400 has a $22 markup over the previous-gen Core i5-12400's $199 debut price, but now you get four e-cores instead of zero. In effect, this makes the 13400 very comparable to the $289 previous-gen Core i5-12600K, which also came with six p-cores and four e-cores.

Perhaps most importantly, you can drop down to the $196 Core i5-13400F and receive the same performance — you just lose the 24-EU integrated UHD Graphics 730 engine, so you'll need a discrete GPU.

The Core i5-13400's p-cores have a 2.5 GHz base clock and boost up to 4.6 GHz on two cores in latency-sensitive work. That's 200 MHz faster than the previous-gen 12400 but 300 MHz slower than the 12600K. The e-cores kick in for background and multi-threaded tasks with a 1.8 GHz base and 3.3 GHz boost, a 300 MHz lower boost than the 12600K. The 13400 has a 65W Processor Base Power (PBP) and 148W Maximum Turbo Power (MTP) rating. As you'll see in our power testing, we didn't see the chip exceed 110W in any of our testing, so the MTP rating is higher than needed.

Most other features, like the supported DDR4-3200 and DDR5-5600 transfer rates, 9.5 MB of L2, and 20 MB of L3 cache, are the same as the previous-gen 12600K, and with good reason — Intel uses repurposed Alder Lake silicon for some of the Core i5-13400 chips, just as it did with the Core i5-12400 processors. The two types of dies come with different core architectures, but are designed to offer the same level of performance.

As such, the Core i5-13400/F doesn't have the same amount of cache per core as the Raptor Lake Core i7 and i9 chips. Instead, each p-core has 1.25 MB of L2 cache, and the quad-core e-core cluster has only 2MB of cache — the same as Alder Lake. That stands in contrast to the higher-end Raptor models that come with 2MB of L2 per p-core and 4MB of L2 for each e-core cluster.

The C0 die is the same silicon that Intel uses for the Alder Lake Core i5 processors, making the Core i5-13400 mostly identical to the Core i5-12600K (our test subject is C0). In contrast, the B0 stepping die comes with the Raptor Cove chip architecture, but has a cut-down cache configuration to match the Alder Lake silicon. 

The Raptor Lake processors have a vastly improved voltage/frequency curve compared to Alder Lake, making them more efficient, and a special microcontroller unit (MCU) firmware that imparts caching and prefetching advantages (deep dive here). Intel says it has tuned the two die to offer the same level of performance, but it is yet to be seen if some of Raptor Lake's finer-grained MCU tuning is present in the B0 stepping chips. We're working to acquire one to put it to the test.

As before, the LGA 1700 chip is compatible with 600-series motherboards, and because the chip has a locked multiplier (no overclocking), the B- and H-series motherboards make the most sense for this class of chip. Unlike the previous-gen 65W Alder Lake processors, there is no workaround for overclocking Raptor Lake's execution cores, as Intel adjusted to prevent a workaround from the motherboard makers.

You are free to overclock the memory on B-series motherboards, but be aware that the System Agent (SA) voltage is locked, which limits DDR4 memory overclocks in the Gear 1 setting (DDR5 still overclocks well).

Gaming Benchmarks on Intel Core i5-13400 and 13400F — The TLDR

• Intel Core i5-13400 DDR5 or DDR4: Corsair H115i 280mm water cooler, power limits removed, Stock DDR5-4800 in Gear 2 / DDR4-3200 in Gear 1
• Intel Core i5-13400 DDR4-3600 / DDR5-6800: Corsair H115i 280mm water cooler, power limits removed, DDR4-3600 CL16, DDR5-6800 XMP 3.0
• AMD PBO Configs: Precision Boost Overdrive (Motherboard), Scalar 10X, DDR5-6000 EXPO

You can find the particulars of our overclock and test setup further below. Here we have the geometric mean of our gaming tests at 1080p and 1440p, with each resolution split into its own chart. We're testing with an Nvidia GeForce RTX 4090 to reduce GPU-imposed bottlenecks as much as possible, and differences between test subjects will shrink with lesser cards or higher resolutions and fidelity. You'll find further game-by-game breakdowns below.

We have testing in the thermal and overclocking section that shows there is little to no performance loss in gaming if you use the 13400's included stock cooler, even with the power limits enforced — we saw no difference with the box cooler in 1080p gaming, and the stock cooler was 1.2% slower at 1440p (close enough to call a tie). To maintain a level playing field, we removed the power limits and used our H115i cooler for these tests.

The 13400's support for DDR4 memory is key in this price range, as it still costs nearly half as much as DDR5. Overall we see a difference of 1 fps between DRR4 and DDR5, with DDR4 coming out on top by a mere 0.6% — meaning the two Core i5-13400 configs offer basically the same performance.

We did get ~3% more performance from DDR5-6800 than we did with DDR4-3600, but the DDR5 kit costs nearly three times as much as the DDR4 kit ($144 more). You could step back to a DDR5-6000 kit that 'only' costs twice as much, but neither of those DDR5 options makes much sense in this price range.

The Core i5-13400 is 7% faster than the previous-gen Core i5-12400, but despite the physical similarities between the 13400 and the 12600K, the latter is still ~3% faster on the strength of its higher clock rates. That's a pretty slim delta, though, considering the 13400's lower pricing and big improvements in productivity applications.

The stock $229 Ryzen 5 7600 is 7% faster than the stock $221 Core i5-13400 with DDR5/DDR4 in the 1080p tests. Overclocking the 7600 increases the delta over the overclocked 13400 DDR5 system to 9%, and 12% over the overclocked DDR4 config. However, the 7600 carries a ~$30 premium over the 13400F. Adding in the higher motherboard and memory costs associated with the AM5 platform also takes a toll, as you can see in our value analysis in the conclusion.

Some have pointed to the $190 Ryzen 7 5700X as an alternative here, but the Core i5-13400 is 7% faster at 1080p. Given the 5700X is only $10 less than the 13400F, comes without a cooler, has lower performance than the 13400F in productivity apps, and uses an older platform, our test data suggests it is really only a contender for an upgrader currently on an AM4 platform.

The deltas in these charts can be slim, and large deltas in individual game titles, like with the 5800X3D, impact cumulative measurements. The competition between AMD and Intel chips can vary based on the title (particularly with DDR4 vs DDR5) and the GPU you use. It's best to make an informed decision based on the types of titles you frequently play, so be sure to check out the individual tests below.

Cyberpunk 2077 on Intel Core i5-13400 and Core i5-13400F

The Ryzen 5 7600 takes a strong lead in Cyberpunk 2077, while the Core i5-13400 trails across the board, even after overclocking the memory. 

The Ryzen 7 5800X3D continues to not provide much performance uplift over the standard Zen 3 models, which is important to remember — not all games benefit from the 3D cache-stacking tech. 

Far Cry 6 on Intel Core i5-13400 and Core i5-13400F

Far Cry 6 is very memory latency-sensitive, and here we can see that the DDR4 Core i5-13400 config in Gear 1 exceeds the DDR5 memory, which runs in the higher-latency Gear 2 mode, despite DDR5's throughput advantage. The DDR4 config even leads after stepping up to the relatively tame DDR4-3600 memory overclock, beating the DDR5-6800 config.

Meanwhile, the Ryzen 5 7600 edges past the Core i5-13400 in stock trim and takes an even bigger lead after overclocking. 

F1 2021 on Intel Core i5-13400 and Core i5-13400F

F1 2021 finds the Ryzen chips again taking substantial leads over the Core i5-13400, but that isn't entirely surprising given their higher pricing.   

Hitman 3 on Intel Core i5-13400 and Core i5-13400F

Here we can see the increased throughput of DDR5 memory having more of an impact in Hitman 3. This title also makes good use of the e-cores, thus delivering a more substantial lead over the previous-gen Core i5-12400. 

Microsoft Flight Simulator 2021 on Intel Core i5-13400 and Core i5-13400F

Microsoft Flight Simulator 2021 obviously benefits tremendously from L3 cache — the Ryzen 7 5800X3D is simply incredible in this title. This large advantage is amazing but doesn't represent the 5800X3D's performance in most titles. It also illustrates how outliers can make the 5800X3D seem more impressive in cumulative measurements.

Back in the more realistic land for mid-range gamers, the Core i5-13400 does much better with DDR4 memory than with DDR5, indicating that this title prefers lower latency accesses over sheer bandwidth. In fact, the Core i5-13400 with DDR4 is faster than the Core i5-12600K, which is tested with DDR5 memory.

Red Dead Redemption 2 on Intel Core i5-13400 and Core i5-13400F

Warhammer 3 on Intel Core i5-13400 and Core i5-13400F

Flipping over to the 1440p charts shows us that higher resolutions are often the great equalizer that levels the playing field for chips with similar accommodations. 

Watch Dogs Legion on Intel Core i5-13400 and Core i5-13400F

Productivity Benchmarks on Intel Core i5-13400 and 13400F — The TLDR:

We can boil down productivity application performance into two broad categories: single- and multi-threaded. These slides show the geometric mean of performance in several of our most important tests in each category, but be sure to look at the expanded results below.

The Core i5-13400 with either DDR4 or DDR5 offers the same performance in our cumulative metrics for single- and multi-threaded work, though there will be certain applications that can take better advantage of DDR5's increased throughput. Conversely, some applications will prefer the lower latency of DDR4.

All configurations of the Core i5-13400 match the Ryzen 5 7600 in single-threaded work, but overclocking the 7600 gives it a 4.5% lead. Meanwhile, the Ryzen 5 7600X stretches its legs and leads by 8%, but that comes with a higher price tag.

Turning to multi-threaded productivity applications, the 13400 with DDR4 and DDR5 is 5.5% faster than the Ryzen 5 7600 at stock settings and keeps a similar 5% lead after overclocking the 13400's memory and the Ryzen 5 7600. Meanwhile, the Ryzen 5 7600X is 2.5% faster than the Core i5-13400 in both stock and overclocked configs. 

The Core i5-13400's e-cores deliver a solid 16% boost in threaded work and a 4% boost in single-threaded tasks over its predecessor, the Core i5-12400. However, the 13400's relatively tame clock speed improvements ensure that it doesn't beat the 12600K, which leads by 7% in both single- and multi-threaded applications.

We saw no decline in gaming performance or single-threaded apps with the 13400's bundled cooler when we limited the chip to the stock 65W/148W power limits, but we did see a difference in multi-threaded applications — the chip was 5% slower with the power limits and stock cooler than it was with the H115i 280mm water cooler. 

However, removing the power limits while still using the boxed fan brought the performance in threaded apps back up to par with the watercooler. Naturally, there could be some long, extended rigorous workloads that might show more of a performance delta, but these will likely not have a meaningful impact on most users. Check our detailed benchmarks in the thermal section for more info.

Rendering Benchmarks on Intel Core i5-13400 and Core i5-13400F

The Raptor Lake Core i5-13400 rivals or beats AMD's like-priced Ryzen 5 7600 in nearly all single- and multi-threaded rendering applications, with the single-threaded Cinebench R23 benchmark being the lone exception. 

Encoding Benchmarks on Intel Core i5-13400 and Core i5-13400F

Most encoders tend to be either heavily threaded or almost exclusively single-threaded — it takes an agile chip to master both disciplines. Handbrake, SVT-HEVC, and SVT-AV1 serve as our threaded encoders, while LAME, FLAC, and WebP are indicative of how the chips handle lightly-threaded engines.   

The Core i5-13400 navigates the threaded workloads well, but trails in many of the lightly-threaded encoders. Again, we have to take pricing into account, as the Ryzen competitors carry higher price tags. 

Adobe, Web Browsing, Office, and Productivity on Intel Core i5-13400 and Core i5-13400F

The Ryzen processors dominate in the Adobe and Office applications across the board. 

Compilation, Compression, AI Chess Engines, AVX-512 Performance on Intel Core i5-13400 and Core i5-13400F

This selection of tests runs the gamut from the exceedingly branchy code in the LLVM compilation workload to the massively parallel molecular dynamics simulation code in NAMD to encryption and compression/decompression performance. Y-cruncher computes Pi with the AVX instruction set, making for an exceedingly demanding benchmark. 

Intel Core i5-13400 and 13400F Thermals, Power Consumption, and Boost Clocks

Our tests show that using the box cooler doesn't meaningfully impact gaming performance, even if you also lock the chip to the recommended 65W/148W power limit. The entry labeled "Core i5-13400" in the above charts represents the chip topped with a Corsair H115i 280mm liquid cooler and all power limits removed, while the "StockPL Box" setup measures performance with the bundled stock "box" cooler and fully-enforced 65W/148W power limits. We saw no difference with the box cooler in 1080p gaming, and the stock cooler was 1.2% slower at 1440p (close enough to call a tie). 

We also measured the difference in productivity applications. Unsurprisingly, there isn't any difference in single-threaded work, regardless of configuration. However, enforcing the power limits with the boxed cooler produces roughly 5% less performance in multi-threaded work.

To determine the source of the disparity, we re-ran the same tests with the box cooler and no power limits (listed as "Core i5-13400 NoPL Box"). Removing the power limits while still using the box fan brought the performance in threaded apps back up to par with the watercooler, showing that the cooler has enough thermal headroom to run heavy threaded apps without power limits imposed. That's an interesting result, so we tested it further. 

To remove thermals as a limitation, we always use the same 280mm Corsair H115i AIO cooler for all our test systems. This time we used the same three test configurations — the 280mm watercooler without power limits, the box fan without power limits (and its original pre-applied TIM), and the box fan with power limits — to measure performance in heavier work. 

We ran Prime95 first, but be aware that this stress test serves as a power virus more than an actual representation of most threaded work; real-world apps simply do not consume as much power or generate as much heat. The chart is a bit of an eye sore, but the takeaway is that even though the box fan without power limits (Box NoPL) ran at the 100C limit during the test with the chip throttling, it only lost 100 MHz of frequency (2.5%) on the p-cores compared to the watercooled configuration. Meanwhile, imposing the 65W/148W power limits dropped clocks to 3.4 GHz, a whopping 21% decline from the results with the liquid cooler. 

Prime95 is a stress test and not very indicative of normal use, so we then ran a looped series of POV-Ray benchmarks to see performance over time with a more realistic workload. Here we can see that the chip didn't exceed 90C with the power limits removed, and that clock rates between the watercooled and box cooler were the same. We then turned to even more standard productivity applications. 

To reiterate the point with a broader selection of real-world benchmarks, we ran our standard series of heavily multi-threaded apps like y-cruncher, HandBrake, Cinebench, Blender, and POV-Ray. 

The first slide shows that the box cooler with no power limits never exceeded 90C during these tests, showing that it is adequate to handle extended real-world workloads. This test shows the chip delivering the same 4.1 GHz all-core frequency as the 280mm AIO configuration. Meanwhile, imposing power limits is often 500 MHz slower. 

The stock cooler isn't silent, so you'll have to accept the extra noise. However, the takeaway is that the box cooler is good enough to handle the chip at the lower 65W/148W power limits and adequate for multi-threaded real-world workloads with the power limits removed. Not that removing the power limits is too taxing — the Core i5-13400 peaked at 110W in our tests, which is far under the 148W official Maximum Turbo Power. 

However, you could run into slightly more throttling if you run a heavy extended-duration heavy workload for an hour or more or have higher ambient temps (our tests were at ~74F ambient). However, the loss would be slight, and you would still come out ahead over running the chip at the recommended 65W power limits. 

We would recommend investing in a $20 tower air cooler to reduce the noise level and ensure 100% operation, but it isn't a steadfast requirement if you're willing to accept the drawbacks. 

Power Consumption on Intel Core i5-13400

The Core i5-13400's power consumption is surprisingly good, especially given that our test chip uses the Alder Lake C0 die. The Ryzen 5 7600 is more power efficient in our renders-per-day metric and consumes less power in every workload, but the deltas aren't large enough to be of concern for most users. The Core i5-13400 is very power efficient in these workloads, largely due to the addition of four e-cores. That confers a generational improvement against the Core i5-12400 in both performance and power consumed. 

Intel Core i5-13400 Overclocking and Test Setup

• Intel Core i5-13400 DDR5 or DDR4: Corsair H115i 280mm water cooler, power limits removed, Stock DDR5-4800 in Gear 2 / DDR4-3200 in Gear 1
• Intel Core i5-13400 DDR4-3600 / DDR5-6800: Corsair H115i 280mm water cooler, power limits removed, DDR4-3600 CL16, DDR5-6800 XMP 3.0
• AMD PBO Configs: Precision Boost Overdrive (Motherboard), Scalar 10X, DDR5-6000 EXPO

There isn't too much to say about overclocking — Intel doesn't allow overclocking the 13400's CPU cores, and a workaround created by the motherboard vendors for the previous-gen chips doesn't work with the new Raptor Lake models. That leaves us with lifting the power limits, which we did as described in the thermal section, and overclocking memory, which Intel allows on B-series motherboards.

We overclocked the DDR5 memory to DDR5-6800 using a single-click XMP 3.0 profile. With DDR4, we couldn't overclock beyond DDR4-3600 CL16 in Gear 1 (Gear 2 is slow) due to Intel's locked System Agent (SA) voltage. This is a small gain over the stock DDR4-3200 transfer rate. 

We did receive decent performance gains in gaming from this overclock, and a cheap DDR4-3600 CL kit only costs about $10 more than a DDR4-3200 kit. Our sources tell us the SA voltage will be unlocked in the Raptor Lake Refresh models, but you'll be stuck with relatively low DDR4 overclocks for the 13400.

Microsoft recently advised gamers to disable several security features to boost gaming performance. As such, we disabled secure boot, virtualization support, and fTPM/PTT for maximum performance. You can find further hardware details in the table at the end of the article. 

The Mid-Range Gaming Champ

The $221 Core i5-13400 and the graphics-less $196 Core i5-13400F build on the gaming performance of the previous-gen model that has long been the go-to mid-range gaming chip, but the addition of four e-cores drastically improves performance in threaded productivity workloads. The 13400's support for DDR4 memory gives it an unmatched value proposition in the $200 price range, making it the go-to chip for mid-range gaming PCs.

Below, we have the geometric mean of our gaming test suite at 1080p and 1440p and a cumulative measure of performance in single- and multi-threaded applications. We conducted our gaming tests with an Nvidia RTX 4090, so performance deltas will shrink with lesser cards and higher resolution and fidelity settings.

The $221 Core i5-13400 is 7% faster than the previous-gen Core i5-12400, a decent generational improvement. However, its main competitor comes in the form of the $229 Ryzen 5 7600, which is 7% faster than the Core i5-13400 at stock and overclocked settings in our 1080p tests. But the pricing between these two chips isn't as close as it appears.

The optimum low-cost build would also use the $196 graphics-less Core i5-13400F — it doesn't have an iGPU but offers the same performance as the normal model — to save another $25. You also have the option to stick with less expensive DDR4 memory with the Core i5-13400/F and get the same gaming performance as DDR5, albeit at the cost of some overclocking headroom with DDR5.

However, the last few percent of gaming performance from DDR5 overclocking is ridiculously expensive. With the Core i5-13400, we gained ~3% more gaming performance from the DDR5-6800 kit over the DDR4-3600 config, but the DDR5 kit costs nearly three times as much ($144 more). Stepping back to a DDR5-6000 kit that 'only' costs twice as much as the DDR4 kit doesn't make much sense in this price range, either. The AMD Ryzen 5 7600 profited more from overclocking with a 6% gain (some of this comes from PBO, though), and since there's no option for DDR4, you'll pay around 21% more over the already-pricey DDR5-5200 kit.

The Core i5-13400/F's DDR4 option combined with the generally much more forgiving Intel motherboard pricing can save ~$135 in total system costs compared to the Ryzen 5 7600.

We typically don't include cost-per-frame analysis due to fickle component pricing, and including several components magnifies the pricing volatility. This won't be applicable after a month or so (be sure to check current pricing), but the above table gives us the needed nuance for the Core i5-13400 vs Ryzen 5 7600 debate.

The DDR4-powered Core i5-13400F system is the absolute lowest-cost solution at $381 compared to the 7600's $514 — you'll pay 35% more for the Ryzen system and ~25% more per frame to get the ~7% increase in 1080p gaming performance.

The $229 Ryzen 5 7600 isn't price-competitive with the Core i5-13400 DDR4 system at all, but it is more attractive if you already plan to shell out the extra cash for DDR5. However, there's a higher pricing bar for entry than with an Intel Core i5-13400 DDR5 system. The Ryzen system costs ~$50 (10%) more than the DDR5-powered Core i5-13400F system, or ~4% more per frame when we factor in performance. That's close enough to make the Ryzen 5 7600 a solid choice for a DDR5 gaming system, provided you have the extra $50 in the build budget.

Thus, DDR5 builders might turn to other factors, like the 7600's performance in applications (it's 5% slower than the 13400 in threaded work), or the fact that the AM5 platform is supported until 2025+ while the Raptor Lake platforms only have one more refresh gen in the cards. Intel's CPU cores also aren't overclockable (memory is), while Ryzen is fully unlocked. Naturally, personal preference also weighs in. 

Both chips are evenly matched in single-threaded productivity applications, but the Core i5-13400 with DDR4 and DDR5 is 5.5% faster than the Ryzen 5 7600 in threaded workloads at both stock and overclocked settings. Again, DDR4 and DDR5 provide similar performance with the 13400 in productivity apps, both at stock and overclocked settings. DDR5 simply isn't worth it.

The 13400's four e-cores provide extra horsepower that helps with streaming — programs like OBS are designed to take advantage — and the additional cores don't add too much power consumption. If you're looking for a more productivity-focused chip, the Core i5-13500 looks very compelling with eight total e-cores; stay tuned for that review.

The Core i5-13400 does suck more power than the Ryzen 5 7600, but it is reasonable, which helps reduce heat output. Both the 13400 and the 7600 come with bundled 'box' coolers, and even though it is surprising, the Intel cooler is serviceable. 

As always, we recommend upgrading to a $20 tower-style air cooler for this class of chip, but it isn't a requirement. The Intel cooler is more than adequate with the recommended power limits and can even handle strenuous work with the power limits fully removed. You could run into throttling if you run a heavy extended-duration heavy workload for an hour or more, but you would still come out ahead over running the chip at the recommended 65/148W power limits.

The Ryzen 5 7600 is a worthy alternative if you plan on upgrading to DDR5, but you will have to stomach the higher price of entry over Core i5-13400 systems that use both DDR4 and DDR5. That doesn't seem as reasonable in this price range.

With DDR5, the Core i5-13400 offers a lower price of entry and higher performance in threaded workloads. Even though it is 7% slower in gaming, it still has a competitive price-to-performance ratio with the competing AMD chips — but you don't need DDR5 to unlock the best performance.

When paired with DDR4, the Core i5-13400 generally provides the same performance as a DDR5 system but delivers a superior price-to-performance ratio in both gaming and productivity applications. Add in the unmatched entry-level platform pricing, and the Core i5-13400 is the best bang for your buck in the ~$200 price range, making it the mid-range gaming CPU leader.

• MORE: AMD vs Intel
• MORE: Zen 4 Ryzen 7000 All We Know
• MORE: Raptor Lake All We Know 

A paper shared on Cell Reports Physical Science suggests coating heatsinks with hydrogels delivers a “significant” benefit to passive cooling. The paper highlights the testing of a hydrogel-based moisture thermal battery (MTB) for passive thermal management of electronics such as a 100W FET and an 84W Intel Core i5-4690 CPU. Test results are encouraging, with the MTB making the electronics components run approximately 15°C cooler than a heatsink alone. Caveats apply, naturally.

In a preamble to the research discussion, the scientists from the University of California remind us that cooling electronics is a necessary but expensive business, with the main costs being energy and water usage and the environmental impact of using those resources. If passive cooling can be improved, it could be of great benefit in trimming such overheads.

Hydrogel coatings cool in a similar way to how an animal sweats. The scientists explain that when the electronics are under load, the hydrogel will release water that evaporates to create a cooling effect. When the electronics are idling, this allows the hydrogel to recharge its moisture content, getting ready for the next workload.

From that description you can probably sense that hydrogel cooling is a great fit for applications with cyclical workloads. The scientists give examples of a 5G cell tower and a data center, both with 12 hour high-load to low-load cycles.

Tests were completed with the aforementioned 100W FET and 84W Intel CPU, and in both cases MTB cooling performance is “significantly higher than previously reported passive cooling methods.” The comparison tests were run at a room temperature of 22°C, and relative humidity (RH) of 70%. The 15°C delta seems a very worthwhile improvement. Humidity helps the MTB recharge between thermal loads being applied. It would thus work particularly well in the tropics — locations like Florida, Hong Kong, Singapore, Taiwan, etc. — where relative humidity can be 80–90%.

In situations where the MTB isn’t sufficient on its own, it can be leveraged in hybrid solutions alongside heat pipes, vapor chambers, or liquid circulation loops. However, there are ways to tune the MTB for your particular use case. For example, the MTB responds positively to having a larger surface area and thinner hydrogel coating, where design constraints allow. Another thing in favor of the MTB is that the scientists claim it is a scalable solution, with a relatively low price of $12 per unit.

The scientists conclude that ideal applications for the MTB are in the “thermal management for 5G chips, power batteries, server/data centers, and optoelectronics.” Device use following the ideal cyclical nature needed for MTB water recharging from humidity might fit with your usage of a device like an HTPC. But this is clearly still in the test phases.

We would have liked to see a chart showing what happens if the workload doesn’t run along a smooth cycle, so the hydrogel is unable to recharge. It would also be interesting to see results at higher power draws, since modern CPUs can easily reach into the 200W and higher range. Would the hydrogel stop delivering its 15°C delta, or would it even break down under a long sustained thermal load? These are perhaps questions for future research.

The longer you own a laptop, the less charge its built-in battery will hold. You could buy one of the best ultrabooks which can go for 12 to 14 hours unplugged and after a couple of years of regular use, it might only endure for 9 hours or less. As your laptop goes through more charge cycles over the years, its battery health gets worse.

Fortunately, for most laptops made in the last 5 to 10 years, there are replacement batteries available for $100 or less, and service manuals that explain how to replace the built-in unit on your own. But before you decide to swap out your old battery for a new one, you should get a battery health report to see just how much capacity it has left. 

Below, we'll show you how to use the built-in battery report in Windows 10 and Windows 11 (our screenshots show Windows 11 but it's the same in 10) to find out just how many charge cycles your battery has experienced and how many mWh (milliwatt hours) it can hold. Note that a charge cycle is a complete discharge and charge of the battery, so many instances of unplugged use could make up a single cycle. For example, if you used the laptop until the battery was at 75% capacity and recharged it to 100%, that would be one quarter of a cycle.

How to Get a Battery Health Report in Windows 10 or 11

1. Launch command prompt as admin. The easiest way to do this is to search for "cmd" in Windows search, right click the top result and select run as admin.

2. Enter the following command at the prompt and hit Enter.

powercfg /batteryreport /output c:\breport.html

You can change the output location and file name to anything you want as long as the file name ends in .html. I've chosen C:\breport.html as mine. 

After you enter this command, you'll get a message saying that the report has been created.

3. Open the battery report HTML file (breport.html) in your browser. You'll see a web page filled with data.

4. Scroll down to see the Design and Full Charge Capacity fields, along with the Cycle Count number. 

The Design Capacity is the amount of mWh that the battery should have had when it was brand new and the Full Charge Capacity is the amount that it can hold now when at 100 percent charge. In the case of my four year old Lenovo ThinkPad X1 Carbon laptop, the difference between Design and Full Charge was about 12 percent when I ran my report. It had run through 309 full charge cycles since I first got it in September 2018.

If you scroll further down the page, you can see a long history of all the times your laptop was operating on battery and on AC power. However, I found this information largely unhelpful as it didn't give me any sense of how much battery life I've been getting in regular use.

Using BatteryInfoView to Get a Battery Report

If you like a more GUI-driven battery report, I recommend trying BatteryInfoView, a freeware utility from NirSoft. Once you've downloaded and unzipped BatteryInfoView, you just run it (it doesn't install). 

There are two main tabs of information: Battery Information and Battery Log. Battery Information, the main view, shows most of the same information as the Windows 10 or 11 built-in battery report does. It has the Design and "Full Charged" capacities, along with the number of cycles. 

By hitting F8 or selecting it from the View menu, you can see the Battery Log view, which updates itself every minute (or a different frequency if you choose) with what is happening while you are running BatteryInfoView. 

The log will show you whether the laptop was plugged in or unplugged and exactly how much charge was left at that moment. You can select "HTML report" from the view menu and have it outputted as an HTML table. You can also clear the log. However, unlike the Windows Battery Report log, this one only shows what happened while it was running. So, if you want to keep a long log, you need to keep BatteryInfoView open at all times.

The 24-core Intel Core i9-13900KS Special Edition processor carries an eye-watering $699 price tag, but it has a feature that Intel feels warrants the price tag — this is the first consumer PC chip to run at 6 GHz without overclocking, extending Intel's lead over AMD's fastest Ryzen 7000 processors. However, with a whopping 250W base power specification, this 13th-Gen Raptor Lake chip is also now officially the most power-hungry desktop CPU in history — its voracious appetite even peaks at 320W in a new Extreme Power Delivery Profile.

Intel's Raptor Lake processors have taken the lead in gaming, productivity, and value over AMD's Zen 4 Ryzen 7000 processors, but the competition remains stiff. The relatively low-profile 13900KS launch comes as AMD preps three new Ryzen 7000X3D processors with its disruptive 3D V-Cache tech that delivers explosive gains in gaming performance. 

The 3D V-Cache tech's first iteration, AMD's Ryzen 7 5800X3D, overtook Intel's Alder Lake as the best CPU for gaming, but Intel later retook the crown with Raptor Lake. AMD's Ryzen 7000X3D chips are poised to come to market in mere weeks, and the Core i9-13900KS is meant to keep Intel's silicon entrenched at the top of the CPU benchmark charts when AMD's new chips arrive. 

That might be a tall order — the Ryzen 7000X3D chips look exceptionally promising due to their exotic performance-enhancing tech. In contrast, the 13900KS is built on the same architecture as the 13900K, just with a higher speed binning to accommodate 200 MHz faster clock rates. Otherwise, it uses the same fundamental design as its standard counterpart, but it comes with a hefty upcharge that destroys any hint of a value proposition for all but the extreme overclocking crowd.  

Naturally, if you're looking to buy a $699 chip, price considerations go out of the window — this is a pure speed play for those with deep pockets. For now, we don't know if the Core i9-13900KS will keep Intel in the top spot when the Ryzen 7000X3D chips arrive next month, but it looks like it will be close. Here's how the Core i9-13900KS stacks up against the current competition.

Intel Core i9-13900KS Specifications and Pricing

The 13900KS is the fastest Intel desktop PC chip, and even at stock settings, it drew up to 328W and ran at 100C to scrape out every last bit of performance. Because the 13900KS is a Special Edition, the company will only produce a limited (but unspecified) number of these processors — hence the $699 price tag. Despite the high pricing, Intel has actually lowered the ceiling for access to its most premium silicon: The previous-gen 12900KS debuted at $739, but Intel later increased the pricing of all its 12th-Gen chips, bringing it to $823 today.

Intel selects its premium-binned 13900K silicon for the 13900KS, so it is guaranteed to be among the very best silicon the company has to offer. That will make the chips very attractive to overclockers, as paying the extra $110 for the KS model improves your odds in the silicon lottery, essentially assuring you're getting a cherry chip. Here's a video that describes Intel's binning process:

 As with the Core i9-13900K (architecture details here), the KS model comes with eight hyperthreaded p-cores and 16 single-threaded e-cores, for a total of 24 cores and 32 threads. The KS model's p-cores have a 3 GHz base clock and reach up to 6 GHz on two cores via Thermal Velocity Boost (TVB) tech. This allows the processor to shift into slightly higher frequencies if it remains under a certain temperature threshold (70C for desktop chips). Meanwhile, the standard Turbo Boost 3.0 clock rates reach 5.8 GHz. We found that the chip easily peaks at 6 GHz, though the impact on workloads can vary. 

The e-cores kick in for background and multi-threaded tasks with a 2.2 GHz base and  4.3 GHz boost, identical to the 13900K. Most other features, like the supported DDR4-3200 and DDR5-5600 transfer rates, 32MB of L2, and 36MB of L3 cache, remain the same. 

As before, the LGA 1700 chip is compatible with 600-series motherboards, but you'll want one of the best-equipped Z-series boards to quench the 13900KS' thirst for power. However, selecting the right motherboard is a bit more complicated for the 13900KS due to the new 320W "Extreme Power Profile."

Intel's power profiles have a few key variables. The PL1 power spec quantifies extended use at base clocks and is also known as the Processor Base Power (PBP). In contrast, PL2 power quantifies when the chip is under full load for short periods (boost) and is also known as the Maximum Turbo Power (MTP). ICCMax denotes the maximum current the chip can pull when under heavy load.

Intel also used to define a Tau variable that specified the duration of the boost/PL2 state before the chip dropped back into PL1. Intel has since stopped defining Tau for K-series models because it assigns PL1 and PL2 as the same value for Alder Lake and following gens in the reference BIOSes that it sends to OEMs — a first. This means the chip always operates in a boosted/PL2 state.

Intel's new policy of sending reference BIOSes with PL1=PL2 settings basically invalidates even having a PL1 value, and the fact that it still exists on the spec sheet as a "Processor Base Power (PBP)" setting — the lower of the two power specs — is misleading.

Intel defined an Extreme Power Profile for its regular 13900K — the 253W PL1/PL2 values remain the same for both the standard and extreme profile, but the extreme profile's ICCMax reaches 400A (up from 307A).

The KS model has the same default profile as the 13900K but also has its own new Extreme Power Profile that allows for a 320W PL1/PL2 and 400A ceiling. You'll need to ensure that your motherboard can deliver the peak current if you want to unleash the full power of the KS, as not all motherboards can for a long period of time. Motherboard vendors allow assigning a higher ICCMax value in the BIOS, typically under settings like "Core/CPU Current Limit" (the name varies by mobo maker), but that doesn't mean the motherboard can actually deliver that amount of current. Obviously, B- and H-series boards don't make the cut.

Intel defines these recommended power profiles but allows motherboard vendors to ignore them completely, and exceeding the default values doesn't void the warranty. Thus, by default, most motherboard makers completely ignore the limits and assign the maximum values for PL1, PL2, and ICCMax, resulting in higher performance and more heat. 

Even at stock settings, the Core i9-13900KS hit up to 328W of power and 100C in our testing. We'll see what power draw, performance, and thermals look like, including gaming and productivity benchmarks, on the following pages.

• MORE: AMD vs Intel
• MORE: Zen 4 Ryzen 7000 All We Know
• MORE: Raptor Lake All We Know 

Intel Core i9-13900KS Thermals, Power Consumption, and Boost Clocks 

Intel's Adaptive Boost Technology (ABT) dynamically boosts to higher all-core frequencies based on available thermal headroom and electrical conditions, so peak frequencies can vary. By design, this tech allows the chip to operate at 100C during normal operation — if the chip runs under the 100C threshold, it will increase its power consumption until it reaches the safe 100C limit, thus providing more performance. However, this feature is only active on the Core i9-13900K/F and the Core i9-13900KS, so other Raptor Lake processors won't exhibit the same behavior.

You can think of ABT much like a dynamic auto-overclocking feature, but because the chip stays within Intel's spec of a safe 100C temperature limit, it is a supported and warrantied feature that doesn't fall into the same classification as overclocking. ABT uplift will vary by chip — much of the frequency uplift depends upon the quality of your chip. Hence, the silicon lottery comes into play, along with cooling and power delivery capabilities.

Remember, AMD's Ryzen 7000 also runs at its limit of 95C at stock settings, so higher temperatures have become the norm for both chipmakers.

To remove thermals as a limitation, we always use the same 280mm Corsair H115i AIO cooler for all our test systems. However, Intel says that it achieved the best results with the KS paired with a 360mm radiator, so we tested the impact by running the same tests with all power limits removed on a 280mm and a 360mm AIO. Our single-threaded test also shows that the 13900KS boosted to 6 GHz frequently, regardless of the cooler.

The chip reached 100C with both coolers during a series of heavily multi-threaded apps, like y-cruncher, Cinebench, Blender, and POV-Ray, just as Intel designed it to do. The 360mm cooler config ran the same 5.6 GHz all-core clock during heavy work as the 280mm setup, yet it also consumed around 20-25W more during some types of work. 

That 360mm's additional cooling capacity enables that extra bit of power consumption, but the ~7% to 9% increase occurs at the top of the chips' voltage frequency curve where increased power consumption is incredibly inefficient: As you near peak power, double-digit percentage increases in power consumption often only yield single-percentage performance gains. That means this slightly higher power will not make much difference in actual benchmarks, which we'll detail below.

We also overclocked the chip via turbo multipliers to 6 GHz on two p-cores and 5.7 GHz when more than two cores are active, while dialing in a 4.4 GHz all-core overclock on the e-cores. The chip only required a 1.29V vCore to sustain these frequencies, indicating it is a cherry chip. The overclocked config uses less power than stock settings, showing that the chips' native power management is inefficient. As you'll see below, the overclocked settings enabled much more performance than the stock settings that consumed more power. 

The chip barely pushed over 300W with standard applications, but switching gears to the strenuous but not-at-all-realistic Prime95 stress test yielded a much higher peak power consumption of 328W. Power consumption leveled off after a short period, so it's possible a custom watercooling loop could allow the chip to consume more for a longer period of time. But, again, this won't result in very meaningful performance improvements in real-world workloads — this chip is tuned to the absolute top of its voltage/frequency curve.

There isn't a huge difference in cooling capacity between our 280mm and 360mm AIOs (we'd see a bigger difference with a 240mm vs 360mm comparison). As you can see in our cumulative performance measurements above, the improved cooling doesn't result in a linear improvement in performance in our gaming or application benchmarks.

The improvement in real-world gaming and productivity applications with the 360mm cooler was around 1%. As such, we used the test results from our 280mm configuration for our gaming and productivity benchmarks.

Our overclock gave us an extra 5% in 1080p gaming and 4% in threaded work, but be aware that the increased memory throughput (we used DDR5-6800 for the overclocked config) is a big contributor here. Additionally, the overclocked vanilla 13900K trailed by only 1% in games and 2.5% in threaded apps. Further tuning, or more luck in the silicon lottery, could narrow that gap.

Temperatures can limit your performance during stock operation, so if you purchase the 13900KS, plan for a powerful cooler to extract the full performance. We think a 280mm AIO would be sufficient, but if you're chasing the last 1% of performance, a 360mm AIO will get you there.

• MORE: AMD vs Intel
• MORE: Zen 4 Ryzen 7000 All We Know
• MORE: Raptor Lake All We Know 

Power Consumption on Intel Core i9-13900KS

These power measurements show that the Core i9-13900KS will rarely reach its 320W power limit with standard applications — there simply isn't enough thermal headroom due to the difficulty of dissipating this amount of heat from such a small area (thermal density). 

More exotic cooling, like custom water loops or sub-ambient, would likely allow the chip to pull more power within its 100C thermal envelope, albeit for little practical performance gain. 

Here we take a slightly different look at power consumption by calculating the cumulative energy required to perform x264 and x265 HandBrake workloads, respectively. We plot this 'task energy' value in Kilojoules on the left side of the chart.

These workloads are comprised of a fixed amount of work, so we can plot the task energy against the time required to finish the job (bottom axis), thus generating a really useful power chart.

Remember that faster compute times, and lower task energy requirements, are ideal. That means processors that fall the closest to the bottom left corner of the chart are the best. 

The 13900KS sucks significantly more power than the 13900K in these tests, but that doesn't deliver much of a performance gain — increasing power consumption when the chip is already near the top of the voltage/frequency curve is an incredibly inefficient way to gain very little extra performance. 

Intel Core i9-13900KS Overclocking and Test Setup

• Intel Core i9-13900KS @ 6.0-5.7: Turbo multiplier: Two cores at 6 GHz, 5.7 GHz all-p-core, 4.4 GHz e-core, 1.29V vCore, DDR5-6800 XMP 3.0
• Intel Core i9-13900K @ 5.6: 5.6 GHz all-core p-core, 4.4 GHz e-core, 1.32V vCore, DDR5-6800 XMP 3.0
• AMD PBO Configs: Precision Boost Overdrive (Motherboard), Scalar 10X, DDR5-6000 EXPO

We used DDR5 memory with all chip testing in this article. Notably, if you opt for DDR4 with the Raptor Lake processors, you'll lose a few percentage points of performance, on average (testing here). The overclocked Ryzen configs (marked as 'PBO' in the charts) use DDR5-6000 memory, Precision Boost Overdrive, and a Scalar 10X setting.

We overclocked the Core i9-13900KS via turbo multipliers to 6 GHz on two p-cores and 5.7 GHz when more than two cores are active, while dialing in a 4.4 GHz all-core overclock on the e-cores. The chip only required a 1.29V vCore to sustain these frequencies. You can see the heat output and more overclocking details in our stress tests in the thermal section on the prior page.

We didn't use Intel's "Extreme Power Profile" for our stock Intel tests; instead, we're sticking with our standard policy of allowing the motherboard to exceed the recommended power limits, provided the chip remains within warrantied operating conditions. This means our power settings exceed the 'Extreme' recommendations — almost all enthusiast-class motherboards ignore the power limits by default anyway. Hence, our completely removed power limits reflect the out-of-box experience. Naturally, these lifted power limits equate to more power consumption and, thus, more heat.

Microsoft recently advised gamers to disable several security features to boost gaming performance. As such, we disabled secure boot, virtualization support, and fTPM/PTT for maximum performance. You can find further hardware details in the table at the end of the article. 

• MORE: AMD vs Intel
• MORE: Zen 4 Ryzen 7000 All We Know
• MORE: Raptor Lake All We Know 

Gaming Benchmarks on Intel Core i9-13900KS — The TLDR 

You can find the particulars of our overclock and test setup further below. Here we have the geometric mean of our gaming tests at 1080p and 1440p, with each resolution split into its own chart. We're testing with an Nvidia GeForce RTX 4090 to reduce GPU-imposed bottlenecks as much as possible, and differences between test subjects will shrink with lesser cards or higher resolutions and fidelity. You'll find further game-by-game breakdowns below.

The Core i9-13900KS is now the fastest gaming CPU in the world, but only by a hair. Using a geometric mean of our 1080p gaming results, the $699 Core i9-13900KS is a mere 1.5% faster than the standard $589 13900K. That's imperceptible to most users and certainly not worth the extra ~20% you'll pay at retail. Further, the performance delta shrinks to a mere 0.5% with 1440p gaming, so you absolutely should not buy the KS model for gaming only.

The Core i9-13900KS contends with the $699 Ryzen 9 7950X in the pricing department, but the standard Core i9-13900K model already holds a strong enough lead in gaming to make the extra $110 for the KS model a non-starter for most of us.

Naturally, the pre-binned Core i9-13900KS will appeal to overclockers — your odds of scoring a cherry chip are increased dramatically. However, you're still at the whims of the silicon lottery with both the KS and the K model, so you could get lucky with either chip.

In fact, our tests didn't show a very big lead for the KS model over the K, and we could have likely dialed in a 5.9 GHz overclock on two of the K model's cores and reduced the delta further. Not that there's a big difference between the chips anyway — the overclocked KS model was only 1% faster than the K-series model after overclocking.

Overall, the KS model doesn't make much of a difference to the positioning of the 13th-Gen processors against the Ryzen lineup, and it is a poor value. This chip will appeal to deep-pocketed individuals either intent on building the highest-end system imaginable or extreme overclockers chasing world records. The Core i7-13700K remains the chip to beat if you're looking for a high-end gaming and productivity chip, as the 13900KS is only 2.5% faster yet costs a whopping $290 more.

The Ryzen 7 5800X3D is the fastest AMD gaming chip by far due to its 3D V-Cache, but its extra 96MB of L3 cache doesn't accelerate all games, and you'll also have to accept much lower performance in just about every other type of productivity application. If you're looking for an exotic 3D V-Cache gaming chip, your best option is to wait until next month when AMD launches the new Zen 4-powered Ryzen 7000X3D models. These chips look like they'll have a much more balanced performance profile and might beat the KS model in gaming, too.

The deltas in these charts can be slim, and large deltas in individual game titles, like with the 5800X3D, impact cumulative measurements. The competition between AMD and Intel chips can vary based on the title and the GPU you use. It's best to make an informed decision based on the types of titles you frequently play, so be sure to check out the individual tests below.

The individual game benchmarks below are predictable — the KS reliably scores one to two percent higher than the standard K model across the board. As such, we won't comment on the individual game results below. 

Cyberpunk 2077 on Intel Core i9-13900KS

Far Cry 6 on Intel Core i9-13900KS

F1 2021 on Intel Core i9-13900KS

Hitman 3 on Intel Core i9-13900KS

Microsoft Flight Simulator 2021 on Intel Core i9-13900KS

Microsoft Flight Simulator 2021 obviously benefits tremendously from L3 cache — the Ryzen 7 5800X3D is a whopping 28% faster than the Core i9-13900K flagship on this test. This large advantage is amazing but doesn't represent the 5800X3D's performance in most titles. It also illustrates how outliers can make the 5800X3D seem more impressive in cumulative measurements.

In either case, this does bode well for the Zen 4 Ryzen 7000X3D models, which should bring all the goodness of the capacious L3 cache with fewer tradeoffs in standard applications.  

Red Dead Redemption 2 on Intel Core i9-13900KS

Warhammer 3 on Intel Core i9-13900KS

Flipping over to the 1440p charts shows us that higher resolutions are often the great equalizer that levels the playing field for chips with similar accommodations. 

Watch Dogs Legion on Intel Core i9-13900KS

• MORE: AMD vs Intel
• MORE: Zen 4 Ryzen 7000 All We Know
• MORE: Raptor Lake All We Know 

Productivity Benchmarks on Intel Core i9-13900KS — The TLDR: 

We can boil down productivity application performance into two broad categories: single- and multi-threaded. These slides show the geometric mean of performance in several of our most important tests in each category, but be sure to look at the expanded results below.

Overall the same general trends that we saw in our gaming benchmarks appear in our application testing — the Core i9-13900KS does nothing to change the competitive positioning against AMD's Ryzen, and while the 13900KS is the fastest chip in the world in both single- and multi-threaded work after overclocking, its high price tag makes the slight gains simply not worth it.  

The Ryzen 9 7950X clings to its lead in multi-threaded work over the Core i9-13900KS by a mere 0.5%, but that's in the noise of the benchmarks. We should consider these two chips tied. Overclocking flips the advantage back in Intel's favor, but this is also close enough to call a tie. We also see slight gains over the Core i9-13900K in multi-threaded work, but this isn't worth the increased pricing. 

Flipping over to single-threaded work reveals that the 13900KS holds a clear win over all contenders — it is 14% faster than the Ryzen 9 7950X and 2% faster than the standard 13900K. Still, that's a relatively slim delta over the standard 13900K.

You'll have to accept severe performance tradeoffs if you opt for AMD's fastest gaming chip, the Ryzen 7 5800X3D. The 5800X3D is significantly slower than the newer chips because it comes with the previous-gen Zen 3 architecture. The extra dollop of L3 cache doesn't impart any meaningful speedups in most standard desktop PC applications. If you're looking for an X3D chip, wait for the 7000X3D chips that launch next month. 

Rendering Benchmarks on Intel Core i9-13900KS

Intel's Raptor Lake rivals or beats AMD's finest in multi-threaded productivity applications at every price point. These results, like many of the results throughout our other application tests below, mirror the established trends of the KS slightly extending the lead relative to the 13900K. As such, we'll have limited commentary for the application benchmarks. 

Encoding Benchmarks on Intel Core i9-13900KS

Most encoders tend to be either heavily threaded or almost exclusively single-threaded — it takes an agile chip to master both disciplines. 

Adobe, Web Browsing, Office and Productivity on Intel Core i9-13900KS

The ubiquitous web browser is one of the most frequently used applications. These latency-sensitive tests tend to be lightly threaded, so a fast response time is critical.

Compilation, Compression, AI Chess Engines, AVX-512 Performance on Intel Core i9-13900KS

This selection of tests runs the gamut from the exceedingly branchy code in the LLVM compilation workload to the massively parallel molecular dynamics simulation code in NAMD to encryption and compression/decompression performance. Y-cruncher computes Pi with the AVX instruction set, making for an exceedingly demanding benchmark. 

• MORE: AMD vs Intel
• MORE: Zen 4 Ryzen 7000 All We Know
• MORE: Raptor Lake All We Know 

Paying the 6 GHz Tax

The $699 Core i9-13900KS Special Edition is the fastest gaming chip in the world with its record-setting 6 GHz peak turbo clock. However, for all but extreme overclockers, collectors, and the well-heeled looking to build the ultimate system at any cost, the costly chip's relatively small performance advantage over the vanilla 13900K isn't worth the $110 upcharge.

Below, we have the geometric mean of our gaming test suite at 1080p and 1440p and a cumulative measure of performance in single- and multi-threaded applications. We conducted our gaming tests with an Nvidia RTX 4090, so performance deltas will shrink with lesser cards and higher resolution and fidelity settings.

While the 13900KS is now the fastest gaming CPU money can buy, it only holds a slim 1.5% advantage over the standard 13900K in our 1080p gaming tests. That delta shrinks to a mere 0.5% when we flip over to 1440p gaming, meaning it is essentially in the noise of the benchmarks. That certainly isn't worth paying ~20% more.

We see a similar trend in multi-threaded workloads — the KS model only offers a slight 1% improvement over the standard 13900K and essentially ties the Ryzen 9 7950X. We see a more noticeable improvement in single-threaded work against Ryzen, as the 13900KS is 14% faster than the Ryzen 9 7950X. However, the KS is only 2% faster than the standard 13900K, so the big upcharge isn't worth it.

The big appeal for the KS is that you're guaranteed to get Intel's best silicon, basically giving you the winning ticket for the silicon lottery. That will undoubtedly make the chip popular with extreme overclockers, especially those chasing world records. However, again, the overclocking advantage can vary, and we only saw a few percentage points of difference in our own head-to-head overclocking against the standard 13900K.

Overall, the KS model doesn't make much of a difference to the positioning of the 13th-Gen processors against the Ryzen lineup, and it will only be sold in limited quantities. Intel now allows pairing its consumer chips with ECC memory on motherboards with the workstation-geared W680 chipset, so the Core i9-13900KS could make for a powerful workstation machine. However, it will still only deliver small gains over the 13900K.

The Core i9-13900KS just isn't a practical buy for the majority of us, especially considering that it will need the priciest of accommodations, like the highest-end motherboard and PSU to pump 320W of power to the processor and a potent cooler to handle the 100C operating temperatures when the chip is under heavy load. All of which is a high price to pay for a few percentage points (at best) of extra performance. 

If you want to invest in the fastest gaming chip, you should wait to see how AMD's Ryzen 7000X3D chips perform when they launch next month. These chips come with 3D V-Cache tech that provided explosive gaming performance gains for Zen 3, and it's possible that it could push Zen 4 into a solid lead in gaming over Raptor Lake. Pricing remains the wild card, but we expect the high-end 16-core Ryzen 9 7950X3D to land in the same range as the 13900KS. 

In the meantime, the Core i9-13900KS represents Intel throwing value and power consumption out the window in an effort to retain the performance crown, particularly in gaming, as Ryzen 7000X3D draws near. The Core i9-13900KS might be pricey and only deliver small gains, but it does set the record as being the first 6 GHz desktop PC chip right out of the box — and that's an incredible milestone. It might even allow Intel to cling to the title of having the fastest gaming chip in the world, a nearly priceless distinction for the 13th-Gen Raptor Lake brand.

Make no mistake: the Special Edition Core i9-13900KS is the fastest desktop PC chip ever built. At least for now. However, AMD's Ryzen 7000X3D is approaching fast. Stay tuned. 

• MORE: AMD vs Intel
• MORE: Zen 4 Ryzen 7000 All We Know
• MORE: Raptor Lake All We Know 

Back when I first reviewed the PlayStation 5, I said that the DualSense controller is "packed with features." With the DualSense Edge, Sony has stuffed it with more. Think of it as the Thanksgiving turkey of controllers: it's too much for most people, but it sure is nice.

The DualSense Edge is Sony's first real take on a pro controller. It has dabbled in attachments to add back buttons, sure, but this adds so much more. The DualSense Edge, adds a pair of mappable back paddles, swappable joysticks with alternate cap options, adjustable triggers, a long braided cable,  a carrying case and more.

That all adds up to a $200 PS5 controller — half the price of a PS5 digital edition and $130 more than a usual DualSense controller.

For most people, this is overkill. But for PlayStation diehards and competitive players who want the best option out there, the DualSense Edge may prove to be worth it, as long as you remember to charge the battery regularly. 

Design and Comfort of the DualSense Edge 

The DualSense Edge, when it launched, seemed heavy at 280 grams. The DualSense Edge is heavier, weighing in at 322 grams on our scale as set up out of the case (Sony didn't respond to a request for an official spec sheet). It feels like a luxury item in the hands. 

At first glance, it looks very much like the existing DualSense, but it has a black trackpad and buttons and a shiny, rather than matte, cover around the joysticks. I personally prefer the matte look of the original controller, because this one attracts fingerprints. But I don't think that will bother most people. If you like the way the existing DualSense feels in your hands, the DualSense Edge isn't making any massive changes, and largely feels better; I'll take that trade.

Of course, there are also a few extras. The two function buttons jut out of the bottom, which let you change control scheme profiles mid-game, or to make adjustments to chat volume. The grips on the inside of the controller's grips are rubberized, which adds tremendously to the comfort.

The controller comes in its hard white carrying case, which includes a passthrough port for charging, as well as all of the accessories: two high-dome and two low-dome stick caps (along with the traditional caps that come pre-attached to the controller); four back buttons (two can be used at a time), including two level-style and two half domes; a long braided USB cable; and a special housing to lock the cable to the controller for those who prefer to play in wired mode. There's also a QR code inside the case that serves as a digital instruction manual

One neat design feature is that you can replace the joysticks without any tools. A small release latch on the front of the controller lets you pop the cover off. There are further levers for each joystick module, and then you can simply swap one out and place another in. Sony is selling replacement modules for $19.99 per stick. It's a tacit admission that, yes, joystick drift can be a problem on Sony's controllers. Part of me wishes that Sony changed the design altogether for this controller. But since it didn't, giving the people who are mostly likely to push their controllers hard the ability to swap out the sticks without junking an entire controller seems like a solid repairability compromise.

My first instinct when I opened the DualSense Edge was to attach a tall dome on the right stick, which is great for aiming in competitive shooters, while leaving the standard cap on the left stick. I initially went for the level-style back buttons, but found that the half domes came up perfectly to my middle fingers for easy pressing. 

I have dabbled with these levers throughout my career and often found them to be in the way, so this felt like a great compromise. Of course, I could always switch (and would be especially likely to switch back to a standard joystick layout for other types of games, like adventure or platformers).  There are fewer back buttons than what you get on the Xbox Elite Controller Series 2, which has four, letting you completely replace the face buttons if you so choose.

The shape of the DualSense Edge is so similar to the DualSense that it can fit in the same Playstation-branded charging cradle and connect to the same headsets with a 3.5 mm jack, so don't worry about any other accessories that you own.

DualSense Edge Specifications

DualSense Edge Setup and Functionality

When you first connect your DualSense Edge, the system will prompt you to take a "tour" of the controller, going over its functions. After that, you can start remapping buttons.

Personally, I’m pretty used to the standard PlayStation layout, but this is the important place to change your back paddles. You can have multiple profiles, and then switch between them with the function menu, even mid-game.

You can also make the same changes you make to the standard controller, including adjusting trigger effect intensity, vibration levels and the brightness of the controller’s built-in lights.

Gaming on the DualSense Edge

I felt like I was better at games with the DualSense Edge. I wasn't, but I felt like it. Maybe $200 buys you confidence.
In general, it feels like you're playing PlayStation. But I set the back buttons to key functions, which at least gave me the ideation that I was improving. In Fortnite, I had one set to jump and another set to the build menu, which greatly reduced the amount of time I spent with my hands off of the sticks (of which I used the high dome on the right side for better aiming). For some, this is where the lack of two more back buttons. When I wanted to switch from a weapon to my pickaxe, I still was reaching up to the triangle button.

In Rocket League, I had one trigger set to jump and another set to boost (I prefer the normal sticks for this game but you do you). In that case, the two buttons were enough for me.

My favorite games are narrative driven. In Spider-Man: Miles Morales, I'm pretty used to the standard controller setup. I did try setting the back buttons to different combat maneuvers, and it worked, but I didn't feel that I was getting the same advantage that I did in competitive play. Spider-Man uses the adaptive triggers, and I found that unless I kept the controller at the default trigger setting, the adaptive functionality went away.

Gaming on PC has some intricacies. When I connected this PS5 controller to a PC, Windows recognized the controller as a DualSense Edge, but Steam saw it as a normal PlayStation 5 controller. But when I dug into Doom Eternal, I was surprised to see that the Edge's back buttons were assigned to the last profile I had used on PS5, which let me jump and rush through enemies with ease.

It would be great if Sony could build an app for the Edge, but it's also possible that developers for games (or clients like Steam) could do that and allow for customizations without going to a PC first. As Sony starts putting out more of its games on PC, it should make sure its flagship controller works as well as possible on the platform.

DualSense Edge Battery Life

The bad news here is that the DualSense Edge has shorter battery life than the standard controller. Back in December, Sony told the Verge:

"The DualSense Edge wireless controller’s operating time is moderately shorter than the original DualSense wireless controller because we’ve included many more features within the same form factor and ergonomic design as the original DualSense controller. We wanted to strike a good balance between wireless operating time and delivering robust, high-performance features. Additionally, the longer USB braided cable is also great for competitive players who prefer playing with a wired connection to avoid wireless interference – this option preserves battery life."

In my testing, which included a mix of Fortnite, Rocket League, Spider-Man: Miles Morales and Doom Eternal on the PC side, I got about 5 hours out of the DualSense Edge. You can likely squeeze more of a charge if you change some settings. I used the out of the box defaults, leaving rumble and haptics on, as well as the controller light set to bright. Turning the light down and turning off the feedback may prolong your time between charges.

In general, this isn't great. My DualSense from launch has already lost some capacity, so we'll have to see how this holds up over time. At least the people who care the most can still fit the Edge in a charging cradle to keep it topped up, and there's a convenient velcro charging passthrough at the top of the case.

Bottom Line

It feels great to play on the DualSense Edge. Diehard PlayStation gamers who love competitive games and are wearing out the joysticks on their existing controllers will get some value out of the Edge, as long as they're willing to charge it slightly more often than the standard controller.

For most people, it's hard to recommend a controller that costs more than twice what a regular DualSense commands. You're halfway to another PlayStation. It's not uncommon for pro controllers, though. Microsoft charges $179 for its pro controller.

If you're playing exclusively on PlayStation, this is probably the best pro controller to get. Those who switch between PlayStation and PC will still be able to use it, but not get all of the features on PC at the moment. If you're PC only, an Xbox Elite Series 2 may be better for you, as most games already have Xbox controller support on Windows.

But if you've been waiting for years for an official pro controller for PlayStation and want a luxury feel at the expense of a bit of battery life (and a lot of cash), there's finally a solid gamepad for you. 

The market for gaming mice has blown up in recent years due to competitive gaming and support on the newest consoles. Almost every gaming peripheral company features a mouse with a honeycomb design to make it as light as possible. That's what we're seeing with the Monoprice Dark Matter Hyper-K Wireless. 

The Monoprice Dark Matter Hyper-K Wireless is a wireless gaming mouse with a lightweight honeycomb design, and it’s fairly affordable with a retail price of $69.99. But even at this low price point, it’s disappointing; its uninspired design just can’t compete with other similarly-priced lightweight gaming mice.

Specs

Design and Comfort of the Dark Matter Hyper-K Wireless 

The Monoprice Dark Matter Hyper-K Wireless continues the ongoing trend of lightweight wireless mice with honeycomb designs and RGB lighting. The honeycomb design has become a cornerstone of lightweight gaming mice, because it provides structure while reducing weight. But I hate it. I have trypophobia, which is a fear of clusters or patterns of small holes — sea coral gives me the chills, for example. So does... honeycomb.

There are a lot of honeycomb mice on the market — the most notable is probably the Glorious Model O Wireless, which retails for around the same price ($79.99) and has software configuration options, unlike the Hyper-K. The Hyper-K Wireless’ DPI cannot be adjusted in any software, so the only options you have are the six baked-in presets that come with the mouse (the Hyper-K Wireless’ maximum DPI is also 16,000 — not up to par with other sensors on the market).

The mouse has six buttons, plus a clickable scroll wheel: right/left mouse buttons, two thumb buttons, one button behind the scroll wheel, and one button next to the sensor on the bottom of the mouse. Because the Hyper-K Wireless has no software support, none of these buttons are programmable. The button behind the scroll wheel is not, as you might expect, for controlling RGB or switching DPI — it’s for the polling rate, which is ridiculous. The mouse has a maximum polling rate of 1,000 Hz, which is pretty normal for a gaming mouse.

The two thumb buttons are set to go forward/backward and the button on the bottom of the mouse is the DPI switch. Also on the bottom of the mouse is a spot to store the 2.4 GHz wireless dongle, which is always an appreciated feature.

The lack of software support means the mouse’s RGB is non-configurable, but it can be toggled on and off with multiple positional options on the power switch. (The top position has both the mouse and RGB on. The middle position has the mouse on but RGB off. The bottom position shuts the mouse off entirely.)

Performance of the Dark Matter Hyper-K Wireless

I’ve been seeing a lot of people post videos on TikTok showing off aim trainer scores, so I decided to try that with the Dark Matter Hyper-K Wireless. After a couple of warm-up rounds with 3D Aim Trainer, I scored 51% accuracy — which could be improved if the Hyper-K skates were of higher quality. Decent PTFE skates can be found on very cheap mice, such as the $30 Cooler Master MM71, so there’s no excuse for Monoprice not to include them here.

The Hyper-K Wireless lacks the bells and whistles of other gaming mice, but it does get the job done. It’s not, however, the highest-quality mouse within its price range. If you use a low DPI, you can hear the brittleness of the internals when you lift it, which worried me when thinking about its long-term durability. That said, it tracked well and clicks were accurate — I didn’t experience any double-clicking issues or drops in wireless connection in my testing.

Battery Life of the Dark Matter Hyper-K Wireless

Like the rest of the mouse, the Dark Matter Hyper-K’s battery life is subpar. Monoprice rates the Hyper-K’s 300mAh battery at 50 hours with RGB disabled — 50 hours isn’t exactly impressive, and that’s with RGB off. Monoprice doesn’t mention whether the 50-hour estimate is over Bluetooth or 2.4 GHz wireless.

Also, unlike most wireless gaming mice, this mouse does not have quick-charging. Instead, you’ll need to wait three hours for this mouse to charge up.

I’m not impressed by the Hyper-K Wireless’ battery life, because there are more reputable mice on the market that have five times the battery life and cost half the price, such as the Logitech G305 Lightspeed (though that doesn't have a rechargeable battery).

Bottom Line

The Monoprice Dark Matter Hyper-K Wireless is one of the least recommendable products I have ever reviewed. For $69.99, I want to see PTFE skates, customizable DPI, and maybe even some creativity. I’ve seen other generic-brand honeycomb mice on Amazon for $17 that have longer battery life and more impressive RGB.

If you’re looking for a wireless honeycomb mouse, I recommend the Glorious Gaming Model O wireless, which I own. If you don’t need something wireless, the Cooler Master MM720 and the Glorious Gaming Model O Minus are two of our favorites.

Apple has quietly refreshed its 14 and 16-inch MacBook Pros with new M2 Pro and M2 Max processors, promising longer battery life and, on the M2 Max versions, up to 96GB of RAM. The new laptops are available for pre-order today, and will launch on January 24.

The 14-inch MacBook Pro will start at $1,999, while the 16-incher will begin at $2,499. Those prices are with the M2 Pro, and prices will scale as you go up to the M2 Max and add more RAM and storage.

Apple didn't announce the laptops in a planned event, but rather subtly dropped a new video and a press release.

While the chassis isn't changing at all, there are some serious changes beneath the hood. The M2 Pro model will start at a 10-core CPU and 16-core GPU (it will go up to 12-core CPU and 19-core GPU), 16GB RAM and 512GB SSD storage. The M2 Max laptops will start with a 12-core CPU and 30-core GPU (it goes up to a 38-core GPU), 32GB of memory and 1TB of storage.

The M2 Pro models have 200 GB/s memory bandwidth, twice of what the normal M2 allows. Apple claims The MacBook Pro with M2 Pro is 20% faster than the previous generation device at rendering in Motion, 25% faster in compiling in XCode and up to 40% in processing images in Photoshop.

Meanwhile, the M1 Max gets 400 GB/s of unified memory, doubling what M2 Pro offers, and goes up to 96GB of unified RAM, which Apple says makes it a graphics powerhouse. Apple claims color grading in DaVinci Resolve is 30% faster than last generation, and that rendering effects in Cinema 4D is as much as 30% faster.

Apple has moved wireless from 6 to 6E, and its "more advanced" HDMI port can support one display at 8K (at 60 Hz) or 4K (at 240 Hz). The laptops can support up to four external displays (one through each Thunderbolt 4 port and HDMI).

Among the parts that aren't changing are the Liquid Retina XDR display, 1080p FaceTime HD camera, six-speaker audio and MagSafe 3.

In addition to the MacBook Pro laptops, Apple is launching the Mac Mini with M2 and M2 Pro, starting at $599 with M2 and $1,299 with M2 Pro. The latter gets four Thunderbolt 5 ports and can support an 8K display. It's also getting Wi-Fi 6E and Bluetooth 5.3.

Valve is not looking to improve performance of the next iteration of Steam Deck, the designers of the portable console told The Verge. Instead, the company will for now try to improve the battery life and display quality of the device. Once computer architectures evolve significantly, Valve will possibly consider a major upgrade.  

"I think we will opt to keep the one performance level for a little bit longer, and only look at changing the performance level when there is a significant gain to be had," said Pierre-Loup Griffais, a Steam Deck designer. 

Many new games do not really run smoothly on Steam Deck's custom-designed Aerith system-on-chip featuring four Zen 2 cores with SMT at 2.40 to 3.50 GHz and an RDNA 2-based GPU with 512 stream processors operating at 1.0 to 1.60 GHz. As a result, demanding gamers (who tend to buy plenty of games) would prefer a Steam Deck with a higher-performing SoC in order to enjoy their latest titles on the go. But for now, Steam Deck designers Lawrence Yang and Pierre-Loup Griffais are looking forward to improving battery life and display in their next Steam Deck iteration.

While they tend to cost quite a bit more than the Steam Deck, there are competing handhelds that may find more of an audience with demanding gamers. The Aya Neo 2 is based on the AMD Ryzen 7 6800U (eight Zen 3+ cores, RDNA 2-powered Radeon 680M iGPU with 768 SPs), which delivers higher performance, but at the cost of battery life. 

"Right now, the fact that all the Steam Decks can play the same games and that we have one target for users to understand what kind of performance level to expect when you are playing and for developers to understand what to target... there is a lot of value in having that one spec," said Griffais. 

Indeed, there are many reasons why it makes sense for Valve to retain the same level of performance with the next iteration of the Steam Deck. If it didn't, it would need to run two separate Steam Deck Compatibility programs for two consoles with different specifications. Secondly, game developers would have to target two hardware configurations, which means longer time-to-market and higher costs for them. Thirdly, some owners of the first-generation Steam Deck would feel left behind once the second-generation launched. Finally, reducing the costs of the Steam Deck to make the hardware profitable makes more financial sense than increasing performance and not making any money on hardware. And as long as the Steam Deck is continuing to sell well, the company has little reason to make major upgrades.

While PC gamers may not exactly feel great about the lack of Steam Deck upgrades in the foreseeable future, it is a normal practice in the console world to keep hardware specifications stable for years. For example, Nintendo could have introduced its 'Switch Pro' based on a more powerful Nvidia SoC by now, but the only upgrade that the Switch console has seen over its five-year lifespan is a better OLED display. 

And if you're a company that's new to the console market and looking to stay there for a long time, you could certainly do worse than looking to Nintendo for some guidance.