GPU overclocker TrashBench set a world overclocking record on an Intel Arc B580 GPU with a relatively simple setup. Instead of employing exotic solutions like liquid nitrogen, used by other overclockers to hit world records, he instead slapped on a pond pump and filled the custom loop with a 50/50 glycol mix of vehicle antifreeze that had been pre-chilled in a freezer. To his surprise, the GPU hit an incredible -17 degrees Celsius (that’s 1.4 degrees Fahrenheit), allowing him to push the performance of Intel’s top-of-the-line consumer GPU. According to the YouTube video, the car-coolant-chilled graphics card achieved a 3DMark Time Spy score of 16,631 — some 12% higher than a stock GPU, breaking the world record for the B580.

The mods TrashBench made to the GPU were rather simple, with the most complicated part being 3D-printing a custom mount to secure the water block. Trashbench notes that antifreeze remains liquid until roughly -25C, and he chilled the liquid in his freezer, which took it down to -17C. From there, he ran a couple of flexible tubes into an open cooler containing the pre-chilled glycol mix and turned on the pump to circulate it, allowing the graphics card to reach sub-zero temperatures.

TrashBench broke a world record in 3DMark Time Spy when the coolant was still cold.

He also tested the Intel Arc B580 in its stock configuration, with the card hitting 2,850 MHz on air cooling alone. This delivered an average of 54, 158, and 107 FPS for Monster Hunter Wilds, Forza Horizon 5, and Cyberpunk 2077, respectively. But with the GPU running on sub-zero cooling, it achieved 3,316 MHz — some 446 MHz over stock performance or a 16.4% higher clock speed. Unfortunately, because the coolant eventually warmed up during testing, TrashBench wasn’t able to maximize its performance, but even so, the card still hit 69, 174, and 120 FPS in the three titles, giving it an average of 16% more FPS.

TrashBench's results show that with just air OC, the B580 achieved 3200 MHz, with the card hitting 60 FPS on Monster Hunter Wilds, 119 FPS on Cyberpunk 2077, and 172 FPS on Forza Horizon 5 at 1080p High settings. Although extreme cooling still seems to offer more FPS, it wasn't a huge gain over what you get without any of the hardware modifications.

AMD's latest RX 9060 XT GPUs continue the saga of bifurcating mainstream options into different memory configurations for market segmentation. According to AMD's Frank Azor, the 8GB flavor of the RX 9060 XT is designed to cater to the needs of the majority of gamers, who play at 1080p.

Gamers and developers have long bemoaned constrained memory capacities with modern GPUs. Even at 1080p, many demanding AAA titles can overwhelm 8GB GPUs, as confirmed by professional testing with cards like the RTX 5060 Ti 8GB. Despite sufficient raw horsepower, the constrained framebuffer often forces you to compromise on graphical fidelity or even resolution for a playable framerate. Nvidia's RTX 5060 GPUs have been at the epicenter of this criticism, to the point where an older RTX 3060 12GB can outperform them in certain scenarios.

There were rumors that AMD might scrap the 8GB model, following the backlash faced by Nvidia. This proved to be inaccurate, as the product was likely too far in the pipeline to be axed. While the GPUs are still almost two weeks away from launch, enthusiasts have already begun raising concerns about the viability of the 8GB model. AMD's Chief Architect of Gaming Solutions and Gaming Marketing, Frank Azor, stepped in to justify their rationale, stating these GPUs target the "majority of gamers," who play at 1080p.

Intel announced the chipmaker's new Arc Pro B50 and Arc Pro B60 at Computex 2025. Maxsun has stitched two of the latter models to produce the Intel Arc Pro B60 Dual 48G Turbo, a dual-GPU solution that provides 48GB of GDDR6 memory designed to address the most challenging AI workloads.

Dual-GPU graphics cards were once popular, but technological advances have made them obsolete. The last consumer-grade dual-GPU graphics card from Nvidia was likely the GeForce GTX Titan Z from 2014, while AMD's was the Radeon Pro Duo from 2016. This means it's been quite a while since chipmakers introduced a dual-GPU product for retail. Although the Arc Pro B60 may not revive the trend, it remains noteworthy since Intel's first foray into the dual-GPU market. Then again, it's an AI graphics card, not a gaming one.

The Intel Arc Pro B60 Dual 48G Turbo essentially combines two Arc Pro B60 graphics cards on one board. It utilizes dual Battlemage BMG-G21 silicon, each accompanied by its own memory. The same silicon powers the Arc B580 but offers double the memory compared to the gaming counterpart.

The GPU inside the Intel Arc Pro B60 Dual 48G Turbo operates according to Intel's reference specifications. It doesn't feature factory overclocks or anything of that sort. Therefore, we're looking at 20 Xe cores, 20 RT units, 160 XMX and 160 Xe vector engines, and a 2,400 MHz clock speed.

Maxsun Intel Arc Pro B60 Dual 48G Turbo Specifications

The Arc Pro B60 has 24GB of GDDR6 memory clocked at 19 GBps—the 192-bit memory interface results in a memory bandwidth of up to 456 GB/s. The biggest selling point is that the Intel Arc Pro B60 Dual 48G Turbo has 48GB of total memory, 50% more than the GeForce RTX 5090 and matching that of the RTX Pro 5000 Blackwell or the Radeon Pro W7900.

With 48GB on one Intel Arc Pro B60 Dual 48G Turbo, users can have multiple graphics cards in a single system to scale the capacity. For example, Intel had a system with two, totaling 96GB, on display. With the correct motherboard, you can easily have four installed for 192GB, akin to Project Battlematrix.

The Intel Arc Pro B60 Dual 48G Turbo is designed to fit into a standard PCIe 5.0 x16 expansion slot; however, there is a catch. Each Arc Pro B60 interacts with your system independently through a bifurcated PCIe 5.0 x8 interface. Thus, it's important to note that the motherboard must support PCIe bifurcation for the PCIe 5.0 slot hosting the Intel Arc Pro B60 Dual 48G Turbo.

Intel has announced its Arc Pro B-series of graphics cards at Computex 2025 in Taipei, Taiwan, with a heavy focus on AI workstation inference performance boosted by segment-leading amounts of VRAM. The Intel Arc Pro B50, a compact card that's designed for graphics workstations, has 16GB of VRAM and will retail for $299, while the larger Intel Arc Pro B60 for AI inference workstations slots in with a copious 24GB of VRAM. While the B60 is designed for powerful 'Project Battlematrix' AI workstations sold as full systems ranging from $5,000 to $10,000, it will carry a roughly $500 per-unit price tag. 

Intel has focused on leveraging the third-party GPU ecosystem to develop its Arc Pro cards, in contrast to its competitors, who tend to release their own-branded cards for the professional segment. That includes partners like Maxsun, which has developed a dual-GPU card based on the B60 GPU. Other partners include ASRock, Sparkle, GUNNR, Senao, Lanner, and Onix. 

Both the B50 and B60 GPUs are now being sampled to Intel partners, as evidenced by a robust display of partner cards and full systems on display, and will arrive on the market in the third quarter of 2025. Intel will initially launch the cards with a reduced software featureset, but will add support for features like SRIOV, VDI, and manageability software in the fourth quarter of the year. 

The Intel Arc Pro B50 has a compact dual-slot design for slim and small-form-factor graphics workstations. It has a 70W total board power (TBP) rating and does not have external power connectors. The GPU wields 16 Xe cores and 128 XMX engines that deliver up to 170 peak TOPS, all fed by 16GB of VRAM that delivers 224 GB/s of memory bandwidth. The card also sports a PCIe 5.0 x8 interface, which Intel credits with speeding transfers from system memory, ultimately delivering 10 to 20% more performance in some scenarios. 

The B50's 16GB of memory outweighs its primary competitors in this segment, which typically come armed with 6 or 8GB of memory. The card also has certified drivers that Intel claims deliver up to 2.6X more performance than the baseline gaming drivers. 

Intel shared a slew of benchmarks against the competing Nvidia RTX A1000 8GB and the previous-gen A50 6GB, but as with all vendor-provided benchmarks, take them with a grain of salt (we included the test notes at the end of the article). In graphics workloads, Intel claims up to a 3.4X advantage over its previous-gen A50, and solid gains across the board against the RTX A1000. It sports similar advantages in a spate of AI inference benchmarks. 

The Intel Arc Pro B60 has 20 Xe cores and 160 XMX engines fed by 24GB of memory that delivers 456 GB/s of bandwidth. The card delivers 197 peak TOPS and fits into a 120 to 200W TBP envelope. This card also comes with a PCIe 5.0 x8 interface.

Intel supports multiple B60 GPUs on a single board, as evidenced by Maxsun's GPU,  with software support in Linux for splitting workloads across both GPUs (each GPU interfaces with the host on its own bifurcated PCIe 5.0 x8 connection). 

Intel's benchmarks again highlighted the advantages of the B60's 24GB of memory vs the competing RTX 200 Ada 16GB and RTX 5060Ti 16GB GPUs, claiming this can impart gains of up to 2.7X over the competition in various AI models. Intel also highlighted the advantages of higher memory capacity in model size, context, and concurrency scaling.

The Intel Arc Pro B60 will primarily come in pre-built inference workstations ranging from $5,000 to $10,000, dubbed Project Battlematrix. The goal is to combine hardware and software to create one cohesive workstation solution. However, the per-unit cost will be in the range of $500 per GPU, depending on the specific model.

Project Battlematrix workstations, powered by Xeon processors, will come with up to eight GPUs, 192GB of total VRAM, and support up to 70B+ parameter models. 

Intel is working to deliver a validated full-stack containerized Linux solution that includes everything needed to deploy a system, including drivers, libraries, tools, and frameworks, that's all performance optimized, allowing customers to hit the ground running with a simple install process. Intel will roll out the new containers in phases as its initiative matures. 

Intel also shared a roadmap of the coming major milestones. The company is currently in the enablement phase, with ISV certification and the first container deployments coming in Q3, eventually progressing to SRIOV, VDI, and manageability software deployment in Q4. 

Intel's partners had multiple Project Battlematrix systems up and running live workloads in the showroom, highlighting that development is already well underway. 

One demo included a system running the full 675B parameter Deepseek model entirely on a single eight-GPU system, with 256 experts running on the CPU and the most frequently used experts running on the GPU. 

Other demos included running and finding bugs in code, an open enterprise platform for building RAGs quickly, and a RAG orchestration demo, among others. 

As noted above, the Intel Arc Pro B50 and Intel Arc Pro B60 will arrive on the market in the third quarter of 2025.

While Intel is gearing up to reveal new Battlemage products at Computex in a few days, X user Haze has shared an image of an Intel ARCade machine featuring a last-generation Alchemist A7 series GPU. Often seen as Intel marketing material for major e-sports events, this machine was reportedly found out of commission and unused at an unnamed Intel campus.

Intel was enthusiastic in the months leading up to the Alchemist launch, as evident in its marketing push for the product. This ranged from custom 60-foot air-conditioned gaming trucks to smaller Arcade machines, like the one we're seeing today. Following the delays associated with Alchemist and inevitable teething problems, Intel has maintained a relatively low profile with Arc since then. The desktop Battlemage launch is proof of this.

The ARCade is an Intel-powered arcade-style machine that has been a recurring presence at events like DreamHack. The controller layout depicts a two-player configuration for fighting games like Street Fighter and Tekken. A quick look inside reveals a GPU at its core, which carries a strong visual resemblance to Intel's Limited Edition models from the Alchemist range, likely the A770. The GPU is presumably housed in an Intel NUC 12 Extreme, but any guess is as good as ours.

An AIB is reportedly preparing a dual-GPU variant of Intel's rumored Arc B580 24GB, totalling 48GB of VRAM on a single board, via VideoCardz. According to the source, this model is slated to be revealed at Computex 2025, which is just days away at this point. Specific details like the AIB, interconnect technology, and price are in the dark, but we can expect more details at the trade show, if the rumors are true.

This is the third leak that references a 24GB edition of the Arc B580, following EEC filings from Maxsun and an earlier slip-up from Sparkle. It's quite surprising that a GPU that's supposed to rival the RTX 4060 is getting a 24GB memory configuration. Many PCs don't even possess that much system memory. These cards are intended as a cost-effective solution for AI/ML developers, where similar capacity cards from AMD and Nvidia carry a steeper price tag. The most affordable Blackwell GPU with 24GB of memory is the RTX Pro 4000, costing over $1,500 based on preliminary listings.

An AIB is apparently doubling down on this approach by building a dual-GPU solution, housing two of these rumored 24GB B580 GPUs on a single PCB. Keep in mind, this product is reportedly a one-time design from the AIB, not a standard reference model from Intel.

Traditionally, multi-GPU setups nowadays, like Nvidia's B200 and Apple's M1 Ultra, rely on their own advanced interconnect solutions like NVLink and UltraFusion. While Intel does have Xe Link, it likely isn't compatible with the B580 and would be too costly for a one-off project. The most probable contender is a PCIe bridge linking the GPU's interfaces, allowing them to communicate through one slot.

Malaysia has reportedly recorded an unprecedented surge in GPU imports as per data shared by Taiwan’s International Trade Administration, highlighted in a post by X user @kakashiii111. Despite requests from the U.S. government for Malaysia to strengthen its monitoring of high-tech exports to China, the month of April saw imports of $2.74 billion in GPUs, a monumental 3,400% jump from 2023.

GPU imports in Malaysia have notably witnessed a sharp increase throughout the year. If we look at historical data, imports in the month of January 2025 totaled $1.12 billion, marking a 700% increase year-over-year. February followed with $627 million in shipments, which is slightly lower but still substantial. March saw a dramatic spike to $1.96 billion, a similar 3,400% increase as April, compared to the same month in 2023. Essentially, the country has roughly imported $6.45 billion worth of GPUs in just the first four months of 2025, surpassing the entire sales for 2024.

This potentially raises some concerns over the fact that certain buyers are rerouting Nvidia-branded GPUs to the Chinese market through Malaysia despite strict norms placed by the U.S. Last month, we reported a similar trend, with customs data showing Taiwan’s exports of computing systems to Malaysia skyrocketing in March to $1.87 billion which is a 366% year-over-year increase and an astonishing 55,117% surge compared to March 2023.

The massive surge in PC components, including GPUs and AI-accelerators, particularly from Nvidia, aligns with the U.S. government’s tighter restrictions on AI and HPC GPU shipments to China. This has naturally led to speculations that Malaysia is either stockpiling hardware for its own cloud AI ambitions or acting as a hub for Chinese buyers trying to bypass the recent sanctions. Notably, the AI Diffusion Rule is set to take effect on May 15, which could make Malaysia a backdoor for restricted tech despite recent efforts to crack down on smuggling rings.

As pointed out by Taiwanese media outlet TechNews, Nvidia may not reveal the actual volume of GPU shipments to Malaysia. This is due to a new reporting method that logs revenue based on the customer’s billing location, rather than the physical destination of the goods. This lack of disclosure does raise questions about transparency and could invite further scrutiny from U.S. regulators, especially amid tightening export controls to China and other Southeast Asian countries.

Just days before Computex, the parent company of Maxsun has registered new Intel Arc B580 models with 24GB of VRAM with the EEC, via Olrak at X, further stoking existing rumors of a high-VRAM Battlemage card. Remember that EEC (Eurasian Economic Commission) submissions aren't definitive confirmations of any product. Manufacturers often register placeholder configurations to cover all future possibilities, many of which never end up seeing the light of day. Therefore, you should treat this leak with a healthy dose of skepticism.

Intel launched its Arc B580 and Arc B570 GPUs in December and January, respectively, bringing its Battlemage architecture to the desktop market just months after Lunar Lake debuted on mobile. It wouldn't be an exaggeration to say these GPUs initially caused a stir in the market, offering an ample 12GB framebuffer, a capacity unheard of in the $250 GPU market. The cheapest B580s on Newegg and Best Buy are now in the $300 territory, which is a shame. That's a trend most new GPUs are facing, including AMD's RX 9000 GPUs and Nvidia's RTX 50 series.

We've been hearing rumors of a 24GB Battlemage Professional GPU since December. Recently, Sparkle accidentally alluded to the existence of an Arc B580 24GB, later retracting their comments, likely due to NDA concerns. Maxsun's parent company has registered several GPUs with the EEC, including the MAXSUN Intel Arc B580 iCraft 24G and its OC variant, matching the exact configuration Sparkle mentioned.

Intel is gearing up to launch new Arc professional GPUs, with a recent teaser strongly suggesting these will be based on the firm's latest Battlemage (Xe2) architecture. Set to debut at Computex, these GPUs should, at minimum, offer a slight increase in memory capacities. Specifications and other technical details remain unknown, but we can expect to learn more from Intel at Taipei.

Intel's Battlemage family currently employs one die, BMG-G21, powering both the Arc B570 and Arc B580 GPUs. Rumors have alluded to several other configurations, with the most persistent being a larger BMG-G31 die, which has surfaced numerous times in shipping manifests. The majority of these shipments are destined for Vietnam, which is home to several OSAT companies that assemble, package, and test silicon. Export data alone isn't enough to confirm whether a BMG-G31-based SKU is still being developed, so we're treating this with significant skepticism.

Rumors of Battlemage-based professional cards with 24GB of VRAM have been afloat since last December. Likewise, a recent slip-up from an OEM indicated that Intel had a 24GB B580 model in the pipeline. If we put two and two together, at least one GPU from the Battlemage professional lineup is expected to employ the BMG-G21 die with twelve 16Gb GDDR6 modules arranged in a clamshell configuration. Other specifications should remain largely similar to the Arc B580.

Intel has announced that its AI-powered upscaling tech XeSS (Xe Super Sampling) is now available in over 200 games, while the newer XeSS 2 has been added to 10 more titles. As promised, the new Assassin’s Creed: Shadows and Civilization VII are getting support for XeSS 2 alongside popular titles including Naraka Bladepoint, Like a Dragon: Pirate Yakuza in Hawaii, and Delta Force: Black Hawk Down.

Notably, Intel lists a total of 13 XeSS 2-enabled games on its website, however, according to PC Gaming Wiki, it is supported in over 20 titles. Introduced alongside the Arc B580 and B570 GPUs late last year, XeSS 2 offered support for only a handful of games up until today. Leveraging the Battlemage architecture, XeSS 2 combines upscaling, frame generation, and latency reduction to compete with rival technologies like Nvidia’s DLSS and AMD’s FSR. As of now XeSS 2 is fully compatible with only Intel’s Arc A-series and B-series discrete GPUs and integrated graphics found in newer Intel Core Ultra processors.

In its announcement, Intel also shared some performance data using its Arc B580 GPU to showcase the capabilities of XeSS 2. According to the company, Diablo IV saw up to a 4x increase in frame rates, jumping from 46 FPS at 1440p native resolution to 186 FPS using XeSS 2. Similarly, games like Assassin’s Creed: Shadows, Delta Force: Black Hawk Down, Hogwarts Legacy, and F1 24 witnessed over 2x improvements in performance.

The tech industry is facing some headwinds. Manufacturers, sellers, and consumers are more concerned than ever about prices and stock as a global trade war upends the supply chain.

But, even as we face these challenges, many companies continue to innovate in ways that surprise and delight us. It’s time to celebrate those who move the ball forward, even in tough times.

That’s why, as we do at this time of year, we present to you the 2025 Tom’s Hardware Innovation Awards: a set of products that set or expand the standard for others. This year’s list includes laptops with rollable screens, the fastest-ever consumer GPU, and a 3D printer that doubles as a laser cutter, among many others.

Nvidia RTX 5090

The Nvidia Blackwell RTX 50-series launch has been both incredibly exciting and horribly disappointing, and nowhere is that more apparent than with the halo GeForce RTX 5090. It comes with all the Blackwell architectural enhancements, which admittedly can feel more than a bit overhyped — looking specifically at Multi Frame Generation (MFG). But the GB202 chip at the heart of the 5090 offers clear generational improvements.

It comes packing 170 Streaming Multiprocessors (SMs), a 33% increase from the prior generation RTX 4090's 128 SMs. It also comes with a 512-bit memory interface and 32GB of VRAM, another 33% increase. Except that memory is now 28 Gbps GDDR7 instead of 21 Gbps GDDR6X memory, which means the net improvement in memory bandwidth is an impressive 78%. To help the GPU reach its full performance potential, the power limit also got a big 28% bump to 575W.

That's all good for performance, but the 5090 isn't without controversy. Deja vu: All that power, through the 16-pin 12V-2x6 connector, has again resulted in some melting adapters. As frustrating and concerning as that might be, the real problem has been a lack of availability and skyrocketing retail prices. Ostensibly carrying a $1,999 starting MSRP, retail cards have routinely been selling for closer to twice that much. Still, it's the dream GPU that gamers would love to own, assuming they could afford it and find it in stock.

—Jarred Walton

AMD Ryzen 9 9950X3D

AMD’s 16-core 32-thread Ryzen 9 9950X3D is designed to provide no-compromise leadership performance in both gaming and productivity application performance, and it delivers. Courtesy of its game-boosting 3D V-Cache technology, the 9950X3D ties the best CPU for gaming, the Ryzen 7 9800X3D, in 1080p gaming performance, and even beats Intel’s flagship Core Ultra 9 285K by an incredible 37%, planting AMD’s flag as the uncontested leader in gaming.

Perhaps most impressively, the 9950X3D also preserves the full performance of the standard models’ performance in productivity applications, a first for the company’s 3D V-Cache models. That allows it to outperform the competing Intel chips in a multitude of workloads.

The Ryzen 9 9950X3D costs a hefty $699, but it justifies the price tag with the performance served up from the 16 Zen 5-powered cores paired with the voluminous 128MB of L3 cache. The chip is also comparatively easy to cool, borne of its reasonable 170W TDP rating. The CPU cores also boost to 5.7 GHz, a surprising peak for this core-heavy model. AMD also enabled its full suite of overclocking bells and whistles for the 9950X3D, a first for its 16-core 3D V-Cache models, giving enthusiasts a plethora of options to tweak performance.

If you’re focused specifically on gaming, the lower-tier $480 Ryzen 7 9800X3D ramians the best option for most users. However, if you’re looking for fantastic performance in heavily-threaded applications and multitasking without compromising on gaming performance, the Ryzen 9 9950X3D is the best option on the market.

Read: AMD Ryzen 9 9950X3D review

—Paul Alcorn

Micron 4600 SSD / Silicon Motion SM2508

Two years is a long time to sit at the top, but that's precisely what Phison's E26 controller that powers some of the best SSDs has done. Now, thanks to drives like the Micron 4600, it finally faces some competition courtesy of the Silicon Motion SM2508 controller (which we previewed last September). It delivers equal to or better performance than the incumbent Crucial T705 in nearly every test we ran. Most critically, it also trims down the average and maximum power consumption by several watts.

The key advantage, for the time being, is that the SM2508 uses TSMC's N7 process node, while Phison's E26 uses the older TSMC 12nm FinFET node. Phison is readying its E28 controller built on TSMC N6/N7, but it's not out quite yet — the company appears to have prioritized its lower power DRAM-less E31T controller.

Competition benefits consumers, and the SM2508 brings a healthy rivalry back to the top-tier SSD realm. The biggest issue right now: There are no retail SM2508 drives available, as the Micron 4600 chiefly targets OEMs and system integrators. That means you can't actually buy the 4600 right now, as far as we're aware.

—Jarred Walton

AMD Radeon RX 9070 XT and RDNA 4

Nvidia may control the lion's share of the GPU market, but the AMD RDNA 4 architecture and Radeon RX 9070 XT promised to bring some real competition to the upper-mainstream market segment. On paper, everything looks good, and our testing definitely showed real potential for AMD to recapture some lost ground. The problem, as with Nvidia's GPUs, has been retail availability and pricing.

The RX 9070 XT carries an official $599 starting MSRP, but such prices haven't been seen in the U.S. Even now, nearly two months after the initial launch, the lowest prices we've seen tend to be in the $850–$900 range — and those often go out of stock. It's not really worse than what we've seen from the competing RTX 5070 Ti, but neither is it clearly better.

AMD finally brought some real improvements to the Navi 48 GPU, with greatly improved ray tracing and AI performance. Nvidia still holds onto a raw performance and support lead in those areas, but FSR 4 looks promising and native RT performance ends up being competitive. Now we just need to see cards selling for closer to $599, rather than the 40% or more 'markups' that have plagued recent GPU launches from all the involved parties.

—Jarred Walton

Lenovo ThinkBook Plus Gen 6

When you’re at home or at the office, you can connect your laptop to one, two or even four external monitors for more screen real estate. But when you’re on the road, you either need to lug one of the best portable monitors or live with a single screen. Lenovo’s ThinkBook Plus Gen 6 offers a different way to get more pixels of workspace.

The ThinkBook Plus Gen 6 features a rollable screen that, with the press of a button, uses a motorized interface to grow from 14 to 16.7 inches, adding a significant amount of additional real estate along the way. You can use those pixels for an additional window or two, but Lenovo also has a utility to help you take advantage of them.

ThinkBook Workspace, an app which comes preloaded, lets you pin widgets such as a to-do list to the bottom of your screen, where that extra space lies. It also allows you to create a virtual display (a window that the OS treats as a separate monitor) or view Smart Copy – a history of your clipboard. You can even put any individual app you want – the browser for example – into a tab in Workspace so it stays pinned to the bottom of your screen while you perform other work at the top.

Read: Lenovo ThinkBook Plus Gen 6 Hands-On

– Avram Piltch

Intel Battlemage and Arc B580

Intel has fallen on hard times, with massive layoffs and underwhelming new CPUs. It's not down and out by any means, but it faces an uphill battle. GPUs have taken over the data centers, powering huge AI clusters, and Intel has finally entered the dedicated GPU market with its Arc graphics cards. Too little, too late? Perhaps, but the second generation Intel Arc Battlemage takes a step in the right direction.

Headlined by the Intel Arc B580, it kicked off the current generation of new GPU architectures in late 2024. Oh, how naive we now look, praising the $249 starting MSRP — that was before we realized just how bad things could get, though there were definitely tremors already. Nearly five months later, the lowest price we can find is the $329 Onix Lumi Arc B580 (that's made by a sibling company to Sapphire, if you're wondering).

Generationally, the Arc B580 delivers an impressive 20% average performance improvement over the Arc A770 16GB, while using about 40W less power on average. That shows real progress, and when coupled with an attractive price, it proved almost too good to be true. It's still faster and costs less than Nvidia's prior-gen RTX 4060, thanks to having 12GB of VRAM. Let's hope Intel can improve the supply and continue to compete in the GPU realm.

—Jarred Walton

Asus ROG Ally X

Perhaps the best way to innovate is to straight up improve. The Asus ROG Ally X took everything its predecessor, the ROG Ally, did, and largely did it better.

While the Ally X uses the same AMD Ryzen Z1 Extreme chip and 7-inch display, it has more, faster RAM to improve graphics performance, adds higher power profiles, a larger battery, and better build quality. All of these in combination make it the best PC gaming handheld that runs Windows out of the box.

The RAM is the biggest deal. Because the Z1 Extreme uses the same RAM for both the system and graphics, it lets the integrated GPU shine, and beat out the regular Ally in all of our gaming tests.

There was no successor to the Z1 Extreme when Asus released the Ally X. It could've let the original continue to sell unchanged. But the Ally X, if you can afford its $799.99 price point, feels so much better in every way.

Read: Asus ROG Ally X review

— Andrew E. Freedman

Mac Mini

Wow, is the Mac Mini tiny. The latest redesign is just 5 x 5 inches wide and deep and a petite 2 inches tall, taking up very little space on a desk. And starting at $599, with an M4, 16GB of RAM and 256GB of storage, this is the first value Mac we've seen in a long time.

Shrinking down the Mac Mini didn't come without issues. The power button is on the bottom of the desktop, which is irksome to some, while others say they never shut their Macs off anyway.

There are some smaller Windows PCs out there, but they're often more expensive and come without RAM, an SSD, or an operating system.

Those looking for a family machine that don't need a laptop have a competitive choice in the Mac Mini, which is small, quiet, and stays out of the way so you can focus on your work.

Read: Mac Mini (M4 Pro) tested: Tiny titan

— Andrew E. Freedman

Tryx Panorama 360 AIO cooler

How do you innovate as an AIO cooler company when the latest generation of CPUs just aren’t as hard to keep cool as previous-gen flagships like the 14900K? If you’re Tryx, you borrow a page from high-end phone makers and slap a massive 6.5-inch curved AMOLED screen on your cooler to make sure it stands out, while also delivering excellent cooling performance (thanks to Asetek’s latest Gen8 pump). Not bad for a company I’d never heard of before seeing this cooler at Computex 2024.

While the screen on your AIO might not be your first priority, our reviewer was quite impressed, noting that the curved screen and absolute black levels created a “striking anamorphic 3D effect” using the pre-loaded visuals. You can of course add your own media using the company’s Kanali software.

A VRM fan is included to help chill components around the CPU, and when we tested it last September, the Tryx Panorama was in most ways the best-performing cooler we’d tested to date. But of course, class-leading performance and a curved screen don’t come cheap. The Tryx Panorama sells for around $350.

Read: Tryx Panorama 360 ARGB review

— Matt Safford

Be Quiet Light Base 900

At a Computex 2024 that was overflowing with multi-sided glass cases, Be Quiet’s Light Base stood out for its tasteful, almost retro-style RGB light strip, and for its ability to be oriented three different ways, horizontally, or vertically with the motherboard on the left or right edge. This is accomplished by some clever internal design and removable feet.

When we got it in for testing, we also appreciated its pre-installed ARGB and PWM hubs, as well as its support for back-connector motherboards. It’s solidly built, as we’ve come to expect from Be Quiet. Its thermal performance wasn’t the best, but we found it to be good enough for the vast majority of PC builders, especially gamers and general-purpose users.

Read: Be Quiet Light Base 900 DX review

— Matt Safford

Lian Li EDGE EG1000 ATX 3.1 power supply

Between rear-connector motherboards, the rise in popularity of dual-chamber setups, and the ever-increasing demands of modern GPUs, PC is still continuing to evolve — which is why we're glad to see products like the Lian Li Edge ATX 3.1 PSU evolving along with it. Designed for dual-chamber cases, its L-shaped, fully modular design gives you much easier access to key (and often tough to connect or disconnect) power cables.

The Edge also has a built-in USB header hub so you can connect more internal accessories, and its internals are well designed, as we saw in testing. It's also quiet until you tax it with a heavy load, and efficient. It won't work well for every case, but for dual-chamber designs where the PSU is mounted vertically, it can make building, cable routing, and maintenance much easier.

Read: Lian Li EDGE EG1000 ATX 3.1 power supply review

— Matt Safford

Corsair EX400U external SSD

After years of limited external storage advances, 2025 has brought us both the first Thunderbolt 5 SSD to land on our test bench, as well as the first truly great USB4 drive we’ve tested, Corsair’s EX400U. And while TB5 drives like LaCie Rugged SSD Pro5 can be faster, unless you own a recent Mac, TB5 ports are exceedingly rare, while USB4 is much more common. That, plus the fact that Corsair’s EX400U is small and designed to magnetically attach to the back of your MagSafe or Qi2-ready phone, makes the Corsair drive feel a lot more innovative.

The EX400U also blasted past every other USB-based SSD in our testing, and sometimes even beat the technically faster Thunderbolt 5 drive in Windows over USB4. In our Crystal DiskMark sequential testing, it edged above 4,000 MB/s reads, and topped 3,700 MB/s writes, while the best USB 3.2 2x2 drives were roughly half as fast. We’re sure to see faster USB4 SSDs before too long, but Corsair’s drive is the first we’ve tested that’s easy to recommend, especially if you want a fast drive that’s designed to connect to your phone.

Read: Corsair EX400U review

— Matt Safford

Mechboards Hyper7 R4 Keyboard

In a world where keyboards keep shrinking their numbers of keys, the Mechboards Hyper7 R4 is a real standout. Loosely based on a 1970s keyboard called the Space Cadet, the Hyper7 has 178 keys and weighs a whopping 5 kg (11 pounds). That makes it three or four times heavier than a typical mechanical keyboard and gives it at least dozens of more keys to play with.

We were fortunate enough to get a late prototype of the Hyper7 R4, of which only about 500 units were made. The unique layout has six blocks of keys, many of which have very unique labels such as “SMOL,” “Hold Input,” ”Suspend,” and “Fun.” A lot of these match commands that no longer matter much in 2025, but the good news is that you can use VIA, a popular keyboard mapping software, to remap the keys into any macro you choose.

The symmetrical shape of the Hyper7 R4 takes some getting used to. Normally, keyboards have their spacebars offset a little to the left, not directly in the center. And if you want to use this with Windows, you’ll need to map yourself a Windows key and a CTRL key. However, once you’ve done the work, you’ll be gifted with more macros than you know what to do with.

Read: Mechboards Hyper7 R4 Review

– Avram Piltch

Asus ProArt PA32UCDM 4K 32-inch Gaming Monitor

The Asus ProArt PA32UCDM is a gaming monitor with a dual mission, appealing to two distinct groups of consumers. On the one hand, it’s a highly capable professional monitor with a 32-inch 4K QD-OLED panel, rich, accurate colors, excellent build quality, Dolby Vision support, and bountiful calibration options.

However, the PA32UCDM also caters to gaming enthusiasts by supporting a heady 240 Hz refresh rate at 4K, Nvidia G-Sync, and AMD FreeSync. We’d also be remiss if we didn’t mention the ultra-low input lag, which is critical for gamers looking for every slight advantage in twitchy online multiplayer games.

As if those fundamentals weren’t enough, the PA32UCDM also features HDMI 2.1 connectivity and Thunderbolt 4 ports, which turns the monitor into a professional-oriented peripheral hub with support for up to 96-watt power delivery.

Read: Asus ProArt PA32UCDM 4K 32-inch Gaming Monitor Review

– Brandon Hill

Asus ZenWiFi BT10 Wi-Fi 7 Mesh Router

If you’re looking for a high-end Wi-Fi 7 mesh router to upgrade your home network, look no further than the Asus ZenWiFi BT10. While it doesn’t sit at the tippy top of Asus’ mesh router lineup (that distinction belongs with the ZenWiFi BQ16 Pro), it delivers scintillating performance, broad connectivity options, a rich web user interface with countless configuration options, and a full-featured smartphone app for those that like to tinker with router settings on the go (the BT10 support remote configuration).

We observed strong performance across 6 GHz, 5 GHz, and 2.4 GHz bands, with the former delivering over 2,600 Mbps throughput at close range. The BT10 supports the full Wi-Fi 7 spec, unlike lesser dual-band Wi-Fi 7 mesh routers, and even allows you to utilize up to a 10 Gbps backhaul between the satellites for improved network performance. When you throw in no-cost software features like AI Protection, which is powered by Trend Micro, support for Apple Time Machine backups, and smartphone cellular tethering in case of an ISP outage, and a street price of around $600, the BT10 is hard to beat.

Read: Asus ZenWiFi BT10 Wi-Fi Mesh Router Review

– Brandon Hill

Logitech G Pro X Superlight 2 Dex

The Logitech G Pro X Superlight 2 Dex isn't necessarily the most innovative mouse we've ever seen, but it's a big innovation in Logitech's Superlight lineup, as it's the first non-ambidextrous Superlight mouse. The Superlight lineup has been highly popular with gamers since it debuted, so it's no surprise Logitech hasn't done much to mess with the style — the most recent G Pro X Superlight 2, which came out in Sept. 2023, was almost identical to its predecessor. But the G Pro X Superlight 2 Dex, which came out last September, is a whole different shape: a contoured, right-handed shape with a hump that's shifted toward the center-left.

It's designed to be more ergonomic and easier to hold, and it feels a little like the Razer DeathAdder V3 Pro, which is one of the best gaming mice we've ever used. While we didn't love the smooth finish as much as we like the DeathAdder V3 Pro's matte texture, it does come with precut grip tape in the box, which is perfect for adding extra friction. It's also superlight (of course) — weighing just 2.12 ounces (60g), while the DeathAdder V3 Pro is 2.22 ounces (63g). The G Pro X Superlight 2 Dex has top-notch specs: it sports Logitech's Hero 2 sensor with a maximum resolution of 44,000 DPI, a maximum speed of 888 IPS, and can handle up to 88 G's of acceleration. It also gets polling rates of up to 8,000 Hz, and up to 95 hours of battery life (with a standard 1,000 Hz polling rate).

Read: Logitech G Pro X Superlight 2 Dex Review

— Sarah Jacobsson Purewal

Bambu Lab H2D

The Bambu Lab H2D is one of the most anticipated (and highly speculated about) 3D printers to come out this year and it does not disappoint. Similar to the popular X1 Carbon, the H2D is an amazing high-speed XY core machine that comes fully enclosed with multi-color and multi-material support — but it's so much more than that.

What makes the H2D better is its bigger build volume, the addition of a dual nozzle hotend, an upgraded heated AMS (Automatic Material System), and of course, a laser and cutter module that turns this 3D printer into every elite hobbyist’s dream machine.

Bambu Lab packed the H2D with great innovative features, such as a dual nozzle system, which makes multi-color printing easier and more efficient. It also saves time because you can print multiple colors, minimizing waste. You can even print using different materials, such as PLA with TPU, at the same time. The upgraded AMS is heated and provides the optimal environment for your filaments, keeping them dry and preventing cracking and storage issues.

And best of all, Bambu Lab has added a laser and cutter module that is easy to install and use, equipped with a laser, blade cutter, and pen plotter. It is available with 10W and 40W laser options. Add in a larger, 350 x 320 x 325mm build volume, and makers can complete bigger and more complex projects faster.

The H2D Laser Combo is available for pre-order at $3,499 (10W) and $4,399 (40W). The H2D AMS Combo is now available for $2,699, while the H2D stand-alone model, priced at $2,399, will be available in June 2025.

Read: Bambu Lab H2D Review

— Anj Bryant

Pimoroni Picade Max

A two-player arcade cabinet is hardly an innovation, but what Pimoroni has done with its latest iteration of the Picade series integrates the Raspberry Pi Pico’s PR2040 into multiple aspects of this DIY arcade cabinet kit.

We start with the controls. The dual arcade sticks and corresponding buttons can take the punishment when trying to land that kick, punch or shot in your favorite arcade game. But linking these controls to your choice of a Raspberry Pi 4, 5 or even an ITX based PC is a custom designed arcade controller to USB adapter.

Powered by the RP2040, Picade Max USB Controller handles 25 inputs for a two player setup, and it worked flawlessly with Recalbox. The Raspberry Pi 5 running Recalbox detected the controls as if it were a typical USB controller. Picade Max USB Controller can also work with StemmaQT devices, which means with some clever coding you could install a spinner for games such as Tempest.

Next is sound, and again the RP2040 powers another add-on board. The Picade Max Audio provides 3W of stereo sound output that really impresses. The “attract” screens, used to entice players to part with their quarters, really grab your attention when the booming audio starts to play. Speaking of the attract screen, the large 19 inch IPS display is a huge upgrade over previous models and it provides plenty of screen for two players to fight it out.

The build process is enjoyable, and it will certainly test your skills with a screwdriver. We had access to a beta unit and provided feedback on the build process which is now part of the final retail instructions. If you’ve got kids, this would be an ideal project to learn how to build electronics projects as the goal is just as exciting as the journey to get there.

This is my third Picade, and it just happens to be my favorite, as now I can play games, shoulder to shoulder with my friends and family, just like the old days.

Read: Picade Max Review

– Les Pounder

The GPU rumor mill has been spinning like crazy this morning. The whole thing kicked off when Sparkle's Bilibili account supposedly confirmed the existence of a 24GB variant of the Intel Arc B580 (via ITHome). A media manager also alluded to a potential release window between May and June. But this was short-lived as Sparkle has officially retracted these comments. Perhaps this is a move to manage fallout after a potential NDA breach? It's hard to say for certain, but this isn't the first time we've heard of a 24GB spinoff of Battlemage.

Intel officially introduced its Arc Battlemage GPU lineup in December of last year. From this family, the Intel Arc B570 and Arc B580 were crafted using the mid-range BMG-G21 die, featuring a 192-bit interface for six memory channels. Shipping documents have suggested the existence of alternate BMG-G31 and BMG-G10 variants, likely for the B700 and B300 families, but Intel has not made any official confirmation yet.

This slip-up seemingly corroborates existing rumor, positioning this GPU under Intel's Arc Pro family, succeeding existing Alchemist-based Flex or Pro-series options, for AI applications. From what we can gather, Intel is allegedly deploying the BMG-G21 die, the same as the B580, with 24GB of memory configured using 12x 16Gb modules in clamshell mode. That's a twofold increase in memory capacity over Intel's current Arc Pro A60 offering.

Architecturally speaking, Battlemage is one generation behind Nvidia and AMD. For tasks demanding raw compute muscle, this Arc B580 24GB, or whatever Intel ends up naming it, will probably not land blows against Nvidia's Blackwell PRO or AMD's upcoming Radeon PRO W9000 GPUs. These can, however, compete in AI training and inference, which necessitate a lot of VRAM in cases like Large Language Models (LLMs), and image/video generation models like Stable Diffusion.

The economics don't favor Intel if they were to create a higher-end GPU. At similar die sizes, Intel's PPA (Performance Per Area) and architectural shortcomings become apparent: the B580 at 272mm² (using N5) is priced at $249, while Nvidia sells its RTX 5070, which uses a 263mm² die (using N4P) at $549. While N4P wafers might be pricier than N5 (likely not double), that's still a small consideration.

Either way, Sparkle's seeming confirmation, followed by a swift retraction, suggests there's potentially more beneath the surface. If this GPU does exist, we can expect to hear more from Intel at Computex, where we might also get an early look at Panther Lake, similar to how Lunar Lake was unveiled last year.

RTX 5070 vs RTX 5060 Ti Introduction

The Nvidia GeForce RTX 5070 and RTX 5060 Ti 16GB are the two most recent GPUs launched by Team Green. Since the beginning of 2025, Nvidia has released five Blackwell RTX 50-series GPUs — six if you want to count the RTX 5060 Ti 8GB as a separate entry.

To say that supplies have been limited and insufficient to keep up with demand would be a gross understatement, but the same applies to any of the best graphics cards right now. In fact, of all the Blackwell RTX GPUs, the 5070 and 5060 Ti are currently the most readily available; just don't expect to find most models at MSRP.

Our GPU benchmarks hierarchy ranks all the graphics cards by performance, and naturally the more expensive cards come out ahead of their less expensive siblings. But how do the RTX 5070 and RTX 5060 Ti 16GB stack up? We'll look at the performance, approximate pricing — because no GPU prices are reliably set in stone right now — as well as other aspects of the cards to help you decide which one might be right for your gaming PC.

Most of the features between the two GPUs will be identical. Both use the Blackwell architecture and support DLSS 4 and MFG, for example. Still, let's start with the specifications to see how the 5070 and 5060 Ti stack up.

RTX 5070 vs RTX 5060 Ti Specifications

As you'd expect, the RTX 5070 offers more raw performance on paper and in practice. It uses a GPU that's 45% larger, with 41.6% more transistors. More importantly perhaps, it has 33% more SMs, tensor cores, RT cores, and related computational units. It also has 50% more memory bandwidth and a 50% wider memory interface, plus 50% more L2 cache.

Memory interfaces don't scale as well as core logic on modern process nodes, and there are other aspects of the Blackwell GPU — the video encoders/decoders, display outputs, etc. — that don't really change, which explains why the chip size and transistor counts don't necessarily scale linearly.

Clock speeds aren't quite the same, at least on paper. The 5060 Ti has a 2572 MHz boost clock and the 5070 has a slightly lower 2512 MHz boost clock. But we're also looking at the 5070 Founders Edition with reference clocks, and a PNY 5060 Ti 16GB OC that has a 2692 MHz boost clock. Plugging in those numbers gives the 5070 30.9 teraflops of FP32 compute and the PNY 5060 Ti has 24.8 TFLOPS. That means the 5070 'only' has 25% more compute on paper — which isn't actually all that accurate.

In our full suite of gaming benchmarks, the 5070 Founders Edition averaged clock speeds of 2826 MHz while the PNY 5060 Ti 16GB OC averaged 2776 MHz. That means despite theoretically 180 MHz lower clocks, in practice Nvidia's 5070 card came out 50 MHz ahead.

That means the 5070 FE offers about 34.7 TFLOPS compared to 25.6 TFLOPS, a 35.5% advantage. Practically speaking, though, the 5070 should be up to 35% faster in games that are compute bound, and up to 50% faster for games that are memory bandwidth bound — and potentially the 5060 Ti could close the gap if VRAM capacity comes into play.

RTX 5070 vs RTX 5060 Ti Gaming Performance

There are multiple facets of performance to discuss. We'll start with gaming performance, as that's what the majority of people looking at the RTX 5070 and RTX 5060 Ti 16GB care about the most. We have an expanded test suite of 21 games, with 15 rasterization benchmarks and six ray tracing benchmarks — slightly more than the 18 games we use in our GPU hierarchy (we're planning to add these into the hierarchy but haven't tested every GPU yet).

The charts group things together by overall performance (geometric mean across all 21 games, where each gets equal weighting), rasterization-only performance, and ray tracing-only performance. We also have separate tables that show the percentage differences below the charts (sorted in reverse order due to the vagaries of Excel).

It's safe to say that neither the RTX 5070 nor the RTX 5060 Ti 16GB are primarily targeted at native 4K ultra gaming. The 1440p result works as a proxy for 4K with DLSS quality mode upscaling, and the 1080p result stands in for 4K with performance mode upscaling, though there would be potential differences. Upscaling does have some overhead, but since we're looking at two Nvidia GPUs we can expect a pretty consistent delta cause by DLSS. CPU bottlenecks would potentially affect performance at lower resolutions as well, which we can see by the results.

At 4K, the RTX 5070 leads the RTX 5060 Ti 16GB by 32% on average, with a performance difference of -5% (Indiana Jones and the Great Circle) to +43% (A Plague Tale: Requiem). Indiana Jones is known to want a lot of VRAM — you can even run the ultra setting on cards with less than 12GB — while various other games appear to want more memory bandwidth.

Dropping to 1440p, the 5070 lead shrinks fractionally to 31% overall, and this time there are no performance deficits. There may be a few other games out there that can push beyond 12GB of VRAM use, but from our test suite — a rather demanding suite overall, we'd say — 12GB remains sufficient in all games for 1440p, and in all but one of the games for 4K.

1080p starts to run into CPU limits, even more so at medium settings, which shrinks the delta between the 5070 and 5060 Ti to 28% at ultra settings and 24% at medium settings. That's still a sizeable gap, with a few of the games showing basically equivalent performance due to the CPU bottleneck (Baldur's Gate 3, Flight Simulator 2020, Spider-Man 2, Spider-Man: Miles Morales, and Stalker 2 end up tied in one or both cases).

Looking at the rasterization and ray tracing performance, overall there's not much difference at the various resolutions and settings. 4K ultra does run into VRAM limits on one of the RT games, so that the gap is slightly narrower there than at 1440p and 1080p, and our RT test suite does end up with lower FPS than the rasterization games at ultra settings. Medium settings with RT enabled actually runs faster on average for our test games, but only Control and Cyberpunk are truly demanding RT games in our title selection.

RTX 5070 vs RTX 5060 Ti Content Creation Performance

Our content creation test suite consists of a variety of test that all loosely fall into the "professional" category. We have multiple AI tests, 3D rendering benchmarks, video transcoding performance courtesy of SPEC Workstation 4.0, and the SPEC Workstation 4.0 Viewport benchmarks (basically the same tests as SPECviewperf).

As expected, the RTX 5070 generally walks away with the content creation crown. It's 45% faster in the Blender 3D rendering benchmarks, 33% faster in MLPerf text generation tokens per second, and 20–43 percent faster in the Procyon and SPEC AI tests that we ran. The 5060 Ti 16GB did have a faster time to first token in MLPerf, though it's possible that's due to updates to the benchmark (we'll recheck the 5070 in the near future, once our test rig isn't busy).

For SPEC Workstation 4.0's Viewport tests, there's another instance of the 5060 Ti 16GB garnering a win — it's 2% faster in Creo. But overall the 5070 still has a 23% lead. Handbrake video transcoding meanwhile ends up as a tie, which is expected as the fixed function encoding and decoding hardware is the same, so the only difference would be GPU clocks during that test.

Neither GPU is specifically marketed as a professional solution, though we anticipate there will be professional variants of both chips (with more VRAM and drivers that enable additional accelerations in some of the SPEC Viewport tests). Mostly, the AI results are interesting as a hobbyist solution, though it's possible future games might leverage the AI hardware more.

RTX 5070 vs RTX 5060 Ti Conclusion

Which is better: RTX 5070 or RTX 5060 Ti 16GB? Price, as we said at the start, will arguably be the biggest factor in deciding between the RTX 5070 and RTX 5060 Ti 16GB. It's also the one thing that we can't really pin down at present. Since the 5060 Ti launched just a few days ago, pricing and availability are more likely to fluctuate in the near term. The 5070 has been out for a month now, and we have a better idea of what to expect.

First, let's talk about MSRPs and performance. On paper, Nvidia gives the 5060 Ti 16GB a $429.99 MSRP and the 5070 has a $549.99 MSRP — so $430 and $550. That means the 5070 is supposed to cost 28% more than the 5060 Ti 16GB. And based on the performance results, that's exactly in line with what you get.

Linear performance scaling generally means you're better off buying the more expensive card — we would normally expect to see diminishing returns. So if you buy an RTX 5060 Ti 16GB at MSRP and get 0.171 FPS/$ across our gaming tests, versus the RTX 5070 with 0.170 FPS/$, we give the win to the 5070. The reason is because the GPU doesn't exist in a vacuum; there's the rest of the PC to consider.

Alternatively, let's talk current street prices. These will change dramatically over time, but we can currently find an RTX 5060 Ti 16GB for $509.98, or an RTX 5070 for $619.98. That's the second pair of tables in the above gallery, which obviously skews in favor of the 5070 at current prices. And that's what you need to look at, first and foremost.

Finally, imagine a PC where the CPU, motherboard, SSD, RAM, PSU, and case together cost $750 for example. Now you're looking at a total cost of $1,180 for the 5060 Ti 16GB compared to $1,300 for the 5070 build (using the base MSRPs for the GPUs). That works out to just 10% more total money for about the same 25~30 percent performance uplift. It's what the last set of tables above show, and again, it favors the 5070 heavily.

Given the 25~30 percent performance advantage of the RTX 5070 over the RTX 5060 Ti 16GB, spending up to 30% more on the RTX 5070 would make sense. It's the clear winner at current online prices, or at the given MSRPs. In fact, we would argue that it's worth spending up to 40% more for the 5070 compared to the 5060 Ti, as we normally expect diminishing returns.

We don't know where prices will end up, in the U.S. or elsewhere, but you can use the above 30–40 percent figure as a guideline. If the RTX 5070 were to cost $550, we wouldn't recommend spending more than $425 on the RTX 5060 Ti 16GB — and $400 would be better. Conversely, if the best price you can find on the RTX 5070 ends up being $700, then the 5060 Ti 16GB would be worth $500 to at most $540. (And you can replace the dollar signs with whatever monetary symbol you choose.)

Long-term, we expect both GPUs will trend toward about a 30% higher price for the RTX 5070 vs RTX 5060 Ti 16GB. Anything less makes the slower card very difficult to justify, even if it does technically offer more VRAM. So unless you specifically need a card with 16GB, perhaps to run a particular LLM, the RTX 5070 offers the better overall value right now, and will probably continue to do so in the future.

The Intel Arc B580 is considered the cheapest modern GPU on the market today, with Tom’s Hardware considering it the $249 GPU champion. However, several reviewers discovered that the budget GPU performs poorly on older processors. Intel has acknowledged the issue and will investigate it.

A moderator on the Intel Community forum started an Intel Arc Graphics and CPU Overhead thread, saying, “Thank you for your patience. We are aware of reports of performance sensitivity in some games when paired with older generation processors. We have increased our platform coverage to include more configurations in our validation process, and we are continuing to investigate optimizations.”

Reviewers first noticed the issue in early January, when they tested the Intel Arc B580 on older and newer CPUs. For example, Hardware Unboxed ran some gaming benchmarks with the GPU using an AMD Ryzen 7 9800X3D and a Ryzen 5 2600. It also used an Nvidia GeForce RTX 4060 as a control, and these are its results.

GPU Benchmarks Introduction

Tom's Hardware exhaustively benchmarks every GPU to find out which are worthy of our list of the best graphics cards. Our GPU benchmarks hierarchy ranks current and previous generation graphics cards by performance. Whether it's playing games, running artificial intelligence workloads, or doing professional video editing, your graphics card typically plays the biggest role in determining performance — even the best CPUs for gaming take a secondary role.

Our 2026 GPU Hierarchy testing spans three generations of Nvidia and AMD graphics cards, as well as Intel's Arc B-series GPUs.

Our testing has been made easier by the fact that no truly new gaming GPUs have been introduced in almost a year. If you haven't already upgraded your graphics card after the GeForce RTX 50-series and Radeon RX 9000-series launches in 2025, well, you're still looking at the exact same products now.

AMD did make its formerly China-only Radeon RX 9070 GRE available globally after Computex 2026, but in our review, we found that $549 product to be too expensive given the level of performance it delivers and the compromises made to hit its price point. Check out that coverage for all the details.

Most of the products we recommend remain at elevated prices compared to their MSRPs, but this is just life in mid-2026. It's admittedly cold comfort, but unless you're shopping for an RTX 5090, graphics card prices haven't risen much more than they already did earlier this year.

Compared to the doubling or tripling of prices we've seen for RAM kits and SSDs in 2026 versus last year, a GPU upgrade remains a relatively affordable (and self-contained) option, either as a boost for an existing PC or part of an all-new parts list.

Let's dive into our ranking of GPUs past and present so you can figure out how all those cards stack up.

Prime Day exceptional graphics card deals

These are a few of the standout deals from Amazon's 2026 Prime Day event, which is currently taking place. Be sure to check out our full list of the best Prime Day graphics card deals, too. The GPU Benchmarks Hierarchy continues below.

Our full GPU hierarchy using traditional rendering (aka rasterization) comes first, and below that we have our separate ray tracing hierarchy. We've also mashed up these results into one overarching ranking for the complete pictures.

The results are all collected at native resolution, without enabling DLSS, FSR, or XeSS upscaling or frame generation.

As a brief refresher of the cards in this version of our hierarchy, Nvidia's cutting-edge Blackwell architecture and its DLSS 4 suite of tech upgrades the quality of both upscaling and ray reconstruction on RTX 50-series GPUs. It also adds Multi Frame Generation support. As of mid-2026, MFG can add anywhere from one to five AI-generated frames in between natively rendered ones.

Even if you're not into framegen, DLSS 4.5 upscaling and its transformer neural network architecture can offer a big boost in image quality at the same output resolution compared to earlier DLSS versions. That tech can benefit all GeForce RTX GPUs going back to the RTX 20-series, albeit at a heavier performance cost on older hardware compared to earlier DLSS versions.

Nvidia's Ada Lovelace architecture powers its prior-gen RTX 40-series cards. Ada introduced DLSS Frame Generation, which can double output frame rates in supported games. Ada cards don't benefit from MFG, though.

AMD's Radeon RX 9000 series cards, powered by its latest RDNA 4 architecture, get a big boost to both ray-tracing and AI capabilities with dedicated RT and matrix math accelerators. AMD uses those capabilities to enable its FSR 4 upscaler and its much-improved image quality in a small but growing range of titles, whether through native support or with driver-level overrides.

The FSR Redstone update last year brought ML Frame Generation to RX 9000-series cards. Like FSR 4.x upscaling, ML Frame Generation can be directly integrated in games or enabled through a driver override.

Meanwhile, the last-gen RDNA 3 architecture powers seven RX 7000-series seven desktop cards. Until very recently, RX 7000-series cards couldn't officially run FSR 4 upscaling, but an official version of that tech will be made available for those cards in July 2026.

Intel's Battlemage-powered Arc B580 and Arc B570 offer major improvements in performance and efficiency compared to its Alchemist architecture. Battlemage only serves the entry-level end of the graphics card market today. Intel introduced a larger Battlemage chip in the form of the Arc Pro B70 earlier this year, but that product is intended for AI and professional visualization, and it's priced like it.

On page two, you'll find our 2025-2026 test data for posterity. Page three has our 2024–2022 benchmark suite, which covers previous-generation GPUs running an older test suite and a Core i9-12900K. Page four has an even older 2020–2021 test suite with only raster games, running on a Core i9-9900K testbed. The legacy tables are no longer being actively updated. There's also the "Ancient Legacy GPU hierarchy" (without benchmarks, sorted by theoretical performance) for reference purposes.

The following tables sort everything solely by our performance-based GPU benchmarks, from fastest to slowest. Factors including real-world pricing, graphics card power consumption, overall efficiency, and features aren't factored into the rankings here. The latest results use an AMD Ryzen 7 9800X3D testbed. Here are the tables and benchmark results — rasterization games first, then ray tracing games, and finally the content creation results.

GPU Benchmarks Hierarchy 2026: The tests

For our latest GPU benchmarks, we test every card at a mix of high and ultra settings, depending on the game. We test across three resolutions: 1080p, 1440p, and 4K.

All the scores are expressed as percentages relative to the top-ranking 1080p ultra card, which is of course the GeForce RTX 5090.

Our 2026 test suite comprises the following games. 11 are raster titles, and of those, four have RT support. We test another four games that either require RT to run (DOOM: The Dark Ages and Indiana Jones and the Great Circle) or look best with RT (Assassin's Creed Shadows and Grand Theft Auto V Enhanced.)

• Black Myth Wukong (+RT)
• Alan Wake II (+RT)
• Apex Legends
• Counter-Strike 2
• Fortnite
• Arc Raiders
• Stalker 2
• DOOM: The Dark Ages (RT required)
• Grand Theft Auto V Enhanced (RT only)
• Marvel's Spider-Man 2 (+RT)
• Indiana Jones and the Great Circle (RT required)
• Marvel Rivals
• Assassin's Creed Shadows (RT only)
• Cyberpunk 2077 (+RT)
• Clair Obscur: Expedition 33

GPU Benchmarks Hierarchy 2026: Raster gaming

The FPS score is the geometric mean (equal weighting) of all 11 games. Note that the specifications column links directly to our original review for the various GPUs.

Rasterization GPU Benchmarks, Key Findings

• Unsurprisingly, the RTX 5090 takes the top spot across the board, but prices remain stratospheric in mid-2026. It's also difficult to fully engage the GB202 GPU for pure raster gaming unless you're playing at 4K.
• Among current cards, the RX 9060 XT 8GB and RTX 5060 duke it out for the best performance per dollar at 1080p.
• The Radeon RX 9070 is the 1440p raster value champ. The recently introduced RX 9070 GRE sits just behind, and the RTX 5070 is a somewhat distant third.
• At 4K, the RX 9070 XT is the cheapest way to get into 4K 60 FPS native gaming. But the RX 9070 and RTX 5070 are also strong options among current GPUs if you're willing to enable a dash of upscaling.

Overall, if you're only interested in native raster gaming, Radeons are a strong choice in 2026. But Nvidia offers superior DLSS 4.5 upscaling and Multi Frame Generation support across all RTX 50-series cards, which makes matching your GPU's performance to your monitor's refresh rate a snap.

FSR 4.x upscaling isn't on par with DLSS 4.5 yet, and FSR ML Frame Generation is limited to a 2x framerate boost where it's available.

Rasterization GPU Benchmarks Rankings 2026

Remember that we're not including any upscaling or framegen results in the above table. DLSS, FSR, and XeSS offer differing image quality, and we want to keep things directly comparable.

Don't buy any of the cards at the top of our hierarchy without a high-refresh-rate 2560x1440 or 4K monitor to match. Even one of our best gaming CPUs, like the Ryzen 7 9800X3D, can only do so much when a powerful card becomes CPU-bound.

In general, if you're still running a 60 Hz fixed-refresh-rate monitor, it's entirely possible that you're not seeing all the frames your graphics card can generate. Practically any current-gen graphics card from the RTX 5060 on up is good enough for high-refresh-rate gaming at 1080p in 2026 even without upscaling or framegen, and the RTX 5070 or RX 9070 will give you the same experience at 1440p.

In this era of upscaling and framegen, a high-resolution monitor is no longer an obstacle to the best gaming experience. Upscalers work best with higher-than-1080p output resolutions, and the advent of the DLSS 4.5 upscaling model especially means that you can get both smooth performance and crisp image quality from surprisingly modest graphics cards.

But if you don't have a high-refresh-rate 1440p or 4K monitor to begin with, you can't take full advantage of the free boost to both performance and image quality that AI-powered upscaling offers, nor can you enjoy the full smoothness boost of framegen. Check out our list of the best gaming monitors for a potential upgrade if you're still rocking a 1080p 60 Hz display from the 2010s.

On to our 2026 raster results. Among currently available graphics cards, Blackwell rules the top three spots. The RX 9070 XT ends up fourth, just behind the potent (but much more expensive, as of June 2026) RTX 5070 Ti.

Moving into the midrange, the RX 9070's strong raster performance gives it the edge over the RTX 5070, but it's close. And the 5070 has the full arsenal of DLSS 4.5 features at its disposal in virtually every modern game.

Given the image quality and smoothness advantage of DLSS 4.5 upscaling and MFG, the 5070 got the nod in our most recent round of best graphics card picks after our 2026 retests.

The recently introduced RX 9070 GRE lands between the 5070 and the lower-end 5060 Ti and 9060 XT, a position for which Nvidia has no current-gen answer (and no easy path to one.) But AMD may have priced the GRE too high at $549, and we're already seeing big price drops on that product that might make it uniquely appealing in a market where the RTX 5060 Ti is its direct price competition.

The formerly midrange RTX 5060 Ti 16GB is hopelessly overpriced for the performance it offers in mid-2026, and the formerly somewhat affordable 8GB model is also selling for elevated prices now, making the entire 5060 Ti family impossible to recommend.

The RX 9060 XT 16GB holds down the entry-midrange 16GB position by itself, and its $459 price tag isn't so absurd as to make it un-recommendable. But as RX 9070 GREs start going on sale for $500-ish, the step up is going to look mighty tempting.

As we move further down the stack, the $300-ish Arc B580 represents Intel's best card right now, but it remains very much a budget part in mid-2026, trading blows with the RTX 3060 Ti and RTX 4060. The only reason it lands as high as it does in our results is that its 12GB of VRAM prevents its performance from totally plummeting at 4K, as it does for those older 8GB cards.

The RTX 5060 is 24% faster than the Arc B580 at 1080p and 17% faster at 1440p, and its prices are proportionately higher. The scarce RX 9060 XT 8GB is a whole 30% faster at 1080p and 23% faster at 1440p, for similarly more money.

We don't think the potential longevity afforded by the B580's 12GB of VRAM is worth the massive performance tradeoff versus current-gen 8GB GPUs. Save up a few more bucks and grab an RTX 5060 or RX 9060 8GB instead (and enjoy the more widely adopted DLSS 4.5 or FSR 4 versus XeSS while you're at it).

The $250 Arc B570 is still a good value in mid-2026, and as the only modern, decent graphics card available for (well) under $300, it's still worth a look if you can't stretch to an RTX 5050. But recent software ills that took months to fix have led us to consider it more of a gamble than a product worthy of a recommendation, even though those issues did eventually get corrected.

Ray Tracing GPU Benchmarks 2026

For 2026, we're testing a range of RT titles that present a progressively more difficult workload to the GPUs under test.

Grand Theft Auto V Enhanced, DOOM: The Dark Ages, and Indiana Jones and the Great Circle are all relatively lightweight RT titles that will run on a wide range of RT-ready hardware. Spider-Man 2, Cyberpunk 2077, and Assassin's Creed Shadows represent greater challenges to compute resources, VRAM, or both. And Black Myth Wukong and Alan Wake II still bring even the most powerful graphics cards to their knees at native resolutions.

Our results for those more demanding games represent more of a jumping-off point for upscaling and framegen rather than native performance. But lighter RT titles mean that you can explore the differences in image quality that ray tracing offers without crushing your GPU flat.

In any case, graphics cards at the top of our RT tests are ready for practically any RT game. But as you'll see, that's still quite the high bar to clear in 2026.

Ray Tracing GPU Benchmarks, Key Findings and Notes

• The RTX 5090 really comes into its own with RT enabled, as expected.
• AMD's RX 9070 XT outperforms the RX 7900 XTX in our RT tests, showing that the ray-tracing improvements in RDNA 4 deliver.
• Even with those improvements, the RX 9070 XT can only manage a tie with the RTX 4070 Ti with RT in play, coming in eighth place overall.
• Maintaining 60 FPS in RT titles (as a good foundation for upscaling and framegen enhancements) is a very tall order. Among recent products, you'll want:
  • an RTX 5060 Ti 16GB, RTX 5070, or RX 9070 GRE to handle ~60 FPS RT at 1080p without running out of VRAM
  • An RTX 4070, an RTX 5070 or RX 9070 at 1440p, or an RX 9070 XT or RTX 5070 Ti for the best possible experience
  • An RTX 5080, RTX 4090, or RTX 5090 at 4K

Ray Tracing GPU Benchmarks Rankings 2026

The RTX 5090 and RTX 4090 lead our RT results. But if you want to enjoy high-end RT without a 4090- or 5090-sized bankroll, the RTX 5070 Ti proves itself as the last card with a reasonable price-to-performance ratio before things get crazy.

The 5070 Ti leads the RX 9070 XT by a wide margin in our RT tests, and it's anywhere from 10%-15% behind the absurdly overpriced RTX 5080 across the board. But if you want the absolute best RT performance possible without stepping up to a 4090 or 5090, the extra cash for a 5080 is your best — and only — option.

That said, you should really explore DLSS 4.5 upscaling and MFG on the 5070 Ti before spending big to step up to a 5080.

The RX 9070 XT remains AMD's fastest RT GPU, outpacing the 7900 XTX across all tested resolutions. It leads the RDNA 3 halo card by about 10% across the board, which is an impressive generational leap considering its 33% deficit in CUs versus the XTX.

But the 9070 XT can only trade blows with the RTX 4070 Ti and RTX 5070 with Nvidia competition in the picture. It beats out the 5070 at 1080p and 1440p, but DLSS 4.5 upscaling and MFG give Blackwell gamers more flexible options for boosting performance with only minimal impact to image quality versus both RDNA 4 (and Ada).

Of course, the RX 9070 XT has FSR 4 upscaling and framegen at its disposal in compatible titles, but support for those features isn't as widespread as DLSS.

Meanwhile, the RX 9070 comes in slightly behind the RTX 5070 at 1080p and 1440p for RT. Even though its 16GB of VRAM prevents the 9070's 4K RT performance from completely nosediving, the average frame rate at that resolution isn't high enough to make it a 4K RT choice.

At 1080p, the RX 9070 GRE presents an intriguing RT option for Radeon gamers, delivering a near-60-FPS average baseline that would make a great FSR 4 foundation. But its 12GB of VRAM quickly becomes a liability at 1440p versus the 9070, and 4K is out of the question.

Overall, our demanding 2026 RT test suite shows that the bar for entry to a good RT experience remains high, especially as elevated prices on the RTX 5060 Ti 16GB mean that the first "real" RT-ready Blackwell card at 1080p and up is the RTX 5070. And on the Radeon side, you really want an RX 9070 GRE or RX 9070 for the best experience.

Test Setup and Hardware for GPU Benchmarks

We've used several different PCs for our testing over the years. The latest update switches to an AMD Zen 5 processor: the unparalleled Ryzen 7 9800X3D and its 3D V-Cache-enhanced performance.

Here are the specifications for our latest GPU test PC.

Tom's Hardware 2026 GPU Benchmarks Testbed

CPU: AMD Ryzen 7 9800X3D
CPU Cooler: Thermalright Phantom Spirit 120 SE
Memory: G.Skill Trident Z5 32GB (2x16GB) DDR5-6000 CL30
Motherboard: Asus TUF Gaming X670E-Plus Wifi
SSD: Inland Performance Plus 4TB
PSU: MSI MPG Ai1600TS 1600W

We test across the three most common gaming resolutions, 1080p, 1440p, and 4K, using a mix of high and ultra settings, depending on the title. Where possible, we use 'reference' cards for all of these tests, like Nvidia's Founders Edition models and AMD's reference designs. Most midrange and lower GPUs don't get reference models, however, and in some cases we only have factory-overclocked cards for testing. We do our best to select cards that are close to the reference specs in such cases.

For each graphics card, we follow the same testing procedure. We run one pass of each benchmark to "warm up" the GPU after launching the game, then perform our actual test runs across each resolution.

We carefully review our test data and check for anomalies. For example, we always expect the RTX 5080 to be faster than the RTX 5070 Ti. If it's not, and we're not in a CPU limited situation, we'll recheck both cards to ensure that our standings our accurate. We also check and retest in cases of subtler issues, as when a transient hitch or frame-time spike causes a large dip in 1% low FPS.

Due to the length of time required for testing each GPU, updated drivers and game patches inevitably come out that can impact performance. We periodically retest a few sample cards to verify our results are still valid, and if not, we go through and retest the affected game(s) and GPU(s). We may also add games to our test suite over time, if one comes out that is popular and conducive to testing. See what makes a good game benchmark for our selection criteria.

GPU Benchmarks: Individual Game Charts

The above tables provide a summary of performance, but for those that want to see the individual game charts, for both the standard and ray tracing test suites, we've got those as well. These charts only cover current-gen GPUs for readability.

These charts are up to date as of June 2026.

GPU Benchmarks — 1080p

GPU Benchmarks — 1440p

GPU Benchmarks — 4K

GPU Benchmarks — Power, Clocks, and Temperatures

Most of our discussion has focused on performance, but for those interested in power and other aspects of the GPUs, here are the appropriate charts. We'll run these from highest to lowest settings, as 4K ultra tends to be the most strenuous workload on most of these GPUs.

If you're looking for the legacy GPU hierarchy, head over to page two! We moved it to a separate page to help improve load times in our CMS as well as for the main website. And if you're looking to comment on the GPU benchmarks hierarchy, head over to our forums and join the discussion!

Choosing a Graphics Card

Choosing a Graphics Card

Choosing a Graphics Card

Which graphics card do you need? To help you decide, we created this GPU benchmarks hierarchy, comprising 48 GPUs from three generations of hardware for Nvidia and AMD and also including Intel's Battlemage cards. Not surprisingly, the fastest cards are using the latest GPU architectures, though they're not always a major upgrade over the prior generation.

Of course, it's not just about playing games. Many applications use the GPU for other work. But a good graphics card for gaming will typically do equally well in complex GPU computational workloads. Buy one of the top cards and you can run games at high resolutions and frame rates with the effects turned all the way up, and you'll be able to do content creation work as needed. Drop down to the middle and lower portions of the list and you'll need to start dialing down the settings to get acceptable performance in regular gameplay and GPU benchmarks.

If your main goal is gaming, you can't forget about the CPU. Getting the best possible gaming GPU won't help you much if your CPU is underpowered or many years out of date. So be sure to check out the Best CPUs for gaming page, as well as our CPU Benchmarks Hierarchy to make sure you have the right CPU for the level of gaming performance you're looking to achieve.

And don't forget about your monitor. Be sure to get a display whose refresh rate range matches the average frame rates of the graphics card that you want in the games that you play. Check out our list of the best gaming monitors for a starting point.

GPU Benchmarks from 2022–2024

Our 'legacy' GPU benchmarks used different hardware. Here are the details for the 2022–2024 configuration that used an Alder Lake 12900K CPU.

Tom's Hardware 2022–2024 GPU Testbed

Intel Core i9-12900K
MSI Pro Z690-A WiFi DDR4
Corsair 2x16GB DDR4-3600 CL16
Crucial P5 Plus 2TB
Cooler Master MWE 1250 V2 Gold
Cooler Master PL360 Flux
Cooler Master HAF500
Windows 11 Pro 64-bit

We have upgraded our test system and changed our test suite for 2025, effectively resetting our benchmarking and rankings. However, the old data — collected using a Core i9-12900K PC — remains valid. We aren't testing new GPUs with the old testbed, as that would double the time required for what is already a lengthy process, but we felt it would be helpful to some to keep the final data available.

We also had an interim system that never quite got wrapped up, which had a Core i9-13900K Raptor Lake CPU. Our original CPU was one of the impacted units that failed over time, which explains in retrospect why so much of the data felt a bit questionable. Thankfully, the new Ryzen 7 9800X3D system seems to be running just fine. Here are the 12900K results.

For these GPU benchmarks, we tested nearly every GPU released between 2016 and 2024, plus a few extras. All graphics cards were tested at 1080p medium and 1080p ultra, and we sorted the table by the 1080p ultra results. Where it made sense, we also tested at 1440p ultra and 4K ultra. All the scores are scaled relative to the top-ranking 1080p ultra card of the time, which in our suite is the RTX 4090 — especially at 4K and 1440p.

The above summary charts show the relative performance of the cards we've tested across the past several generations of hardware at 1080p ultra — swipe through the above gallery if you want to see the 1080p medium, 1440p, and 4K ultra images. There are a few missing options (e.g., the GT 1030, RX 550, and several Titan cards), but otherwise it's basically complete. Note that we also have data in the table below for some of the other older GPUs.

The eight games used for our standard GPU benchmarks hierarchy are Borderlands 3 (DX12), Far Cry 6 (DX12), Flight Simulator (DX11 Nvidia, DX12 AMD/Intel), Forza Horizon 5 (DX12), Horizon Zero Dawn (DX12), Red Dead Redemption 2 (Vulkan), Total War Warhammer 3 (DX11), and Watch Dogs Legion (DX12). The fps score is the geometric mean (equal weighting) of the eight games. Note that the specifications column links directly to our original review for the various GPUs.

*: GPU couldn't run all tests, so the overall score is slightly skewed at 1080p ultra.

While the RTX 4090 does technically take first place at 1080p ultra, it's the 1440p and especially 4K numbers that impress. It's less than 2% faster than the RTX 4080 Super at 1080p ultra, but that increases to 9% at 1440p and then 25% at 4K. Also note that the fps numbers in our table incorporate both the average and minimum fps into a single score — with the average given more weight than the 1% low fps.

Again, keep in mind that we're not including any ray tracing or DLSS results in the above table, as we use the same test suite with the same settings on all current and previous generation graphics cards. Since only RTX cards support DLSS (and RTX 40-series if you want DLSS 3), that would drastically limit which cards we could directly compare. You can see DLSS 2/3 and FSR 2 upscaling results in our RTX 4070 review if you want to check out how the various upscaling modes can help.

The RTX 4090 comes at a steep price, though on paper it's not that much worse than the previous generation RTX 3090. In fact, we'd say it's a lot better in some respects, as the 3090 was only a minor improvement in performance compared to the 3080 at the time of launch, but with more than double the VRAM. Nvidia pulled out all the stops with the 4090, increasing the core counts, clock speeds, and power limits to push it beyond all contenders. There are two problems with the 4090, however: It's not available at MSRP any longer, due to demand from the AI sector — it often costs $2,000 or more — and there are still concerns with pulling 450W of power over the 16-pin connector.

Stepping down from the RTX 4090, the RTX 4080 Super and RX 7900 XTX trade blows at higher resolutions, while CPU bottlenecks come into play at 1080p. We'll be switching our testbed in the near future, with the current results from our 13900K testing in the charts at the bottom of the page.

Outside of the latest releases from AMD and Nvidia, the RX 6000- and RTX 30-series chips still perform reasonably well and if you're using such a card, there may not be any need to upgrade at present. Intel's Arc GPUs also fall into this category and are something of a wild card.

We've been testing and retesting GPUs periodically, and the Arc chips running the latest drivers now complete all of our benchmarks without any major anomalies. (Minecraft was previously a problem, though Intel has finally sorted that out.) They're not great on efficiency, but overall performance and pricing for the A750 is quite good.

Turning to the previous generation GPUs, the RTX 20-series and GTX 16-series chips end up scattered throughout the results, along with the RX 5000-series. The general rule of thumb is that you get one or two "model upgrades" with the newer architectures, so for example the RTX 2080 Super comes in just below the RTX 3060 Ti, while the RX 5700 XT basically matches the newer and less expensive RX 6600 XT.

Go back far enough and you can see how modern games at ultra settings severely punish cards that don't have more than 4GB VRAM. We've been saying for a few years now that 4GB was just scraping by, and these days we'd avoid buying anything with less than 8GB of VRAM — 12GB or more is the minimum we'd want with a mainstream GPU, and 16GB or more for high-end and above. Old cards like the GTX 1060 3GB and GTX 1050 actually failed to run some of our tests, which skews their results a bit, even though they do better at 1080p medium.

Now let's switch over to the ray tracing hierarchy.

Ray Tracing GPU Benchmarks 2022–2024

Enabling ray tracing, particularly with demanding games like many of those we're using in our DXR test suite, can cause framerates to drop off a cliff. We're testing with "medium" and "ultra" ray tracing settings. Medium generally means using the medium graphics preset but turning on ray tracing effects (set to "medium" if that's an option; otherwise, "on"), while ultra turns on all of the RT options at more or less maximum quality.

Because ray tracing is so much more demanding, we're sorting these results by the 1080p medium scores. That's also because the RX 6500 XT, RX 6400, and Arc A380 basically can't handle ray tracing even at these settings, and testing at anything more than 1080p medium would be fruitless.

The five ray tracing games used are Bright Memory Infinite, Control Ultimate Edition, Cyberpunk 2077, Metro Exodus Enhanced, and Minecraft — all of these use the DirectX 12 / DX12 Ultimate API. The fps score is the geometric mean (equal weighting) of the five games, and the percentage is scaled relative to the fastest GPU in the list, which again is the GeForce RTX 4090.

If you want to see what the future may hold with ray tracing, check out our Alan Wake 2 benchmarks where the full path tracing barely manages playable performance even with upscaling on non-Nvidia GPUs. However, and this is a critical point, the number of games where RT truly makes a striking difference in visuals is extremely limited. For most games, we still feel running pure rasterization rendering makes more sense.

If you felt the RTX 4090 performance was impressive at 4K in our standard test suite, just take a look at the results with ray tracing. Nvidia put even more ray tracing enhancements into the Ada Lovelace architecture, and those start to show up here. There are still further potential performance improvements for ray tracing with SER, OMM, and DMM — not to mention DLSS 3, though that ends up being a bit of a mixed bag, since the generated frames don't include new user input and add latency.

If you want a real kick in the pants, we also ran many of the faster ray tracing GPUs through Cyberpunk 2077's RT Overdrive mode, which implements full "path tracing" (full ray tracing, without any rasterization) — as well as Alan Wake 2, which uses path tracing at higher settings, and Black Myth: Wukong that supports full ray tracing. Those games provide a glimpse of how future games could behave, and why upscaling and AI techniques like frame generation are here to stay.

Even at 1080p medium, a relatively tame setting for DXR (DirectX Raytracing), the RTX 4090 roars past all contenders and leads the previous generation RTX 3090 Ti by 41%. At 1080p ultra, the lead grows to 53%, and it's nearly 64% at 1440p. Nvidia made claims before the RTX 4090 launch that it was "2x to 4x faster than the RTX 3090 Ti" — factoring in DLSS 3's Frame Generation technology — but even without DLSS 3, the 4090 is 72% faster than the 3090 Ti at 4K.

AMD continued to relegate DXR and ray tracing to secondary status, focusing more on improving rasterization performance — and on reducing manufacturing costs through the use of chiplets on the new RDNA 3 GPUs. As such, the ray tracing performance from AMD isn't particularly impressive. The top RX 7900 XTX basically matches Nvidia's previous generation RTX 3080 12GB, which puts it barely ahead of the RTX 4070 — and that's not even in all DXR games. There are some minor improvements for RT performance in RDNA 3, though, as the 7800 XT for example ends up basically tied with the RX 6800 XT in rasterization performance but is 10% faster in DXR performance.

Intel's Arc A7-series parts show a decent blend of performance in general, with the A750 coming in ahead of the RTX 3060 overall. With the latest drivers (and with vsync forced off in the options.txt file), Minecraft performance also looks much more in line with the other Arc DXR results.

You can also see what DLSS Quality mode did for performance in DXR games on the RTX 4090 in our review, but the short summary is that it boosted performance by 78% at 4K ultra. DLSS 3 frame generation improved framerates another 30% to 100% in our testing, though we recommend exercising (extreme) caution when looking at FPS with the feature enabled. It can boost framerates in benchmarks, but when actually playing games it often doesn't feel much faster than without the feature.

Overall, with DLSS 2, the 4090 in our ray tracing test suite is nearly four times as fast as AMD's RX 7900 XTX. Ouch. AMD's FSR 2 and FSR 3 can help as well, and AMD continues to work on increasing the rate of adoption, but it still trails DLSS both in the number of games supported and in the overall image quality. Only two of the games in our DXR suite have FSR2 support. By comparison, all of the DXR games we're testing support DLSS2 — and one also supports DLSS3.

Without FSR2, AMD's fastest GPUs can only clear 60 fps at 1080p ultra, while remaining decently playable at 1440p with 40–50 fps on average. But native 4K DXR remains out of reach for just about every GPU, with only the 3090 Ti and above breaking the 30 fps mark on the composite score — and a couple of games still come up short on the 3090 Ti.

AMD also has FSR 3 frame generation. Like DLSS3, it adds latency, and AMD requires the integration of Anti-Lag+ support in games that use FSR 3. But Anti-Lag+ only works with AMD GPUs, which means non-AMD cards will likely incur a larger latency penalty. We've tested it in Avatar: Frontiers of Pandora and found it worked pretty well, but that was not the case in Forspoken and Immortals of Aveum. It has since gained a lot more traction, though quality and latency remain quite variable — it can look and run well in one game, and then fall flat in another.

The midrange GPUs like the RTX 3070 and RX 6700 XT basically manage 1080p ultra and not much more, while the bottom tier of DXR-capable GPUs barely manage 1080p medium — and the RX 6500 XT can't even do that, with single digit framerates in most of our test suite, and one game that wouldn't even work at our chosen "medium" settings. (Control requires at least 6GB VRAM to let you enable ray tracing.)

Intel's Arc A380 ends up just ahead of the RX 6500 XT in ray tracing performance, which is interesting considering it only has 8 RTUs going up against AMD's 16 Ray Accelerators. Intel posted a deep dive into its ray tracing hardware, and Arc seems reasonably impressive, except for the fact that the number of RTUs severely limits performance. The top-end A770 still only has 32 RTUs, which proves sufficient for it to pull ahead (barely) of the RTX 3060 in DXR testing, but it can't go much further than that. Arc A750 and above also ends up ahead of AMD's RX 6750 XT in DXR performance, showing just how poor AMD's RDNA 2 hardware is when it comes to ray tracing.

It's also interesting to look at the generational performance of Nvidia's RTX cards. The slowest 20-series GPU, the RTX 2060, still outperforms the newer RTX 3050 by a bit, but the fastest RTX 2080 Ti comes in a bit behind the RTX 3070. Where the 2080 Ti basically doubled the performance of the 2060, the 3090 delivers about triple the performance of the 3050.

Test System and How We Test for GPU Benchmarks

For each graphics card, we follow the same testing procedure. We run one pass of each benchmark to "warm up" the GPU after launching the game, then run at least two passes at each setting/resolution combination. If the two runs are basically identical (within 0.5% or less difference), we use the faster of the two runs. If there's more than a small difference, we run the test at least twice more to determine what "normal" performance is supposed to be.

We also look at all the data and check for anomalies, so for example RTX 3070 Ti, RTX 3070, and RTX 3060 Ti all generally going to perform within a narrow range — 3070 Ti is about 5% faster than 3070, which is about 5% faster than 3060 Ti. If we see games where there are clear outliers (i.e. performance is more than 10% higher for the cards just mentioned), we'll go back and retest whatever cards are showing the anomaly and figure out what the "correct" result would be.

Due to the length of time required for testing each GPU, updated drivers and game patches inevitably will come out that can impact performance. We periodically retest a few sample cards to verify our results are still valid, and if not, we go through and retest the affected game(s) and GPU(s). We may also add games to our test suite over the coming year, if one comes out that is popular and conducive to testing — see our what makes a good game benchmark for our selection criteria.

GPU Benchmarks: Individual Game Charts

The above tables provide a summary of performance, but for those that want to see the individual game charts, for both the standard and ray tracing test suites, we've got those as well. We're only including more recent GPUs in these charts, as otherwise things get very messy. These are also using a newer test suite and a 13900K CPU, which changes the performance slightly from the above table, simply because our newest tests are more relevant (but haven't been run on a lot of the older GPUs shown in the tables).

These charts were up to date as of November 11, 2024.

GPU Benchmarks — 1080p Medium

GPU Benchmarks — 1080p Ultra

GPU Benchmarks — 1440p Ultra

GPU Benchmarks — 4K Ultra

GPU Benchmarks — Power, Clocks, and Temperatures

Most of our discussion has focused on performance, but for those interested in power and other aspects of the GPUs, here are the appropriate charts.

Our even older 2020–2021 'legacy' GPU benchmarks used another set of hardware with different games. Here are the details for the Coffee Lake 9900K and Z390 PC.

Tom's Hardware 2020–2021 GPU Testbed

Intel Core i9-9900K
Corsair H150i Pro RGB
MSI MEG Z390 Ace
Corsair 2x16GB DDR4-3200
XPG SX8200 Pro 2TB
Windows 10 Pro (21H1)

The results below combine results from nine games with six resolution and setting combinations. All of the scores are combined (via a geometric mean calculation) into a single overall result, which tends to penalize the fastest and slowest GPUs — CPU bottlenecks come into play at 1080p medium, while VRAM limitations can kill performance at 4K ultra.

These results have not been updated since early 2022, when we added the RTX 3050 and RX 6500 XT to the list. We won't be adding future GPUs to this table, so there's no RTX 40-series, RX 7000-series, Arc, 3090 Ti, 6950 XT, 6750 XT, or 6650 XT, but it does help to provide a look at a slightly less demanding suite of games, where 6GB or more VRAM isn't generally required at 1080p ultra settings. You can use these older results to help inform your purchase decisions, if you don't typically run the latest games at maxed out settings.

Ancient Legacy GPU Hierarchy

Below is our legacy desktop GPU hierarchy dating back to the late 1990s. We have not tested most of these cards in many years, driver support has ended on most models, and the relative rankings are pretty coarse. Note that we also don't factor in memory bandwidth or features like AMD's Infinity Cache or Nvidia's larger L2 cache on Ada Lovelace. The list below is mostly intended to show relative performance between architectures from a similar time period.

We sorted the table by the theoretical GFLOPS, though on architectures that don't support unified shaders, we only have data for "Gops/s" (giga operations per second). That's GeForce 7 and Radeon X1000 and earlier — basically anything from before 2007. We've put an asterisk (*) next to the GPU names for those cards, and they comprise the latter part of the table. Comparing pre-2007 GPUs against each other should be relatively meaningful, but trying to compare those older GPUs against newer GPUs gets a bit convoluted.

These results are, at best, merely theoretical and we don't have any recent benchmarks for most of the GPUs. As one recent example, AMD's RX 7900 GRE ranks above the RTX 4070 Ti Super, even though Nvidia's card nearly matches the RX 7900 XT in rasterization performance and easily beats even the 7900 XTX in ray tracing performance. Take the following with a healthy dose of skepticism and a liberal sprinkling of salt, in other words, but it does contain a list of just about every major desktop GPU from the past 25 years.