Tom’s Hardware has tested dozens of processors from AMD and Intel to find the best CPU for gaming. Our list of 2026 CPU gaming benchmarks currently comprises 17 of the most demanding titles available on the market, which we run each gaming processor through to see the chips that come out on top. We select our picks based on the data in our CPU benchmark hierarchy, so all of the CPUs below are backed by hundreds of hours of real-world, hands-on testing where we gather extensive data on how a CPU performs and behaves while gaming. If you want a broader look at the CPU market, our AMD vs. Intel article shows you where the current CPU duopoly stands, while our CPU buying guide can help you narrow down the best processor for you.

Now in the back half of the year, we don’t expect major new releases from AMD or Intel. AMD has been on a tear with refreshes, particularly among X3D CPUs. We recently reviewed the Ryzen 9 9950X3D2 Dual Edition, AMD’s first CPU with 3D V-Cache on both CCDs, and it’s the most powerful chip from Team Red currently available. It’s earned a spot on this list, though the (much cheaper) Ryzen 9 9950X3D isn’t far behind in overall performance.

We’ve also seen the Ryzen 7 9850X3D, which is technically the fastest gaming processor on the market, as you can see in our Ryzen 7 9850X3D review. However, we’ve kept the base Ryzen 7 9800X3D as our top recommendation for the best CPU for gaming due to its price. It’s only marginally behind the refreshed model (about 3% on average), and much cheaper. For most gamers, the Ryzen 7 9800X3D makes more sense.

Intel has seen a recent boost in gaming performance with Arrow Lake Refresh, and both the Core Ultra 5 250K Plus and Core Ultra 7 270K Plus have earned spots on our list. AMD dominates in gaming at the moment, however. Our sights are set on Intel’s next-gen Nova Lake chips for a big gaming boost from Team Blue, as well as the rumored ‘Raptor Lake Next’ lineup that’s supposedly arriving early next year.

For the rest of the year, we have the Ryzen 7 7700X3D and Ryzen 7 5800X3D 10th Anniversary Edition to look forward to, which are arriving in July and June, respectively. We don’t anticipate they’ll make a major impact on our rankings here, but we plan on reviewing both CPUs as soon as they’re available.

In addition to the fastest CPUs from AMD and Intel, we’ve included a few DDR4 options on this list. The price of DRAM and NAND flash has made building even a budget PC prohibitively expensive, so DDR4 platforms are a great way to save money. Vendors are signaling a shift back toward DDR4 platforms at the moment, so we may reconsider some older CPUs for our rankings as pricing and availability allows.

Here are the gaming CPUs we recommend buying. We have a shortlist of the top options and some alternatives below, but you can click the ‘More’ links to read our thoughts about a particular CPU and where it stands in the current market.

Best CPU for Gaming in 2026 at a glance (more info below):

The list below is for the best CPUs for gaming, while the best budget CPUs can help you find a cheap chip. Processors benefit from the best thermal paste, so check out our guide if you're shopping for a new processor. But if you're after the best CPU for gaming, you're in the right place.

Best CPU for Gaming Benchmarks

We rank all the Intel and AMD processors based on our in-depth CPU benchmarks hierarchy. You can see some of those numbers in the charts above, including CPU overclock performance results (marked as PBO for AMD processors). We're currently retesting all of these processors with the Nvidia RTX 5090, but only the first four slides have that testing. The remainder are historical testing results with the RTX 4090, which we'll remove once we have fully retested all of the gaming CPUs with the RTX 5090 for our benchmarks. This group of results comprises only the chips that have passed through our newest test suite. Additionally, the tables in our CPU benchmark hierarchy include rankings based on past CPU benchmarks and breakdowns of single- and multi-threaded performance in productivity applications across a broad spate of processors. Finally, be aware that the pricing in the charts above can fluctuate.

Quick Shopping Tips

When choosing the best CPU for gaming in 2026, consider the following:

• You can't lose with AMD or Intel: As noted in our AMD vs. Intel feature, AMD tends to make the best all-around CPU for gaming for mainstream PCs lately, but both offer compelling performance options at any given price point.
• Eight cores is sufficient for gaming: If you’re looking at a pricey flagship, you’re likely wasting some money if gaming is your primary focus. You can game on as little as a quad-core CPU, but performance scaling really falls off past eight cores.
• Budget platform costs: You never want to pair a strong CPU with a weak GPU, RAM, and storage. Right now, it’s especially important to consider platform costs, however. DDR5 prices are peaking, and you’ll need to factor in the cost of DDR5 and a new motherboard if you’re coming from an older socket like AM4.
• Overclocking isn’t for everyone, but if you follow our How to Overclock a CPU guide, you can scrape out extra performance gains.

Best CPU for Gaming 2026 - $300 to $400

The Ryzen 7 9800X3D technically isn’t the fastest gaming chip on the market any more. That title goes to the new Ryzen 7 9850X3D, though the victory is marginal. As you can read in our Ryzen 7 9850X3D review, AMD’s latest X3D offering pushes ahead by 3.3% on average. Despite a minor uplift, we’re still recommending the Ryzen 7 9800X3D. With prices as they currently are, the Ryzen 7 9850X3D is only 3.3% faster despite costing around 6% more than the Ryzen 7 9800X3D.

This chip really has no peer in the market outside of the Ryzen 7 9850X3D — the Ryzen 7 9800X3D delivers outstanding gaming performance, beating Intel's fastest gaming chip, the $469 Core i9-14900K, by 30% in our test suite. The 9800X3D is also almost unbelievably 35% faster than the current-gen Intel flagship, the $560 Core Ultra 9 285K. The stock Ryzen 7 9800X3D's 1% low frame rates (a good smoothness indicator) also deliver an exceptionally smooth gaming experience, benefiting gamers even in GPU-limited scenarios.

The Ryzen 7 9800X3D has eight cores and 16 threads that operate at a 4.7 GHz base and 5.2 GHz boost clock rate. The chip employs AMD's 3D V-Cache tech with a new spin, which places a 3D-stacked SRAM chiplet underneath the die to deliver an incredible 96MB of L3 cache to great effect. AMD moved the L3 cache chiplet from the top to the bottom of the compute die this generation. That gives the integrated heat spreader (IHS) direct access to the compute die, allowing for more thermal headroom, and in turn, higher clock speeds. The end result is a comparatively low-power chip that delivers incredible gaming performance and comparable productivity performance to other eight-core models on the market.

3D V-Cache previously came with trade-offs in the productivity department, but that’s not the case with the Ryzen 7 9800X3D. Still, 3D V-Cache doesn’t provide a performance benefit in every game, and the performance benefit is less pronounced as your display resolution climbs.

The Ryzen 7 9800X3D has much lower power consumption than the Intel competition, making it a far cooler processor that won't require as expensive accommodations, like a beefy cooler, motherboard, and power supply. It also takes particularly well to undervolting, which is easy to accomplish with AMD’s Curve Optimizer. That means the 9800X3D delivers top-notch gaming performance and a cooler, quieter, and less expensive system than you'll get with an Ultra 9 or Core i9.

Read: AMD Ryzen 7 9800X3D Review

The $305 Ryzen 7 9700X had a rough initial product launch, but AMD's targeted firmware and operating system improvements have changed the picture tremendously, allowing the chip to place much higher on our CPU benchmark hierarchy (head there for the most up-to-date gaming benchmarks). Combined with lower-than-launch pricing, the Ryzen 7 9700X is a strong contender, tying Intel's Core i9-14900K in gaming and beating the Core i7-14700K. That's not to mention that it beats Intel's entire lineup of Arrow Lake processors as well. Now, all of those processors offer faster performance in heavily-threaded productivity applications than the 9700X, but when it comes to a pure gaming experience, the 9700X either ties or beats all current Intel competitors.

The Ryzen 7 9700X has eight Zen 5 cores with 16 threads that operate at a 3.8 GHz base and 5.5 GHz boost clock. The chip has a 65W TDP, though AMD retroactively added a 105W TDP option you can select in the BIOS that helps boost performance in productivity applications. It's covered by the warranty, as well. With either setting, the 9700X has comparatively tame power consumption, so it is an easy chip to cool. You'll have to buy your own cooler for the processor, though.

The Ryzen 7 9700X drops into socket AM5 motherboards, and B-series motherboards make the most sense for this class of chip. B850 and B840 motherboards get AMD's latest chipset with features like mandatory PCIe 5.0 support on the top M.2 slot and better availability for features like Wi-Fi 7. However, the Ryzen 7 9700X will still work with the older B650 chipset if you can find a board on sale.

Read: Ryzen 5 9700X Review

The $350 Core Ultra 7 270K Plus performs like a flagship CPU, but it costs about half as much. In games, it narrowly outclasses the Core i7-14700K and offers a 2.4% boost over the competing Ryzen 7 9700X. AMD’s last-gen Ryzen 7 7800X3D still offers around a 10% boost over the Core Ultra 7 270K Plus, but it’s also around $50 to $80 more expensive depending on sales.

It’s a solid gaming CPU, and certainly a better recommendation than the Core i7-14700K given prices right now. Compared to the Ryzen 7 9700X, things are tighter. The Core Ultra 7 270K Plus gains an edge with productivity performance. Short of the 9950X, it’s at the top of our multithreaded performance rankings, more than doubling the performance of the Ryzen 7 9700X.

On the gaming front, it supports Intel’s new Binary Optimization Tool, which offers an average of an 8% improvement in gaming performance based on our testing. It’s only available in a limited number of games at the moment, but Intel says it plans to support the feature with updates in the future.

For specs, the Core Ultra 7 270K Plus is close to the 285K. It comes with 24 cores and threads, split across eight Lion Cove P-cores and 16 Darkmont E-cores. The P-cores boost up to 5.4 GHz and the E-cores can climb to 4.7 GHz. Across the CPU, you get a total of 76 MB of combined L2 and L3 cache. It comes with a 125W TDP and 250W MTP. Critically, the Core Ultra 270K Plus also comes with a 900 MHz boost in die-to-die frequency and 400 MHz boost in fabric frequency compared to stock Arrow Lake chips.

The Core Ultra 7 270K Plus slots into existing 800-series motherboards with the LGA 1851 socket. This is an unlocked chip, so if you want to get the full benefits of overclocking, you’ll need a Z890 board. However, it’ll still work with H- and B-series motherboards, just without CPU overclocking support.

Read: Core Ultra 7 270K Plus Review

Mid-Range Best CPU for Gaming - $200 to $300

Intel has returned to gaming prominence with its Arrow Lake Refresh CPUs, and nowhere is that clearer than with the $220 Core Ultra 5 250K Plus. It’s priced like a budget CPU at $220, but it can perform as well (and sometimes even better) than chips that cost twice as much. It doesn’t dominate the gaming charts in the same way as AMD’s X3D offerings, but at this price, it doesn’t need to. It offers marginally better performance than AMD’s competing six-core Ryzen 5 9600X in games while running the tables with application performance.

On average, the Core Ultra 5 250K Plus is 1% faster than the Ryzen 5 9600X at 1080p, and 9% faster than the 245K. It’s functionally identical, but Intel’s new iBOT feature allows the chip to hold some solid leads in certain titles. For instance, it’s 10% ahead of the 9600X in Cyberpunk 2077. Even in a non-iBOT title like Doom: The Dark Ages, the 250K Plus leads by 12%. There are still some games that struggle with the unique Arrow Lake architecture like F1 2024, but the losses are less pronounced with the souped-up Arrow Lake Refresh chips compared to the stock offerings.

The application performance is what really stands out with the 250K Plus, however. With 18 cores, it outpaces the Core i7-13700K, nearly matches the Core i7-14700K, and more than doubles the performance of the Ryzen 5 9600X in multithreaded applications. In single-threaded applications, it beats the Ryzen 5 9600X by 6%.

Although you get 18 cores, they’re split between six Lion Cove performance cores and 12 Darkmont efficient cores. The P-cores climb up to 5.3 GHz, while the E-cores top out at 4.6 GHz. The CPU comes with a combined 60 MB of L2 and L3 cache, along with a TDP of 125W and a MTP of 159W. Like all Arrow Lake chips, it doesn’t support Hyper-Threading, so you get 18 total threads.

The Core Ultra 5 250K Plus slots into existing motherboards with the LGA 1851 socket. It’s unlocked for overclocking, so a Z-series motherboard is an ideal pairing. However, Intel increased the die-to-die frequency and the fabric frequency out of the box, and you’ll see those improvements in action on B- and H-series motherboards, as well. It’s locked to DDR5 memory, unlike Raptor Lake and Alder Lake platforms, and it officially supports speeds up to 7200MT/s.

Read: Core Ultra 5 250K Plus Review

The $230 Ryzen 5 7600X3D is currently the best value gaming CPU you can get right now, though it trades performance in other areas to reach that status. It's just 4.5% slower than the Ryzen 7 7800X3D based on our testing, giving you most of the performance of AMD's coveted 3D V-Cache in games without the extra cost.

In games, it outclasses more expensive CPUs with ease, including the Ryzen 7 9700X, and averaged just 65W of power draw during our gaming tests. Outside of games, however, the Ryzen 5 7600 X3D struggles. The Core Ultra 5 250K Plus is more than twice as fast in multithreaded performance, and in single-threaded performance, even the base Ryzen 5 7600X is around 13% faster.

The lagging productivity performance makes sense. The Ryzen 5 7600X3D is a six-core / 12-thread chip, so it has limited multithreaded potential, and it only clocks up to 4.7 GHz. The limited specs give AMD room to cram 102 MB of combined L2/L3 cache on the die, however, which comes with a sizeable boost in gaming performance. Compared to the base Ryzen 5 7600X, the X3D version is 22% faster despite coming in at lower peak clocks and power draw.

You can slot the Ryzen 5 7600X3D into socket AM5, which is available on 600- and 800-series motherboards, though the latter may require a BIOS update. Memory and CPU overclocking is available on both B- and X-series chipsets; however, the Ryzen 5 7600X3D has a locked multiplier, so the only overclocking you can access is through AMD's Precision Boost Overdrive, or PBO.

Read: Ryzen 5 7600X3D review

Highest Performance Best CPU for Gaming - $400+

How do you improve upon a CPU that already claims a dominating position in gaming and productivity workloads? You add more cache, of course. The Ryzen 9 9950X3D2 is powerful, expensive, and hungry for wattage, but it’s the best of the best if you want top-shelf gaming and application performance. It throws value out the window, and it’s only marginally better than the Ryzen 9 9950X3D, but it is still better.

Based on our testing, it’s about 3.9% ahead of the Ryzen 9 9950X3D in multithreaded applications, and in lockstep in gaming at 1080p. Compared to Intel’s Core Ultra 7 270K Plus, the Ryzen 9 9950X3D2 is 9% ahead in multithreaded performance and 23% ahead in average gaming performance. The Ryzen 7 9800X3D delivers a better value on the gaming front, and the Core Ultra 7 270K Plus is a monster productivity chip at a third of the price of the 9950X3D2. But the magic trick of this chip is that it can do both without breaking a sweat.

Under the hood, the Ryzen 9 9950X3D2 is similar to the Ryzen 9 9950X3D. It’s a 16-core / 32-thread chip packing AMD’s Zen 5 architecture, and it tops out with a 5.6 GHz boost clock; just 100MHz behind the 9950X3D. As the name suggests, this processor is unique because it uses AMD’s 3D V-Cache on both CCDs. Both eight-core CCDs have 32 MB of onboard cache, plus an additional 64 MB chunk placed under the cores, giving you a total of 192 MB of L3 cache.

The extra cache slightly accelerates multithreaded performance overall, though only by around 4%. There are specific workloads where the advantage is more present, with some data science workloads showing performance gains in the realm of 26% over the Ryzen 9 9950X3D. Those specific workstation-class workloads are where the Ryzen 9 9950X3D2 earns its stripes.

Otherwise, it’s the chip to buy because you simply want the best, no matter what the cost or how marginal the improvements are. It slots into existing AM5 motherboards, and it’s best suited for newer 800-series chipsets. AMD officially supports memory speeds up to DDR5-5600, though we find that DDR5-6000 is the sweet spot for Zen 5 CPUs.

Prices have dropped since release, though the 9950X3D2 is still expensive. It launched at $1,000, but you can find the chip for around $900 now.

Read: AMD Ryzen 9 9950X3D2 Review

The prior-gen $469 Core i9-14900K is now selling for all-time low pricing, primarily because the newer $560 Core Ultra 9 285K has arrived to take its place. However, the Core Ultra 9 285K is actually slower than the 14900K in gaming, so it isn't a suitable replacement. The new Core Ultra 7 270K Plus is marginally slower based on our testing, as well. Even in the face of Intel’s 200S Boost update, which was meant improve gaming performance, the competitive landscape remains unchanged. In our testing, the Core Ultra 9 285K gained an average of 7% from the update, which means it’s still slower than the Core i9-14900K.

You should be aware that the much more economically-priced 14700K (listed above) is only 2% slower than the 14900K in gaming but costs over $100 less. The Ryzen 7 9700X, also listed above, is also less expensive and effectively ties the 14900K in gaming.

However, there are Intel fans willing to pay extra for the absolute most gaming performance they can get from an Intel platform. Also, the 14900K does offer more multi-threaded horsepower than the 14700K and 9700X, which could be useful if you game, stream, and record simultaneously or do other heavy multi-tasking while gaming. Just make sure that your use case justifies the extra cost.

The 14900K sports leading-edge connectivity, supporting DDR4-3200 or up to DDR5-5600 memory, along with 16 lanes of PCIe 5.0 and an additional four lanes of PCIe 4.0 from the chip for M.2 SSDs.

The chip comes with eight P-cores that support Hyper-Threading and 16 single-threaded E-cores for a total of 32 threads. The P-cores have a 3.2 GHz base, and peak frequencies reach an amazing 6.0 GHz with Turbo Boost Max 3.0 (this feature is only active on P-cores). Meanwhile, the E-cores have a 2.4 GHz base and stretch up to 4.4 GHz via the standard Turbo Boost 2.0 algorithms. The chip also has 36MB of L3 cache and 32MB of L2.

This 14900K has a 125W PBP (base) and 253W MTP (peak) power rating, but we recorded considerably lower power consumption than its prior-gen counterpart. You'll need to buy a capable cooler for the chip, and you'll also need either a 700-series or 600-series motherboard. Like other Raptor Lake Refresh chips, you can find DDR4 and DDR5 motherboards, though you’ll need to go with a DDR5 board for the highest performance.

The lower price of DDR4 might entice some gamers, but you'll lose anywhere from 5-8% of gaming performance with higher-end Intel chips. You can step up to the much more expensive DDR5 if you need access to more memory throughput and, thus, every bit of performance possible.

Beyond specs, the Core i9-14900K was at the center of a years-long controversy concerning instability. An error in the microcode (CPU firmware) meant the Core i9-14900K would degrade faster than expected, starting with instability in games. Intel has rectified the issue with microcode 0x12F, so make sure you update your BIOS immediately if you pick up Intel’s last-gen flagship

Read: Intel Core i9-14900K Review

Best Budget CPU Pick - $100 to $150

The $164 Ryzen 5 7600X is an attractive budget CPU at its new price, forced down by Intel's new Arrow Lake Refresh chips. It’s marginally slower than the Ryzen 5 9600X, but also marginally cheaper – the Ryzen 5 7600X offers about 90% of the performance of the Ryzen 5 9600X for 94% of the price. It’s a slightly worse value, but it’s still a good option to keep in mind, especially if you find it on sale. The Ryzen 5 7600, sans X, is available at around the same price. We’ve yet to see it drop below the Ryzen 5 7600X, however.

With the 7600X, you get six cores and 12 threads based on the Zen 4 architecture, clocked at 4.7GHz with boost speeds up to 5.3GHz. Unlike the Ryzen 5 9600X, the Zen 4-based version comes with a TDP of 105W. Cooling it shouldn’t be an issue, and you’re free to run in AMD’s 65W Eco mode through the Ryzen Master software.

The Ryzen 5 7600X slots into AM5 motherboards, including 600- and 800-series chipsets, and it supports PCIe 5.0. DDR5 is required, which is a tough pill to swallow at this bang-for-your-buck price point, but it’s hard to avoid soaring RAM prices.

Read: Ryzen 5 7600X review

The $135 AMD Ryzen 5 5600 delivers a solid blend of performance in both gaming and productivity applications, bringing a new level of value to the Zen 3 lineup. If you're fine sticking with a previous-gen AM4 motherboard, the Ryzen 5 5600 makes a great base for a budget build. The primary trade-off for the AM4 platform is that you're limited to DDR4, and you don't have access to PCIe 5.0. You also have a limited runway for upgrades, as the fastest gaming CPUs on AM4 – the Ryzen 7 5700X3D and Ryzen 7 5800X3D – have reached end of life.

The Ryzen 5 5600 also makes an absolutely unbeatable budget chip if you're updating a first-gen Ryzen system. The 5600 unseats the Ryzen 5 5600X, a long-time favorite. The 5600X is only a mostly imperceptible ~1% faster in gaming and multi-threaded PC work than the non-X model, but provides a 4% advantage in single-threaded work.

Our testing shows that the Ryzen 5 5600 generally matches the gaming performance of its more expensive sibling, the ~$230 Ryzen 7 5800X. That makes the 5600 an incredibly well-rounded chip that can handle gaming well, from competitive-class performance with high refresh rate monitors to multi-tasking gaming workloads like streaming, while also serving up more than enough performance for day-to-day productivity apps. As with all AMD CPUs for gaming, you can fully overclock the chip.

The Ryzen 5 5600 has a 3.7 GHz base and 4.6 GHz boost clock. The chip also has a 65W TDP rating, so it runs cool and quiet. Existing AMD owners with a 500-series motherboard will be happy, as the 5600X drops right into existing 500-, 400-, and 300-series motherboards. If you need a new motherboard to support the chip, AMD's AM4 motherboards are plentiful and relatively affordable, with the B-series lineup offering the best overall value for this class of chip.

Prices for the Ryzen 5 5600 have drifted upward as stock depletes, but that’s offset by platform costs. In addition to low prices on AM4 motherboards, the Ryzen 5 5600 is limited to DDR4. High DDR5 prices are a significant roadblock to opting for a newer chip, as prices continue to surge. So it’s hard to recommend a newer budget CPU, even if it’ll net you higher performance. In the event you already have a kit of DDR5, the $189 Ryzen 5 7600 is a compelling option, and it includes AMD’s Wraith Stealth cooler.

Read: AMD Ryzen 5 5600 Review

Entry-Level Best CPU for Gaming - For gaming on integrated GPUs

The $268 Ryzen 7 8700G, AMD's flagship desktop APU, delivers the fastest socketed performance on the market from integrated graphics, bringing passable 1080p gaming to the desktop PC without a discrete graphics card, but its high price point relegates it to a niche audience.

In contrast, the $191 Ryzen 5 8600G delivers 90% of the 8700G's performance but for ~$80 less, making it a solid alternative for gaming systems that don't use a discrete GPU.

Naturally, you'll have to accept lower fidelity settings and be realistic about which titles can play at 1080p resolution. Still, AMD's Hyper-RX suite of features, which includes in-driver Radeon Super Resolution upscaling tech, frame generation with AMD Fluid Motion Frames (AFMF), Anti-Lag+, and Radeon Boost, helps boost performance at a slight cost to image quality. This new feature set, a first for AMD's iGPUs, is a boon for budget gamers.

The Ryzen 5 8600G has six Zen 4 CPU cores and the RDNA 3 GPU engine with eight CUs. The Ryzen 5 8600G drops into the AM5 platform, with value-focused B650 and A620 motherboards being the obvious best combination. These systems offer a new level of connectivity for AMD's APU processors, which were previously on the aging AM4 platform but require DDR5 memory. That adds some cost, so do a value analysis before selecting this processor. If you're looking for the lowest entry price possible with an APU, the Ryzen 5 5600G listed below slots in as the value alternative.

The Ryzen 7 8600G only supports 16 usable lanes of PCIe 4.0 connectivity, while other processors on the AM5 platform support PCIe 5.0. However, we don't feel this will impact this class of system.

More: AMD Ryzen 7 8700G and Ryzen 5 8600G Review

The Ryzen 5 5600G steps into the arena as the value champ for APUs, which are chips with strong enough integrated graphics that they don't require a discrete GPU for light gaming—just be sure you're willing to accept lowered quality settings.

The Ryzen 5 5600G gives you 96% of the gaming performance on integrated graphics than its more expensive sibling, the Ryzen 7 5700G, but for 25% less cash. Our testing shows that its level of performance makes it the best value APU on the market. As long as you're willing to sacrifice fidelity and resolution and keep your expectations in check, the Ryzen 5 5600G's Vega graphics have surprisingly good performance in gaming.

The 5600G's Vega graphics served up comparatively great 1280x720 gaming across numerous titles in our tests, but options become more restricted at 1080p. Of course, you can get away with 1080p gaming, but you'll need to severely limit the fidelity settings with most titles.

With eight cores and 16 threads that operate at a 3.9 GHz base and boost up to 4.4 GHz, the Ryzen 5 5600G also offers solid performance for its price point in standard desktop PC applications. The chip also comes with a bundled Wraith Stealth cooler, sweetening the value prop, and drops into existing 500-series and some 400-series motherboards, though support on the latter will vary by vendor.

The Ryzen 5 5600G is nearly five years old, and as a result, it’s hard to find it in stock at a reasonable price. AMD updated this model with the Ryzen 5 5600GT in early 2024, which features identical silicon and a slight boost to clock speed, and you’ll generally find it for less at around $150.

If your budget is tight and you're looking to build a system for modest gaming, you should check out our Best Cheap CPU feature. Some of those chips can deliver passable gaming performance without a graphics card, and their prices start at just $55 (£40).

Read: AMD Ryzen 5 5600G Review

• MORE: CPU Benchmark Hierarchy
• MORE: AMD vs Intel
• MORE: Best Cheap CPUs
• MORE: Best CPUs for Workstations

A Chinese techtuber has designed a liquid-cooled PC system that hooks up to a household air conditioner unit instead of a PC-mounted radiator. The video creator, oddly named 'Electrolytic sodium carbonate' (ESC), showcases a rather normal-looking PC tower - but next to it, there is a 12,000 BTU air conditioning unit to supercharge cooling. The result is that the GPU runs at just 20 degrees Celsius under stress testing.

This project is ready to be the "perfect companion" (machine translation) for an upcoming Core i9-14900K with Nvidia GeForce RTX 5090 system, says ESC. However, as they are still waiting for a water block release for their graphics card these first tests have been completed using a Core i9-13900K PC system fitted with an RTX 4090.

This isn't ESC's first work in integrating powerful household air conditioning units with personal computer systems. The video host shows an established 'whole house' liquid cooling system that appears to have five loops available for PCs. We have pictures of the inside and outside parts of this cooling setup in the gallery.

Next up, the ESC video host talks about the increased TDP of the RTX 5090, which moves the power needle a not insignificant amount beyond the expected load of an RTX 4090. We are then introduced to the Xiaomi KFR-35GW air conditioner, which gets a nice bit of Nvidia – AMD – Intel branding, though we are not aware of any AMD components in the test system. This AC unit offers a quoted 12,000 BTU cooling power and typically consumes about 1.2kW. It measures 765 x 268 x 550mm, so is quite a lot bigger than the PC tower it is cooling, and a lot bigger than a 360mm PC cooling radiator.

We are then guided through the modification of the Xiaomi KFR-35GW. Some welding and pipe management later, and we see the liquid cooler reservoir ready inside the AC unit.

How well does this AC unit cool the test PC? After 40 minutes of "roasting" this PC by stress testing it with popular tools such as FurMark and the AIDA System Stability Test the system CPU cores are said to be only a couple of degrees hotter than when the test started.

ESC then took the AC unit outdoors, where it would normally be installed, and noted that the GPU temperature was just 2 degrees Celsius, with a 12 degrees Celsius hotspot. Running stress tests again saw these GPU temperatures rise to 20 and 36 degrees Celsius, respectively. Sadly, the video doesn't include charts for inside and outdoors, CPU and GPU, idle and load – those would have been very interesting.

During the video the host says that they are happy with the capability of the air conditioner-cooled PC as demonstrated, and believe it is ready for both the 14900K and RTX 5090 components, which are waiting in the wings, and any upcoming RTX 6090.

Intel’s Core i9-13900K desktop processor has hit its lowest price on Amazon for Black Friday. This processor launched as the Raptor Lake desktop flagship CPU in October 2022 and remains a potent force two years later. The big difference is that at $399.61, it is now half the money people would have had to splash out to grab one when it first hit retail, according to three camels.

So, what do you get with the Core i9-13900K? This CPU comes packing eight Raptor Cove cores doing duty as the performance cores (P-cores) plus 16 Gracemont efficiency cores (E-cores). Users thus benefit from 24 cores and 32 threads of processing power. P-core base/boost frequencies ranged from 3.0 to 5.8 GHz. Intel also fitted 68MB of total cache, and the CPU ran at 125W Base Turbo Power and up to 253W using Thermal Velocity Boost and Turbo Boost 3.0 tech.

The Core i9-13900K was crowned the fastest gaming chip in 2022. However, any recommendation had specific warnings about high power consumption and cooling requirements. We would later learn that Intel’s powerful Raptor Lake chips (and 14^th Gen refresh parts) used microcodes that elevated CPU voltage levels beyond safe limits. However, starting in September this year, Intel identified, outlined, and issued mitigations for the root cause of Raptor Lake / Refresh issues and failures dubbed Vmin Shift Instability.

An attractive quality of a CPU like the Core i9-13900K in 2024 is that it can be partnered with a wide range of existing socket LGA 1700 motherboards. You can also build a platform that uses either DDR4 or DDR5 RAM, which could save money for DIY desktop upgraders. It might be most attractive to those currently using an Alder Lake CPU or a lower-tier processor from the Raptor Lake (Refresh) family used as a stopgap.

However, this lowest-ever-priced Core i9-13900K might also appeal to those considering a new build. There are plenty of great Black Friday bargains for SSDs, Monitors, and more around now to assemble completely new PC systems capable of high-end gaming fun and accelerated content creation tasks.

Dylan Browne, an Unreal Engine Supervisor and Feature Film VFX at the ModelFarm visual effects studio, posted on X that his company is experiencing a 50% failure rate for systems powered by Intel's Core i9-13900K and 14900K processors. As a result, the company is deploying AMD's as-yet-unreleased Zen 5 Ryzen 9 9950X processors in place of Intel-powered solutions, with Browne praising AMD's single-thread performance.

The report represents yet another piece of unwelcome news for Intel, which announced yesterday that it had found the root cause of the issues and will issue a microcode mitigation in mid-August. (This isn't a 'fix' for CPUs experiencing the issue — impacted processors are irreversibly damaged and must be replaced.)

The problems with Unreal Engine aren't entirely unexpected, as early reports of the Intel crashes revolved around the Oodle compression used with the game engine. The news that ModelFarm is dropping Intel CPUs follows game studio Alderon Games' announcement that Intel desktop CPUs have a 100% crash rate, and laptop chips of the same generation are also affected. However, Intel has now disputed the claims of laptop failures.

Browne claimed that two brand-new processors immediately exhibited instability, while a few others took some time to exhibit symptoms. The computers were all focused on Unreal Engine work, which works best with multi-core systems.

Browne is "fairly sure" that most of the unstable systems use Asus ROG motherboards but will provide an update with more specifics later. However, the Intel chips exhibited instability even with lower power limits. Browne said the motherboards of the affected CPUs have already been tweaked, but that didn’t seem to help with the problems the Core i9-14900Ks and 13900Ks were experiencing.

Intel has acknowledged the issue and announced a solution to the instability problem yesterday. However, the microcode patch to address the problem isn’t expected to arrive until mid-August, so we’re unsure if this will truly stop the crashes. Nevertheless, businesses cannot wait that long for a system that doesn’t suffer from this issue, and Browne said that any new machines for the studio he’s working at will use AMD Ryzen 9950X chips.

The Ryzen 9000 series is expected to become available on store shelves on July 31, but some retailers have already added them to their online stores. Nevertheless, some organizations likely already have early samples of the chips for testing under NDA, and ModelFarm seems to be among them. 

When another X user asked the 9950X's single-threaded performance, Dylan said that chips were “Pretty damn good, especially the newer ones.” While he did not give specific details, we can extrapolate from his response that Browne is likely talking about the latest AMD processors. Browne responded with a follow-up Tweet to the question, saying that his “5950X is good for single-threaded [performance], and they get better every [generation], not to mention having all core being equal unlike most of Intel’s chips.”

Intel says its patch will address the instability issues in August. Regardless, it has a serious scandal on its hands that could put some extra wind in AMD’s sails right as it launches the Ryzen 9000 series at the end of July.

Last December, AMD launched the Radeon RX 7900 XTX and 7900 XT. At the time, we noted that the price gap was far too narrow to warrant getting the "lesser" card. Since then, online prices have dropped over $100, with the Sapphire RX 7900 XT Pulse we're looking at today being one of the prime options. The RX 7900 XT ranks as one of the best graphics cards, and Sapphire looks to improve on the reference card with a larger form factor and better cooling... or at least quieter cooling.

All of the core specs remain unchanged, so you get the same AMD RDNA 3 GPU architecture, with a healthy 20GB of GDDR6 20Gbps memory. That's 67% more VRAM than Nvidia's similarly priced RTX 4070 Ti. Performance, as you might expect, remains largely unchanged. That's true of most third-party graphics cards, since at most you typically get a 3–5 percent factory overclock. In the case of the RX 7900 XT Pulse, it's a 2.1% factory overclock, with a 5.1% increase in TBP (total board power).

Here's the quick overview of the specs table for the Sapphire card along with some competing GPUs.

We already noted the only differences in core specs between the Sapphire RX 7900 XT Pulse and the reference 7900 XT: GPU boost clock and TBP/TGP. Those are the paper specs, however, and there are other differences in terms of card design, aesthetics, and other aspects.

Right now, Sapphire is one of the three least expensive RX 7900 XT cards available, the others being the XFX RX 7900 XT that uses AMD's reference design — an MBA or "Made By AMD" card — and the ASRock RX 7900 XT Phantom Gaming OC. They all cost $779.99, though Sapphire and ASRock cards require an instant rebate at Newegg to get there.

Several other 7900 XT models are available for around $800, while the most expensive options right now can cost as much as an RX 7900 XTX (around $950). Unless you really love the aesthetic of one of those premium cards, we'd recommend stepping up to the faster AMD XTX GPU rather than paying $200 extra, but that's ultimately an individual choice.

Looking at the primary competition, the Nvidia RTX 4070 Ti only comes with 12GB of VRAM, but it also has a lower TGP and gives you the typical Nvidia extras: DLSS, Frame Generation, tensor cores, and improved ray tracing hardware. We'll get to the performance results shortly, but in our current test suite, it's the usual story: Similar overall native resolution performance, AMD leads in rasterization and Nvidia leads in ray tracing.

Sapphire RX 7900 XT Pulse

Sapphire's Pulse cards represent the base model designs from the company, with no extras like RGB lighting. You do get a support bracket for the card, though this sort of bracket tends to be far less user-friendly than the "kickstands" provided with some other GPUs. On the other hand, the support bracket screws into the case, so it won't shift around if you happen to carry your PC into a different room.

The Sapphire RX 7900 XT Pulse is a relatively large card at 313x134x53 mm. It's a triple-slot card for all intents and purposes — even if it's technically only a "2.7-slot" design, you can't really fit anything in that third overlapped slot. It weighs 1417g as well, which isn't too bad given the dimensions. For comparison, the reference AMD 7900 XT card measures 276x113x51.5 mm and weighs 1490g.

AMD's card is quite a bit more compact, but that also tends to impact cooling performance and noise levels, which is arguably the biggest draw for Sapphire's Pulse model. We'll get into the details on temperatures and noise later in the the review.

As you'd expect for a >300W graphics card, you get triple fans to go along with the large dimensions. Sapphire's latest designs use "angular velocity blade" fans, which are supposed to move more air and last longer than previous Sapphire designs. Sapphire notes that the inclusion of the outer rim helps to improve downward air pressure, by up to 44%, with 19% more airflow. The fans are also 95mm in diameter, compared to AMD's reference design that uses 78mm fans.

The radiator has six heatpipes to help carry heat away from the GPU core and VRAM, which again represents a typical design for graphics cards with a 300W or higher TBP rating. The only real improvement would be to opt for a full vapor chamber instead of heatpipes, and that's something the Made By AMD cards use.

Video connectivity consists of dual HDMI 2.1 ports and dual DisplayPort 2.1 UHBR13.5 ports. The HDMI ports have a peak bandwidth of 48 Gbps while the DP2.1 ports offer up 54 Gbps of bandwidth. Uncompressed resolution support for AMD's latest RDNA 3 GPUs tops out at 4K and 229 Hz on DP2.1, while DSC (Display Stream Compression) can provide "visually lossless" 4K at up to 480 Hz. Do note that we're nowhere near getting 480 Hz 4K displays, however, with the fastest options right now being 240 Hz models.

Sapphire RX 7900 XT Test Setup

Our graphics card test system was updated earlier this year to use Intel's fastest Core i9-13900K, with all the bells and whistles. We tested the Sapphire RX 7900 XT card using AMD's most recent 23.5.2 drivers, while the other AMD cards were tested with 23.3.2 through 23.5.1. All the cards in this review have been retested in the past month or two, without Windows 11's Virtualization Based Security enabled.

Our initial review of the RX 7900 XT looked at performance across all resolutions and settings that we normally test (1080p medium, 1080p ultra, 1440p ultra, and 4K ultra). It also had professional and AI benchmarks. For third-party cards like this Sapphire Pulse, we'll limit our testing to 1440p and 4K ultra, as that's the primary target for graphics cards that cost $800 or more. Results in other tests should have similar margins to the gaming results.

Our test PC includes an MSI Z790 DDR5 motherboard, G.Skill DDR5 memory, and a Sabrent 4TB SSD — which we promptly filled to about half its total capacity. be quiet! also supplied us with its new 1600W Dark Power Pro 13 80 Plus Titanium rated power supply. That would have been overkill, back before cards like the RTX 4090 were a thing.

We're including a collection of current and previous generation GPUs, with the Nvidia results from the RTX 40-series (but skipping the significantly more expensive RTX 4090). If you want to see how the RX 7900 XT stacks up against other GPUs, check out our GPU benchmarks hierarchy.

We have Nvidia PCAT v2 (Power Capture and Analysis Tool) hardware as well, which means we can grab real power use, GPU clocks, and more during all of our gaming benchmarks. We'll have most of the details for power testing in a few pages.

• MORE: Best Graphics Cards
• MORE: GPU Benchmarks and Hierarchy
• MORE: All Graphics Content

We have 1440p and 4K ultra benchmarks for this Sapphire RX 7900 XT review. Given the $799 typical price (with sales bringing it down to $779), we'll start with the 1440p ultra performance before moving on to the 4K gaming results.

Here's the overview, first with the 15-game full suite, and with separate rasterization-only and DXR-only (i.e. ray tracing) charts. We'll discuss the individual game results further down the page.

First, overall performance ends up exactly tied (a 0.1 fps difference) with the reference RX 7900 XT. It's interesting, because we've seen other tests of Sapphire's card where it's a couple of percent faster. That's what we'd normally expect, since it has a higher power limit and a minor factory overclock, but there's always a bit of variability between GPUs.

Maybe our sample card was a bit of a dud, or maybe our reference card is a bit better than typical. It's not drivers, either, as we previously tested the card with the same drivers that we used for our most recent reference GPU testing. Regardless, there's really nothing to worry about as no one would notice a less than 10% difference without running benchmarks.

The Sapphire card also lands between the RX 6950 XT and RTX 4070 Ti in performance, but here we need to look at the rasterization and ray tracing charts as well. AMD's GPU ends up slightly ahead of the 4070 Ti in rasterization performance with a 6% lead, but conversely ends up trailing by 19% in the DXR (DirectX Raytracing) test suite.

That's expected in general, and how much you want to weight rasterization versus ray tracing performance is a contentious subject. Some will argue ray tracing is just technical snake oil, others hail it as the next big thing for graphics. There are currently close to 150 games that have some form of ray tracing. That's the "good" news.

Subjectively, however, we'd say probably only fifteen or so rate as reasonably popular games where the RT effects make a noticeable difference. Not coincidentally, six of those games are in our test suite. We'd also say that good upscaling (DLSS, XeSS, or one of the better FSR2 implementations) often ends up being more useful than token ray tracing support.

Across our 15-game test suite, the Sapphire and reference 7900 XT cards are basically equivalent, with a spread of -2.2% to +4.3% for the Sapphire card. Forza Horizon 5 shows the biggest change, and that could be due to drivers or a recent game patch. Everything else is within margin of error.

Switching gears to the Sapphire versus RTX 4070 Ti, we'll split that up into rasterization and ray tracing games. The spread for rasterization ends up ranging from -7.2% (Total War: Warhammer 3) to +25.9% (Borderlands 3) in favor of Sapphire. DXR meanwhile is a clean sweep for Nvidia, with anywhere from a 10.3% (Metro Exodus Enhanced) to a 57.1% (Minecraft) lead for Nvidia. Not surprisingly, but the most demanding DXR games are the ones that favor Nvidia the most. Those are also the games where ray tracing makes the most noticeable visual difference.

• MORE: Best Graphics Cards
• MORE: GPU Benchmarks and Hierarchy
• MORE: All Graphics Content

The standings don't change much with our 4K ultra test results, though AMD does close the gap with the RTX 4070 Ti just a touch. Overall, Sapphire ends up 3.4% slower than the RTX 4070 Ti, with a slightly larger lead in rasterization performance and a slightly worse loss in DXR performance. Specifically, it's 10.6% faster in our rasterization suite, and 21.2% slower in the DXR suite.

If you ever want a good reason for upscaling algorithms, 4K gaming will suffice. Not only can  you get quite good results when doing a modest 2x upscaling (e.g. going from 2715x1527 to 3840x2160). That's because the more data you have, the 'easier' it is for AI and conventional algorithms to fill in the gaps.

Put another way, let's look at the extreme. Imagine a 2x upscale from 5x5 pixels to 7x7 pixels. The algorithm would need to interpolate from 25 total pixels to 49 total pixels, but there's simply not as much detail to start with. Conversely, when the target ends up being 8.3 million pixels interpolated from 4.1 pixels, there's a wealth of information available.

From a performance perspective, upscaling to hit 4K also has some great benefits. We explained why 4K requires so much more VRAM than 1440p or 1080p, and while games can do things in various ways, the end result is that rendering at 2K and then upscaling to 4K can result in some excellent performance gains with a relatively minimal loss in image fidelity. You're also not likely to be CPU limited when targeting 4K, which isn't always the case with upscaling to 1080p.

Native 4K performance in our rasterization suite ranges from 48.9 fps in A Plague Tale: Requiem to 107.1 fps in Far Cry 6. Only two of the games failed to average 60 fps or more (though Flight Simulator barely got there). But with the Sapphire RX 7900 XT, or the RTX 4070 Ti, every rasterization game is very much playable, particularly if you drop from ultra to high settings.

For our DXR suite, things aren't quite so nice. Only Spider-Man: Miles Morales and Metro Exodus Enhanced can even break 30 fps at native 4K, and Control nearly gets there with 29.5 fps. But the other three games are in the high teens or low 20s. Even with FSR2 upscaling, which only three of the six games support, getting to 60 fps might be a stretch.

You could of course use a different suite of games for testing to make the results look better, or use lower settings (particularly in the DXR games) to close the gap between AMD and Nvidia GPUs. Or you could enable upscaling where supported, in which case DLSS remains more widely used than FSR2. But ultimately, the RX 7900 XT and RTX 4070 Ti do deliver relatively similar overall experiences, even though AMD's card has 20GB while Nvidia's GPU only has 12GB of VRAM.

• MORE: Best Graphics Cards
• MORE: GPU Benchmarks and Hierarchy
• MORE: All Graphics Content

Our power, clocks, and temperature testing now utilizes the same test suite as our gaming benchmarks, as the PCAT v2 hardware and FrameView software lets us collect this data alongside frametimes. We're also using our updated Core i9-13900K platform, so we're less likely to have CPU or platform limitations playing a role.

We have 1440p and 4K charts for power, GPU clocks, and temperatures below. Then we do a test using Metro Exodus at whatever "demanding for the GPU being tested" means — 1440p ultra in the case of the Sapphire RX 7900 XT — and after letting the game run for 15 minutes or more, we check noise levels. This basically represents the worst-case workload and maximum noise, though in practice a lot of games will end up at similar levels.

We'll present additional tables and information about efficiency (FPS/W) and value (FPS/$) at the bottom of the page.

The Sapphire RX 7900 XT does as advertised and uses slightly more power across our test suite than the reference 7900 XT, at both 1440p and 4K. In both cases, the Sapphire card averaged 322W of power use, while the reference card used 307W and 308W. Nvidia's RTX 4070 Ti meanwhile needed just 246W on average at 1440p, and 262W at 4K.

This is an ongoing issue with AMD's RDNA 3 architecture. If you look at RDNA 2 and Ampere, the previous generation GPUs, AMD generally held an advantage in efficiency. Part of that was because Nvidia used Samsung 8N (a 10nm-class node) while AMD used TSMC N7. Now, they're basically on the same 5nm-class node (TSMC 4N and N5, for Nvidia and AMD, respectively). It seems like Nvidia's architecture is overall simply more efficient.

There's also the question of how much extra power AMD had to use by opting for GPU chiplets instead of a monolithic die. We may never know for certain, but certainly there's room for improvement from AMD. Sapphire for its part managed to provide equivalent performance while drawing more power, which definitely isn't the best way of doing things.

GPU clock speeds on their own don't mean too much, unless you're comparing within the same architecture and even GPU. And that's where things get a bit interesting with the Sapphire RX 7900 XT. On paper, it has a higher 2450 MHz boost clock while the reference card 'only' has a 2400 MHz boost clock. But we know from experience that AMD's latest generation GPUs are likely to exceed that conservative boost clock.

With the reference RX 7900 XT, across our 15-game suite, we measured average GPU clocks of 2573 MHz — 173 MHz above the advertised boost clock. The Sapphire model meanwhile only averaged 2488 MHz, 38 MHz above its advertised speed. That's a 1440p, incidentally. The clocks drop quite at bit at 4K, where the reference card still averaged 2494 MHz, but the Sapphire card dropped to just 2290 MHz.

Looking at these results, we can't help but wonder if our Sapphire card was a bit of a lemon. We've since returned the card to the company, but other sites' results don't seem to corroborate with our results. There's a good chance a different Pulse card would have performed better than what we've shown here.

Finally, we have to look at temperatures and noise levels as a linked pair. Higher fan speeds can drop noise levels, and conversely lower fan speeds can result in higher temperatures but less noise. There's a balancing act that needs to be maintained, and generally speaking GPU core temperatures of less than 80C aren't a concern.

Looking just at the thermals, the Sapphire RX 7900 XT ran about 3–5C hotter than the reference card. Neither one averaged more than 70C, so temperatures aren't a concern, but we can only get a better idea of how the cards compare by also looking at fan speeds and/or noise levels.

We check noise levels using an SPL (sound pressure level) meter placed 10cm from the card, with the mic aimed right at the center of the middle fan (on the Sapphire card). This helps minimize the impact of other noise sources like the fans on the CPU cooler. The noise floor of our test environment and equipment is less than 32 dB(A).

After running Metro Exodus for over 15 minutes, the Sapphire RX 7900 XT settled at a fan speed of 38% with a noise level of 40.0 dB(A). While not the quietest GPU we've tested — some of the really large cards run at 37–38 dB(A) — it's equal to or better than most other competing GPUs. The reference AMD 7900 XT on the other hand had a 48.7 dB(A) result, with the fans running at 63%.

There's no question in my mind that the Sapphire cooling solution ends up performing better overall, even if the actual temperatures are slightly higher. The reference cards get relatively loud, while Sapphire's fans and cooler generally do a better job while keeping noise levels in check.

We also tested with a static fan speed of 75%, which caused the Sapphire RX 7900 XT to generate 56.2 dB(A) of noise. That's similar to a lot of other GPUs, but not hugely important as you shouldn't normally see the fans spinning at more than 50% or so.

GPU Value and Efficiency

This final gallery of images shows the full performance test suite, along with the above power, clocks, and temperature information. Latency is also provided, at least in some of the games (depending on the GPU and drivers used).

We've calculated efficiency in FPS/W for the various games, plus value in FPS/$ using the best current online prices we could find (usually at Newegg or Amazon, though B&H and Best Buy were also checked). We've summarized those results in the above table (based on 1440p performance and power), sorted by overall value.

As expected, even though cards like the RTX 4060 Ti have a (perhaps deservedly) bad rap, their lower prices still make them a "better value" than the more expensive GPUs. It's the typical diminishing returns in practice. However, we should note that we're only factoring in the cost of the graphics card. If you were to include a decent high-end PC, that changes the equation a lot. That's what the "PC FPS/$" column at the far right.

For the system cost, we used a Core i7-13700K, Arctic Freezer II 240 cooler, ASRock Z790 PG Lightning motherboard, 32GB G.Skill DDR5-6400 CL32 memory, Solidigm P44 Pro 2TB SSD, Phanteks Eclipse P400A case, and Phanteks Revolt Pro 850W gold PSU. Combined, those cost $1,074 at the time of writing. For a complete PC, the "best value" (in quotes because there's an element of subjectivity involved) switches to the most expensive RTX 4080 first, then the RTX 4070 Ti, followed by the RX 7900 XTX and RX 7900 XT. Food for thought if you're planning a complete system update!

Efficiency also puts the RTX 4080 in first, followed by the RTX 4070, RTX 4060 Ti, RTX 4070 Ti, and then finally the RX 7900 XT (reference card). Sapphire's GPU ranks two steps lower, just behind the RX 7900 XTX, thanks to its higher power draw. There's still a decent gap in efficiency between it and the RX 6800, however, which was the highest efficiency GPU from AMD's RX 6000-series.

• MORE: Best Graphics Cards
• MORE: GPU Benchmarks and Hierarchy
• MORE: All Graphics Content

The Sapphire RX 7900 XT Pulse is a bit of a curiosity, as far as our test results go. On paper, it should be faster than the reference 7900 XT. Other sites show it as being a few percent faster. Our particular sample, though? Not quite so great, with higher power draw and at best equivalent performance. We suspect most Pulse cards will fare slightly better, as it's a bit surprising that the card didn't clearly beat the reference design in performance.

It also used about 5% more power on average while delivering equivalent performance, and temperatures were a few degrees Celsius higher under load. But the good news is that the fans and cooling are clearly superior, with noise levels peaking at 40 dB(A) compared to the reference card that gets noticeably louder at 49 dB(A).

If you're looking for an inexpensive RX 7900 XT and you don't care about RGB lighting or flashy designs, the Sapphire Pulse is a great option. If you prefer something with more bling, the ASRock Phantom Gaming OC has some RGB to brighten your day, for the same $779 asking price. And that's probably the best news overall: RX 7900 XT cards can now be had for $120 less than the original (too high) launch MSRP.

That price change was pretty much required, as performance in general ends up being roughly equivalent to the RTX 4070 Ti. Rasterized performance at native rendering is faster, but ray tracing is slower, and DLSS still provides better quality upscaling, in more games, than FSR 2.

AMD does provide you with a lot more VRAM, 20GB versus Nvidia's competing card with just 12GB. If you're worried about games starting to push into requiring 16GB or more VRAM, it's potentially a safer bet. Or you can just recognize that HD texture packs often do very little for image fidelity and stick with settings that won't exceed 12GB (unless a game is poorly coded, which unfortunately seems to be the case with a lot of recent ports).

Which of these high-end cards ends up being the better pick will ultimately come down to the games you want to play, and what settings you want to use. If you want all the bells and whistles turned on, you might think that the 20GB VRAM trumps 12GB, and it does in some cases. But then you also have to consider that ray tracing will almost always favor Nvidia's RTX 4070 Ti, sometimes by a lot, even if it doesn't have as much VRAM.

We're also still waiting for AMD's lower cost RX 7800 XT and RX 7700 XT GPUs to launch. That will likely happen in the next month or two, though we're not sure on pricing just yet. We're also not sure where performance will land, and it could end up that a $600 RX 7800 XT will only roughly match the existing $600-ish RX 6950 XT.

This generation of GPUs increasingly feels like a step sideways from AMD, moving to GPU chiplets and prepping for future generation architectures while cutting costs, rather than a significant step forward in performance. Yes, the top RX 7900 XTX and 7900 XT are faster than the previous generation, but we could see diminishing gains further down the RX 7000-series product stack.

For top-shelf performance from AMD, the RX 7900 XTX can still beat the RX 7900 XT by 18% at 4K ultra. That's a pretty big jump, accompanied by a 21% increase in price (for the least expensive cards right now). The Sapphire RX 7900 XT Pulse meanwhile also beats the previous generation RX 6950 XT by 20% at 4K, for a 31% increase in price (using current online prices). Prices are fluctuating quite a lot right now, likely due to the impending Navi 32 launch, so keep that in mind.

The Sapphire RX 7900 XT ultimately ends up as a fine option for those shopping in the upper range of AMD GPUs. It's not exceptional, but it's better than some of the other alternatives, depending on your perspective. If you're planning to sit this generation out, it will probably be another 16–18 months before we start seeing the next generation GPUs that might have something more to say about performance, hopefully at better prices (though we wouldn't count on that).

• MORE: Best Graphics Cards
• MORE: GPU Benchmarks and Hierarchy
• MORE: All Graphics Content

We showed benchmarks of the PNY GeForce RTX 4070 Verto Dual in our initial Nvidia RTX 4070 Founders Edition launch review. Then a bunch of other stuff came along, like the RTX 4060 Ti and RX 7600 launches, but now we're getting around to various third-party reviews. These are some of the best graphics cards, for gaming and other purposes, and the RTX 4070 looks to hit the sweet spot in terms of specs and pricing.

At its core, the RTX 4070 performs a lot like the previous generation RTX 3080, despite having 2GB more VRAM but a much narrower 192-bit memory interface. That's thanks to the Nvidia Ada Lovelace architecture, and specifically the inclusion of up to 8MB of L2 cache per 32-bit chunk of memory interface. The additional memory can help in a few instances, but higher resolutions also tend to prefer the increased raw bandwidth of the 3080's 320-bit interface.

Let's start with the usual specs table before getting into the specifics of the PNY card.

The above specs table shows the ten GPUs we'll be using for this review. The PNY RTX 4070 Verto Dual and Nvidia RTX 4070 Founders Edition have identical specs, since both use the reference clocks. There's not much more we can add.

The RTX 4070 lands at a $599.99 MSRP, and there are even cards periodically available for less than that — like right now, you can pick up the RTX 4070 from Asus, Gigabyte, MSI, or Zotac with a $50 instant rebate using code "ZIPDADS" (for Fathers' Day, if that wasn't obvious). PNY's card unfortunately isn't part of that deal. It's also "while funds last," so you may or may not be able to take advantage of that offer.

That sort of deal does speak volumes about graphics card sales in general. The past several Black Fridays and Prime Days, for example, had very few if any noteworthy GPU deals, but that's a straight $50 discount from Newegg on a graphics card that released less than two months ago. It's a safe bet that Nvidia could also drop the prices on all of its GPUs $50 or so without incurring too much pain.

Short-term sales aside, we can't help but feel the RTX 4070 should be selling for a lot less than its MSRP. It's faster than the previous generation RTX 3070, but it also comes with a substantially smaller die size. 294.5 mm^2 on TSMC 4N compared to 392.5 mm^2 on Samsung 8N likely means the AD104 chips do cost more, but Nvidia should still get about 190 chips per wafer. Even at $19,000 per 4N wafer, that's only $100 per chip (and Nvidia probably pays less than that due to the number of wafers it orders).

PNY GeForce RTX 4070 Verto Dual

PNY uses a nondescript box for its base model RTX 4070. That's fine, as most people are far more interested in the performance and graphics card than in the packaging. Inside the box sits the card in an anti-static bag... and that's pretty much it.

There's no support stand here, which, considering the card weighs 687g, shouldn't be a problem. The card's dimensions are 247x120x40 mm. Nvidia's RTX 4070 Founders Edition by way of comparison measures 244x111x40, so it's just a touch shorter in both length and height, though it's quite a bit heavier at 1021g.

PNY uses a dual-fan configuration, as implied by the model name. Each fan is 85mm in diameter, so smaller than some cards. More importantly, there's no integrated rim to improve static pressure and airflow. That gives Nvidia's reference card a double advantage: Its 91mm fans have more area, and they include a rim around the outside diameter.

The radiator also carries this same approach of not going hog-wild on cooling capacity. There are plenty of heatsink fins, but only three heatpipes running through them. The RTX 4070 only has a 200W TGP (Total Graphics Power), so the cooling should be sufficient, but as we'll see later, this modest setup ultimately results in higher fan speeds and noise levels.

As with most recent Nvidia GPUs, you get three DisplayPort 1.4a outputs and a single HDMI 2.1 port. With DSC (Display Stream Compression), you can get "visually lossless" 4K at 240Hz. Some might quibble about how good the image truly looks, but having tested it, we don't have any complaints — and most of the time, at 4K you won't be pushing framerates anywhere near the 240 Hz maximum refresh rate.

PNY RTX 4070 Test Setup

Our graphics card test system was updated earlier this year to use Intel's fastest Core i9-13900K, with all the bells and whistles. We tested the PNY RTX 4070 card using Nvidia's 531.42 launch drivers, though as far as we can tell nothing substantial has changed with the latest 535.98 that came out on May 30. All of the cards in this review have been retested in the past month or two, without Windows 11's Virtualization Based Security enabled.

Our initial review of the RTX 4070 looked at performance across all resolutions and settings that we normally test (1080p medium, 1080p ultra, 1440p ultra, and 4K ultra). It also had professional and AI benchmarks, and performance results with upscaling including DLSS 3 (see how to enable DLSS 3). For third-party cards like this PNY RTX 4070, we'll limit our testing to 1080p/1440p/4K ultra, without looking at the other workloads — everything should be relatively similar margins as the gaming results.

Our test PC includes an MSI Z790 DDR5 motherboard, G.Skill DDR5 memory, and a Sabrent 4TB SSD — which we promptly filled to about half its total capacity. be quiet! also supplied us with its new 1600W Dark Power Pro 13 80 Plus Titanium rated power supply. That would have been overkill, back before cards like the RTX 4090 were a thing.

We're including a collection of current and previous generation GPUs, with the Nvidia results all from the RTX 40-series. If you want to see how the RTX 4070 stacks up against other GPUs, check out our GPU benchmarks hierarchy.

We have Nvidia PCAT v2 (Power Capture and Analysis Tool) hardware as well, which means we can grab real power use, GPU clocks, and more during all of our gaming benchmarks. We'll have most of the details for power testing in a few pages.

PNY RTX 4070 Overclocking

Our standard approach to overclocking graphics cards is to use MSI Afterburner and then max out the power slider, but the PNY card is locked to 100%. So we left that alone and played with the GPU and GDDR6X clocks. We ultimately achieved a stable (for our benchmark suite) +200 MHz GPU and +1750 MHz GDDR6X overclock. The memory overclock suggests that this is 24 Gbps Micron memory, even though it's marked as standard 21 Gbps memory.

The thing about the overclock is that, without an increased power limit, it's not as effective as with other GPUs. We've got overclocked results from the 4070 Founders Edition in the charts as well (+225 MHz GPU, +1350 MHz GDDR6X), and the increased power limit seems to be the deciding factor.

Regardless, there's a pretty large gap between the RTX 4070 and the RTX 4070 Ti. With 5888 CUDA cores versus 7680 cores, you'd need to run the RTX 4070 at about 3.4 GHz just to match the reference 4070 Ti. That's not going to happen, possibly not even with liquid nitrogen. All told, our overclock on the PNY RTX 4070 yielded a 6% improvement in performance, while the RTX 4070 Ti ends up being around 25% faster than the stock 4070.

• MORE: Best Graphics Cards
• MORE: GPU Benchmarks and Hierarchy
• MORE: All Graphics Content

We have 1080p, 1440p, and 4K ultra benchmarks for this PNY review. Because it has a $599 MSRP, we feel that starting with 1440p ultra is most appropriate. We'll cover the 1080p results next, and finish with 4K gaming results (which are mostly academic, as you'd need DLSS to get good performance in many games at 4K ultra).

Here's the overview, first with the 15-game full suite, and then with separate rasterization-only and DXR-only (i.e. ray tracing) charts. We'll discuss the individual results further down the page.

There's very little to add to the RTX 4070 story when looking at the PNY card. It's effectively tied with the reference RTX 4070 Founders Edition, with all the results being within the expected margin of error.

Against other GPUs, the RTX 4070 and RX 6950 XT trade blows, with the 4070 leading in ray tracing games and mostly trailing in rasterization performance (Total War: Warhammer III being the sole exception on the latter). DLSS upscaling will generally favor Nvidia's new card, because it's both more widely available and it looks better, with Frame Generation being the potential smooth (as in, frame smoothing) cherry on top.

Across our 15-game test suite, the PNY and Founders Edition 4070 cards are within 2% of each other in every test. We'll need to look to other aspects to find some differentiating factors.

One interesting point is just how much the gap has grown between the various Nvidia GPUs this generation. The 4070 ends up being 33% faster than the 4060 Ti, while the 4070 Ti in turn delivers 24% higher performance. Compared to the previous generation, the RTX 3070 was only about 12% faster than the RTX 3060 Ti, while the RTX 3070 Ti tacked on just 7% more performance than the 3070.

That mostly applies to other 30-series versus 40-series comparisons as well, though the 4080 and 4070 Ti at least are about the same 20~25 percent difference as the previous generation. We'll have to see where the upcoming RTX 4060 and, eventually, RTX 4050 land. The 3060 Ti was 34% faster than the vanilla 3060, despite having less VRAM. On paper at least, the 4060 Ti has 46% more compute than the 4060.

• MORE: Best Graphics Cards
• MORE: GPU Benchmarks and Hierarchy
• MORE: All Graphics Content

The PNY RTX 4070, and RTX 4070 in general, does a bit better at 1080p ultra than at 1440p ultra. That's partly because the large L2 cache helps more at lower resolutions, and partly because CPU bottlenecks also come into play. Overall, the 4070 comes in just a bit ahead of the RX 6950 XT, where it was just a bit behind at 1440p.

We could point out that it's fair to expect any $600 graphics card to easily handle 1080p gaming. After all, even the $350 RX 6750 XT averages over 60 fps, though as usual, AMD's GPUs struggle a bit more in some of the ray tracing games. Likewise, AMD's GPUs do better overall in rasterization games. Anyway, let's move on.

The spread in the individual games, when comparing the PNY and Founders Edition 4070 cards, is meaningless. That's as you'd expect, given they have the same reference clock speeds.

The RX 6950 XT remains the closest competition, with a slightly higher price right now. PNY's card just edges past it, by 1% overall, with a 20% lead in the DXR test suite and a 9% loss in the rasterization games.

What the above performance charts don't show (and we'll get to this in a couple of pages) is that the RTX 4070 basically matches the RX 6950 XT in performance while using 178W on average at 1080p, while the RX 6950 XT uses 307W. That means you need a better PSU if you want to run AMD's previous gen halo card, and that's still not accounting for DLSS upscaling or frame generation.

• MORE: Best Graphics Cards
• MORE: GPU Benchmarks and Hierarchy
• MORE: All Graphics Content

Wrapping up with the 4K ultra results, as you'd expect, the PNY RTX 4070 starts to struggle. Several of the rasterized games still break 60 fps, but only Spider-Man: Miles Morales breaks 30 fps in our DXR tests.

4K definitely needs help, and if you enable DLSS upscaling, performance improves a lot. We could debate about whether or not AI-based upscaling to 4K counts, but unless you're using a substantially more potent GPU or running lighter games, it's really the only way 4K is viable on GPUs like the 4070.

Overall, the RX 6950 XT now leads the PNY 4070 by 3%, which again doesn't mean a lot. It's still the same split between DXR games favoring Nvidia and rasterization games favoring AMD. And while we haven't said much about some of the other GPUs, skipping 4K and getting a card like the RX 6750 XT certainly makes sense if you're on a budget.

Native 4K in the rasterized games ranges from 35 fps in A Plague Tale: Requiem to 81 fps in Forza Horizon 5. So all nine of those games are at least playable, even without upscaling. Of the games that don't hit 60 fps, it's also worth mentioning that four of the six games support DLSS, and two others also support Frame Generation.

The DXR games are another matter. Even with DLSS support in all six games, most aren't going to get to 60 fps or more without Performance mode upscaling, and possibly Frame Generation as well. But you could drop to high settings instead of ultra and probably make up the difference.

• MORE: Best Graphics Cards
• MORE: GPU Benchmarks and Hierarchy
• MORE: All Graphics Content

Our power, clocks, and temperature testing now utilizes the same test suite as our gaming benchmarks, as the PCAT v2 hardware and FrameView software lets us collect this data alongside frametimes. We're also using our updated Core i9-13900K platform, so we're less likely to have CPU or platform limitations playing a role.

We have 1080p, 1440p, and 4K charts for power, GPU clocks, and temperatures below. Then we do a test using Metro Exodus Enhanced at whatever "demanding for the GPU being tested" means, and after letting the game run for 15 minutes or more, we check noise levels. We'll present additional tables and information about efficiency (FPS/W) and value (FPS/$) at the bottom of the page.

Nvidia's Ada Lovelace architecture proves very efficient overall, and PNY locks down power use more than Nvidia's reference card, so even when overclocked power use doesn't break 200W. The maximum average power use we saw was actually just 192W when overclocked, versus 198W running stock (though we didn't test the overclocked card at 4K ultra or 1080p medium).

The PNY RTX 4070 uses less power than AMD's RX 6800 while delivering performance closer to that of the RX 6950 XT. Even if we drop the ray tracing results, it would still be equal to the RX 6800 XT in rasterized performance while using over 100W less power. Like we said, it's an efficient GPU.

GPU clock speeds on their own don't mean too much, but it's interesting how much clocks have shot up with the latest Nvidia RTX 40-series. Taking a page from AMD's RDNA 2 book, Nvidia focused on improving clocks and is now routinely hitting 2.6–2.8 GHz, where the previous generation RTX 30-series usually topped out at around 1.9 GHz (without serious manual overclocking).

Here's where we also start to see some of the low-level limitations of the PNY RTX 4070. It's nearly as fast as the Founders Edition, but clocks are typically slightly lower, and the gap widens at more demanding settings. Overclocking also hits power limits on the PNY card, so that it runs about 100 MHz slower than the FE card at 1440p.

Temperatures are the other part of the equation, and it's possible to tune cards for higher clocks by cranking up fan speeds and noise levels. The PNY RTX 4070 doesn't run hot, but it does run warmer than the reference card by about 4C–5C. Let's look at noise levels to see what's going on...

We check noise levels using an SPL (sound pressure level) meter placed 10cm from the card, with the mic aimed right at the center of one of the fans (the right fan on the PNY card). This helps minimize the impact of other noise sources like the fans on the CPU cooler. The noise floor of our test environment and equipment is less than 32 dB(A).

After running Metro Exodus for over 15 minutes, the PNY RTX 4070 settled in at a fan speed of 57% and a noise level of 42.7 dB(A). That's okay in terms of noise, but it's still 3 dB(A) higher than the Founders Edition. It's also interesting that both the FE and PNY cards are at relatively high fan speed percentages, meaning there's not as much room for additional cooling.

We also tested with a static fan speed of 75%, which caused the PNY RTX 4070 to generate 51.4 dB(A) of noise. So the stock noise levels are pretty average, with larger cards typically making less noise. Conversely, at 75% fan speed this is one of the quieter results... though the RTX 4070 FE still does better.

GPU Value and Efficiency

This final gallery of images shows the full performance test suite, along with the above power, clocks, and temperature information. Latency is also provided, at least in some of the games (depending on the GPU and drivers used).

We've calculated efficiency in FPS/W for the various games, plus value in FPS/$ using the best current online prices we could find (usually at Newegg or Amazon, though B&H and Best Buy were also checked). We've summarized those results in the above table (based on 1440p performance and power), sorted by overall value.

It shouldn't be too surprising that the RX 6750 XT is the best overall value out of the GPUs we've tested, followed by the RTX 4060 Ti and RX 6800 XT. Lower prices will go a long way toward making a GPU a better value, and those are the least expensive cards we've included in this review. The Nvidia and PNY RTX 4070 cards are tied for fourth, followed closely by the RX 6950 XT. Note also that the spread in FPS/$ for a lot of the cards is very narrow, with the RX 7900 XT to RTX 4070 all in the 0.111–0.122 range.

Meanwhile, the RTX 4080 ranks as the most efficient GPU we've tested to date, at least for 1440p ultra, with the PNY RTX 4070 coming in second — just edging out the RTX 4070 FE. All of the Nvidia RTX 40-series GPUs have better efficiency than any of the AMD GPUs, with a pretty sizeable drop going from the RTX 4070 Ti (0.367 FPS/W) to the RX 7900 XT (0.281 FPS/W).

• MORE: Best Graphics Cards
• MORE: GPU Benchmarks and Hierarchy
• MORE: All Graphics Content

The PNY RTX 4070 doesn't try to do anything wild. There's no RGB lighting, no factory overclocks, and not even an option to crank up the power limit for manual overclocking. It's not a card targeted at enthusiasts, in other words, but instead something that OEMs and system integrators might find useful. It will happily chug along at less than 200W, providing good performance.

If you want a flashier RTX 4070 card, or something quieter, there are plenty of options. Depending on the week or month, some of those cards might even cost less than the PNY RTX 4070 Verto Dual. That's fine, and there's room for all types of cards. Some will undoubtedly like the no-frills aesthetic that's on tap.

But if we're being frank ("Hi, I'm Frank..."), we want graphics cards to either beat the Nvidia reference card on price, features, or performance. PNY doesn't do any of those. It's close enough to call performance a tie, but the RTX 4070 Founders Edition will still be a better card for most people.

Unless you hate the superfluous 12VHPWR 16-pin connector and the adapter cable that comes with Nvidia's card. And there are definitely good reasons to dislike it! In a typical case, you end up with a stiff adapter cable that sticks out several inches above the top of the card. Then you have to make the side panel of your case fit on properly, without putting too much strain on the cable.

If you have a high-end ATX 3.0 power supply, maybe it won't matter. But how many people are actually buying high-end PSUs with native 12VHPWR support for a 200W graphics card? We still don't have a good explanation for why Nvidia has pushed so hard on use of the 16-pin connector (or the 12-pin precursor for it's RTX 30-series Founders Editions). It's not like they're provably better, and the melting cable issue still hasn't really ended.

Is that one item a big enough reason to opt for the PNY RTX 4070 over a different model? Perhaps! But if so, know that there are MSI, Asus, Zotac, and Gigabyte models that will also suffice. In fact, it's only the higher-end third-party 4070 cards that have 16-pin connectors, and not even all of those.

PNY has other RTX 4070 options as well, with some RGB lighting if that's what you're after. It's a good GPU, going toe to toe with the previous generation RTX 3080 while cutting the price, increasing features, and improving efficiency. That's not the most exciting description of a new GPU, but you could do far worse.

• MORE: Best Graphics Cards
• MORE: GPU Benchmarks and Hierarchy
• MORE: All Graphics Content

The Core i9-13900K and Core i9-13900KS, two of the best CPUs for gamers, will look somewhat different the next time you see them on a store shelf. Intel has notified its customers that the chipmaker is altering the retail packaging for the two 13th Generation Raptor Lake processors.

The packaging change will apply to both boxed Core i9-13900K and Core i9-13900KS universal and China SKUs. The reason for the modification is that Intel wants to decrease the volumetric storage of the processors for shipping. It'll help increase the units per pallet to lower shipping costs. Although pallet quantities vary by region, Intel estimates that this new packaging will help boost the number of units from 324 to 1,620, a whopping fourfold improvement.

Intel has downgraded the Core i9-13900K and Core i9-13900KS packaging from a "Tier 2" box to a "Tier 4" box. The company will replace the current packaging with a standard folding carton. While Intel provided a visual guide to the new packaging, the company didn't share its dimensions in the Product Change Notification (PCN) document. Judging by the renders, the new packaging looks thinner.

Since there's no preview of the silicon wafer-inspired plastic case, we can only assume that Intel will deliver the Core i9-13900K and Core i9-13900KS in a regular paperboard box, like the other Core i7 and lower SKUs. The initial hype for Core i9-13900K and Core i9-13900KS is long gone, so there's no more need to commercialize the chips with premium packaging. Depending on how fast retailers move their inventory, customers will likely see a combination of old and new packaging.

Logically, the changes are superficial, so they won't affect the Core i9-13900K or Core i9-13900KS operation. The package makeover benefits Intel because it'll help the company save a few bucks. However, we don't expect Intel to issue price drops on the Core i9-13900K or Core i9-13900KS. Intel processors typically hold their value very well unless there's a new chip in town. Even so, they retain a lot of their value.

A change in promotional packaging usually conveys that something new is coming. Intel has a history of simplifying the flagship Core i9 K-series packaging following up to the release of a successor. It's a practice that Intel has kept, one that we've seen with the Core i9-10900K (Comet Lake), Core i9-11900K (Rocket Lake), and Core i9-12900K (Alder Lake). Intel is likely doing the same for Raptor Lake.

Assuming that there is some credence to the recent rumors that Intel has allegedly canceled desktop Meteor Lake, the chipmaker may be paving the way for the rumored Raptor Lake Refresh. In the past, Intel has revamped the packaging on Core i9 K-series processors around one to two months before a new processor launch. If we abide by the traditional cadence, Intel may talk about Raptor Lake Refresh very soon.

Apple's M2 Ultra is a mighty processor with 24 general-purpose cores and up to 76 integrated GPU cores. It is also probably faster than Intel's 28-core Xeon W found in the 2019 Mac Pro, so new Mac Pro workstation owners will feel a significant performance uplift. But according to a purported benchmark, this CPU cannot beat competitors from AMD and Intel in Geekbench 5 (via @VadimYuryev) as they boast very high clocks or feature a formidable core count.

Heavy-duty workstation-grade processors are different from desktop and server CPUs in that they should deliver both very responsive performance (like all client processors) and consistently high performance under heavy workloads. This means that they must feature high instruction-per-clock performance, high clocks, high core count, support for loads of memory, and feature loads of PCIe lanes. AMD's Ryzen Threadripper Pro W5995X and Intel's Xeon W9-3495X comply with these requirements with their 64 and 56 cores, respectively. In a nutshell, they use server silicon configured to offer extremely high clocks when needed. 

When it comes to Apple's M2 Ultra, it consists of two M2 Max system-on-chips that were designed primarily for MacBook Pro and Mac Studio machines. Those workstations have moderate power consumption and do not support expandability. M2 Max was not exactly architected for high clocks or for expandability, as you can only install so much memory and so much storage in a compact PC. Instead of boosting clocks to extremes when a computationally heavy workload emerges, M2 uses built-in special-purpose accelerators. It was also not designed for an extreme core count because of power and cooling limitations. While two M2 Maxes look formidable on paper, they cannot boast clocks of Intel's Core i9-13900K or the core count of AMD's Ryzen Threadripper Pro W5995WX. 

This is why the benchmark result shows Apple's M2 Ultra cannot beat Intel's Core i9-13900K in single-thread workloads and even fall behind in multi-core workloads in Geekbench 5. Intel's desktop offering supports simultaneous multi-threading and can process up to 32 threads instantly. Compared to its actual rivals — AMD's Ryzen Threadripper Pro W5995X and Intel's Xeon W9-3495X — it easily beats them in single-thread workloads but is dramatically slower when more cores are needed.

Some may argue that Geekbench 5 is a synthetic benchmark that does not reflect performance in real-world applications, which is a fair argument. But it gives a sense of what to expect from CPUs regarding their compute capabilities without any special-purpose accelerators. And this brings us to the fact that Apple's M2 SoCs have plenty of accelerators inside. Therefore, it may not need to have high clocks or extreme core count to offer great performance in many workstation-grade workloads. 

Then again, with so much compute horsepower under the hood, AMD's and Intel's workstation processors are designed to handle even the most demanding workloads. That said, it remains to be seen whether Apple's Mac Pro can actually beat workstations based on AMD's and Intel's CPUs in workstation applications.

Seby9123, one of the world's top professional overclockers, set a new DDR5 SDRAM data transfer record of 11,240 MT/s at G.Skill's overlocking competition at Computex last week. Using premium hardware, liquid nitrogen, and a lot of magic, he has managed to increase the stock speed of G.Skill's DDR5-7800 module by 44%.

The new world record of 11,240 MT/s set by Seby9123 is just slightly higher than the second-highest result of 11,233 MT/s set by Hicookie on the same day. Meanwhile, the third highest result is noticeably higher than 11,130 MT/s set by lupin_no_museme back in October 2022, so the progress made in a few months by DDR5 is more or less obvious.

To set his memory overclocking record, Seby9123 used Asustek's ROG Maximus X790 Apex motherboard, Intel's Core i9-13900KS processor with only two cores enabled (albeit working at 5.0 GHz), a lot of liquid nitrogen, and a 16 GB G.Skill Trident Z5 RGB DR5-7800 memory module in a single-channel mode. The actual frequency was 5620.1 MHz, and timings were set to CL62 126-126-127-127-2. 

Unfortunately, there is no word about DRAM voltage for the memory, but we suspect it is even higher than 1.45V, which is already over 30% higher compared to the standard DDR5 voltage set by JEDEC.

In addition to setting the DDR5 speed record, the world's top extreme overclockers assembled by G.Skill at its contest set several other overclocking records at Computex.

The AMD Radeon RX 7600 brings the latest-generation RDNA 3 architecture to the sub-$300 realm. It may even qualify as a "budget" graphics card, depending on how you want to stretch that definition. Originally planned for a $299 launch price, AMD dropped the official RX 7600 MSRP to $269 over the past weekend, likely thanks to Nvidia's RTX 4060 $299 announcement. Numerous leaks over the past few weeks have already hinted at what we'll see today, but let's get to it.

Will the RX 7600 rank among the best graphics cards? We'll give it a resounding maybe. Ultimately, it will come down to street prices over the coming months. Right now, plenty of existing RX 6650 XT, RX 6600 XT, and RX 6600 cards largely fill the same role, but with lower pricing. Those originally had significantly higher MSRPs, thanks to the GPU shortages and mining that was going on in 2021–2022, but now you can pick up one of the XT models for under $250, and the vanilla RX 6600 starts at just $199.

What does the new RDNA 3 architecture bring to the party? Reworked compute units (CUs) with a big boost in theoretical number crunching prowess, improved FP16 (AI) functionality, and better silicon utilization, according to AMD. In addition to our normal gaming, power, and thermal tests, we also ran the card through some stable diffusion tests. Here's the overview of the specifications.

On paper, there appear to be some major upgrades. For example, the dual compute units have twice as many ALUs (arithmetic logic units) as before, so the RX 7600 has a massive increase in FP32 teraflops: 21.5 versus just 10.8 on the RX 6650 XT, or basically double the compute performance. FP16 throughput is also doubled.

That sounds good in theory, but in practice, the real-world gains seem to be far more limited. For example, the RX 7900 XT has 51.6 teraflops of FP32 while the RX 6950 XT only has 23.7, so a 118% increase in compute. In our benchmarks, particularly for games, the 7900 XT is about 20% faster on average. If the same pattern holds with Navi 33 (spoiler: it mostly does), gaming performance won't be significantly faster.

Elsewhere, you don't get any extra VRAM, and both the RX 7600 and RX 6650 XT use 18 Gbps GDDR6. Even the process node hasn't been radically altered, as TSMC's N6 is a refined variant of the N7 node used for Navi 23. Density did improve, though, from around 47 million transistors per square millimeter (Mtrans/mm^2) to 65 Mtrans/mm^2. However, that's still a far cry from the ~122 Mtrans/mm^2 that Nvidia is getting on its Ada Lovelace RTX 40-series GPUs, or the 152 Mtrans/mm^2 for the Navi 31 (RX 7900-series) GCD.

Besides doubling the maximum FP32/FP16 throughput from the shader cores, one of the biggest architectural changes with AMD's RDNA 3 architecture is the move to GPU chiplets. Except Navi 33 doesn't do that, sticking with a traditional monolithic die approach. That's not really surprising, as GPU chiplets are primarily about cost savings, and adding in extra interconnects between a GCD and multiple MCDs likely only makes sense with wider memory interfaces and stacked L3 caches.

AMD and Nvidia have similar marketing surrounding their new 8GB graphics cards. They're good for 1080p, and lots of people use that resolution (or lower). Where AMD and Nvidia diverge is in the particulars of some of the hardware, as well as memory configurations and pricing.

Nvidia has had tensor cores in its RTX GPUs since 2018, and these can help accelerate deep learning and AI workloads — they're part of the special sauce behind DLSS, for example. Nvidia has also put more effort into making fast ray tracing hardware.

In contrast, AMD has generally focused more on delivering improved rasterization performance, at least for comparable price points. There's also less proprietary hardware and APIs. AMD does have "AI Accelerators" in RDNA 3, which share the FP16 compute functionality of the shaders. They're not quite the same as the full-blown tensor core implementations found in AMD's CDNA/CDNA 2 architectures for data centers, but they do provide some optimizations in throughput compared to accessing the same hardware via shaders. (We'll see some of this in our Stable Diffusion testing later.)

Depending on what product tier you're looking at, AMD also offers potentially more memory. However, that's not the case with the RX 7600, as it's basically near the bottom of the RDNA 3 stack and comes with the same 128-bit interface and 8GB as Nvidia's RTX 4060 Ti and the upcoming RTX 4060. Of course, the RX 7600 also costs less than either of those. The mainstream and high-end AMD RDNA 3 solutions (for desktops) haven't been officially announced yet, but it looks like AMD will probably offer a 192-bit interface and 12GB of VRAM with the 7700-class parts, and then a 256-bit interface and 16GB of VRAM with 7800-class parts. We'll have to wait and see how those stack up in performance against Nvidia's offerings.

In case you're interested, here's the full slide deck from AMD's RX 7600 presentation for reviewers and analysts. There are some performance charts and various talking points, like the fact that the GTX 1060 is still one of the most popular GPUs, even though it's now seven years old.

The thing is, people using a seven-year-old GPU probably aren't even playing the latest games. Some of them might be using laptops as well. If someone hasn't upgraded in seven years, it's doubtful that the launch of a new $269 graphics card will suddenly spur tons of people to upgrade. But if you're in the market, it's good to have options, and that's exactly what AMD and Nvidia are doing with the RX 7600 and (upcoming) RTX 4060.

• MORE: Best Graphics Cards
• MORE: GPU Benchmarks and Hierarchy
• MORE: All Graphics Content

AMD Radeon RX 7600 Reference Design

In a nice change of pace, AMD has decided to create its own reference design for the RX 7600. We don't know how widely available it will be, but it follows the same aesthetic and design language as the RX 7900 XTX / XT. But, of course, this is more of a budget offering, or at least a mainstream GPU, so there's no RGB lighting or other frills.

That's not necessarily a bad thing, though! We're well past the saturation point for RGB effects, and sometimes you just want a card that does what it's supposed to do without screaming, "Look at me!" If you're building a compact mini-ITX or micro-ATX PC to go in a bedroom, for example, having zero lights would be a good thing.

AMD's MBA ("Made By AMD") design is black with red highlights, basically like a miniature take on the RX 7900 XT MBA design. There are two fans with integrated rims, the same 78mm diameter as the 7900 XT fans. (The 7900 XTX has larger 84mm fans, if you're wondering.)

AMD's reference RX 7600 measures 204x109x40 mm and weighs 752g. It's a pretty dense feeling card, in other words, but it seems like it should keep the GPU cool. By way of comparison, the PowerColor RX 6650 XT measures 220x132x45mm and weighs 740g, while the XFX RX 6600 is 243x114x39 mm and weighs 615g. As we'll see later, however, while the cooler looks sufficient, it's not quite as robust as some AIB cards.

As you'd expect from AMD, there's a single 8-pin power connector — no 16-pin silliness here. It's a bit interesting that AMD has a 165W TBP (total board power) rating for the vanilla RX 7600, considering the previous generation RX 6600 had a 132W TBP. Still, either way, a single 8-pin connector should be more than sufficient.

AMD's card has the usual three DisplayPort 2.1 UHBR13.5 (54 Gbps) ports plus a single HDMI 2.1 port (48 Gbps). This is one area where AMD's latest generation hardware exceeds Nvidia's Ada design, which still uses DisplayPort 1.4a. Naturally, you'll need a DisplayPort 2.1 monitor or TV hooked up before the added functionality matters, and considering the cost, it likely won't matter to most RX 7600 buyers. Still, it's there should you ever need to power a 4K 240Hz monitor without using DSC (Display Stream Compression) — or you could potentially use a 4K 480Hz display with DSC.

Our test setup incidentally uses a Samsung Odyssey Neo G8 32-inch monitor, which supports 4K and 240Hz via DSC. It only supports DP1.4a, but it looks great and works fine on all the latest-generation GPUs, in my experience.

• MORE: Best Graphics Cards
• MORE: GPU Benchmarks and Hierarchy
• MORE: All Graphics Content

We updated our GPU test PC at the end of last year with a Core i9-13900K, though we continue to also test reference GPUs on our 2022 system that includes a Core i9-12900K for our GPU benchmarks hierarchy. (We'll be updating that later today, once the embargo has passed.) Our RX 7600 review will use the 13900K results for the gaming tests, which ensures, as much as possible, that we're not CPU limited. We also use the 2022 PC for AI tests and the professional workloads.

Multiple games have been updated over the past few months. We retested most of the cards for the RTX 4070 review, and we've retested additional midrange cards for this and the 4060 Ti review. We're using preview drivers provided by AMD for the RX 7600, which only has support for that GPU. The other AMD GPUs used 23.2.1–23.4.3 drivers, depending on when they were last tested. Intel's Arc A770 cards were (re)tested with the latest 4369 drivers, and Nvidia's RTX 4060 Ti was tested with Nvidia preview 531.93 drivers, while other Nvidia cards were tested with 531.41–531.79 drivers.

Our current test suite consists of 15 games. Of these, nine support DirectX Raytracing (DXR), but we only enable the DXR features in six of the games. At the time of testing, five games support FRS 2. In contrast, 12 of the games support DLSS 2, and five support DLSS 3. Because only five of the games support FSR2, we're just going to include those results (for 1080p/1440p ultra) in our main charts. Speaking of which...

We tested the RX 7600 at 1080p (medium and ultra), and 1440p ultra — ultra being the highest supported preset if there is one, and in some cases maxing out all the other settings for good measure (except for MSAA or super sampling). We ran the rasterization tests at 4K as well, but we won't show those results here as they're not super meaningful — the overall average was 29 fps, if you're wondering.

Our PC is hooked up to a Samsung Odyssey Neo G8 32, one of the best gaming monitors around, allowing us to potentially experience some of the higher frame rates that might be available on the fastest GPUs. G-Sync and FreeSync were enabled, as appropriate. As you can imagine, most games can't get anywhere close to the 240 Hz limit of the monitor, especially not with the RX 7600 trying to push 4K.

As far as the GPUs we'll be including in our charts; we've elected to put in 15 cards that range in current price from $200 and $500. We're only looking at current and previous generation hardware (so no RTX 20-series or RX 5000-series parts), but if you want to have a decent idea of where those GPUs land, check out our full GPU benchmarks hierarchy. Since the RX 7600 falls at the lower end of that price range, it's fair to expect it to land in the lower portion of the charts as well. We could have included RX 6500 XT or other slow GPUs, but we generally don't recommend those as good gaming solutions in the first place.

When we assembled the new test PC, we installed all the then-latest Windows 11 updates. We're running Windows 11 22H2, and we've used InControl to lock our test PC to that major release for the foreseeable future (critical security updates still get installed each month).

Our new test PC includes Nvidia's PCAT v2 (Power Capture and Analysis Tool) hardware, which means we can grab real power use, GPU clocks, and more during all of our gaming benchmarks. We'll cover those results on our page on power use.

Finally, because GPUs aren't purely for gaming these days, we've run some professional application tests, and we also ran some Stable Diffusion benchmarks to see how AI workloads scale on the various GPUs.

 AMD RX 6700 Overclocking?

Unfortunately... we have run out of time. Yesterday's RTX 4060 Ti review took its toll, so we'll see about adding overclocking test results in the next day or two once we've recovered and slept.

• MORE: Best Graphics Cards
• MORE: GPU Benchmarks and Hierarchy
• MORE: All Graphics Content

The Radeon RX 7600 definitely qualifies as a 1080p gaming solution, and not just because AMD says so. Where Nvidia's $399 RTX 4060 Ti seems like it ought to be more of a 1440p card (outside of ray tracing, perhaps), we wouldn't expect a $269 card to primarily target 1440p.

This isn't to say that you can only game at 1080p on the RX 7600. Obviously, less demanding games will run perfectly fine at 1440p, and even 4K in some cases. There will also be super demanding games that won't even run well at 1080p with maxed-out settings. But overall, it makes sense to consider mainstream GPUs as 1080p solutions.

Like Nvidia, AMD points to the Steam Hardware Survey, where 65% of surveyed PCs still have 1080p as their primary monitor's resolution — and another 12% use lower resolutions (probably laptops for a lot of those). Paying less than $300 for a mainstream 1080p gaming GPU sounds much better than paying $400 for the same privilege, even if the Nvidia card is faster.

We'll start with the global look at 1080p performance, using the geometric mean of all 15 games in our test suite. That includes both the ray tracing and rasterization test suites, and you can probably guess that AMD's GPUs will struggle quite a bit with some of the ray tracing games. If you don't care about ray tracing, that's fine as well, and we'll have separate charts looking at just the rasterization performance (nine games) or just the ray tracing performance (six games). We'll also have charts for the individual games.

Some will say our test suite is too heavy on ray tracing games. Perhaps. But those are the most demanding games these days, so if a GPU can handle something like Cyberpunk 2077 at RT-Ultra settings, it's a safe bet it can run just about any other game. Besides, ray tracing has been the holy grail for graphics for decades, and we're coming up on five years since DirectX Raytracing (DXR) was first revealed as the next Microsoft gaming API. So it's useful to see whether modern mainstream solutions are getting to the point where they can handle ray tracing games without choking.

We have two charts, 1080p ultra and 1080p medium. The RX 7600 arguably doesn't fully belong to either of those, but we have to draw the line somewhere. By testing at medium and ultra, you'll at least have a reasonable idea of where high would land: roughly midway between the two. Let's talk about what the overall charts tell us.

First, the RX 7600 doesn't offer a new level of performance. There was a chance it might offer improved performance at a lower price point, but even that doesn't generally hold water. The RX 7600 falls right between the existing RX 6650 XT and the RX 6700 10GB cards, which was already a narrow gap. Based on our testing, the 6700 10GB is about 5% faster than the 6650 XT at 1080p medium, and 10% faster at 1080p ultra. So AMD is basically wedging the RX 7600 between the two, at least until supplies of the previous generation parts disappear.

The RX 7600 also matches or beats the RTX 3060 in overall performance. It's around 5% in our medium tests and tied at 1080p ultra. And the 3060 is still a good card, better in some ways than its future replacement since it has a 192-bit memory interface and 12GB of VRAM, but that's a topic for another day.

We're glad to see that the RX 7600 can at least surpass the RX 6650 XT because when AMD first started telling us about the new GPU, we thought it might come up short. At the same time, the RX 6650 XT now sells for under $250, so getting 4% more performance for 10% more money isn't exactly an awesome bargain. Also, the RX 6700 10GB costs the same $269 and gives you 2GB more VRAM and slightly higher performance; it just doesn't have hardware AV1 encoding.

Looking at just rasterization performance, AMD's GPUs move up the charts relative to their Nvidia and Intel competition. The RX 7600 beats the RTX 3060 by 21% at medium settings and 17% at ultra. Interestingly, it also comes out a touch ahead of the RX 6700 at medium settings (though it's basically margin of error). Considering we're using a factory overclocked RX 6650 XT, having a 4% lead with a reference clocked card also means third-party RX 7600 models might widen the gap a bit.

It's still pretty much a tie, though. We normally say that the threshold for truly noticing the difference in performance between graphics cards is around 20%, though some people might be able to register as little as a 10% difference. The RX 7600 versus RX 6650 XT lands way down in the single digits, and only benchmarks will really show which card is (barely) faster.

Across the nine rasterization games, the difference between the RX 7600 and the previous generation RX 6650 XT is anywhere from -11% (Forza Horizon 5) to at most 14% (Total War: Warhammer 3). We just retested those two games last week on the RX 6650 XT running the 23.4.3 drivers, so some of that might simply be the newer drivers on the RX 7600, or perhaps those two games are just more susceptible to the architectural changes with RDNA 3.

Except that doesn't really make sense. We could see a more compute-heavy game (TWW3) being faster, but we don't have a good explanation for why FH5 would be that much slower, other than drivers. Too bad the preview drivers are only for the RX 7600.

Our ray tracing test suite turns the table on AMD GPUs, and now all of the red bars drop further down the charts. This was expected, though RDNA 3 does have some enhancements that can potentially help it in ray tracing. Looking at the RX 7600 and RX 6650 XT numbers, though, any architectural upgrades relative to ray tracing don't seem to matter much.

We should also note that the current AMD drivers seem to have problems with Control Ultimate Edition. At times, we'll get massive slowdowns and single digit framerates. This has started happening on all AMD GPUs to various degrees, but it was especially noticeable on the RX 7600. We've notified AMD of the problem and it will likely get fixed in a future driver.

Nvidia's RTX 3060 clearly pulls ahead of the RX 7600 in our DXR test suite. VRAM can also be a bigger factor with ray tracing, which explains why the RX 6700 ends up 10% faster at 1080p ultra. But most of the GPUs in our ray tracing charts are striving more for adequate than actually good performance, particularly in the 1080p ultra chart.

At basically maxed out settings, with demanding ray tracing games, the RX 7600 can fall below 30 fps. And if you're wondering, the RX 7600 basically offers the same level of performance (give or take) as a PlayStation 5. If a game were to offer an equivalent to our DXR Ultra settings, it would also struggle to run well on the current generation consoles.

Of the six DXR games we tested, three dropped below 30 fps at 1080p ultra on the RX 7600: Minecraft (a very demanding ray tracing game), Cyberpunk 2077, and the Bright Memory Infinite Benchmark (which is way more graphically intensive than the actual game, but it makes for a demanding benchmark). But if you drop to "medium" RT settings — which sort of defeats the purpose of RT, but whatever — most games will at least be playable.

• MORE: Best Graphics Cards
• MORE: GPU Benchmarks and Hierarchy
• MORE: All Graphics Content

AMD isn't pushing 1440p gaming as the primary goal for the RX 7600, but there will still be plenty of games that run reasonably well at this resolution. We're also using ultra settings, however, which can be a problem in some games for GPUs with 8GB of memory. If you're actually hoping to play games at 1440p, you'll definitely benefit by tweaking the settings. Or just use the high preset in most games, and that will get you 90% of the way there.

The aggregate performance from the RX 7600 at 1440p ultra ends up at just 35.5 fps, and all of the ray tracing games in our test suite fall below the 30 fps mark. Some (Cyberpunk 2077) even drop into the single digits.

The RX 6650 XT still delivers very comparable performance, meaning the RX 7600 is less about upgrading that product tier and more about a sideways step with a few extras like AV1 encoding support and a slightly smaller chip.

The RTX 3060 ranks ahead of the RX 7600 now, thanks at least in part to its 12GB VRAM configuration. (We feel bad for anyone that got conned into buying the RTX 3060 8GB variant.) The RX 6700 10GB also starts to pull away, as the extra 2GB (and 11% more bandwidth) can help.

Rasterization performance is still pretty good, especially if you're willing to massage your settings a bit. The RX 7600 averages 60 fps across our nine game test suite, with A Plague Tale: Requiem and Total War: Warhammer 3 being the two lowest results. Using a combination of medium and high settings would probably get both of those closer to 60 fps.

We haven't talked much about upscaling, so let's look at that as well, since 1440p is a great time to start using it on a card like the RX 7600. It's too bad more games don't support FSR 2, as Flight Simulator ran 43% faster at 1440p with FSR 2 Quality mode upscaling. Red Dead Redemption 2 and Forza Horizon 5 don't scale nearly as well, but that seems to be something with their respective engines — DLSS upscaling doesn't help as much in those games either.

Native 1440p ultra with ray tracing on the RX 7600 is just asking for trouble. Only Metro Exodus Enhanced and Spider-Man: Miles Morales come anywhere near being playable. But going along with what we just said above, FSR 2 at least can push the RX 7600 well above 30 fps — Quality mode improved performance by 63%.

The only other ray tracing game in our test suite that supports FSR 2 is Cyberpunk 2077, and it provides a massive 150% boost to performance. But that's because the RX 7600 was absolutely choking while trying to run 1440p native. Anyway, the less said about 1440p ray tracing performance on the RX 7600, the better.

• MORE: Best Graphics Cards
• MORE: GPU Benchmarks and Hierarchy
• MORE: All Graphics Content

We were still collecting results for some of the GPUs on our AI / professional apps tests when the RTX 4060 Ti review went live, but we've managed to catch up (mostly). Back to back new GPU launches aren't fun, in case you were wondering.

Modern GPUs aren't just about gaming, and they can be used for video encoding, AI tasks like Stable Diffusion and Whisper, and other professional applications. We've tested the RX 7600 in these areas as well, to see how it stacks up.

Radeon RX 7600: AI Performance

For our AI benchmarks, we're running Automatic1111's Stable Diffusion version for the Nvidia cards, and Nod.ai's Shark variant for AMD GPUs. We used the automatic build version 20230521.737 for testing, launched with "--iree_vulkan_target_triple=rdna3-7900-windows" as recommended by AMD, or "rdna2-unknown-windows" for the RX 6000-series (that's the default). The Nvidia GPUs were tested after replacing the default CUDA DLL files with newer versions, as recommended by Nvidia.

There are other AI text to image tools available, and even looking at different variations of Stable Diffusion, results can vary quite a bit. But this is the best solution we've got right now, so let's see the results. We generated batches of 20 images and timed how long each GPU took, then converted the result to images per minute. We test at both 512x512 and 768x768 resolutions, where possible, though the latter isn't yet optimized to work well on AMD's GPUs.

We have some good news and some bad news. The bad news is that AMD's GPUs get absolutely demolished in this particular test. The fastest AMD card in our charts, the RX 6800, is 34% slower than even the rather lackluster RTX 3050. The good news, if you're the silver lining type of person, is that RDNA 3 shows some clear performance improvements relative to RDNA 2.

Except we don't know if that's thanks to RDNA 3 targeted optimizations, or if it's because of the doubling of ALUs per GPU shader. We'd like to think it's the latter, but that may or may not be correct. Regardless, in our 512x512 Stable Diffusion tests, the RX 7600 is 40% faster than the RX 6650 XT. We never came anywhere near that big of a gap in the gaming tests.

Perhaps even more interesting is that the 768x768 results have the RX 7600 topping even the RX 6800. Again, that could be targeted optimizations, but theoretically the RX 7600 has 43.5 teraflops of FP16 compute performance available, while the RX 6800 only has 33.3 teraflops of FP16. So, as unlikely as it might seem based on the gaming results, these charts may not actually be a fluke.

SPECviewperf 2020 consists of eight different benchmarks, and we use the geometric mean from those to generate an aggregate "overall" score. Note that this is not an official score, but it gives equal weight to the individual tests and provides a nice high-level overview of performance. Few professionals use all of these programs, however, so it's typically more important to look at the results for the application(s) you plan to use.

Across the eight tests, AMD's RX 7600 lands between the RX 6700 10GB... and the RX 6600. Wait, what? Yeah, we know, the RX 6600 seems to be overachieving here, or else the RX 6650 XT and RX 6600 XT are underachieving. All three of those GPUs were tested with the same 23.4.3 drivers in the past day or so, and the results were consistent at the time (we ran the whole test suite twice). We'll see about retesting after doing a driver clean in the next day or so.

For now, AMD's RX 7600 places about where it should relative to the other AMD GPUs, and in the overall score AMD beats all of the RTX 3070 Ti and lower GPUs. That's because AMD GPUs score particularly well in snx-04, the eighth slide in the above gallery. AMD GPUs also generally score higher in catia-06, creo-03, energy-03, and medical-03, while Nvidia GPUs do better in 3dsmax-07 — with maya-06 and solidworks-07 being more neutral.

If you use any of those applications on a regular basis, that could be enough to sway your GPU purchasing decision. But if you're using Catia or 3ds Max on a regular basis, you're probably not looking at purchasing a mainstream $269 GPU expressly for that purpose.

Moving on to 3D rendering, Blender is a popular open-source rendering application, and we're using the latest Blender Benchmark, which uses Blender 3.50 and three tests. Blender 3.50 includes the Cycles X engine that leverages ray tracing hardware on AMD, Nvidia, and even Intel Arc GPUs. It does so via AMD's HIP interface (Heterogeneous-computing Interface for Portability), Nvidia's CUDA or OptiX APIs, and Intel's OneAPI — which means Nvidia GPUs have some performance advantages due to the OptiX API.

The RX 7600 takes up its familiar spot between the RX 6700 10GB and the RX 6650 XT. However, the Classroom scene actually puts the RX 7600 ahead of the 6700. Nvidia's GPUs still easily surpass AMD, and even the RTX 3050 (and Arc A770) outperform the fastest AMD card. (Also, RX 6800 seems to be underperforming a bit.)

• MORE: Best Graphics Cards
• MORE: GPU Benchmarks and Hierarchy
• MORE: All Graphics Content

We've been working on this for a while, retesting all the GPUs on our new test PC using an Nvidia PCAT v2 device, and we're finally making the switch from the Powenetics hardware and software we've previously used. The benefit is that we're now using data from our full gaming suite, rather than just a couple of specific tests. The charts are the geometric mean across all 15 games, though we'll also have full tables showing the individual results — some games are more taxing than others, as you'd expect.

Our new testing approach also provides data on GPU clock speeds and temperatures (but not fan speeds), and we'll have separate charts for 1080p ultra, 1080p medium, and 1440p ultra. Besides the power testing, we also check noise levels using an SPL meter. This is done at a distance of 10cm from in order to minimize noise pollution from other sources (like the CPU cooler).

AMD's official TBP (total board power) for the RX 7600 is 165W, and as you can see, on average it's about 10–15 watts below that mark. Of course, being an average that means there are games where it can slightly exceed the official TBP, and other games where it's well below the limit.

While they're not direct competitors, it's interesting to note that Nvidia's RTX 4060 TI actually uses less power than the RX 7600 while delivering better performance. That's proof again of just how efficient Nvidia's Ada Lovelace architecture can be, though TSMC's 4N node certainly deserves a chunk of the credit. RX 7600 uses the prior generation TSMC N6, which is still better than the N7 node on RX 6000-series GPUs and the RTX 30-series Samsung 8N node.

It's worth pointing out that the RX 6650 XT ends up using about 15–20 watts more power than the RX 7600. Even if performance hasn't changed much, it does show that the Navi 33 GPU is about 15% more efficient than Navi 23. Full details of the power use in the individual games can be found in the gallery further down the page, along with information on FPS/W (i.e. efficiency).

AMD helped usher in a new era in GPU clocks with RDNA 2, and Nvidia has joined the party with the RTX 40-series. Previous generation GPUs were only likely to break the 2GHz barrier with extreme overclocking, but now everything passes that mark with ease.

AMD lists a "Game Clock" of 2250 MHz for the RX 7600, with a boost clock of 2625 MHz. The game clock is obviously very conservative, with every game in our test suite averaging at least 2480 MHz. There are even a few games, like Spider-Man, where the RX 7600 averaged around 2.85 GHz.

Comparing the RX 7600 and RX 6650 XT, there's almost no difference in GPU clocks, and both have 18Gbps GDDR6 memory. Perhaps future games will be better optimized for the RDNA 3 architecture, but right now it seems as though the enhancements don't make more than a few percent difference.

Last we have GPU temperatures, and it's worth pointing out that many graphics cards choose to prioritize lower noise over temperatures. The RX 7600 ranks near the bottom of the thermals charts, keeping company with the RX 6700 XT, but thermals on their own don't mean much.

Also remember that this is a compact GPU that could fit in a mini-ITX case quite easily. The RX 6600 XT, RX 6650 XT, and RX 6700 10GB by contrast are larger AIB designs, with a triple-fan cooler on the ASRock RX 6600 XT. The other two are more modest dual-fan designs at least, though they're still larger than AMD's RX 7600 reference design.

Let's finish up with a look at noise levels, as that will also help put the thermals into context.

We check noise levels using an SPL (sound pressure level) meter placed 10cm from the card, with the mic aimed right at the center of one fan: the center fan if there are three fans, or the right fan for two fans. This helps minimize the impact of other noise sources like the fans on the CPU cooler. The noise floor of our test environment and equipment is around 31–32 dB(A).

After leaving Metro Exodus running for over 15 minutes, the AMD RX 7600 settled in at a fan speed of 56% and a noise level of 45.2 dB(A). That's not particularly quiet, especially for a GPU drawing around 150W. GPU temperatures were also 77C, so despite moderately high fan speeds and noise levels, AMD's reference cooler doesn't seem to be doing all that well.

This is in an open testbed as well. We can only imagine what would happen if you put the RX 7600 into a cramped mini-ITX build. It would probably run fine, but further restrictions to the airflow would cause the fans to have to spin even faster. That's not ideal for a card that otherwise looks like a great option for tiny PC builds.

We also tested with a static fan speed of 75%, and the RX 7600 registered 52.7 dB(A). There's still some headroom left for AMD to ramp up fan speeds, and you can use Radeon Settings to tweak the fan curve if you'd prefer a cooler running card. But for most people, the default fan curve should work fine.

Here's the full rundown of all of our test results, including performance per watt and performance per dollar columns. We won't dig into the details much here, but we used the lowest priced card we could find online for the tables — for a reputable model from a reputable online site. You could find used cards on eBay for a lot less money in some cases, but buying a used graphics card has more associated risks.

The best value right now, if you're not worried about power use, is the Intel Arc A750. Intel dropped the price to $249 a few months back, but now the Arc A750 costs $199 at Newegg. We don't know if that's a permanent price drop (until the cards sell out), or just a temporary sale. Besides that, the RX 6600 comes as the second best value, followed very closely by the RX 6650 XT, RX 6600 XT, and RX 6700 10GB. The RX 7600 ranks sixth overall.

Efficiency is the other item you can see here, represented as FPS/W. Nvidia's RTX 4060 Ti takes top honors, with 0.95 FPS/W at 1080p medium and 0.60 FPS/W at 1080p ultra. Second place goes to the RX 6800 at 0.60 FPS/W for 1080p medium and 0.377 FPS/W at 1080p ultra, and then the RX 7600 comes in third and only slightly behind the 6800 with 0.595 FPS/W and 0.361 FPS/W.

• MORE: Best Graphics Cards
• MORE: GPU Benchmarks and Hierarchy
• MORE: All Graphics Content

Yesterday, we had the Nvidia RTX 4060 Ti review. Today, AMD joins the party with a lower-priced Radeon RX 7600 offering. Neither one is particularly exciting, as we've already seen what the new architectures from AMD and Nvidia bring to the high-end and extreme segments. Pricing for both the AMD RX 7600 and the RTX 4060 Ti isn't awesome either, though at least AMD drops below the $300 mark.

We end up with a similar perspective on both of the new graphics cards. They're okay, and we're glad to have new options, but they're not must-have cards. Given a choice between one of the new RX 7600 or RTX 4060 Ti cards and an existing card with similar performance for the same price, getting the latest generation card makes more sense. However, things get a lot murkier if pricing starts to favor the older cards, and unfortunately, that's what we're likely to see.

If AMD only had Nvidia as competition, the RX 7600 would probably look pretty good. Ray tracing performance won't brighten your day, but rasterization performance beats Nvidia's RTX 3060, and you can save $40 at the same time. The problem is that AMD also has to compete with itself, meaning the previous generation RX 6000-series cards.

When you see the $269 RX 7600 barely beating the $239 RX 6650 XT and trailing the $269 RX 6700 10GB, you have to decide what it is that you really want. DisplayPort 2.1 is basically a checkbox feature, especially on the RX 7600 — because who will hook up a DP2.1 display to such a modest card? AV1 support is fine, but are you actually going to use it? Maybe if you're a streamer, but otherwise, you could use HEVC encoding and get roughly the same quality and compressions ratio. For gaming purposes, getting a bit more VRAM and performance would make more sense.

Let's also point out that the Radeon RX 7600 is, as far as we're aware, the "full-fat" implementation of AMD's Navi 33 GPU. There's not going to be a faster RX 7600 XT using the same chip. The previous generation RX 6600 XT and 6650 XT were also fully enabled variants of the Navi 23 GPU. Why did AMD drop the XT suffix from the RX 7600? If we were to venture a guess, it was so that AMD could compare performance with the RX 6600 and claim a bigger generational improvement.

What we can't say for certain is where the RX 7600 pricing will end up. AMD still hasn't officially changed the MSRP for its RX 6600-series GPUs, but street pricing has dropped by over $100. While the RX 6600 had a $329 launch MSRP, prices back in mid-2021 were all kinds of messed up, so that reasoning doesn't really hold much sway now that Ethereum mining is dead and buried.

Could the RX 7600 end up selling at closer vanilla RX 6600 prices? Possibly, but it will likely stay pretty close to the official MSRP for a while. We'd instead expect AMD to release a lesser card, like an RX 7500 XT, that uses a trimmed down Navi 33 GPU with a $199 MSRP. That would be a massive upgrade relative to the 6500 XT, but then the same issues apply: Call it an RX 7500 XT, but a trimmed down Navi 33 GPU would actually be an RX 6600 replacement.

In a couple of months, we'll also get Nvidia's RTX 4060 as a $299 card. If AMD is going to drop the price on the RX 7600, just before the RTX 4060 debut would be the time to do so. Either way, you still get an 8GB / 128-bit interface graphics card. That's not a deal-breaker if you're only running a 1080p monitor, but it does mean any card bought for 1080p could end up feeling very inadequate if you happen to upgrade your display. You'll probably also need to drop texture quality a notch in some games, even if you're only running at 1080p.

Plenty of people have been hating on Nvidia's RTX 40-series, especially the pricing aspect. Some say the new RTX 4060 Ti is a terrible and overpriced card. But here's the thing: It's over 30% faster than the RX 7600 in rasterization games, and roughly twice as fast in demanding ray tracing games — and that's without including DLSS upscaling or frame generation. So it's fair to say it's about 50% faster than AMD's RX 7600 for a bit less than 50% more money, and it even uses less electricity.

We're not saying people should forget about the RX 7600 and jump up to the $400 RTX 4060 Ti. We're just pointing out that their value propositions (FPS per dollar) aren't all that different. That's a big part of why we ultimately landed at the same 3.5-star rating. These two new cards are okay — not amazing, but just okay. If you're in the market for a new sub-$300 card, the RX 7600 is a decent option. However, for $400, the RTX 4060 Ti does offer some tangible benefits like DLSS 3 support.

But if you already have a graphics card that does what you need it to do, just stick with that and wait out the current series of new GPUs. Alternatively, the Radeon RX 6600 is still an excellent deal at $199, and the same goes for the Intel Arc A750. You'll probably have fewer driver concerns with AMD's GPUs, though.

The AMD Radeon RX 7600 will go on sale tomorrow, May 25. We expect the usual gamut of custom overclocked AIB cards as well as branded variations of the reference design. Reference models should start at $269, while factory overclocked cards with lots of RGB lighting will probably push prices north of $299. Now all we need is for AMD to launch its RX 7700- and 7800-series parts.

• MORE: Best Graphics Cards
• MORE: GPU Benchmarks and Hierarchy
• MORE: All Graphics Content

The AMD Ryzen 7 7800X3D vs Intel Core i9-13900K vs Intel Core i7-13700K showdown is a matchup of one upper mid-range processor against two nimble and competent competitors for the title of the best CPU for gaming. 

The $448 Ryzen 7 7800X3D is now the fastest gaming CPU in the world. The 7800X3D recently debuted to critical acclaim, arriving with eight cores and 16 threads and AMD's Second-Gen 3D V-Cache that deliver ever-so-slightly more gaming performance than the flagship $699 Ryzen 9 7950X3D. The 3D V-Cache tech fuses a 3D-stacked cache chiplet atop the processor to deliver market-leading gaming performance at the expense of performance in productivity applications, but while it delivers explosive uplift in some games, not all titles benefit. 

In contrast, the $570 Core i9-13900K wields 24 cores and 32 threads with Intel's 13th-Gen Raptor Lake x86 hybrid architecture. The Intel Core i9-13900K is Intel's fastest gaming processor, and while it isn't quite as fast as AMD's X3D chips in gaming, it serves up a more predictable blend of performance in both gaming and productivity apps.

Intel also has its Core i7-13700K, which offers nearly the same level of performance as the 13900K in gaming, but at a much more amenable $417 price point that is more competitive with the Ryzen 7 7800X3D. While the 13700K isn't as fast in productivity work as the 13900K, it also offers far more performance than the 7800X3D in the types of apps. 

This makes for a complex array of choices if you're looking for a flagship gaming experience with your next rig, and that's before we even factor in pricing. Below we've taken the Ryzen 7 7800X3D vs Core i9-13900K vs Core i7-13700K rivalry through a six-round faceoff to see which chips take the crown in our gaming and application benchmarks, along with other key criteria like power consumption and pricing. Let's see how they stack up. 

Features and Specifications: AMD Ryzen 7 7800X3D vs Intel Core i9-13900K and Core i7-13700K

AMD’s $449 Ryzen 7 7800X3D comes packing eight Zen 4 cores with the company’s second-gen 3D V-Cache tech that unlocks 96MB of total L3 cache. This combo delivers the fastest gaming performance available on the market. The 7800X3D has the same underlying design as the Ryzen 7 7700X with a central I/O die and one eight-core compute chiplet that provides a total of 8 cores and 16 threads.

Unlike the standard Ryzen 7000 chips, the 7800X3D's compute chiplet has an additional 3D-stacked L3 cache. However, this tech results in a lower boost clock of 5.0 GHz than the standard single compute chiplet models, which can reach up to 5.4 GHz. Despite the lower frequency, the cache-equipped chiplet provides the best gaming performance. In contrast, AMD's 7950X3D and 7900X3D come with two compute chiplets, requiring a new driver to steer threads for different types of workloads to the correct chiplet. In contrast, the 7800X3D is a plug-and-play processor that doesn't require special software.

The Ryzen 7 7800X3D has all the accommodations we expect from a modern chip, including DDR5, PCIe 5.0, and the latest USB standards. The chip also has an integrated graphics engine, but it isn't suitable for gaming. You can see our iGPU testing here.

The 7800X3D is incredibly power efficient, but AMD assigned a base TPD of 120W and a max 162W PPT, 15/20W higher than the 105W/142W rating for the standard 7700X. As you'll see in the power testing section, the higher TDP rating doesn't make much sense. Regardless, you'll still need to plan for a 240mm to 280mm AIO (or better) for the best performance.

The Ryzen 7 7800X3D requires an AM5 motherboard, so DDR5 memory is the only option. Unfortunately, DDR5 continues to carry a premium over DDR4 memory. Granted, that pricing pressure isn't as much of a consideration for the highest-end systems, but you should factor it in if you're struggling to cobble together a system in this price range. The AM5 platform is new, and the Ryzen 7000 chips aren't backward compatible with older AM4 motherboards. However, AMD will support the AM5 platform until 2025+.

The 7800X3D supports the auto-overclocking Precision Boost Overdrive (PBO), Curve Optimizer, and fabric and memory overclocking. Unfortunately, direct frequency overclocking isn't possible due to a voltage limitation for the 3D tech.

Intel's $589 sixteen-core Core i9-13900K comes with eight hyper-threaded P-cores and 16 single-threaded E-cores, for a total of 32 threads. That's an increase of eight additional E-cores over the previous-gen Core i9-12900K, equating to more threaded horsepower. You can also save some cash with the $564 Core i9-13900KF, which has identical specs but lacks the integrated GPU.

The P-cores come with a 3.0 GHz base, but, more importantly, a 5.8 GHz boost that's a whopping 600 MHz increase over the prior gen. Meanwhile, the E-cores now have a 3.0 GHz base (+600 MHz) and stretch up to 4.3 GHz (+400 MHz). The Core i9-13900K also has 32MB of L3 and 36MB of L2 cache.

The $417 Core i7-13700K comes with eight p-cores and eight e-cores, for a total of 16 cores and 24 threads. The reduced number of e-cores means this chip doesn't provide as much threaded horsepower as the 13900K, but its 5.4 / 2.4 GHz boost/base clocks keep performance snappy in latency-sensitive work, like gaming, while the e-cores are also effective in threaded work with a 4.2 / 2.5 GHz boost/base. 24MB of L2 and 30MB of L3 round out the package. You can also save some cash with the graphics-less $384 Core i7-13700KF.

The Raptor Lake chips use a newer revision of the 'Intel 7' node to push clock speeds higher while improving power efficiency. Intel's chips now have a Processor Base Power (PBP) value instead of a TDP metric, and a secondary Maximum Turbo Power (MTP) value that denotes the highest power level during boost activity. Be aware that modern Intel chips can always operate at the MTP when under load, though the actual power use will vary with application intensity. The 13900K and 13700K both come with 125W PBP (base) and 253W MTP (peak) power ratings, but in practice, the 13700K sucks less power.

The Raptor Lake chips support DDR4 or DDR5 memory, which can save you money on both the motherboard and memory. Both Raptor Lake chips drop into existing 600-series motherboards or the new 700-series models, but you'll need to ensure a 600-series board has the right BIOS or BIOS flashback.  Intel allows full overclockability, but only when you pair the chips with a Z-series motherboard.

✅ Winner: Tie

The Core i9-13900K vs Ryzen 9 780X3D vs Core i7-13700K battle yields a tie in this category. All three chips support DDR5 and the PCIe 5.0 interface, meaning none have an inherent connectivity advantage.

Intel has the advantage of backward compatibility for DDR4 and the 600-series chipset, enabling lower-cost memory and motherboards for value seekers (caveats apply). Intel also has its Raptor Lake Refresh chips on the horizon, meaning there will be one more generation of upgrades for the LGA 1700 platform.

AMD's new AM5 platform generally carries a premium over the Intel motherboard ecosystem, but it does afford extended forward compatibility — AMD plans on supporting the AM5 platform until 2025+.

Given its new hybrid architecture with two types of cores, Intel's core counts aren't directly comparable to AMDs due to their different capabilities. Our performance results will dictate the performance value of the overall designs.

Gaming Benchmarks and Performance: AMD Ryzen 7 7800X3D vs Intel Core i9-13900K vs Core i7-13700K

Remember, this article is an overview of our much more in-depth testing in our AMD Ryzen 7 7800X3D Review. Below is the geometric mean of our gaming tests with the Core i7-13700K vs AMD Ryzen 7 7800X3D vs Intel Core i9-13900K at 1080p and 1440p, with each resolution split into its own chart. As usual, we're testing with an Nvidia GeForce RTX 4090 to reduce GPU-imposed bottlenecks as much as possible, but differences will shrink with lesser cards or higher resolutions. 

The 3D V-Cache tech puts AMD’s X3D chips in a league of their own in gaming, so the $449 7800X3D contends with the $699 16-core Ryzen 9 7950X3D for the lead — a victory it wins by a mere 1% in its stock config. The 7800X3D increases its lead over the 7950X3D to 2.5% at 1440p, but this is still a very close contest that falls into the imperceptible range. The price tag and number of cores is the biggest difference between these two chips; the 7950X3D costs $250 more and has twice the number of cores, which will help in productivity workloads.

The Ryzen 7 7800X3D is 12% faster in 1080p gaming than the $580 Core i9-13900K, and overclocking only narrows that to 8%. The Core i9-13900K requires exceptionally aggressive accommodations for overclocking. In contrast, the Ryzen 7 7800X3D never exceeded 90W in our testing. Overall, the 7800X3D will run faster and cooler than the 13900K in every gaming scenario, but the 13900K is far more potent in productivity-oriented tasks.

The $417 Core i7-13700K trails the 7800X3D by 14%, but it costs ~$30 less, and its more balanced performance in productivity apps makes it attractive if you're looking for a better all-rounder. The 13700K really takes the shine off the 13900K for gaming-focused rigs — the 13900K is only 1.5% faster in gaming but costs 39% more cash (an extra $163).

Moving over to 1440p pushes the bottleneck more towards the GPU, so the difference between the chips shrinks. For example, the Ryzen 7 7800X3D is 9% faster than the Core i9-13900K at 1440p with both stock and overclocked settings. It's logical to assume that most high-end gaming systems will use higher-resolution monitors, so the differences could be slight, but the 7800X3D will give you more headroom for future GPU upgrades.

As a reminder, the 3D V-Cache doesn’t accelerate all titles equally. Therefore, you should also look at the individual game benchmarks in the above album to assess the overall trend. As always, the AMD vs Intel competition can vary based on the title and the GPU you use, so it's best to make an informed decision based on the types of titles you play frequently.

Be aware that large deltas in individual game titles, as we see with the Ryzen 7 7800X3D, impact cumulative measurements. You can see examples of that in the above album in titles like Microsoft Flight Simulator 2021, Far Cry 6, and F1 2021. Our test suite heavily favors the improvements from 3D V-Cache, so we also included AMD’s own benchmarks in a broad range of games at the end of the above album to give you an idea of performance gains in other titles. As with all vendor-provided benchmarks, take AMD's slides with a grain of salt.

✅ Winner: AMD Ryzen 7 7800X3D | All-Rounder Alternative — Core i7-13700K
It may be by the slimmest of margins over its own sibling, but on average, the Ryzen 7 7800X3D is the fastest gaming chip money can buy, dethroning the 7950X3D at a much friendlier price point. It is also much faster than both of the Intel contenders.

In our test suite, the Ryzen 7 7800X3D is 12% faster than the Core i9-13900K at 1080p gaming and 9% faster at 1440p. After overclocking, those deltas shrink to ~9% at 1080p and 1440p.

Ultimately, despite its lower-than-suggested pricing of $580, we can't give the 13900K the nod in this competition — instead, the Core i7-13700K gets an honorable mention. The 13700K delivers 98.5% of the 13900K's gaming performance despite the fact that the 13900K costs $163 more (39%).

The Ryzen 7 7800X is 14% faster than the 13700K, but you will pay higher up-front pricing to build a 7800X3D system. The 13700K is competitive with the 7800X3D in our system-level and fps-per-dollar pricing calculations below, leaving it as a competent alternative.

Productivity Performance: AMD Ryzen 7 7800X3D vs Intel Core i9-13900K vs Core i7-13700K

We can boil down performance in productivity apps into two broad categories: single- and multi-threaded. The first slide above shows the geometric mean of performance in several of our most important tests in the single-threaded category, but be sure to look at the expanded results in the above album.

The Core i9-13900K beats the 7800X3D easily: it's 32% faster in single-threaded work, while the Core i7-13700K is 22% faster. Naturally, those performance deltas won't carry over to every type of workload. However, flipping through the album of test results reveals that the Intel processors excel in single-threaded work, thus providing more responsive performance than the 7800X3D in light applications.

The first slide in the above album contains a geometric mean of performance in several of our multi-threaded benchmarks. In multi-threaded work, the $580 Core i9-13900K is a whopping 89% faster than the $449 Ryzen 7 7800X3D at stock settings, while the $417 Core i7-13700K is 51% faster. That's an easy win for the Intel chips.

✅ Winner: Intel Core i9-13900K and Core i7-13700K

The Core i9-13900K and 13700K have a more balanced profile in both single- and multi-threaded productivity applications than the 7800X3D, delivering solid performance in both disciplines. In contrast, the Ryzen 7 7800X3D suffers mightily in these workloads compared to the Intel silicon, highlighting that this is truly a highly specialized chip for gaming that won't offer as much performance in standard workloads as the regular Ryzen 7000 processors. 

The implication here is clear — if performance in productivity workloads is important to you, select the Intel processor that best matches your needs in terms of both performance and pricing. The Ryzen 7 7800X3D is specifically designed for gaming-centric rigs, so you should steer clear if productivity apps are a big part of your daily usage.

Overclocking: AMD Ryzen 7 7800X3D vs Intel Core i7-13700K vs Core i9-13900K

We have long since reached the land of diminishing returns for overclocking the highest-end chips from AMD and Intel, like the Ryzen 9 and Core i9 families. Because AMD has pulled Intel into a dogfight for performance superiority on the high end, much of the overclocking frequency headroom is rolled into standard stock performance. However, there are still at least some gains to be had, and both Intel and AMD expose tunable parameters and sophisticated software overclocking utilities, like Intel's XTU and AMD's Ryzen Master.

Like the rest of AMD's new X3D chips, the Ryzen 7 7800X3D doesn't support overclocking via the CPU multiplier, so you can't change the core clocks or voltages directly. However, AMD now allows changing the power limits (PPT, TDC, EDC) through the auto-overclocking Precision Boost Overdrive (PBO) feature, and you can also now undervolt with the Curve Optimizer. These features weren’t supported on the previous-gen models. You’ll find plenty of examples of PBO+UV performance in our Ryzen 7 7800X3D review. As before, you can overclock the memory and fabric, too, though we've seen limited gains with the 7800X3D from these approaches.

Meanwhile, the Intel Core i9-13900K and Core i7-13700K expose every knob possible for tweaking and tuning. However, the 13900K's aggressive out-of-the-box power and Adaptive Boost Technology already consume most of the overclocking headroom you could achieve with conventional cooling. The 13700K leaves slightly more room than the 13900K for manual tuning due to its less-aggressive tech, but you shouldn't expect massive improvements. Sub-zero overclockers with exotic cooling solutions can reach much higher with the Intel chips than the 7800X3D, but that isn't as valuable to the overwhelming majority of users. 

As usual, performance gains from overclocking can boil down to your luck in the silicon lottery. As you can see above, overclocking with conventional cooling for both the Intel and AMD processors results in fairly minimal gains in applications. In fact, the 13900K actually suffers from lower performance in single-threaded work because its 5.6 GHz overclock is lower than its stock 5.8 GHz boost. The Core i9-13900K does gain 6% in gaming, a higher percentage gain than the 13700K and 7800X3D, but it still lags behind the 7800X3D's peak gaming performance.

✅ Winner: Tie

The payoff for overclocking is extremely limited with today's high end chips. Still, both platforms have plenty of knobs for enthusiasts, their respective overclocking advantages, and a suite of auto-overclocking and software utilities. Of course, dedicated tuners could find larger performance boosts by overclocking the core, fabric, and memory more aggressively than we have, but bear in mind that you're always at the whims of the silicon lottery when it comes to overclocking.

AMD's 3D V-Cache technology prevents direct voltage and frequency manipulations, but AMD has unlocked memory, fabric, Precision Boost Overdrive, and undervolting via the Curve Optimizer, all of which combine to provide plenty of avenues for boosting performance via various tuning techniques. In fact, the 7800X3D often delivers a similar amount of performance uplift as the Core i9-13900K, showing the value of its somewhat unconventional approaches. Additionally, some motherboard vendors, like MSI, have baked in new tuned overclocking profiles directly into the motherboard firmware, promising up to a 10% improvement in performance. But, again, your mileage will vary.

Meanwhile, Intel offers more overclocking functionality and the highest peak frequencies from direct tuning. However, much of that capability is already exposed right out of the box via the company's aggressive boosting technologies and sky-high power limits, thus reducing its value to casual overclockers. 

Power Consumption, Efficiency, and Cooling: Intel Core i7-13700K vs AMD Ryzen 7 7800X3D vs Core i9-13900K

The Ryzen 7 7800X3D is incredibly power efficient, drawing far less power than the competing Core i9-13900K or 13700K, thus delivering superb power efficiency that ranks among the best we’ve ever tested. A quick glance at the HandBrake renders-per-day-per-watt metric shows that the efficiency competition isn't even close — the 7800X3D is the hand's down winner.

In fact, the Ryzen 7 7800X3D appears to fall into the standard 65W profile that specs a maximum of 88W of power consumption (PPT). AMD has spec'd a higher 120W TDP rating for the 7800X3D, but it appears to be running closer to the 65W TDP envelope during normal use — given our measurements, the Ryzen 7 7800X3D's 120W TDP rating seems far overprovisioned. This means the chip is basically operating in Eco Mode, which confers massive power consumption and efficiency benefits.

In contrast, Intel's chips consume far more power in every type of workload, resulting in higher thermal output.

✅ Winner: Ryzen 7 7800X3D

Intel has reduced its power consumption metrics, but the Ryzen 7 7800X3D is still far more power efficient. The 7800X3D consumes much less peak power and also delivers more work per unit of power consumed. That results in a win in power consumption, efficiency, and thermal output, ultimately giving you a cooler and quieter system.

None of these chips comes with a bundled cooler, but Intel's higher power consumption means you'll need more aggressive, and thus more expensive, cooling to extract the peak performance. In contrast, we don't think most users will have problems cooling the Ryzen 7 7800X3D with the recommended 280mm or greater cooler. 

Pricing: Intel Core i9-13900K vs AMD Ryzen 7 7800X3D vs Core i7-13700K

Tallying up the pricing numbers to compare the Core i7-13700K vs Ryzen 7 7800X3D vs Core i9-13900K can get incredibly complex once we start factoring in different classes of motherboards, so we included as much information as possible (links in the text below).

The Ryzen 7 7800X3D stands out as the fastest gaming chip money can buy and as the best gaming value in this price range. At $2.00 per frame, the 7800X3D is a far better value than the 13900K, which works out to $2.90 per frame. Of course, the Core i7-13700K is slower than the 7800X3D, but it is within striking distance in our value calculation at $2.11 per frame.

As you can see when we calculate the total platform cost above, plenty of other cost factors are in play, like memory and motherboards. In the end, provided you ensure it has adequate power delivery (which is easy for the 7800X3D), you can drop the 7800X3D into a B650 board and pay $2.97 per frame. That's the best value of the potential configurations. The 13700K with DDR4 memory on a B760 motherboard is close at $3.21 per frame, but it still lags slightly at $3.21.

Surprisingly, after equipping both chips with DDR5, we basically have a tie with the 13700K on a Z790 board and the 7800X3D on an X670. The Core i9-13900K simply isn't worth the cash if you're aiming solely for gaming, but it has its own attractions on the productivity side.

We used both the B650 and X670 chipsets for modeling the Ryzen 7 7800X3D's total system price, and tallied different price points based on DDR4, DDR5, Z790, and B760 motherboards for the Intel processors. Note that we didn't include a B760/13900K combo because that doesn't make sense with that class of chip, and you'll need to be careful to ensure adequate VRMs when pairing the 13700K with a B760 board. You also can't overclock the cores with a B-series board. Finally, we reduced the fps value for the Intel chips by 4% for the DDR4 calculations, as our testing shows that to be the average performance loss with the higher-end chips from stepping back from DDR5. You will retain the full performance in productivity applications, though.

The graphics-less $551 Core i9-13900KF saves you $30, and the $392 Core i7-13700KF saves you $25. We did not include those reductions in our calculations. The Core i9-13900K and 13700K are also compatible with previous-gen motherboards if they have the right BIOS, and performance will be the same between 600- and 700-series if the power delivery is identical.

AMD's X670/E motherboards currently start at roughly $80 more than Intel's Z790, but AMD B650 boards start at $126. However, you'll only find a single B650 board for around $130, while seven Intel B760 boards are either at or below $130. This is due to the lower pricing for Intel's DDR4-equipped motherboards; we've priced those differences into our full-system measurements.

Memory pricing is also tricky. DDR4 memory isn't an option with the AM5 platform that houses the 7800X3D, and while the deltas between DDR5 and DDR4 are becoming smaller over time, they still exist: You'll pay ~$60 bucks at a minimum for 32GB of DDR4, but pricing starts at $90 for 32GB of DDR5. That gap increases with higher capacities — it's $100 for the least expensive 64GB DDR4 kit but $195 for a 64GB DDR5 kit. That's nearly twice the price. How much weight you should give to the difference in memory pricing will be impacted by the speed and capacity of the kit you select. Additionally, perhaps the best memory kit is the one you don’t have to buy — many upgraders likely already have a DDR4 kit, which is something to consider if you have a kit on hand. 

✅ Winner: Ryzen 7 7800X3D and Core i7-13700K

The Ryzen 7 7800X3D has a lower price tag than the $580 Core i9-13900K, but much higher performance in games. Overall, the 7800X3D has a commanding lead in both raw and system-level fps-per-dollar metrics, thus giving it the nod in the pricing department.

The Core i9-13900K is a superior offering if you're interested in performance in productivity applications. Still, the Core i7-13700K offers the lion's share of the 13900K's performance at a lower price point. In gaming, the Core i7-13700K offers similar raw and system-level pricing to the 7800X3D when you factor in the fps-per-dollar, thus also earning it a win in the pricing category.

The higher cost of AMD's motherboards generally adds more cost to your build than Intel's platforms, and the price premium for DDR5 also continues to add cost: At higher capacities, you'll pay nearly twice as much for DDR5 memory as DDR4 memory, and DDR5 is a strict requirement for AMD's 7800X3D.

However, the Ryzen 7 7800X3D has the performance lead in gaming by a large enough margin that it holds the lead in terms of overall value for builds focused strictly on gaming, an advantage that is most apparent against the 13900K. Additionally, the 7800X3D is the fastest gaming chip on the market, a privilege that most are typically willing to pay a premium for, yet it still beats the more value-centric 13700K, albeit by smaller margins, in our pricing efficiency metrics. 

Bottom Line: AMD Ryzen 7 7800X3D vs Intel Core i9-13900K vs Core i7-13700K

We have two separate battles in this contest: The Ryzen 7 7800X3D vs Core i7-13700K ends in a tie, with both chips getting the nod in five categories. Meanwhile, the Ryzen 7 7800X3D vs Core i9-13900K matchup tallies up to a 5-to-3 advantage in the 7800X3D's favor. 

For gaming-centric rigs, the Core i9-13900K doesn't offer enough gaming performance to justify its price premium over the 7800X3D, and its other deficiencies in pricing and power consumption hurt its broader overall score. However, the 13900K vs 7800X3D contest isn't as much of a blowout win as the score implies. In the end, both of these processors have their own distinct advantages and disadvantages depending on your target use case.

If you're concerned about building a gaming-focused rig and everything else is a secondary concern, the Ryzen 7 7800X3D is by far the superior chip over the 13900K in terms of both outright performance and value. However, the Core i9-13900K is a better option than the 7800X3D if you use productivity applications frequently, and ensuring the highest level of performance is paramount. You should also consider that AMD has both the Ryzen 9 7950X3D and Ryzen 97900X3D on offer, both of which offer more cores for the productivity minded while maintaining 13900K-beating levels of gaming performance. Just be prepared to pay for the privilege of faster gaming performance.

When it comes to competing in gaming, the 13900K's biggest foe comes in the form of the Core i7-13700K — the 13900K is only 1.5% faster in gaming but costs 39% more cash (an extra $163). That makes the 13700K the obvious Intel pick for gaming-focused rigs. Additionally, although the 7800X3D beats the 13700K by 14% in gaming, the 13700K is competitive in terms of both raw and system-level fps-per-dollar pricing metrics, though you might have to make a few reasonable tradeoffs.

The Ryzen 7 7800X3D beats the 13700K by 14% in gaming but suffers from much lower performance than the standard Ryzen 7000 models in productivity applications, so you'll need to be aware of the tradeoffs. If you're looking for strong gaming performance paired with the ability to cut through heavier workloads, the Intel Core i7-13700K is a solid all-rounder at $417. For instance, the 13700K is 22% faster in single-threaded and 51% faster in multi-threaded work than the 7800X3D, and you can opt for the graphics-less $390 13700KF to save some extra cash.

AMD's 3D V-Cache tech doesn't accelerate all games evenly, so the lead could vary depending on your gaming selection. In either case, the Ryzen 7 7800X3D is the current gaming champ, and it often isn't a close contest. That's especially true in some game titles, like Microsoft Flight Simulator 2021 and F1 2021. The chip also drops into AM5 motherboards that support the latest connectivity tech, like PCIe 5.0 and DDR5, and AMD plans to support the platform until 2025+. In contrast, Intel’s current round of motherboards will only support one more generation of chips, so upgradeability remains an advantage for AMD.

The 7800X3D also holds the advantage in all power consumption metrics, including peak power and efficiency, ranking among the best we’ve ever tested. That results in more forgiving cooling requirements and a cooler and quieter system. On the other end of the spectrum, overclocking is becoming increasingly less interesting on the high end for most enthusiasts — you'll get a better payoff in the mid-range and low-end of the market. The Intel and AMD processors all provide at least some uplift after overclocking, making that contest a wash.

Given AMD's history of lower-than-MSRP pricing several months after launch, we think the Ryzen 7 7800X3D will eventually become another great chip for value seekers, particularly as platform-level costs continue to recede. If you're after an all-rounder, the Core i7-13700K is the better choice. If you're after the fastest gaming chip on the market, the Ryzen 7 7800X3D offers a lower-cost pathway to the very best that AMD's 3D V-Cache has to offer than you'll pay with the Ryzen 9 X3D chips. If you're willing to accept the tradeoffs, the Ryzen 7 7800X3D is the fastest gaming chip money can buy, earning a leading spot on our list of the best CPUs for gaming.

• MORE: AMD vs Intel
• MORE: Zen 4 Ryzen 7000 All We Know
• MORE: 13th-Gen Intel Raptor Lake All We Know 

Amazon may have started off selling books beforee becoming the world’s largest online retailer, selling just about everything. But part of its ascent includes the introduction of its Amazon Basics line, designed to provide commonly used items like batteries, trash bags or towels at value prices.

With the success of the brand, it grew to include basic electronics like USB cables and wall chargers, creeping ever closer to the realm of PC hardware. And then finally this year, Amazon partnered with Cooler Master to expand that lineup even further, with an air cooler that’s listed on the site, in true convoluted keyword fashion, as the Amazon Basics Computer Cooling Fan with Cooler Master Technology.

At $28.49 USD, this basic air cooler costs less than a tank of gas – but is it actually any good? We’ll have to put it through testing to see if it performs well enough to make our best coolers list, but first let's take a look at the specifications from Amazon.

Cooler Specifications

If it wasn’t obvious from the Amazon Basics Computer Cooling Fan with Cooler Master Technology name, the Amazon Basics CPU cooler is effectively a rebranded Cooler Master product. The user manual included lists the model number as RR-H410-20PC-AS, which is nearly identical to the model number of Cooler Master’s Hyper H410R RGB. The specifications are identical to that product, so it’s basically a more-affordable rebrand of the H410R RGB.

Packing and Included Contents

The Amazon Basics CPU Cooler arrives in a very basic small brown box and it is efficiently packaged. Taking off the top reveals the accessories box and removing that reveals the cooler protected by molded foam.

Included with the package are the following:

• 240 mm radiator and CPU block
• Heatsink
• Mounting for modern AMD & Intel Platforms
• RGB adapter
• Thermal paste
• Installation guide
• Safety manual

LGA 1700 Installation

Installation of the Amazon basics cooler was fairly simple, but not necessarily easy.

1. Attach the Intel mounting bars to the cooler’s base
 

 2. Apply thermal paste.

3. Take the backplate and hold it securely against the motherboard. While doing this, press the heatsink against the motherboard and secure it by screwing it into the backplate. 

4. Snap the fan against the heatsink and connect the PWM and RGB cords.

Features of the Amazon Basics CPU Cooler

SFF Compatible

With a height of only 136.5mm, the Amazon Basics CPU cooler can fit in tight builds and is ideal for Micro ATX and Mini ITX systems.

Four Direct Contact Heatpipes

The Amazon Basics CPU cooler has four copper heatpipes that make direct contact with the CPU for maximum efficiency of thermal dissipation.

Hardware RGB Controller

The lighting on this cooler uses the older 4-pin RGB standard, which some modern motherboards no longer support. Included with the cooler is a simple three-button hardware lighting controller, which allows you to change the brightness, color, and lighting patterns.

Stacked Aluminum fin design

The cooler heatsink is comprised of 40 fins, precisely stacked to increase the static pressure of airflow.

DF0922512RFHN 92mm Fan

There’s more to a cooler than just the heatsink or radiator, as the fan(s) impact both cooling capacity and noise levels. Included with Amazon’s cooler is a black 92mm fan with clear blades, listed as model DF0922512RFHN on its label.

Testing Methodology

While it was fairly easy with previous generations of CPUs for coolers to keep the flagship i9 processor well under TJ max (the maximum temperature a CPU can sustain without throttling) in tough workloads, this is no longer realistically possible on current generation CPUs (and the 13900K especially) without extreme cooling (or enabling power limits).

In the past, a CPU hitting its peak temperature was cause for concern. But enthusiasts are going to have to learn to accept high temperatures as “normal” while running demanding workloads with Raptor Lake and Ryzen 7000 CPUs. Modern AMD and Intel CPUs are designed to run fairly hot without any problems – up to 95 degrees Celsius for AMD Ryzen 7000 CPUs, and up to 100 C for Intel’s Core i9-13900K. Similar behavior has been standard in laptops for years due to cooling limitations in tight spaces.

Furthermore, Intel’s Core i9-13900K supports Adaptive Boost Technology (ABT) which allows Core i9 processors to dynamically boost to higher all-core frequencies based on available thermal headroom and electrical conditions. This allows multi-core loads to operate at up to 5.5 GHz if the necessary amount of thermal dissipation is there. This feature works in a way that actively seeks high temperatures: If the chip sees that it is running underneath the 100-degree C threshold, it will increase its performance and power consumption until it reaches the safe 100 C limit, thus sustaining higher clocks (and providing better performance) for longer periods.

The increased cooling challenges posed by Raptor Lake mean that we’ve had to change some of the ways we test coolers. Some coolers were able to pass Cinebench R23 multicore testing with Intel’s 12th Gen i9-12900K when power limits were removed (although only the strongest models were able to pass that test). Most liquid coolers and all air coolers I’ve tested “failed” that test because the CPU reached TJ max in this scenario. 

With Raptor Lake’s 13900K, not a single cooler tested has been able to keep the CPU under TJ max in this test – because as we pointed out, the chip is designed to dial up performance and power until it reaches that thermal result. We’ll compare performance instead by looking at total benchmark scores and clock speeds maintained. 

I’ll be testing Intel’s i9-13900K CPU using Asus’ TUF Gaming Z690 Gaming Plus WIFI motherboard and Cooler Master’s HAF 700 Berserker computer case, with case fans limited to 35% speeds. The motherboard’s default fan curve is used for the CPU Cooler’s fans. 

In addition to testing Cinebench without power limits enforced, we’ll also be showing results when the CPU’s power consumption is limited to a more reasonable 200W. We’ll also show results at 125W for those who prefer whisper-quiet cooling, at the cost of some performance. For both of these results, we’ll show traditional delta over ambient temperature results.

We’ll provide noise level measurements recorded using a PSPL25 Sound Meter for all three power levels tested to compare how much noise each cooler makes in different scenarios. We expect most coolers to run effectively silently at 125W.

LGA1700 Socket Bending

Note there are many factors other than the CPU cooler that can influence your cooling performance, including the case you use and the fans installed in it. A system's motherboard can also influence this, especially if it suffers from bending, which results in poor cooler contact with the CPU.

In order to prevent bending from impacting our cooling results, we’ve installed Thermalright’s LGA 1700 contact frame into our testing rig. If your motherboard is affected by bending, your thermal results will be worse than those shown below. Not all motherboards are affected equally by this issue. I tested Raptor Lake CPUs in two motherboards. And while one of them showed significant thermal improvements after installing Thermalright’s LGA1700 contact frame, the other motherboard showed no difference in temperatures whatsoever! Check out our review of the contact frame for more information.

Testing Configuration

 MORE: How to Buy the Right CPU Cooler

MORE: How to Check CPU Temperature

MORE: All CPU Cooling Content

No Power Limits Cinebench Results

With Intel's i9-13900K pushing speeds of 5.5 GHz or higher, even the strongest of coolers hit TJ Max while running Cinebench R23 and other demanding scenarios. As the 13900K is designed to aim for its top safe temperature, we’ll be comparing the overall noise levels and total watts cooled instead. 

The results below are for a 10-minute testing run. But to be sure that was sufficiently long to tax the cooler, we also retested both Thermalright’s Assassin X 120 R SE and DeepCool’s LT720 with a 30-minute Cinebench test. The results didn’t change much at all with the longer test: The average clock speeds maintained dropped by 29 MHz on DeepCool’s LT720 and 31 MHz on Thermalright’s Assassin X 120 R SE. That’s an incredibly small 0.6% difference in clock speeds maintained, a margin of error difference that tells us that the 10-minute tests are indeed long enough to properly test the coolers.

Because of the difficulty of cooling the i9-13900K in this workload, I feel the best way to compare coolers here is to record the average power consumption of the CPU. As Cinebench R23’s multi-core benchmark will push coolers to their limits, it’s also a great test for recording the worst-case scenario of fan noise levels.

When it comes to total cooling capacity, the Amazon Basics air cooler performed true to it’s name and only provided basic cooling capacity. At 198W total, you’re not going to win any overclocking competitions with this cooler – but it’s plenty sufficient for most common tasks, and should have no problems taming your average Core i5 or Ryzen 5 CPU.

The average noise levels of all coolers tested is in this test 48.7 dBA, and many coolers will run loudly at full speeds. That’s not a problem with the Amazon Basics cooler, which runs quietly in all scenarios. With total system noise measuring at only 41.9 dBA, this is the second-quietest CPU cooler I’ve tested. Only Fractal’s $200 360mm AIO was quieter.

You might notice that our noise charts start at 36 decibels, which is the noise floor of our testing environment. This makes 36dB our baseline measurement, as we’re unable to measure noise levels below this threshold. Keep in mind that noise measurements are logarithmic, meaning the differences between the noise levels of the coolers will be more perceptible than these graphs would suggest.

200W Cinebench Results

The Amazon Basics CPU cooler doesn’t technically pass our 200W test. In our default testing environment with a 23C ambient temperature, it’s only able to cool an average of 198W, but it’s still not bad considering the price.

Because of how close this cooler came to passing this test, I wanted to see if a lower ambient temperature would allow the cooler to “pass” this test. It was able to get it to handle 200W, but only when the ambient was 20C or lower. So if you want to drive this cooler to its limits, make sure your climate control is also on point.

125W Cinebench Results

The lowest power limit I test at is 125W, mainly because this is also the lowest level where I can reliably measure noise measurements using Cooler Master’s HAF 700 Berserker. Lower power consumption causes the noise of many CPU coolers to fall below the noise created by the system fans (even while restricted to 35% speed).

Thermal results aren’t very important in this scenario, because even the weakest SFF coolers should be able to cool the CPU below TJ Max in this test. The thermal results of the Amazon Basics cooler here are not impressive at 50C over ambient. Noise levels are more important for the evaluation of performance in this scenario. The results here are excellent for the Amazon Basics air cooler, tied with our best results for low noise levels.

Conclusion

The Amazon Basics CPU cooler is true to it’s name, offering just 198W of cooling capacity when paired with Intel’s i9-13900K. While that’s not chart topping cooling capacity, it’s more than enough for the needs of common users who aren’t running a Core i9 and don’t plan to overclock. And the cooler runs whisper silent in all workloads.

If you’re looking for quiet cooling performance that doesn’t cost much or take up much space in your case, the Amazon Basics CPU Cooler will provide what you need. Just know that it’s not the easiest cooler to install.

 MORE: How to Buy the Right CPU Cooler

MORE: How to Check CPU Temperature

MORE: All CPU Cooling Content

Corsair’s come a long way since its 1994 debut selling L2 cache modules for OEMs. Today the company sells a wide variety of components and peripherals, like the Corsair 5000X case and CX750M Power Supply. Cooling has also been a staple of the company's lineup for a long time, with its flagship iCUE H170i Elite LCD XT proving to be the strongest cooler we’ve tested thus far with Intel’s i9-13900K.  

Along with that cooler, Corsair also sent us a more ‘budget’ option, the 240mm H100i Elite Capellix XT. The H170i earned a spot on our best AIO coolers list, but does the H100i Elite also have what it takes? We’ll have to put it through testing to find out. But first, here are the specifications from Corsair.

Cooler Specifications

Packing and Included Contents

Corsair’ iCUE H100i arrives in a medium-sized box, not dissimilar in size or shape from what expect a shoe purchase to come in.

Included with the package are the following:

• 240 mm Radiator and CPU block
• Mounting hardware and alternative CPU block plates
• Two 120 mm fans
• Mounts for all modern CPU sockets (including AM5 & LGA1700)
• Safety and warranty information
• Lighting and PWM hub
• Pre-applied thermal paste

Installation

The installation of Corsair’s AIO is fairly simple, much like many other coolers on the market.

1.  Secure the backplate.

2. Secure the standoffs.

3. Secure the radiator and fans to the top of the case.

4.  Secure the CPU block to the standoffs.

5. Secure the fan control hub. Connect the CPU block and fans to it.

Features of Corsair iCUE H170i ELITE LCD XT

CPU Block with interchangeable plates

Included with the cooler are three CPU plates with different designs, allowing for limited customization. This cooler can also be upgraded with a 2.1” IPS LCD by purchasing Corsair’s upgrade kit – but you’d be better off purchasing the iCUE H100i Elite LCD XT instead if that interests you, as the upgrade kit is $100 whereas the Elite LCD version of this cooler is only $80 more expensive than the base version.

Full Copper CPU plate

The CPU contact plate on Corsair’s iCUE H100i Elite Capellix XT is made of copper and comes with thermal paste pre-installed in the triangle pattern you’ll find on many of its current cases.

Hardware RGB and PWM Hub

Corsair’s cooler includes a hardware RGB & PWM fan hub, capable of supporting up to 4 additional fans in addition to those that come with the cooler (six devices supported in total).

27mm thick 240 mm radiator

The radiator included is 240 mm with 27 mm thickness. This isn’t the thickest (or thinnest) radiator we’ve seen, but it should be compatible with almost any case that supports a 240mm AIO.

2x AF RGB ELITE PWM 120mm fans

There’s more to a cooler than just the heatsink or radiator, as the fans impact both cooling capacity and noise levels. Corsair includes a pair of 120mm AF RGB ELITE fans with the H100i Elite. The company advertises up to 65.57 CFM of airflow at peak 2,100 RPM and quiet, long-lasting operation thanks to a fluid dynamic bearing. There’s also eight individually addressable RGB LEDs per fan, with the lighting customizable by the company’s comprehensive iCUE software.

By default, the pump and fan curves of Corsair’s iCUE H100i Elite Capellix XT are based on the temperature of the CPU coolant instead of the CPU’s temperature. This method has advantages and drawbacks. Most will appreciate how it reduces noise levels in common loads and eliminates bursty fan behavior. If you’d rather have the fan curve tied to the CPU temperature – or another sensor – this behavior can be changed using Corsair’s iCUE software. 

To demonstrate how the out-of-the-box settings cause the fans to slowly ramp up in speed over time. I’ve included a screenshot below from the iCUE software while running a steady, sustained load.

The advantage to this method is that common loads will not cause the cooler’s fans to reach louder levels. Even when I opened 50 tabs in a row in Mozilla Firefox, the cooler ran as silently as an idle load. 

The disadvantage to this method is that semi-intensive, sustained, workloads (like those we use for testing) will cause the cooler to run as loudly as in a fully unrestricted workload. The fans will also spin at higher speeds for a few minutes once a long-term workload has completed, rather than instantly returning to low speeds. This happens because it takes a short time to bring the AIO’s coolant temperatures down after a sustained workload. The screenshot below demonstrates the fan behavior after an intensive workload has been completed.

Corsair’s iCUE software provides comprehensive customization for its coolers, components and peripherals. I was pleasantly surprised to see my motherboard’s RGB options detected by the software, allowing easy control over all system RGB functions. This is due cooperation between Asus and Corsair that was first announced in 2020. Other motherboard brands are not currently supported by iCue.

The software allows for in-depth controls of the LEDs; you can change the settings for each individual LED on the included AF RGB ELITE fans.

Testing Methodology

While it was fairly easy with previous generations of CPUs for coolers to keep the flagship i9 processor well under TJ max (the maximum temperature a CPU can sustain without throttling) in tough workloads, this is no longer realistically possible on current generation CPUs (and the 13900K especially) without extreme cooling (or enabling power limits).

While in the past a CPU hitting its peak temperature was cause for concern, enthusiasts are going to have to learn to accept high temperatures as “normal” while running demanding workloads with Raptor Lake and Ryzen 7000 CPUs. Modern AMD and Intel CPUs are designed to run fairly hot without any problems – up to 95 degrees Celsius for AMD Ryzen 7000 CPUs, and up to 100 C for Intel’s Core i9-13900K. Similar behavior has been standard in laptops for years due to cooling limitations in tight spaces.

Furthermore, Intel’s Core i9-13900K supports Adaptive Boost Technology (ABT) which allows Core i9 processors to dynamically boost to higher all-core frequencies based on available thermal headroom and electrical conditions. This allows multi-core loads to operate at up to 5.5ghz if the necessary amount of thermal dissipation is there. This feature works in a way that actively seeks high temperatures: If the chip sees that it is running underneath the 100-degree C threshold, it will increase its performance and power consumption until it reaches the safe 100 C limit, thus sustaining higher clocks (and providing better performance) for longer periods.

The increased cooling challenges posed by Raptor Lake mean that we’ve had to change some of the ways we test coolers. Some coolers were able to pass Cinebench R23 multicore testing with Intel’s 12th Gen i9-12900K when power limits were removed (although only the strongest models were able to pass that test). Most liquid coolers and all air coolers I’ve tested “failed” that test because the CPU reached TJ max in this scenario. 

With Raptor Lake’s 13900K, not a single cooler tested has been able to keep the CPU under TJ max in this test – because as we pointed out, the chip is designed to dial up performance and power until it richest that thermal result. We’ll compare performance instead by looking at total benchmark scores and clock speeds maintained. 

I’ll be testing Intel’s i9-13900K CPU using Asus’ TUF Gaming Z690 Gaming Plus WIFI motherboard and Cooler Master’s HAF 700 Berserker computer case, with case fans limited to 35% speeds. The motherboard’s default fan curve is used for the CPU Cooler’s fans.

In addition to testing Cinebench without power limits enforced, we’ll also be showing results when the CPU’s power consumption is limited to a more reasonable 200W. We’ll also show results at 125W for those who prefer whisper-quiet cooling, at the cost of some performance. For both of these results, we’ll show traditional delta over ambient temperature results.

We’ll provide noise level measurements recorded using a PSPL25 Sound Meter for all three power levels tested to compare how much noise each cooler makes in different scenarios. We expect most coolers to run effectively silently at 125W.

LGA1700 Socket Bending

Note there are many factors other than the CPU cooler that can influence your cooling performance, including the case you use and the fans installed in it. A system's motherboard can also influence this, especially if it suffers from bending, which results in poor cooler contact with the CPU. 

In order to prevent bending from impacting our cooling results, we’ve installed Thermalright’s LGA 1700 contact frame into our testing rig. If your motherboard is affected by bending, your thermal results will be worse than those shown below. Not all motherboards are affected equally by this issue. I tested Raptor Lake CPUs in two motherboards. And while one of them showed significant thermal improvements after installing Thermalright’s LGA1700 contact frame, the other motherboard showed no difference in temperatures whatsoever! Check out our review of the contact frame for more information.

Testing Configuration

 MORE: How to Buy the Right CPU Cooler

MORE: How to Check CPU Temperature

MORE: All CPU Cooling Content

No Power Limits Cinebench Results

With Intel's i9-13900K pushing speeds of 5.5 GHz or higher, even the strongest of coolers hit TJ Max while running Cinebench R23 and other demanding scenarios. As the 13900K is designed to aim for its top safe temperature, we’ll be comparing the overall noise levels and total watts cooled instead. 

The results below are for a 10-minute testing run. But to be sure that was sufficiently long to tax the cooler, we also retested both Thermalright’s Assassin X 120 R SE and DeepCool’s LT720 with a 30-minute Cinebench test. The results didn’t change much at all with the longer test: The average clock speeds maintained dropped by 29 MHz on DeepCool’s LT720 and 31 MHz on Thermalright’s Assassin X 120 R SE. That’s an incredibly small 0.6% difference in clock speeds maintained, a margin of error difference that tells us that the 10-minute tests are indeed long enough to properly test the coolers.

Because of the difficulty of cooling the i9-13900K in this workload, we feel the best way to compare coolers here is to record the average power consumption of the CPU. As Cinebench R23’s multi-core benchmark will push coolers to their limits, it’s also a great test for recording the worst-case scenario of fan noise levels. 

The average noise levels of all coolers tested is 48.7 dBA in this test. Cosair’s h100i measures at 48 dBA in this worst case scenario, placing it right in the middle of our results. While certainly audible, this isn’t loud and is what I consider a good noise level in this scenario.  

You might notice that our noise charts start at 36 decibels, which is the noise floor of our testing environment. This makes 36dB our baseline measurement, as we’re unable to measure noise levels below this threshold. Keep in mind that noise measurements are logarithmic, meaning the differences between the noise levels of the coolers will be more perceptible than these graphs would suggest.

When it comes to total cooling capacity, Corsair’s iCUE H100i Elite Capellix XT performed slightly behind our expectations – running behind DeepCool’s AG620 air cooler and Silverstone’s VIDA 240 Slim AIO in total cooling capacity.

200W Cinebench Results

Looking at CPU temperatures when restricting power consumption to a more reasonable 200W, Corsair’s cooler performs slightly behind the 360mm AIOs we’ve tested it against here.  

The acoustic results are a little louder than most other coolers I’ve tested in this scenario, but the noise level is still acceptable. It’s important to note here that in shorter term loads of the same wattage, Corsair’s AIO will run much quieter than competing coolers for the first minute or two, whereas other coolers will reach their peak noise levels quickly.

125W Cinebench Results

The lowest power limit I test at is 125W, mainly because this is also the lowest level where I can reliably measure noise measurements using Cooler Master’s HAF 700 Berserker. Lower power consumption causes the noise of the CPU cooler to fall below the noise created by the system fans (even while restricted to 35% speed). 

In this scenario, Corsair’s cooler was in the top half of our results, cooling the CPU to 36 degrees over ambient. While not noisy in any sense of the word, noise levels at the end of testing were louder than the competition because steady, sustained loads cause the AIOs coolant temperature to rise. It’s important again to note that the noise levels shown here are the result of a steady, sustained load. In common tasks, the cooler will run much quieter.

Conclusion

Corsair’s iCUE H100i Elite Capellix XT did well, cooling loads of 270W+ with Intel’s i9-13900K. If you need a 240mm cooler that doesn’t suffer from bursty fan behavior as speeds spin drastically up and down, Corsair’s iCUE H100i Elite Capellix XT is what you’re looking for. Unlike other coolers, the noise levels of Corsair’s cooler is tied to the CPU Coolant temperature by default – resulting in quieter operation during most common tasks and the elimination of bursty fan behavior.

 MORE: How to Buy the Right CPU Cooler

MORE: How to Check CPU Temperature

MORE: All CPU Cooling Content

The AMD Ryzen 9 7950X3D vs Intel Core i9-13900K rivalry is a battle of flagships for the highest-end of the gaming market, but the chips take drastically different approaches to serving up leading performance. The Core i9-13900K wields Intel's 13th-Gen Raptor Lake x86 hybrid architecture to square off against AMD's chiplet-based Zen 4 Ryzen 7000 chips, but AMD has a trick up its sleeve — the Second-Gen 3D V-Cache tech, which fuses on a 3D-stacked cache chiplet atop the processor to deliver market-leading gaming performance. These two fundamentally different approaches have shaken up our list of the best CPUs for gaming, but each approach has its respective strengths.

AMD's $699 Ryzen 9 7950X3D is geared specifically for gamers looking to blast through CPU-limited games while still having the threaded heft of 16 cores that remain competitive in the heaviest of productivity workloads. The Ryzen 9 7950X3D does lag competitors in some productivity apps due to the tradeoffs associated with the 3D chip-stacking tech, and it doesn't accelerate all games equally. However, one thing is certain — it is the fastest gaming chip on the market.

The $589 Intel Core i9-13900K is Intel's fastest gaming processor, and while it isn't quite as fast as AMD's Ryzen 9 7950X3D in gaming, it has a much friendlier price tag and serves up ultra-competitive performance in both gaming and productivity apps. Intel also continues to support DDR4 memory to offer a less-expensive path for builders, lowering the bar for entry to flagship performance even further. Meanwhile, AMD's decision to support only DDR5 has proven to be a pricing pain point.

The different strengths of these two chips make for a difficult decision if you're looking to pick one for your next high-end build. Below we've taken the Core i9-13900K vs Ryzen 9 7950X3D rivalry through a six-round faceoff to see which chips take the crown in our gaming and application benchmarks, along with other key criteria like power consumption and pricing. Let's see how they stack up. 

Features and Specifications: AMD Ryzen 9 7950X3D vs Intel Core i9-13900K

AMD’s $699 Ryzen 9 7950X3D comes packing sixteen Zen 4 cores and the company’s second-gen 3D V-Cache tech that unlocks 128MB of L3 cache, a combo that delivers the fastest gaming performance available on the market. The 7950X3D has the same underlying design as the Ryzen 9 7950X with a central I/O die and two eight-core compute chiplets that provide a total of 16 cores and 32 threads.

Unlike the standard 7950X, one of the 7950X3D's compute chiplets comes with an additional 3D-stacked L3 cache and boosts to 5.25 GHz. Meanwhile, the 'standard' compute chiplet without cache boosts to 5.7 GHz, the same as the standard 7950X (details here). Despite the lower frequency, the cache-equipped chiplet provides the best performance for games, while the faster chiplet is best for productivity applications. As such, AMD designed an innovative new driver that helps steer threads for different types of workloads to the correct chiplet.

The Ryzen 9 7950X3D brings all of the amenities we would expect from a modern processor, including support for DDR5, PCIe 5.0, and the latest USB connectivity standards. AMD has also added integrated graphics for a basic display out for the first time, which is a plus if you need to troubleshoot. You can see our iGPU testing here.

The 7950X3D sips power — it has a base TPD of 120W and a max 162W PPT, which is 68W lower than the 170W/230W rating for the standard 7950X. The lower power threshold will result in less performance in heavy applications, but you'll still need to plan for a 240mm to 280mm AIO (or better) for the best performance.

The Ryzen 9 7950X3D requires an AM5 motherboard, so pricey DDR5 memory is the only option. Unfortunately, DDR5 continues to be much more expensive than DDR4 memory. Granted, that pricing pressure isn't as much of a consideration at the high end, but you should factor it into your purchasing decision. The AM5 platform is new, so the Ryzen 7000 chips aren't backward compatible with older AM4 motherboards. However, AMD will support the AM5 platform until 2025+.

AMD only allowed overclocking the memory and Infinity Fabric for the previous-gen 5800X3D, but the 7950X3D also adds support for both the auto-overclocking Precision Boost Overdrive (PBO) and the Curve Optimizer. Unfortunately, AMD still doesn’t allow direct frequency overclocking due to a voltage limitation for one of the chiplets.

Intel's $589 sixteen-core Core i9-13900K comes with eight hyper-threaded P-cores and 16 single-threaded E-cores, for a total of 32 threads. That's an increase of eight additional E-cores over the previous-gen Core i9-12900K, equating to more threaded horsepower. You can also save some cash with the $564 Core i9-13900KF, which has identical specs but lacks the integrated GPU.

Intel still uses the 'Intel 7' process node for Raptor Lake but leveraged a newer revision of the silicon to push clock speeds higher while improving power efficiency. The P-cores come with a 3.0 GHz base, but, more importantly, a 5.8 GHz boost that's a whopping 600 MHz increase over the prior gen. Meanwhile, the E-cores now have a 3.0 GHz base (+600 MHz) and stretch up to 4.3 GHz (+400 MHz), meaning that not only do you get more E-cores, but you also get more out of each core.

The Core i9-13900K is also equipped with 32MB of L3 cache and 36MB of L2 and drops into existing 600-series motherboards or the new 700-series models. The chip comes with the integrated UHD Graphics 770 engine with 32 EUs that boost up to 1.65 GHz.

Intel's chips now have a Processor Base Power (PBP) value instead of a TDP metric, and a secondary Maximum Turbo Power (MTP) value that denotes the highest power level during boost activity. The 13900K comes with 125W PBP (base) and 253W MTP (peak) power ratings, but be aware that the chip can always operate at the 253W MTP when it is under load, though the actual power use will vary with application intensity. Intel allows full overclockability, but only when you pair the chip with a Z-series motherboard.

Winner: Tie

The Core i9-13900K vs Ryzen 9 7950X3D battle yields a tie in this category. Both chips support DDR5 and the PCIe 5.0 interface, meaning neither has an inherent connectivity advantage.

Intel has the advantage of backward compatibility for DDR4 and the 600-series chipset, enabling lower-cost memory and motherboards for value seekers (caveats apply). Intel also has its Raptor Lake Refresh chips on the horizon, meaning there will be one more generation of upgrades for the LGA 1700 platform.

AMD's new AM5 platform only supports pricey DDR5 and carries a premium over the Intel motherboard ecosystem, but it does afford extended forward compatibility — AMD plans on supporting the AM5 platform until 2025+.

Given its new hybrid architecture with two types of cores, Intel's core counts aren't directly comparable to AMDs due to their different capabilities. Our performance results will dictate the performance value of the overall designs.

Gaming Benchmarks and Performance: AMD Ryzen 9 7950X3D vs Intel Core i9-13900K

Remember, this article is an overview of our much more in-depth testing in our AMD Ryzen 9 7950X3D Review. Below you can see the geometric mean of our gaming tests with the AMD Ryzen 9 7950X3D vs Intel Core i9-13900K at 1080p and 1440p, with each resolution split into its own chart. As usual, we're testing with an Nvidia GeForce RTX 4090 to reduce GPU-imposed bottlenecks as much as possible, but differences will shrink with lesser cards or higher resolutions. 

The Ryzen 9 7950X3D lives up to its billing as the fastest gaming chip in the world, beating out Intel’s flagship by impressive margins in some titles. We also see the 7950X3D take a big lead in the geometric mean of our test suite, but be aware that the 3D V-Cache doesn’t accelerate all titles equally. Therefore, you should also look at the individual game benchmarks in the above album to assess the overall trend.

In our testing, the $699 Ryzen 9 7950X3D is 12% faster than the $589 Core i9-13900K at 1080p gaming at stock settings, and 9% faster when the chips are overclocked. The latter score is discouraging for Intel — the 13900K consumes far more power yet still can't beat the 7950X3D. As a reminder, Intel released its limited-edition Core i9-13900KS as a preemptive strike against the 7950X3D, but despite its impressive 6 GHz boost clock and big price tag, it offers basically the same gaming performance as the $589 Core i9-13900K. As such, it also falls behind the 7950X3D.

While the 7950X3D is 12% faster than the 13900K in this selection of titles, it also costs 19% more. As mentioned above, you can also equip the 13900K with DDR4 to save even more cash, but be aware that you'll lose a few percentage points of gaming performance on average (see our testing here).

Moving over to 1440p pushes the bottleneck more towards the GPU, so the difference between the chips shrinks. For example, the Ryzen 9 7950X3D is 8% faster than the Core i9-13900K at 1440p, and that shrinks to a 3% difference after both chips are overclocked. It's logical to assume that most high-end gaming systems will use higher-resolution monitors, so the differences could be slight, but the 7950X3D will give you more headroom for future GPU upgrades.

Be aware that large deltas in individual game titles, as we see with the Ryzen 9 7950X3D, impact cumulative measurements. You can see examples of that in titles like Microsoft Flight Simulator 2021, Far Cry 6, and F1 2021. Our test suite heavily favors the improvements from 3D V-Cache, so we also included a table below with results from an additional five games that we don’t normally test. You’ll also find AMD’s own benchmarks in a broad range of titles at the end of the above album to give you an idea of performance gains in other titles. As with all vendor-provided benchmarks, take AMD's slides with a grain of salt.

As always, the AMD vs Intel competition can vary based on the title and the GPU you use, so it's best to make an informed decision based on the types of titles you play frequently. 

Winner: AMD Ryzen 9 7950X3D

On average, the Ryzen 9 7950X3D is the fastest gaming chip money can buy. In our test suite, the Ryzen 9 7950X3D is 12% faster than the Core i9-13900K at 1080p gaming and 8% faster at 1440p. After overclocking, those deltas shrink to 9% and 3% at 1080p and 1440p, respectively.

These gaming results come with a few caveats: The 7950X3D doesn't accelerate all games, and the Core i9-13900K costs 19% less before we factor in its much lower overall platform costs. However, you'll always pay a disproportionate premium for leading performance, and at the end of the day, a win is a win.

As always, we don't recommend chips in this price range if you're only interested in gaming — you'll find lower-priced alternatives from both AMD and Intel that are a far better value for gaming. However, there will always be those determined to build the absolutely fastest gaming system regardless of cost, and the Ryzen 9 7950X3D is the fastest gaming chip currently available for those high-end builds.

Meanwhile, the rest of us should probably wait to see how the eight-core $449 Ryzen 7 7800X3D fares when it arrives in April. This chip will likely offer nearly the same gaming performance as the Ryzen 9 7950X3D, but at a much lower price. 

Productivity Performance: AMD Ryzen 9 7950X3D vs Intel Core i9-13900K

We can boil down performance in productivity apps into two broad categories: single- and multi-threaded. The first slide above shows the geometric mean of performance in several of our most important tests in the single-threaded category, but be sure to look at the expanded results in the above album.

The Core i9-13900K is 17% faster than the standard Ryzen 9 7950X3D configuration in single-threaded work. We have two overclocked configs here for the 7950X3D — the auto-overclocked PBO, and PBO combined with undervolting (PBO UV). Notably, the PBO UV config narrows the 13900K’s lead in single-threaded work to 11%. 

Naturally, those performance deltas won't carry over to every type of workload. However, flipping through the album of test results reveals that the Core i9-13900K excels in single-threaded work, which would ultimately provide snappier performance than the 7950X3D in light applications. The Ryzen 9 7950X3D does lead in the AVX-512 y-cruncher benchmark, but Intel leads all other lightly-threaded benchmarks. 

The first slide in the above album contains a geometric mean of performance in several of our multi-threaded benchmarks. In multi-threaded work, the $589 Core i9-13900K is 4% faster than the $699 Ryzen 9 7950X3D at stock settings, and a mere 1% faster after overclocking. Both of these chips serve up plenty of performance in this type of work, but if gaming isn't as important and you're looking for the utmost performance in threaded apps, the standard Ryzen 9 7950X is the better choice.

Winner: Intel Core i9-13900K

The Core i9-13900K offers a more balanced profile in both single- and multi-threaded productivity applications than the 7950X3D, delivering solid performance in both disciplines. In contrast, while the Ryzen 9 7950X3D remains within striking distance in threaded applications, it loses quite a bit of steam in single-threaded work. Not only does that result in slower performance in lighter workloads and less responsiveness, but it can also hamstring performance in typical real-world productivity workloads — these often include both lightly-threaded and multi-threaded portions in the workflow, and significantly slower performance in either category hampers overall performance.

AMD's Ryzen 7950X3D does benefit from hardware acceleration to speed AVX-512 workloads, an advantage over the 13900K in an increasing number of heavily-threaded tasks, but these continue to be fairly rare for the majority of users. Given its overall superior blend of performance in productivity applications, the Core i9-13900K takes the win. 

Overclocking: AMD Ryzen 9 7950X3D vs Intel Core i9-13900K

We have long since reached the land of diminishing returns for overclocking the highest-end chips from AMD and Intel, like the Ryzen 9 and Core i9 families. Because AMD has pulled Intel into a dogfight for performance superiority on the high end, much of the overclocking frequency headroom is rolled into standard stock performance. However, there are still at least some gains to be had, and both Intel and AMD expose tunable parameters and sophisticated software overclocking utilities, like Intel's XTU and AMD's Ryzen Master.

The Ryzen 9 7950X3D doesn't support overclocking via the CPU multiplier, so you can't change the core clocks directly. You also cannot adjust CPU voltage directly. However, you can overclock the memory and fabric. AMD has also added support for altering the power limits (PPT, TDC, EDC) through the auto-overclocking Precision Boost Overdrive (PBO) feature, and you can also now undervolt the X3D chips using the Curve Optimizer. These features weren’t supported on the previous-gen model. You’ll find plenty of examples of the PBO and PBO+UV settings in our 7950X3D review.

Meanwhile, the Intel Core i9-13900K exposes every knob possible for tweaking and tuning. However, its aggressive out-of-the-box power and boosting algorithms already consume most of the overclocking headroom you could achieve with conventional cooling. Sub-zero overclockers with exotic cooling solutions can reach much higher with the 13900K than the 7950X3D, but that isn't as valuable to the overwhelming majority of users. 

As usual, overclocking performance gains can boil down to your luck in the silicon lottery. As you can see above, overclocking with conventional cooling for both the Intel and AMD processors results in fairly minimal gains in applications. In fact, the 13900K actually suffers from lower performance in single-threaded work because its 5.6 GHz overclock is lower than its stock 5.8 GHz boost. The Core i9-13900K does gain 5.5% in gaming, a higher percentage gain than the 7950X3D, but it still lags behind the 7950X3D's peak gaming performance.

Winner: Tie

The payoff for overclocking is extremely limited with today's high end chips. Still, both platforms have a wealth of tunable parameters for enthusiasts, their respective overclocking advantages, and a suite of auto-overclocking and software utilities. Of course, dedicated tuners could find larger performance boosts by overclocking the core, fabric, and memory more aggressively than we have, but bear in mind that you're always at the whims of the silicon lottery when it comes to overclocking.

AMD's 3D V-Cache technology prevents direct voltage and frequency manipulations, but AMD has unlocked memory, fabric, Precision Boost Overdrive, and undervolting via the Curve Optimizer, all of which combine to provide plenty of avenues for boosting performance via various tuning techniques. In fact, the 7950X3D often delivers a similar amount of performance uplift as the Core i9-13900K, showing the value of its somewhat unconventional approaches.

Meanwhile, Intel offers even more overclocking functionality and the highest peak frequencies you can get from direct tuning. However, much of that capability is already exposed right out of the box via the company's aggressive boosting technologies and sky-high power limits, thus reducing its value to casual overclockers. 

Power Consumption, Efficiency, and Cooling: AMD Ryzen 9 7950X3D vs Intel Core i9-13900K

The Ryzen 9 7950X3D is incredibly power efficient, drawing far less power than either the standard 7950X or the competing Core i9-13900KS, thus delivering superb power efficiency that ranks among the best we’ve ever tested. A quick glance at the HandBrake renders-per-day-per-watt metric shows that the efficiency competition isn't even close — the 7950X3D is the hand's down winner.

AMD's Ryzen chips have excellent power and efficiency metrics, but the Ryzen 9 7950X3D is even more efficient because it has a lower 120W TDP rating. This means the chip is similar to a 7950X operating slightly above the 105W Eco Mode. In contrast, Intel's chip consumes far more power in every type of workload, resulting in higher thermal output.

Winner: AMD

Intel has made plenty of progress in reducing its power consumption metrics, but the Ryzen 9 7950X3D is much more power efficient. Not only does the 7950X3D consume far less peak power, but it also delivers more work per unit of power consumed. That results in a win in power consumption, efficiency, and thermal output, ultimately giving you a cooler and quieter system.

Neither of these chips comes with a bundled cooler, but Intel's higher power consumption means you'll need more aggressive, and thus more expensive, cooling to extract the peak performance. 

Pricing: AMD Ryzen 9 7950X3D vs Intel Core i9-13900K

The Core i9-13900K has impressive performance for its $598 price tag, and you can even go graphics-less to save some cash with the $564 Core i9-13900KF model. That's exceptionally competitive pricing against AMD's $699 Ryzen 9 7950X3D, but it does lag in peak gaming performance. Normalizing the performance values to a dollars-per-fps metric shows that they are pretty closely matched in terms of gaming value, but as you can see when we factor in the total platform cost above, there are other factors at play. 

Both the 13900K and the 7950X3D are best suited for high-end motherboards, but AMD's X670/E motherboards currently start at roughly $80 more than Intel's Z790. The Core i9-13900K is also compatible with previous-gen motherboards if they have the right BIOS, and performance will be the same between 600- and 700-series if the power delivery is identical. That means you could choose a DDR5 Z690 motherboard that offers nearly all the same amenities as the Z790 and have plenty of lower-cost options that save you even more cash. 

Pricing for a suitable memory kit could also be a sticking point. To save some cash, you can go with DDR4 with the 13900K and sacrifice a few percentage points of gaming performance while retaining the full performance in productivity applications. The savings are steep: You'll pay around $60 bucks to start with 32GB of DDR4, but pricing starts at $105 for 32GB of DDR5. That DDR5 tax is unavoidable for Ryzen 9 7950X3D users. Additionally, perhaps the best memory kit is the one you don’t have to buy — many upgraders likely already have a DDR4 kit, which is a plus.

Winner: Intel Core i9-13900K

The $589 Core i9-13900K has a lower price tag than the $699 Ryzen 9 7950X3D, but the fps-per-dollar contest is a wash if we only factor in chip pricing. 

However, the high cost of AMD's X-series motherboards adds extra cost to your build compared to Intel's Z790 platform. Additionally, you can opt for a previous-gen Z690 to save even more cash on a 13900K build (be sure to assure BIOS compatibility).

The high price premium for DDR5 also continues to be a sticking point even for a higher-tier Ryzen 9 build. You'll pay nearly twice as much for DDR5 memory than DDR4 memory, and DDR5 is a strict requirement for AMD's 7950X3D. Overall, these factors combine to make the Ryzen 9 7950X3D a much pricier build, giving the Core i9-13900K the lead in the pricing category.   

Bottom Line: AMD Ryzen 9 7950X3D vs Intel Core i9-13900K

The Core i9-13900K vs Ryzen 9 7950X3D battle ends in a four-to-four tie, but the contest isn't as close as the score implies. In the end, each of these processors has its own distinct advantages and disadvantages depending on your target use case.

If you're building a no-holds-barred gaming rig and pricing isn't a concern, the $699 Ryzen 9 7950X3D is your chip. The 7950X3D is 12% faster than the Core i9-13900K in our selection of gaming benchmarks. Of course, that comes with the caveat that its 3D V-Cache tech doesn't accelerate all games evenly, so the lead could vary depending on your gaming selection. In either case, this chip is the current gaming champ, and it often isn't a close contest. That's especially true in some game titles, like Microsoft Flight Simulator 2021 and F1 2021.

The $589 Core i9-13900K isn't as attractive for a gaming-focused build if you're going with DDR5 memory, but it becomes much more compelling with DDR4 if you're looking to access flagship-class performance at a lower price point. You will sacrifice a few percentage points of gaming performance with DDR4, but you could save quite a bit of money for other system additives. However, if you're strictly focused on flagship-level gaming at a more affordable price, the less-expensive $409 Core i7-13700K offers nearly the same gaming performance as the 13900K.

The Ryzen 9 7950X3D is competitive in most types of apps, but its lower performance in lightly-threaded tasks could be a sticking point if you're focused on productivity work with complex workflows. For productivity-focused systems, or if you're generally looking for a solid all-rounder, the Core i9-13900K is the better choice.

The 7950X3D holds the advantage in all power consumption metrics, including peak power and efficiency, ranking among the best we’ve ever tested. That results in more forgiving cooling requirements and a cooler and quieter system. On the other end of the spectrum, overclocking is becoming increasingly less interesting on the high end for most enthusiasts — you'll get a better payoff in the mid-range and low-end of the market. The Intel and AMD processors both provide at least some uplift after overclocking, making that contest a wash.

On the pricing front, the $699 Ryzen 9 7950X3D carries a hefty premium that, combined with the high motherboard and memory pricing, limits it to the highest-end gaming machines. But there will always be performance addicts willing to pay top dollar for the absolute best; for those, the Ryzen 9 7950X3D offers the fastest gaming performance money can buy. We also expect that the 7950X3D will see steep discounts over time. Meanwhile, the Core i9-13900K slots in as a more affordable option and a better all-rounder in productivity applications.

We typically don't recommend spending this much on a chip if gaming is your primary goal — you can find better lower-cost options from both camps. Instead, gamers should wait to see how the eight-core $449 Ryzen 7 7800X3D pans out when it arrives on April 6, 2023. AMD says it will nearly match the 7950X3D while delivering Core i9-13900K-beating performance at a much lower price point. 

• MORE: AMD vs Intel
• MORE: Zen 4 Ryzen 7000 All We Know
• MORE: 13th-Gen Intel Raptor Lake All We Know 

Remember back when Windows 11 launched and there was a concern about how the default of enabling Virtualization Based Security (VBS) and HyperVisor-Enforced Code Integrity (HVCI) might impact performance? There was a lot of noise made, benchmarks were run... and then we all moved on. Flash forward to 2023, and I recently discovered that sometime in the past few months, the PC I use for the GPU benchmarks hierarchy had it turned on, which could have been the result of mucking about with Windows Subsystem for Linux while trying to get Stable Diffusion benchmarks on various GPUs. (We have an article on how to disable VBS should you want to).

Correction (3/22/23): An earlier version of this article incorrectly stated that a Windows update had enabled VBS on my PC. However, we have since learned that Windows updates do not enable VBS, so it must have been something else I installed or enabled. Whatever the case, most new PCs ship with VBS on and clean Windows 11 installs have it enabled also.

This defaulting to VBS on, everywhere, worried me, because I'm already in the middle of retesting all the pertinent graphics cards for the 2023 version of the GPU hierarchy, on a new testbed that includes a Core i9-13900K CPU, 32GB of DDR5-6600 G.Skill memory, and a Sabrent Rocket 4 Plus-G 4TB M.2 SSD. Needless to say, you don't put together best-in-class parts only to run extra features that can hurt performance.

Except... I did. When I put together the new testbed for the coming year back in November, just before the RTX 4080 and RX 7900 XTX/XT launches, I was under a time crunch. I got Windows 11 installed and updated, downloaded the rather massive 1.5TB of games that I use for testing onto the SSD, and got to work — all with VBS enabled. Having now caught my breath and with a bit of extra time, I belatedly realized my error, if you can call it that.

So I set about testing, and retesting, performance of the fastest graphics card, the GeForce RTX 4090, with and without VBS enabled. After all, we're now two new CPU generations beyond what we had at the Windows 11 launch, and with faster CPUs and new architectures, perhaps VBS has even less of an impact than before. At the same time, we're also using new GPUs that deliver substantially more performance than the RTX 3090, which was the fastest GPU back in 2021, which could make CPU bottlenecks and extras like VBS more of a hindrance than before.

You can see our test PC hardware, using Nvidia's 528.49 drivers (which have now been superseded, thrice). Let's get straight to the results, with our updated test suite and settings that consist of a battery of 15 games, at four different settings/resolution combinations. We're going to summarize things in a table, split into average FPS on the left and 1% low FPS (the average FPS of the bottom 1% of frametimes) on the right.

To be clear, all of the testing was done on the same PC, over a period of a few days. No game updates were applied, no new drivers were installed, etc. to keep things as apples-to-apples as possible. The one change was to disable VBS (because it was on initially, the Windows 11 default).

Each test was run multiple times to ensure consistency of results, which does bring up the one discrepancy: Total War: Warhammer 3 performance is all over the place right now. I don't recall that being the case in the past, but sometime in February or perhaps early March, things seem to have changed for the worse. I'm still investigating the cause, and I'm not sure if it's the game, my system, or something else.

Taking the high level view of things, perhaps it doesn't look too bad. Disabling VBS improved performance by up to 5% overall, and that dropped to just 2% at 4K ultra. And if you're running this level of gaming hardware, we think you're probably also hoping to run 4K ultra. But even at our highest possible settings, there are still some noteworthy exceptions.

The biggest improvement overall comes in Microsoft Flight Simulator, which makes sense as that game tends to be very CPU limited even with the fastest possible processors. Turning off VBS consistently improved performance in our RTX 4090 testing by around 10%, and the 1% lows increased by as much as 15%.

Not coincidentally, Flight Simulator is also one of the games that absolutely loves AMD's large 3D V-Cache on the Ryzen 9 7950X3D. Our CPU tests use a different, less demanding test sequence, but even there the AMD chips with large caches are anywhere from about 20% (Ryzen 7 5800X3D) to 40% (7900X3D) faster than the Core i9-13900K. Perhaps VSB would have less of an impact on AMD's X3D CPUs, but I didn't have access to one of those for testing.

Another game that tends to bump into CPU bottlenecks at lower settings is Far Cry 6, and it also saw pretty consistent 5% or higher increases in performance — noticeable in benchmarks, but less so in actual gaming. Interestingly, Cyberpunk 2077 with ray tracing enabled also still saw about 5% higher performance. That's perhaps because the work of building the BVH structures for ray tracing calculations happens on the CPU; many of the other ray tracing games also showed 5% or higher increases.

What about games where VBS didn't matter much if at all? Bright Memory Infinite (the standalone benchmark, not the full game) showed almost no change, and Minecraft only showed a modest improvement at 1080p with our more taxing settings (24 RT render chunk distance). A Plague Tale: Requiem, Borderlands 3, Forza Horizon 5, and Red Dead Redemption 2 also showed less impact, though in some cases the minimum FPS may have changed more.

(And again, I'm not really saying anything about Total War: Warhammer 3 as performance fluctuated far too much. Even after more than 20 runs each, with and without VBS, there was no clear typical result. Instead of a bell curve, the results fell into three clumps at the low, mid, and high range, with the 1% lows showing even less consistency. Removing TWW3 from our geometric mean only changes the 1% low delta by less than two percent, though, so I left it in.)

The biggest deltas are generally at 1080p, and it didn't seem to matter much whether we were running "medium" or "ultra" settings. That's probably because ultra settings often hit the CPU harder for other calculations, so it's not just a case of higher resolution textures or shadows.

But the question remains: to VBS or not to VBS? Especially for my GPU testing. The good news is that it's pretty much a never-ending process, since new drivers and game patches seem to routinely invalidate older results. I could switch at some point to having VBS off, and maybe I will. But that retesting is also the bane of GPU benchmarks.

Windows VBS: The Bottom Line

So, should you leave VBS on or turn it off? It's not quite that clear cut of a question and answer. The actual security benefits, particularly for a home desktop that doesn't go anywhere, are probably minimal. And if you're serious about squeezing every last bit of performance out of your hardware — via improved cooling, overclocking, and buying more expensive hardware — losing 5% just to some obscure "security benefits" probably isn't worth doing, so you could disable VBS.

Still, having VBS turned on is now the default for new Windows installations (and I'm pretty sure one of the various Windows Updates that came out in late 2022 may have also switched it back on if it was disabled). So you can argue that Microsoft at least thinks it's important and it should be left on. However, the fact that Microsoft also has instructions on how to go about disabling it indicates the performance impact can be very real.

It's also worth noting that the 5~10 percent drop in performance remains consistent with what we measured way back in 2021 when Windows 11 first launched. Nearly two years of upgraded hardware later, sporting some of the most potent components money can buy, and we're still looking at a 5% loss on average in gaming performance. For a top-tier gaming setup, that's almost as much of a performance gain from a traditional CPU architecture update — though Raptor Lake and Zen 4 provided much bigger increases than in the past.

For a lot of people, particularly those with less extreme hardware, the performance penalty while gaming will more likely fall into the low single digit percentage points. But if you're trying to set a performance record, it could certainly hold you back. And now we're left wondering what new vulnerabilities and security mitigations will come next, and how much those may hurt performance. Progress isn't always in a single direction, unfortunately.

AMD has finally shared its own official gaming benchmarks of the hotly-anticipated $449 Ryzen 7 7800X3D going toe-to-toe with Intel's flagship $589 Core i9-13900K and winning by an average of 20%, and up to 24% in one title - but for $150 less. The new benchmarks also give us a limited comparison of the Ryzen 7 7800X3D with the Ryzen 9 7950X3D, too, implying that the less expensive chip offers nearly the same level of gaming performance as the pricey flagship - but for $250 less. 

AMD has already released its flagship 3D V-Cache-powered $699 Ryzen 9 7950X3D to critical acclaim, with the new chip taking the crown as the best CPU for gaming while maintaining a commanding presence on our CPU benchmark hierarchy. The company also low-key launched the $599 Ryzen 9 7900X3D without sampling press (you can see our review at the link).

However, AMD has held back launching its $449 eight-core Ryzen 7 7800X3D chips until April and has been stingy with the benchmarks, having only compared it to the previous-gen Ryzen 7 5800X3D in gaming and not the entirely more relevant Core i9-13900K and Ryzen 9 7950X3D. 

That might be because the 7800X3D is too good for its own good: the eight-core, 16-thread Ryzen 7 7800X3D stands out as the go-to model for gamers with this generation, so it'll be the big volume seller. AMD’s reps tell us the 78000X3D will offer the lion’s share of 3D V-Cache’s gaming performance, but at a much more affordable price point than the other X3D models. So naturally, if the 7800X3D offers nearly the same level of gaming performance for $250 less than the flagship, that won't leave much reason to buy the pricier models once the 7800X3D arrives.

We’ll have to see things pan out in our own benchmark testing, but it looks like the Ryzen 7 7800X3D will take the lead over Intel’s Raptor Lake when it arrives next month. AMD's new benchmarks below certainly paint that picture, too. 

As with all vendor-provided benchmarks, take these with a grain of salt. (We've included the test notes below with the test setup particulars, but the '5' footnote doesn't list the 13900K details; those are in the '2' footnote).

We've seen the bottom half of the chart before — it compares the Ryzen 7 7800X3D to the Ryzen 7 5800X3D. But the top half is new with four gaming benchmarks versus the Core i9-13900K. 

AMD claims the Ryzen 7 7800X3D is 20% faster, on average, in gaming than the Core i9-13900K across these gaming benchmarks. The Ryzen 7 7800X3D carves out a 24% win in Horizon Zero Dawn, a 23% win in Red Dead Redemption 2, an 18% win in Total War: Three Kingdoms, and a 13% win in Rainbow Six Siege.

This is a limited selection of games, and they tend to be the types of titles that benefit from the gaming benefits of the 3D V-Cache, whereas some games don't benefit quite as much. Conversely, some other games benefit from the tech far more, so this list isn't entirely skewed. 

The zoomed-out version includes the Ryzen 9 7950X3D vs Core i9-13900K comparison. Here we see that AMD tested two of the same games against the Core i9-13900K with both the 7950X3D and the 7800X3D, which helps us make at least a small comparison between the two AMD chips. 

In Horizon Zero Dawn (not "down," as listed in the slide), we can see that the 7800X3D is a mere 3% slower than the 7950X3D, whereas the two chips tie in Rainbow Six Seige.

Yes, two benchmarks don't give us much to work with, but it certainly implies that the 7800X3D will do just what we've suspected — provide within 5% of the performance of the 7950X3D, but for $250 less. We'll have to see how that pans out across a larger gamut of gaming benchmarks in our review, of course. 

Thoughts

AMD's Ryzen 7 7800X3D has plenty of promise, and we're sure that it will be this generation's Ryzen 7 5800X3D — the overwhelming choice for gamers that are solely interested in gaming.

The 12-core and 16-core Ryzen 7000X3D bring 3D V-Cache tech to AMD's multi-compute-chiplet processors for the first time (deep dive here), but they require special handling via AMD's drivers to ensure the highest performance, while the Ryzen 7 7800X3D has a single chiplet and will operate more like a traditional CPU. AMD's special driver sauce does a great job of delivering exceptional gaming performance for multi-chiplet designs without user intervention, but the 7800X3D will have a simpler plug-and-play implementation.

AMD was also uncharacteristically silent about the 12-core Ryzen 9 7900X3D, but for different reasons. AMD shared zero official benchmarks for the 7900X3D and didn't sample it to the press. We can see why; in our Ryzen 9 7900X3D review, we saw that the price-to-performance ratio was less than inspiring. If you're buying on the high end, there's no reason not to plunk down the comparatively small extra amount to get the flagship Ryzen 9 7950X3D.  

In contrast, it appears that the Ryzen 7 7800X3D will be the complete opposite; this chip could severely cannibalize the sales of AMD's own higher-end models. We'll see soon enough, the Ryzen 7 7800X3D comes to market on April 6, 2023, and you can be sure that we'll have our review published in the same time frame. 

While the company started out selling L2 cache modules for OEMs, today Corsair sells a wide variety of components and peripherals, like the Corsair 5000X case and CX750M Power Supply. Cooling has also been a staple of the company's lineup for a long time; most recently we tested Corsair’s iCUE H100i Elite 240mm with Intel’s i9-12900K. 

But with Raptor Lake raising the bar of cooling difficulty, Corsair has sent us its strongest AIO yet – the 420mm iCUE H170i Elite LCD XT. Can it tame Intel’s 13900K and earn a spot on our best AIO coolers list? It definitely has a size advantage, as well as a high $309.99 MSRP, but we’ll have to put it through testing to find out how it performs. First, here are the specifications from Corsair.

Cooler Specifications

Packing and Included Contents 

420mm is far from small, and as such the Corsair iCUE H170i Elite LCD XT arrives in a package that’s almost as long as my arm.

Included with the package are the following:

• 420 mm Radiator
• CPU block with 2.1-inch IPS Display
• 3x 140 mm fans
• Mounts for all modern CPU sockets (including AM5 & LGA1700)
• Safety and warranty information
• Hardware lighting and PWM hub
• Pre-applied thermal paste

Installation

The installation of Corsair’s AIO is fairly simple.

1. Secure the backplate.

2. Secure the standoffs.

3. Secure the radiator and fans to the top of the case.

4. Secure the CPU block to the standoffs.

5. Secure the fan control hub, then connect the CPU block and fans to it.

Features of Corsair iCUE H170i Elite LCD XT

2.1-inch 480x480 IPS Display

On top of the CPU block is a vivid 480x480 IPS display that can show the sensor readings of your choice with a variety of display themes, or be loaded with a custom background or animated GIF of your choosing.

Full Copper CPU plate

The CPU contact plate on Corsair’s iCUE H170i Elite LCD XT is made of copper and comes with thermal paste pre-installed.

Hardware RGB & PWM Hub

The H170i Elite comes with a hardware RGB & PWM hub, capable of supporting up to three additional extra fans, not including those that come with the cooler (6 devices supported in total).

27mm thick 420 mm radiator

The S36 Prisma features a 420 mm radiator that’s 27 mm (1.06 inches) thick. This isn’t the thickest (or thinnest) radiator we’ve seen, but it should be compatible with any case that supports a 420mm AIO.

3x AF RGB Elite PWM 140mm fans

There’s more to a cooler than just the heatsink or radiator. The bundled fans have a significant impact on cooling and noise levels. Included with the iCUE H170i Elite LCD XT are three 140mm AF RGB Elite fans. The versions of these fans included with the cooler are slightly stronger than the standard AF RGB Elite fans, with a maximum speed of 1700RPM vs 1600 on the standard models. Corsair advertises powerful airflow, up to 89 CFM of peak airflow, a low-noise rating (10-33.8 dBA), and durability thanks to fluid dynamic bearings. Here are the fan’s specs:

Fan Curve tied to AIO Coolant temperature, rather than CPU temperature.

The pump and fan curves of the iCUE H170i Elite LCD XT are based on the temperature of the CPU coolant instead of the CPU. This has advantages and drawbacks, but many will appreciate how it reduces noise levels in common loads and eliminates bursty fan behavior. 

The disadvantage to this method is that semi-intensive, but sustained, workloads will cause the cooler to run just as loudly as it would in a fully unrestricted workload. The fans will also spin at higher speeds for a moderate period of time once the workload has completed, rather than instantly returning to low speeds. This happens because it takes a few moments to bring the AIO’s coolant temperatures down after a sustained workload.

Comprehensive Software Suite

Corsair’s iCUE software provides comprehensive customization for its coolers, components and peripherals. I was pleasantly surprised to see my motherboard’s RGB options detected by the software, allowing easy control over all system RGB functions. This is due to cooperation between the two companies that was first announced in 2020. Other motherboard brands are not currently supported by iCue.

The software allows for in-depth controls of the cooler’s LEDs. You can change the settings for each individual LED on the included AF RGB Elite fans.

You are also able to change what is shown on the CPU block’s small screen via the software. You can set it to show the time, coolant temperature, or another sensor of your choice. The background can be changed from a variety of customizable presets, or you can also upload an image or GIF file.

Testing Methodology

While it was fairly easy with previous CPU generations for coolers to keep the flagship i9 processor well under TJ max (the maximum temperature a CPU can sustain without throttling) in tough workloads, this is no longer realistically possible on current generation CPUs (and the 13900K specifically) without extreme cooling (or enabling power limits).

While in the past a CPU hitting its peak temperature was cause for concern, enthusiasts are going to have to learn to accept high temperatures as “normal” while running demanding workloads with Raptor Lake and Ryzen 7000 CPUs. Modern AMD and Intel CPUs are designed to run fairly hot without any problems – up to 95 degrees Celsius and for AMD Ryzen 7000 CPUs and up to 100 C for Intel’s Core i9-13900K. Similar behavior has been standard in laptops for years due to cooling limitations in tight spaces.

Furthermore, Intel’s Core i9-13900K supports Adaptive Boost Technology (ABT) which allows Core i9 processors to dynamically boost to higher all-core frequencies based on available thermal headroom and electrical conditions. This allows multi-core loads to operate at up to 5.5 GHz if the necessary amount of thermal dissipation is there. This feature works in a way that actively seeks high temperatures: If the chip sees that it is running underneath the 100-degree C threshold, it will increase its performance and power consumption until it reaches the safe 100 C limit, thus maintaining higher clocks (and providing better performance) for longer periods.

The increased cooling challenges posed by Raptor Lake mean that we’ve had to change some of the ways we test coolers. Some coolers were able to pass Cinebench R23 multicore testing with Intel’s 12th Gen i9-12900K when power limits were removed (although only the strongest models were able to pass that test). Most liquid coolers and all air coolers I’ve tested “failed” that test because the CPU reached TJ max in this scenario.

With Raptor Lake’s 13900K, not a single cooler tested has been able to keep the CPU under TJ max in this test – because as we pointed out, the chip is designed to dial up performance and power until it reaches that thermal result. For these intense loads, we’ll test performance by comparing the total amount of watts cooled and noise levels.

I’ll be testing Intel’s i9-13900K CPU using Asus’ TUF Gaming Z690 Gaming Plus WIFI motherboard and Cooler Master’s HAF 700 Berserker PC case, with case fans limited to 35% speeds. The motherboard’s default fan curve is used for the CPU Cooler’s fans.

In addition to testing Cinebench without power limits enforced, we’ll also be showing results when the CPU’s power consumption is limited to a more reasonable 200W. We’ll also show results at 125W for those who prefer whisper-quiet cooling, at the cost of some performance. For both of these results, we’ll show traditional delta over ambient temperature results.

We’ll provide noise level measurements recorded using a PSPL25 Sound Meter for all three power levels tested to compare how much noise each cooler makes in different scenarios. We expect most coolers to run quietly at 125W.

LGA1700 Socket Bending

Note there are many factors other than the CPU cooler that can influence your cooling performance, including the case you use and the fans installed in it. A system's motherboard can also influence this, especially if it suffers from bending, which results in poor cooler contact with the CPU.

In order to prevent bending from impacting our cooling results, we’ve installed Thermalright’s LGA 1700 contact frame into our testing rig. If your motherboard is affected by bending, your thermal results will be worse than those shown below. Not all motherboards are affected equally by this issue. I tested Raptor Lake CPUs in two motherboards. And while one of them showed significant thermal improvements after installing Thermalright’s LGA1700 contact frame, the other motherboard showed no difference in temperatures whatsoever! Check out our review of the contact frame for more information.

Testing Configuration

 MORE: How to Buy the Right CPU Cooler

MORE: How to Check CPU Temperature

MORE: All CPU Cooling Content

No Power Limits Cinebench Results

With Intel's i9-13900K pushing speeds of 5.5 GHz or higher, even the strongest of coolers hit TJ Max while running Cinebench R23 and other demanding scenarios. As the 13900K is designed to aim for its top safe temperature, we’ll be comparing the overall benchmark score and the CPU’s clock speeds instead.

The results below are for a 10-minute testing run. But to be sure that was sufficiently long to tax the cooler, we also retested both Thermalright’s Assassin X 120 R SE and DeepCool’s LT720 with a 30-minute Cinebench test. The results didn’t change much at all with the longer test: The average clock speeds maintained dropped by 29 MHz on DeepCool’s LT720 and 31 MHz on Thermalright’s Assassin X 120 R SE. That’s an incredibly small 0.6% difference in clock speeds maintained, a margin of error difference that tells us that the 10-minute tests are indeed long enough to properly test the coolers.

Because of the difficulty of cooling the i9-13900K in this workload, we feel the best way to compare coolers here is to record the average power consumption of the CPU. As Cinebench R23’s multi-core benchmark will push coolers to their limits, it’s also a great test for recording the worst-case scenario of fan noise levels.

Looking at noise levels, you’ll see Corsair’s cooler reached 52.6 dBA in this test. This does place it on the louder end of the results shown here, but it’s still in the range of what I consider an acceptable noise level.

You might notice that our noise charts start at 36 decibels, which is the noise floor of our testing environment. This makes 36dB our baseline measurement, as we’re unable to measure noise levels below this threshold. Keep in mind that noise measurements are logarithmic, meaning the differences between the noise levels of the coolers will be more perceptible than these graphs would suggest.

When it comes to total cooling capacity, Corsair’s iCUE H170i Elite LCD XT achieved the best result we’ve ever seen paired with Intel’s i9-13900K – cooling on average over 325W in Cinebench testing.

200W Cinebench Results

Looking at CPU temperatures when restricting power consumption to a more reasonable 200W, Corsair’s iCUE H170i Elite LCD XT maintains its lead over all other coolers tested in comparison – if only by a single degree.

However, the acoustic results are not as impressive due to the unique way in which Corsair’s AIOs operate, tying coolant temperature to fan speed rather than CPU temperature. This causes a sustained 200W load to run just as loudly as a 300W+ load.

It’s important to note here that in shorter term loads of the same wattage, Corsair’s AIO will run much quieter than competing coolers for the first minute or two, whereas other coolers will reach their peak noise levels sooner.

125W Cinebench Results

The lowest power limit I test at is 125W, mainly because this is also the lowest level where I can reliably measure noise using Cooler Master’s HAF 700 Berserker. Lower power consumption causes the noise of the CPU cooler to fall below the noise created by the system fans (even while restricted to 35% speed).

In this scenario, Corsair’s cooler was tied with the best 360mm AIOs we have tested for best overall CPU temperature. While the cooler was not noisy in any sense of the word, the acoustic levels at the end of testing were louder than the competition again because steady, sustained loads cause the AIOs coolant temperature to rise. In common bursty tasks, the cooler will run much quieter.

Conclusion

Corsair’s iCUE H170i Elite LCD XT is the strongest AIO we’ve tested with Intel’s i9-13900K, cooling over 325W in our most intensive thermal tests. Unlike other coolers, the noise level of Corsair’s cooler is tied to the CPU coolant temp, resulting in quieter operation during most common tasks and the elimination of bursty fan behavior, but noisier operation under the kinds of sustained loads that we use for testing.

There’s no question. The Corsair iCUE H170i Elite LCD XT is one of the best coolers currently available. It’s also quite expensive at $310. But if you don’t mind going without the display, the company also sells an iCUE H170i Elite Capellix XT that the company says features the same radiator and pump for the same cooling performance at a lower $240 price. That’s still no small amount of money. But if your cooling needs are extreme enough to warrant a 420 mm radiator, you can probably afford to spend a bit extra on keeping your powerful CPU as cool as it can be under load.

 MORE: How to Buy the Right CPU Cooler

MORE: How to Check CPU Temperature

MORE: All CPU Cooling Content

AMD's Ryzen 7000X3D V-Cache processors, which will aim to take the number one spot on the list of best CPUs for gaming, are almost among us. However, before the rumor mill grinds to a stop, new benchmarks of the Ryzen 9 7950X3D reportedly illustrate the flagship Zen 4's supremacy over the Intel Core i9-13900K.

The gaming benchmarks (via chi11eddog) have the Ryzen 9 7950X3D pairing up with the Nvidia GeForce RTX 3090 in five different titles. Additionally, the user seemingly tested the Ryzen 9 7950X3D's gaming performance at a 1080p (1920x1080) resolution with maximum graphics settings. Granted that it's a small sample size of games and being leaked benchmarks from an unidentified source, throw some salt over the results.

Overall, the Ryzen 9 7950X3D outperformed the Core i9-13900K by 11%. However, the margins were more pronounced in some games than others. For example, the Ryzen 9 7950X3D delivered up to 35% higher performance in Assassin's Creed Valhalla. 

In other games, such as Far Cry 5 and Shadow of the Tomb Raider, Ryzen 9 7950X3D was only 12% and 11% faster than the Core i9-13900K, respectively. The Ryzen 9 7950X3D's dominance in Metro Exodus dropped to 5%. The only victory for the Core i9-13900K was in Star Control, where the Raptor Lake barely edged out the Ryzen 9 7950X3D by a 2% margin.

Ryzen 9 7950X3D Gaming Benchmarks

The Ryzen 9 7950X3D, on average, posted 18% better gaming performance than the Ryzen 9 7950X, one of the two highest-performance chips for gaming. Again, Assassin's Creed Valhalla heavily favored the Ryzen 9 7950X3D, as evidenced by the 39% delta between it and the Ryzen 9 7950X. The Ryzen 9 7950X3D also excelled in Far Cry 5, outperforming its vanilla counterpart by 32%.

Regarding Shadow of the Tomb Raider and Metro Exodus, the Ryzen 9 7950X3D pumped out 16% and 4% higher frame rates, respectively. Furthermore, unlike the case with the Core i9-13900K, the margin between the Ryzen 9 7950X3D and Ryzen 9 7950X was almost non-existent.

The Ryzen 9 7950X3D launches on February 28 for $699, the original MSRP for the Ryzen 9 7950X. However, much has changed, and the Ryzen 9 7950X currently retails for as low as $576. If you trust the benchmarks, the Ryzen 9 7950X3D offers 18% higher gaming performance for 21% more money. Comparing the Ryzen 9 7950X3D to the Core i9-13900K suddenly doesn't look like a good deal, though. The Core i9-13900K debuted at $599, but discounts have brought the Raptor Lake down to $570. Therefore, the Ryzen 9 7950X3D only provides gamers with 11% faster gaming performance at a 23% higher price tag.

However, this will come with some tradeoffs in creativity applications, which could impact the chips' ranking on our CPU benchmarks hierarchy. 

The review embargo for the Ryzen 7000X3D V-Cache processors allegedly lifts on February 27, one day before the chips hit the retail market. We'll see what the Ryzen 9 7950X3D brings to the table soon enough, so put your final purchase decision on hold until the reviews come out.

If you're looking to get your hands on a shiny new 13th Gen Raptor Lake CPU, then how about a powerful Core i9-13900K for only $559 from eBay (via Antonline)? This is the lowest price I've seen for this CPU and could be the heart of your new gaming rig. This processor also has great single-core performance. Have a look at our review of the i9-13900K for more details. 

Prefer gaming on a laptop? There's an excellent discount today on this RTX 3080-powered Asus ROG Strix Scar 15 gaming laptop for $1,669 from Newegg. This portable gaming machine also comes with an AMD Ryzen 9 5900HX CPU, 16GB of RAM, and 1TB of storage. 

Supporting 12th/13th Gen Intel processors with its LGA 1700 socket, the ASRock Z690 Steel Legend is only $171 at Newegg. This motherboard also supports DDR4 memory complete with white accents and would look good in a color-themed build. 

TL;DR — Today’s Best Deals

• Intel Core i9-13900K: now $559 at eBay (was $669)
• Asus ROG Strix Scar 15 Gaming Laptop: now $1,669 at Newegg (was $2,199)
• ASRock Z690 Steel Legend ATX Intel Motherboard: now $171 at Newegg (was $249)

Today’s best deals in detail

Looking for more deals?

IceBerg Thermal's IceSleet G6 Stealth is a unique air cooler that features a radiator that wraps around the included fan. With today’s high-end CPUs kicking out more heat than ever, does the Iceberg Thermal IceSleet G6 Stealth have what it takes to cool Intel’s i9-13900K and earn a spot on our best CPU coolers list?

We’ll have to put it through testing to say for sure, but it does sport some impressive specs and stats. IceBerg claims the fan's lifespan is rated at 17.5 years, and the company backs that up with a lengthy 10-year warranty. The big question is whether the cooler can stand out against options that are far cheaper.

We'll find out shortly in testing. But first, here are the cooler’s specifications, direct from Iceberg Thermal.

Cooler Specifications

Packing and Included Contents 

Iceberg Thermal’s IceSleet G6 Stealth is uniquely packaged. The inner contents consist of two cardboard boxes, one small and one larger. The smaller box contains the accessories for the unit, and the larger box unfolds to reveal the CPU cooler.

Included with the package are the following:

• CPU Heatsink
• 1x 140 mm fan
• Mounts for all modern CPU sockets (including AM5 & LGA1700)
• Quick Start Guide
• Thermal paste
• Small Screwdriver

Installation

Installing the G6 Stealth on an Intel system was fairly simple. To begin, you attach the mounting studs to the backplate and secure them with the included O-rings. Then, press the backplate to the motherboard, slide on the standoffs, install the mounting bars and secure them using the included screws.

Once the thermal paste has been placed on the CPU, you push the heatsink against the mounting bars and secure it with a long screwdriver (which isn't included). The last step is to attach the fan to the top, and secure it using the included screws and small screwdriver. 

The installation on an AMD system is slightly simpler, as it utilizes the default backplate.

Features of Iceberg Thermal’s IceSleet G6 Stealth

Impressive 10-year warranty

Iceberg Thermal backs the IceSleet G6 Stealth with an impressive 10-year warranty. This is the longest warranty we have ever seen on a CPU cooler, and an indication that IceBerg is confident in the quality of its products.

Nickel-plated copper CPU block

The CPU contact plate on the IceSleet G6 Stealth is copper, plated with Nickel. The base is a bit smaller than those we’ve seen on other coolers, so it will be interesting to see how well it performs in our testing.

All black, nickel-plated

The entire cooler is plated in black nickel. This helps prevent corrosion and should also keep it looking good for several years.

Unique wrap-around radiator design

The IceSleet G6 Stealth offers a unique design, with the radiator engulfing a single 140mm fan that sits in the middle of the unit. The tower has 6 groups of fins, 57 fins in total. Each group is shaped differently than the group above or below it, to improve the overall cooling performance. The lowest set of fins is recessed, to allow for better RAM compatibility, supporting sticks up to 56 mm tall. 

One 140mm fan

There’s more to a cooler than just the heatsink or radiator. The bundled fan(s) impact both cooling and noise levels. Included with the IceSleet G6 Stealth is a single 140 mm fan, which is installed in the center of the unit. 

The fan features Fluid Dynamic Bearings, and is two-plane balanced, to support a long lifespan. IceBerg claims the fan will last for over 150,000 hours of operation. Converted into more traditional terms, they claim the fan should last an impressive 17.5 years.

The fan also supports a unique feature designed to reduce noise levels in low heat workloads called “Auto Stop/Start.” Not unlike the zero-RPM features seen on many graphics cards and processors, this keeps the cooling fan turned off for truly silent cooling until the PWM signal rises to 21%.

Testing Methodology

It was fairly easy with previous-gen CPUs for coolers to keep flagship processors well under TJ max (the maximum temperature a CPU can sustain without throttling) in tough workloads. But this is no longer realistically possible on current-generation CPUs (and the 13900K especially), without extreme cooling (or enabling power limits).

Enthusiasts are going to have to learn to accept high temperatures as “normal” while running demanding workloads with Raptor Lake and Ryzen 7000 CPUs. Modern AMD and Intel CPUs are designed to run fairly hot without any problems – up to 95 degrees Celsius for AMD Ryzen 7000 CPUs and up to 100 C for Intel’s Core i9-13900K. Similar behavior has been standard in laptops for years due to cooling limitations in tight spaces. 

Intel’s i9-13900K supports Adaptive Boost Technology (ABT), which allows the chip to dynamically boost to higher all-core frequencies based on available thermal headroom and electrical conditions. This lets multi-core loads to operate at up to 5.5 GHz if the required of thermal dissipation is available. This feature works in a way that actively seeks high temperatures: If the chip sees that it is running underneath the 100 C threshold, it will increase its performance and power consumption until it reaches the safe 100 C limit, sustaining higher clocks (and providing better performance) for longer periods.

The increased cooling challenges posed by Raptor Lake mean that we’ve had to change some of the ways we test coolers. Some coolers were able to pass Cinebench R23 multicore testing with Intel’s 12th Gen i9-12900K when power limits were removed (although only the strongest models were able to pass that test). Most liquid coolers and all air coolers I’ve tested 'failed' that test because the CPU reached TJ max in this scenario. 

With Raptor Lake’s 13900K, not a single cooler tested has been able to keep the CPU under TJ max in this test – because as we pointed out, the chip is designed to dial up performance and power until it reaches that thermal threshold. We’ll compare performance instead by looking at total benchmark scores and clock speeds maintained in our testing. 

I’ll be testing Intel’s i9-13900K CPU using Asus’ TUF Gaming Z690 Gaming Plus WIFI motherboard and Cooler Master’s HAF 700 Berserker computer case, with case fans limited to 35% speeds. The motherboard’s default fan curve is used for the CPU Cooler’s fans.

In addition to testing Cinebench without power limits enforced, we’ll also be showing results when the CPU’s power consumption is limited to a more reasonable 200W. We’ll also show results at 125W for those who prefer whisper-quiet cooling, at the cost of some performance. For both of these results, we’ll show traditional delta over ambient temperature results.

We’ll provide noise level measurements recorded using a PSPL25 Sound Meter for all three power levels tested, to compare how much noise each cooler makes in different scenarios. We expect most coolers to run effectively silently at 125W.

LGA1700 Socket Bending

Note there are many factors other than the CPU cooler that can influence your cooling performance, including the case you use and the fans installed in it. A system's motherboard can also influence this, especially if it suffers from bending, which results in poor cooler contact with the CPU.

In order to prevent bending from impacting our cooling results, we’ve installed Thermalright’s LGA 1700 contact frame into our testing rig. If your motherboard is affected by bending, your thermal results will be worse than those shown below. Not all motherboards are affected equally by this issue. I tested Raptor Lake CPUs in two motherboards, and while one of them showed significant thermal improvements after installing Thermalright’s LGA1700 contact frame, the other motherboard showed no difference in temperatures whatsoever. Check out our review of this contact frame for more information.

Testing Configuration

 MORE: How to Buy the Right CPU Cooler

MORE: How to Check CPU Temperature

MORE: All CPU Cooling Content

No Power Limits: Cinebench Results

With Raptor Lake’s i9-13900K pushing speeds of 5.5 GHz or more, even the strongest of coolers hit TJ Max while running Cinebench R23 and other demanding scenarios. As the 13900K is designed to aim for its top safe temperature, we’ll be comparing the overall benchmark score and the CPU’s clock speeds instead. 

The results below are for a 10-minute testing run. To be sure that was sufficiently long to tax the cooler, we also retested both Thermalright’s Assassin X 120 R SE and DeepCool’s LT720 with a 30-minute Cinebench test. The results didn’t change much with the longer test: The average clock speeds maintained dropped by 29 MHz on DeepCool’s LT720 and 31 MHz on Thermalright’s Assassin X 120 R SE. That’s an incredibly small 0.6% difference in clock speeds maintained, a margin of error difference that tells us that the 10-minute tests are indeed long enough to properly test the coolers.

Because of the difficulty of cooling the i9-13900K in this workload, we feel the best way to compare coolers here is to record the average power consumption of the CPU. When it comes to total cooling capacity, the IceSleet G6 Stealth was able to cool roughly 227 watts, slightly higher than the company's official rating of 225W. This places it in exactly the middle of our air cooling results. Three air coolers performed better and three performed worse.

As Cinebench R23’s multi-core benchmark will push coolers to their limits, it’s also a great test for recording the worst-case scenario of fan noise levels. You might notice that our noise charts start at 36dB. This is the noise floor of our testing environment.

This makes 36dB our baseline measurement, as we’re unable to measure noise levels below this threshold. Keep in mind that noise measurements are logarithmic, meaning the perceived differences between the noise levels of the coolers will be more noticeable to the average ear than these graphs would suggest. 

When it comes to acoustic levels, the IceSleet G6 is also right in the middle of our results in this test. Five of the coolers we've tested ran quieter than the G6, while six were louder.

We’re also showing the total Cinebench R23 benchmark scores to show how little is lost  – or gained – with different cooling solutions. Between the weakest and strongest coolers, there is a total benchmark score variance of roughly 13%. 

The benchmark score of Iceberg Thermal’s IceSleet G6 Stealth here is, again, middle-of-the-road for our air cooling test results, scoring a total of 36.65K points. Three comparison air coolers performed worse than the G6 Stealth, and three performed better.

200W Cinebench Results

When restricting power consumption to a more reasonable 200W, the IceSleet G6 Stealth’s performance is impressive - second only to DeepCool’s AG620 among our tested air coolers. But that’s only part of the story; how loud does it get when the CPU is set to 200W?

Looking at acoustic levels when restricted to 200W, the IceSleet G6 Stealth’s results aren’t quite as impressive as the thermal results. At 47.4 dBA, the results here aren’t expressly bad. But this result is louder than most of the competing coolers tested here.

125W Cinebench Results

The lowest power limit I test at is 125W, mainly because this is also the lowest level where I can reliably measure noise measurements. Lower power consumption causes the noise of most CPU cooler fans to fall below the sound created by the system fans (even while restricted to 35% speed).

Looking at thermal results alone, the IceSleet G6 Stealth again does well, running at lower temperatures than any other air cooler tested. However, noise levels arguably more important than temps at 125W, and in this metric, the IceSleet G6 Stealth does very well — it was inaudible over the sound of the restricted system fans. This is an achievement accomplished by only half of the coolers we’ve tested on this platform.

Conclusion

Iceberg Thermal’s IceSleet G6 Stealth is a good cooler on a number of levels. It runs particularly quietly in low TDP loads, due to the Auto Stop/Start feature. Its fans have an impressive 17.5-year rated lifespan, and it comes with an unheard-of 10-year warranty.

This being said, it's tough to recommend due to its high MSRP. If you can find it on sale well below $80, it's worth considering. But even with the phenomenal warranty and long-rated lifespan, this coolers price is hard to swallow when options like Thermalright's excellent Peerless Assasin 1210 SE sell for less than $40.

 MORE: How to Buy the Right CPU Cooler

MORE: How to Check CPU Temperature

MORE: All CPU Cooling Content

The 24-core Intel Core i9-13900KS Special Edition processor carries an eye-watering $699 price tag, but it has a feature that Intel feels warrants the price tag — this is the first consumer PC chip to run at 6 GHz without overclocking, extending Intel's lead over AMD's fastest Ryzen 7000 processors. However, with a whopping 250W base power specification, this 13th-Gen Raptor Lake chip is also now officially the most power-hungry desktop CPU in history — its voracious appetite even peaks at 320W in a new Extreme Power Delivery Profile.

Intel's Raptor Lake processors have taken the lead in gaming, productivity, and value over AMD's Zen 4 Ryzen 7000 processors, but the competition remains stiff. The relatively low-profile 13900KS launch comes as AMD preps three new Ryzen 7000X3D processors with its disruptive 3D V-Cache tech that delivers explosive gains in gaming performance. 

The 3D V-Cache tech's first iteration, AMD's Ryzen 7 5800X3D, overtook Intel's Alder Lake as the best CPU for gaming, but Intel later retook the crown with Raptor Lake. AMD's Ryzen 7000X3D chips are poised to come to market in mere weeks, and the Core i9-13900KS is meant to keep Intel's silicon entrenched at the top of the CPU benchmark charts when AMD's new chips arrive. 

That might be a tall order — the Ryzen 7000X3D chips look exceptionally promising due to their exotic performance-enhancing tech. In contrast, the 13900KS is built on the same architecture as the 13900K, just with a higher speed binning to accommodate 200 MHz faster clock rates. Otherwise, it uses the same fundamental design as its standard counterpart, but it comes with a hefty upcharge that destroys any hint of a value proposition for all but the extreme overclocking crowd.  

Naturally, if you're looking to buy a $699 chip, price considerations go out of the window — this is a pure speed play for those with deep pockets. For now, we don't know if the Core i9-13900KS will keep Intel in the top spot when the Ryzen 7000X3D chips arrive next month, but it looks like it will be close. Here's how the Core i9-13900KS stacks up against the current competition.

Intel Core i9-13900KS Specifications and Pricing

The 13900KS is the fastest Intel desktop PC chip, and even at stock settings, it drew up to 328W and ran at 100C to scrape out every last bit of performance. Because the 13900KS is a Special Edition, the company will only produce a limited (but unspecified) number of these processors — hence the $699 price tag. Despite the high pricing, Intel has actually lowered the ceiling for access to its most premium silicon: The previous-gen 12900KS debuted at $739, but Intel later increased the pricing of all its 12th-Gen chips, bringing it to $823 today.

Intel selects its premium-binned 13900K silicon for the 13900KS, so it is guaranteed to be among the very best silicon the company has to offer. That will make the chips very attractive to overclockers, as paying the extra $110 for the KS model improves your odds in the silicon lottery, essentially assuring you're getting a cherry chip. Here's a video that describes Intel's binning process:

 As with the Core i9-13900K (architecture details here), the KS model comes with eight hyperthreaded p-cores and 16 single-threaded e-cores, for a total of 24 cores and 32 threads. The KS model's p-cores have a 3 GHz base clock and reach up to 6 GHz on two cores via Thermal Velocity Boost (TVB) tech. This allows the processor to shift into slightly higher frequencies if it remains under a certain temperature threshold (70C for desktop chips). Meanwhile, the standard Turbo Boost 3.0 clock rates reach 5.8 GHz. We found that the chip easily peaks at 6 GHz, though the impact on workloads can vary. 

The e-cores kick in for background and multi-threaded tasks with a 2.2 GHz base and  4.3 GHz boost, identical to the 13900K. Most other features, like the supported DDR4-3200 and DDR5-5600 transfer rates, 32MB of L2, and 36MB of L3 cache, remain the same. 

As before, the LGA 1700 chip is compatible with 600-series motherboards, but you'll want one of the best-equipped Z-series boards to quench the 13900KS' thirst for power. However, selecting the right motherboard is a bit more complicated for the 13900KS due to the new 320W "Extreme Power Profile."

Intel's power profiles have a few key variables. The PL1 power spec quantifies extended use at base clocks and is also known as the Processor Base Power (PBP). In contrast, PL2 power quantifies when the chip is under full load for short periods (boost) and is also known as the Maximum Turbo Power (MTP). ICCMax denotes the maximum current the chip can pull when under heavy load.

Intel also used to define a Tau variable that specified the duration of the boost/PL2 state before the chip dropped back into PL1. Intel has since stopped defining Tau for K-series models because it assigns PL1 and PL2 as the same value for Alder Lake and following gens in the reference BIOSes that it sends to OEMs — a first. This means the chip always operates in a boosted/PL2 state.

Intel's new policy of sending reference BIOSes with PL1=PL2 settings basically invalidates even having a PL1 value, and the fact that it still exists on the spec sheet as a "Processor Base Power (PBP)" setting — the lower of the two power specs — is misleading.

Intel defined an Extreme Power Profile for its regular 13900K — the 253W PL1/PL2 values remain the same for both the standard and extreme profile, but the extreme profile's ICCMax reaches 400A (up from 307A).

The KS model has the same default profile as the 13900K but also has its own new Extreme Power Profile that allows for a 320W PL1/PL2 and 400A ceiling. You'll need to ensure that your motherboard can deliver the peak current if you want to unleash the full power of the KS, as not all motherboards can for a long period of time. Motherboard vendors allow assigning a higher ICCMax value in the BIOS, typically under settings like "Core/CPU Current Limit" (the name varies by mobo maker), but that doesn't mean the motherboard can actually deliver that amount of current. Obviously, B- and H-series boards don't make the cut.

Intel defines these recommended power profiles but allows motherboard vendors to ignore them completely, and exceeding the default values doesn't void the warranty. Thus, by default, most motherboard makers completely ignore the limits and assign the maximum values for PL1, PL2, and ICCMax, resulting in higher performance and more heat. 

Even at stock settings, the Core i9-13900KS hit up to 328W of power and 100C in our testing. We'll see what power draw, performance, and thermals look like, including gaming and productivity benchmarks, on the following pages.

• MORE: AMD vs Intel
• MORE: Zen 4 Ryzen 7000 All We Know
• MORE: Raptor Lake All We Know 

Intel Core i9-13900KS Thermals, Power Consumption, and Boost Clocks 

Intel's Adaptive Boost Technology (ABT) dynamically boosts to higher all-core frequencies based on available thermal headroom and electrical conditions, so peak frequencies can vary. By design, this tech allows the chip to operate at 100C during normal operation — if the chip runs under the 100C threshold, it will increase its power consumption until it reaches the safe 100C limit, thus providing more performance. However, this feature is only active on the Core i9-13900K/F and the Core i9-13900KS, so other Raptor Lake processors won't exhibit the same behavior.

You can think of ABT much like a dynamic auto-overclocking feature, but because the chip stays within Intel's spec of a safe 100C temperature limit, it is a supported and warrantied feature that doesn't fall into the same classification as overclocking. ABT uplift will vary by chip — much of the frequency uplift depends upon the quality of your chip. Hence, the silicon lottery comes into play, along with cooling and power delivery capabilities.

Remember, AMD's Ryzen 7000 also runs at its limit of 95C at stock settings, so higher temperatures have become the norm for both chipmakers.

To remove thermals as a limitation, we always use the same 280mm Corsair H115i AIO cooler for all our test systems. However, Intel says that it achieved the best results with the KS paired with a 360mm radiator, so we tested the impact by running the same tests with all power limits removed on a 280mm and a 360mm AIO. Our single-threaded test also shows that the 13900KS boosted to 6 GHz frequently, regardless of the cooler.

The chip reached 100C with both coolers during a series of heavily multi-threaded apps, like y-cruncher, Cinebench, Blender, and POV-Ray, just as Intel designed it to do. The 360mm cooler config ran the same 5.6 GHz all-core clock during heavy work as the 280mm setup, yet it also consumed around 20-25W more during some types of work. 

That 360mm's additional cooling capacity enables that extra bit of power consumption, but the ~7% to 9% increase occurs at the top of the chips' voltage frequency curve where increased power consumption is incredibly inefficient: As you near peak power, double-digit percentage increases in power consumption often only yield single-percentage performance gains. That means this slightly higher power will not make much difference in actual benchmarks, which we'll detail below.

We also overclocked the chip via turbo multipliers to 6 GHz on two p-cores and 5.7 GHz when more than two cores are active, while dialing in a 4.4 GHz all-core overclock on the e-cores. The chip only required a 1.29V vCore to sustain these frequencies, indicating it is a cherry chip. The overclocked config uses less power than stock settings, showing that the chips' native power management is inefficient. As you'll see below, the overclocked settings enabled much more performance than the stock settings that consumed more power. 

The chip barely pushed over 300W with standard applications, but switching gears to the strenuous but not-at-all-realistic Prime95 stress test yielded a much higher peak power consumption of 328W. Power consumption leveled off after a short period, so it's possible a custom watercooling loop could allow the chip to consume more for a longer period of time. But, again, this won't result in very meaningful performance improvements in real-world workloads — this chip is tuned to the absolute top of its voltage/frequency curve.

There isn't a huge difference in cooling capacity between our 280mm and 360mm AIOs (we'd see a bigger difference with a 240mm vs 360mm comparison). As you can see in our cumulative performance measurements above, the improved cooling doesn't result in a linear improvement in performance in our gaming or application benchmarks.

The improvement in real-world gaming and productivity applications with the 360mm cooler was around 1%. As such, we used the test results from our 280mm configuration for our gaming and productivity benchmarks.

Our overclock gave us an extra 5% in 1080p gaming and 4% in threaded work, but be aware that the increased memory throughput (we used DDR5-6800 for the overclocked config) is a big contributor here. Additionally, the overclocked vanilla 13900K trailed by only 1% in games and 2.5% in threaded apps. Further tuning, or more luck in the silicon lottery, could narrow that gap.

Temperatures can limit your performance during stock operation, so if you purchase the 13900KS, plan for a powerful cooler to extract the full performance. We think a 280mm AIO would be sufficient, but if you're chasing the last 1% of performance, a 360mm AIO will get you there.

• MORE: AMD vs Intel
• MORE: Zen 4 Ryzen 7000 All We Know
• MORE: Raptor Lake All We Know 

Power Consumption on Intel Core i9-13900KS

These power measurements show that the Core i9-13900KS will rarely reach its 320W power limit with standard applications — there simply isn't enough thermal headroom due to the difficulty of dissipating this amount of heat from such a small area (thermal density). 

More exotic cooling, like custom water loops or sub-ambient, would likely allow the chip to pull more power within its 100C thermal envelope, albeit for little practical performance gain. 

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

These workloads are comprised of a fixed amount of work, so we can plot the task energy against the time required to finish the job (bottom axis), thus generating a really useful power chart.

Remember that faster compute times, and lower task energy requirements, are ideal. That means processors that fall the closest to the bottom left corner of the chart are the best. 

The 13900KS sucks significantly more power than the 13900K in these tests, but that doesn't deliver much of a performance gain — increasing power consumption when the chip is already near the top of the voltage/frequency curve is an incredibly inefficient way to gain very little extra performance. 

Intel Core i9-13900KS Overclocking and Test Setup

• Intel Core i9-13900KS @ 6.0-5.7: Turbo multiplier: Two cores at 6 GHz, 5.7 GHz all-p-core, 4.4 GHz e-core, 1.29V vCore, DDR5-6800 XMP 3.0
• Intel Core i9-13900K @ 5.6: 5.6 GHz all-core p-core, 4.4 GHz e-core, 1.32V vCore, DDR5-6800 XMP 3.0
• AMD PBO Configs: Precision Boost Overdrive (Motherboard), Scalar 10X, DDR5-6000 EXPO

We used DDR5 memory with all chip testing in this article. Notably, if you opt for DDR4 with the Raptor Lake processors, you'll lose a few percentage points of performance, on average (testing here). The overclocked Ryzen configs (marked as 'PBO' in the charts) use DDR5-6000 memory, Precision Boost Overdrive, and a Scalar 10X setting.

We overclocked the Core i9-13900KS via turbo multipliers to 6 GHz on two p-cores and 5.7 GHz when more than two cores are active, while dialing in a 4.4 GHz all-core overclock on the e-cores. The chip only required a 1.29V vCore to sustain these frequencies. You can see the heat output and more overclocking details in our stress tests in the thermal section on the prior page.

We didn't use Intel's "Extreme Power Profile" for our stock Intel tests; instead, we're sticking with our standard policy of allowing the motherboard to exceed the recommended power limits, provided the chip remains within warrantied operating conditions. This means our power settings exceed the 'Extreme' recommendations — almost all enthusiast-class motherboards ignore the power limits by default anyway. Hence, our completely removed power limits reflect the out-of-box experience. Naturally, these lifted power limits equate to more power consumption and, thus, more heat.

Microsoft recently advised gamers to disable several security features to boost gaming performance. As such, we disabled secure boot, virtualization support, and fTPM/PTT for maximum performance. You can find further hardware details in the table at the end of the article. 

• MORE: AMD vs Intel
• MORE: Zen 4 Ryzen 7000 All We Know
• MORE: Raptor Lake All We Know 

Gaming Benchmarks on Intel Core i9-13900KS — The TLDR 

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

The Core i9-13900KS is now the fastest gaming CPU in the world, but only by a hair. Using a geometric mean of our 1080p gaming results, the $699 Core i9-13900KS is a mere 1.5% faster than the standard $589 13900K. That's imperceptible to most users and certainly not worth the extra ~20% you'll pay at retail. Further, the performance delta shrinks to a mere 0.5% with 1440p gaming, so you absolutely should not buy the KS model for gaming only.

The Core i9-13900KS contends with the $699 Ryzen 9 7950X in the pricing department, but the standard Core i9-13900K model already holds a strong enough lead in gaming to make the extra $110 for the KS model a non-starter for most of us.

Naturally, the pre-binned Core i9-13900KS will appeal to overclockers — your odds of scoring a cherry chip are increased dramatically. However, you're still at the whims of the silicon lottery with both the KS and the K model, so you could get lucky with either chip.

In fact, our tests didn't show a very big lead for the KS model over the K, and we could have likely dialed in a 5.9 GHz overclock on two of the K model's cores and reduced the delta further. Not that there's a big difference between the chips anyway — the overclocked KS model was only 1% faster than the K-series model after overclocking.

Overall, the KS model doesn't make much of a difference to the positioning of the 13th-Gen processors against the Ryzen lineup, and it is a poor value. This chip will appeal to deep-pocketed individuals either intent on building the highest-end system imaginable or extreme overclockers chasing world records. The Core i7-13700K remains the chip to beat if you're looking for a high-end gaming and productivity chip, as the 13900KS is only 2.5% faster yet costs a whopping $290 more.

The Ryzen 7 5800X3D is the fastest AMD gaming chip by far due to its 3D V-Cache, but its extra 96MB of L3 cache doesn't accelerate all games, and you'll also have to accept much lower performance in just about every other type of productivity application. If you're looking for an exotic 3D V-Cache gaming chip, your best option is to wait until next month when AMD launches the new Zen 4-powered Ryzen 7000X3D models. These chips look like they'll have a much more balanced performance profile and might beat the KS model in gaming, too.

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

The individual game benchmarks below are predictable — the KS reliably scores one to two percent higher than the standard K model across the board. As such, we won't comment on the individual game results below. 

Cyberpunk 2077 on Intel Core i9-13900KS

Far Cry 6 on Intel Core i9-13900KS

F1 2021 on Intel Core i9-13900KS

Hitman 3 on Intel Core i9-13900KS

Microsoft Flight Simulator 2021 on Intel Core i9-13900KS

Microsoft Flight Simulator 2021 obviously benefits tremendously from L3 cache — the Ryzen 7 5800X3D is a whopping 28% faster than the Core i9-13900K flagship on this test. This large advantage is amazing but doesn't represent the 5800X3D's performance in most titles. It also illustrates how outliers can make the 5800X3D seem more impressive in cumulative measurements.

In either case, this does bode well for the Zen 4 Ryzen 7000X3D models, which should bring all the goodness of the capacious L3 cache with fewer tradeoffs in standard applications.  

Red Dead Redemption 2 on Intel Core i9-13900KS

Warhammer 3 on Intel Core i9-13900KS

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

Watch Dogs Legion on Intel Core i9-13900KS

• MORE: AMD vs Intel
• MORE: Zen 4 Ryzen 7000 All We Know
• MORE: Raptor Lake All We Know 

Productivity Benchmarks on Intel Core i9-13900KS — The TLDR: 

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

Overall the same general trends that we saw in our gaming benchmarks appear in our application testing — the Core i9-13900KS does nothing to change the competitive positioning against AMD's Ryzen, and while the 13900KS is the fastest chip in the world in both single- and multi-threaded work after overclocking, its high price tag makes the slight gains simply not worth it.  

The Ryzen 9 7950X clings to its lead in multi-threaded work over the Core i9-13900KS by a mere 0.5%, but that's in the noise of the benchmarks. We should consider these two chips tied. Overclocking flips the advantage back in Intel's favor, but this is also close enough to call a tie. We also see slight gains over the Core i9-13900K in multi-threaded work, but this isn't worth the increased pricing. 

Flipping over to single-threaded work reveals that the 13900KS holds a clear win over all contenders — it is 14% faster than the Ryzen 9 7950X and 2% faster than the standard 13900K. Still, that's a relatively slim delta over the standard 13900K.

You'll have to accept severe performance tradeoffs if you opt for AMD's fastest gaming chip, the Ryzen 7 5800X3D. The 5800X3D is significantly slower than the newer chips because it comes with the previous-gen Zen 3 architecture. The extra dollop of L3 cache doesn't impart any meaningful speedups in most standard desktop PC applications. If you're looking for an X3D chip, wait for the 7000X3D chips that launch next month. 

Rendering Benchmarks on Intel Core i9-13900KS

Intel's Raptor Lake rivals or beats AMD's finest in multi-threaded productivity applications at every price point. These results, like many of the results throughout our other application tests below, mirror the established trends of the KS slightly extending the lead relative to the 13900K. As such, we'll have limited commentary for the application benchmarks. 

Encoding Benchmarks on Intel Core i9-13900KS

Most encoders tend to be either heavily threaded or almost exclusively single-threaded — it takes an agile chip to master both disciplines. 

Adobe, Web Browsing, Office and Productivity on Intel Core i9-13900KS

The ubiquitous web browser is one of the most frequently used applications. These latency-sensitive tests tend to be lightly threaded, so a fast response time is critical.

Compilation, Compression, AI Chess Engines, AVX-512 Performance on Intel Core i9-13900KS

This selection of tests runs the gamut from the exceedingly branchy code in the LLVM compilation workload to the massively parallel molecular dynamics simulation code in NAMD to encryption and compression/decompression performance. Y-cruncher computes Pi with the AVX instruction set, making for an exceedingly demanding benchmark. 

• MORE: AMD vs Intel
• MORE: Zen 4 Ryzen 7000 All We Know
• MORE: Raptor Lake All We Know 

Paying the 6 GHz Tax

The $699 Core i9-13900KS Special Edition is the fastest gaming chip in the world with its record-setting 6 GHz peak turbo clock. However, for all but extreme overclockers, collectors, and the well-heeled looking to build the ultimate system at any cost, the costly chip's relatively small performance advantage over the vanilla 13900K isn't worth the $110 upcharge.

Below, we have the geometric mean of our gaming test suite at 1080p and 1440p and a cumulative measure of performance in single- and multi-threaded applications. We conducted our gaming tests with an Nvidia RTX 4090, so performance deltas will shrink with lesser cards and higher resolution and fidelity settings.

While the 13900KS is now the fastest gaming CPU money can buy, it only holds a slim 1.5% advantage over the standard 13900K in our 1080p gaming tests. That delta shrinks to a mere 0.5% when we flip over to 1440p gaming, meaning it is essentially in the noise of the benchmarks. That certainly isn't worth paying ~20% more.

We see a similar trend in multi-threaded workloads — the KS model only offers a slight 1% improvement over the standard 13900K and essentially ties the Ryzen 9 7950X. We see a more noticeable improvement in single-threaded work against Ryzen, as the 13900KS is 14% faster than the Ryzen 9 7950X. However, the KS is only 2% faster than the standard 13900K, so the big upcharge isn't worth it.

The big appeal for the KS is that you're guaranteed to get Intel's best silicon, basically giving you the winning ticket for the silicon lottery. That will undoubtedly make the chip popular with extreme overclockers, especially those chasing world records. However, again, the overclocking advantage can vary, and we only saw a few percentage points of difference in our own head-to-head overclocking against the standard 13900K.

Overall, the KS model doesn't make much of a difference to the positioning of the 13th-Gen processors against the Ryzen lineup, and it will only be sold in limited quantities. Intel now allows pairing its consumer chips with ECC memory on motherboards with the workstation-geared W680 chipset, so the Core i9-13900KS could make for a powerful workstation machine. However, it will still only deliver small gains over the 13900K.

The Core i9-13900KS just isn't a practical buy for the majority of us, especially considering that it will need the priciest of accommodations, like the highest-end motherboard and PSU to pump 320W of power to the processor and a potent cooler to handle the 100C operating temperatures when the chip is under heavy load. All of which is a high price to pay for a few percentage points (at best) of extra performance. 

If you want to invest in the fastest gaming chip, you should wait to see how AMD's Ryzen 7000X3D chips perform when they launch next month. These chips come with 3D V-Cache tech that provided explosive gaming performance gains for Zen 3, and it's possible that it could push Zen 4 into a solid lead in gaming over Raptor Lake. Pricing remains the wild card, but we expect the high-end 16-core Ryzen 9 7950X3D to land in the same range as the 13900KS. 

In the meantime, the Core i9-13900KS represents Intel throwing value and power consumption out the window in an effort to retain the performance crown, particularly in gaming, as Ryzen 7000X3D draws near. The Core i9-13900KS might be pricey and only deliver small gains, but it does set the record as being the first 6 GHz desktop PC chip right out of the box — and that's an incredible milestone. It might even allow Intel to cling to the title of having the fastest gaming chip in the world, a nearly priceless distinction for the 13th-Gen Raptor Lake brand.

Make no mistake: the Special Edition Core i9-13900KS is the fastest desktop PC chip ever built. At least for now. However, AMD's Ryzen 7000X3D is approaching fast. Stay tuned. 

• MORE: AMD vs Intel
• MORE: Zen 4 Ryzen 7000 All We Know
• MORE: Raptor Lake All We Know 

MSI is one of the best-known brands in the PC market, producing a robust lineup of laptops, desktops, GPUs, and motherboards. They also make a variety of other components and peripherals, like keyboards and CPU coolers. 

MSI has three lines of AIO Coolers that mirror its motherboard sub-brands: The MPG, MEG and MAG series. The MAG series is MSI’s budget-oriented offering, and features a pump which is integrated into the radiator. Today we’re looking at thelargest unit in the lineup, the MAG CoreLiquid P360. 

With today’s CPUs raising the bar for cooling difficulty, does MSI’s MAG CoreLiquid P360 have what it takes to cool Intel’s Core i9-13900K and earn a spot on our best AIO coolers list? We’ll have to put it through testing to say for sure. But first, here are the cooler’s specifications, direct from MSI.  

Cooler Specifications

Packing and Included Contents

MSI’s CoreLiquid P360 comes in a fairly large box that uses plastic, foam, and cardboard to protect the inner contents.

Included with the package are the following:

• 360 mm Radiator + CPU block
• Three 120 mm fans
• Mounts for all modern CPU sockets (including AM5 & LGA1700)
• Fan splitter
• PSU Molex 4-Pin to 4-pin (2 Pin Wired) Cable
• Noise reduction adapter
• Thermal paste

Installation

Installing the P360 was fairly simple. To begin, press the backplate to the motherboard, and secure it by screwing in the standoffs. Next, you’ll need to secure the fans to the radiator, and then mount the radiator to your computer case. Apply the included thermal paste from the syringe to your CPU. Afterwards, all that’s left is to press the CPU block against the standoffs and then secure them with the included thumbscrews.

Many coolers come with pre-applied thermal paste. This won't be a big deal to experienced builders (they may have a different paste they prefer to use, anyway), but it's rarer these days to see an AIO without pre-applied thermal paste.

Features of MSI’s MAG CoreLiquid P360

Radiator Pump Design

Most liquid coolers today are based on Asetek designs, which have the pump included in the CPU block. However, MSI’s MAG CoreLiquid P360 features a pump that’s integrated within the radiator, powered by a three-phase motor. MSI claims two advantages from this design: longer life expectancy due to the pump being removed from the CPU’s direct heat, and reduced noise levels.

Next to the pump, below the radiator tubing, is a refill port. This allows a user to service and refill the AIO without having to send it to the manufacturer for repair (or more likely, just buying a new cooler).

Full Copper CPU plate

The CPU contact plate on MSI’s CoreLiquid P360 is made of pure copper.

Medium thickness 360 mm radiator

MSI’s CoreLiquid P360 features a 360mm radiator that’s 27 mm thick. This isn’t the thickest (or thinnest) radiator we’ve seen, but it should be compatible with most cases that have space for three 120 mm fans.

3x 120 mm fans

There’s more to a cooler than just the heatsink or radiator. The bundled fans have a significant impact on cooling and noise levels. Included with the MSI CoreLiquid P360 are three 120 mm fans, which are not available for purchase individually. 

Testing Methodology

While it was fairly easy with previous generations of CPUs for coolers to keep the flagship i9 processor well under TJ max (the maximum temperature a CPU can sustain without throttling) in tough workloads, this isn't realistic on current generation high-end CPUs (the 13900K especially) without extreme cooling (or enabling power limits).

While in the past a CPU hitting its peak temperature was cause for concern, enthusiasts are going to have to learn to accept high temperatures as “normal” while running demanding workloads with Raptor Lake and Ryzen 7000 CPUs. Modern AMD and Intel CPUs are designed to run fairly hot without any problems – up to 95 degrees Celsius for AMD Ryzen 7000 CPUs, and up to 100 C for Intel’s Core i9-13900K. Similar behavior has been standard in laptops for years due to cooling limitations in tight spaces. 

Furthermore, Intel’s i9-13900K supports Adaptive Boost Technology (ABT) which allows Core i9 processors to dynamically boost to higher all-core frequencies based on available thermal headroom and electrical conditions. This allows multi-core loads to operate at up to 5.5 GHz if the necessary amount of thermal dissipation is there. This feature works in a way that actively seeks high temperatures; If the chip sees that it is running underneath the 100 C threshold, it will increase its performance and power consumption until it reaches the safe limit, sustaining higher clocks (and providing better performance) for longer periods.

The increased cooling challenges posed by Raptor Lake mean that we’ve had to change some of the ways we test coolers. Some coolers were able to pass Cinebench R23 multicore testing with Intel’s 12th Gen i9-12900K when power limits were removed (although only the strongest models were able to pass that test). Most liquid coolers and all air coolers I’ve tested “failed” that test because the CPU reached TJ max in this scenario. 

With Raptor Lake’s 13900K, not a single cooler tested has been able to keep the CPU under TJ max in this test – because as we pointed out, the chip is designed to dial up performance and power until it richest that thermal result. We’ll compare performance instead by looking at total benchmark scores and clock speeds maintained. 

I’ll be testing Intel’s i9-13900K CPU using Asus’ TUF Gaming Z690 Gaming Plus WIFI motherboard and Cooler Master’s HAF 700 Berserker computer case, with case fans limited to 35% speeds. The motherboard’s default fan curve is used for the CPU Cooler’s fans.

In addition to testing Cinebench without power limits enforced, we’ll also be showing results when the CPU’s power consumption is limited to a more reasonable 200W. We’ll also show results at 125W for those who prefer whisper-quiet cooling, at the cost of some performance. For both of these results, we’ll show traditional delta over ambient temperature results.

We’ll provide noise level measurements recorded using a PSPL25 Sound Meter for all three power levels tested to compare how much noise each cooler makes in different scenarios. We expect most coolers to run effectively silently at 125W.

LGA1700 Socket Bending

There are many factors other than the CPU cooler that can influence your cooling performance, including the case you use and the fans installed in it. A system's motherboard can also influence this, especially if it suffers from bending, which results in poor cooler contact with the CPU. 

In order to prevent bending from impacting our cooling results, we’ve installed Thermalright’s LGA 1700 contact frame into our testing rig. If your motherboard is affected by bending, your thermal results will be worse than those shown below. Not all motherboards are affected equally by this issue. I tested Raptor Lake CPUs in two motherboards. And while one of them showed significant thermal improvements after installing Thermalright’s LGA1700 contact frame, the other motherboard showed no difference in temperatures whatsoever! Check out our review of this contact frame for more information.

Testing Configuration

 MORE: How to Buy the Right CPU Cooler

MORE: How to Check CPU Temperature

MORE: All CPU Cooling Content

No Power Limits Cinebench Results

With Raptor Lake’s i9-13900K pushing speeds of 5.5 GHz or more, even the strongest of coolers hit TJ Max while running Cinebench R23 and other demanding scenarios. As the 13900K is designed to increase clocks and power until it reaches its top safe temperature, we’ll be comparing the overall benchmark score and the CPU’s clock speeds instead. 

The results below are for a 10-minute testing run. But to be sure that was sufficiently long to tax the cooler, we also retested both Thermalright’s Assassin X 120 R SE and DeepCool’s LT720 with a 30-minute Cinebench test. The results didn’t change much at all with the longer test: The average clock speeds maintained dropped by 29 MHz on DeepCool’s LT720 and 31 MHz on Thermalright’s Assassin X 120 R SE. That’s an incredibly small 0.6% difference in clock speeds maintained, a margin of error difference that tells us that the 10-minute tests are indeed long enough to properly test the coolers.

Because of the difficulty of cooling the i9-13900K in this workload, the best way to compare coolers here is to record the average power consumption of the CPU. When it comes to total cooling capacity, MSI’s CoreLiquid P360 is capable of cooling roughly 287W. This puts it a bit behind the other 360 mm AIOs we’ve tested, but the P360 also has a lower price than the competition.

As Cinebench R23’s multi-core benchmark will push coolers to their limits, it’s also a great test for recording the worst-case scenario of fan noise levels. You might notice that our noise charts start at 36 dB. This is the noise floor of our testing environment; we’re unable to measure noise levels below this threshold. Keep in mind that noise measurements are logarithmic, meaning the differences between the noise levels of the coolers will be more noticable than these graphs would suggest. 

When it comes to acoustic levels, the P360 is one of the louder units we’ve tested if the fans are running at full speed. It is quieter than the competing DeepCool and Enermax AIOs, but noisier than other coolers here.

We’re also showing the total Cinebench R23 benchmark scores to show illustrate little is lost  – or gained – with different cooling solutions. Between the weakest and strongest coolers, there is a total benchmark score variance of roughly 13%. 

MSI’s MAG CoreLiquid P360 does very well here, with a total score of 39,292, which makes it the third-strongest showing of the coolers we’ve tested with Raptor Lake.

200W Cinebench Results

When restricting power consumption to a more reasonable 200W, MSI’s MAG CoreLiquid P360 does reasonably well, tying with SilverStone’s VIDA 240 for the fourth strongest result. But that’s only part of the story: How loud does it get when set to 200W?

Looking at acoustic levels when restricted to 200W, the P360’s results are roughly middle of the road. It’s quieter than six noisier coolers here, and the results are a tie with DeepCool’s LT720.

125W Cinebench Results

The lowest power limit I test at is 125W, mainly because this is also the lowest level where I can reliably measure noise measurements. Lower power consumption causes the noise of the CPU cooler to fall below the noise created by the system fans (even while restricted to 35% speed). 

Looking at thermal results alone here, MSI’s P360 does well, taking fourth place in cooling performance. However, I feel that noise levels are more important at 125W. And in this metric, the P360 doesn’t do very well in comparison to other units. It’s not “loud” in any sense of the word, but it was noisier than most of the competing coolers we’ve tested.