Today is the day! AMD has allowed reviewers to start publishing their Zen 4 Ryzen 7000 Series processor reviews. The four new CPUs were detailed just a few weeks ago, now we can share everything with you. With Zen 4, AMD has made significant improvements to nearly all parts of their platform. It starts with the new Zen 4 architecture, which improves instruction-per-clock by well over 10% and adds support for AVX-512 and acceleration for AI workloads. Also new is that compatible motherboards use the swanky "Socket AM5," which does away with the pins on the processor and uses an LGA package, similar to what Intel has been using for nearly two decades now. The new socket has support for all the latest technologies: PCI-Express 5.0 for graphics and NVMe storage; DDR5 memory, and also higher power delivery capability. Last but not least, AMD is finally including integrated graphics across the board, even with their non-APU processors, which will be extremely useful when building a rig that will only see light activity, such as office productivity, light browsing, or a media PC.

In this review we've examined the Ryzen 9 7950X, the company's new flagship featuring a 16-core/32-thread configuration. The chip runs at a base clock of 4.5 GHz and boosts up to 5.7 GHz—an 800 MHz increase over the 5950X. AMD has also increased the TDP from 105 W to 170 W, which brings with it more headroom for boosting to higher clocks, but also increases cooling requirements, more on that later.

We've spent months working on a new and improved CPU benchmarking suite, with tons of new apps and games, and every processor was retested with the latest Windows 11 updates, game patches and drivers. Averaged over our 45 application tests we can confirm that the Ryzen 9 7950X is a mighty processor and the fastest CPU we ever tested. It runs 25% faster than last generation's 5950X flagship, 10% faster than 7900X, 20% faster than 7700X—that's in our mix of highly-threaded, mostly-threaded, lightly-threaded and single-threaded workloads. If you look at parallelized apps only, the differences will be even bigger, and you probably should check out each of our application tests if you feel you're in the market for a 7950X. While performance is often impressive, there are also many scenarios in which the 16 cores can really make no noteworthy difference over a 7900X, or even 7700X for that matter, but that is expected—what's important is that you are aware of this. Rendering applications are a perfect match for the 7950X, but this processor isn't "just for rendering"—we've also seen impressive results in web hosting, databases, media encoding and scientific simulation tasks. Compared to Intel's Core i9-12900K, the 7950X is 15% faster on average, in some workloads it's a bloodbath however, and AMD's offering achieves 30-50% higher performance.

In gaming, Zen 4 can also deliver. At 1080p, the 7950X is the second-fastest Zen 4 processor for gaming. Only the Ryzen 7 7700X is a tiny bit faster. The underlying reason is that the 7900X and 7950X use a dual CCD design, where the processor's computation ability is spread across two silicon dies, while the 7600X and 7700X use only a single CCD. This single CCD configuration helps make a difference in gaming, because things are better localized and there's no inter-CCD communication penalties. The differences are small though, the other Ryzens are still formidable CPUs for gaming. However, considering the cost for the 7950X, and assuming that gaming is your primary focus, then the 7700X is your best choice. The Ryzen 7 5800X3D is a fantastic gaming processor, too, but on average, for our 12-game mix, the 7950X can pull ahead. Where the 5800X3D will still have an edge is in specific highly-CPU bound games that can benefit from its huge cache (such as Borderlands 3). Gaming performance on Intel is still a tiny bit better than Zen 4, at least with our selection of games, as long as you have an Intel CPU with P-Cores and E-Cores—the P-Core-only Intel 12600 and 12400F in our test group do lose to Zen 4 in gaming. In early summer, way before we had any Zen 4 samples, I picked titles based on their relevance and popularity and it seems the game list tends to run a bit better on Intel than what a general average over many more titles would end up at. I have plans to test Zen 4 on a wider range of games, 50 is the target, to get a better statistical feel for this. Both AMD Zen 4 and Intel Alder Lake are fantastic processors for gaming and in a subjective test without FPS counter you wouldn't be able to tell their performance apart. For gaming targeting the highest FPS, the processor is secondary, you should rather spend as much money as possible on the graphics card and pair it with a more affordable processor, instead of focusing on brute CPU performance at the cost of GPU budget.

I also did a gaming benchmark run at DDR5-6000 with CL30 + AMD EXPO activated as opposed to our standard DDR5-6000 CL36 setting, which we used on all platforms, just to get a feel for how much performance difference we can expect from tighter timings. The performance results show that there's some minor gains to be had, but these are pretty small, around 1%, which makes it pretty hard to justify the extra cost for the faster memory. We'll be looking into this in more detail in an upcoming "Zen 4 memory scaling" article.

AMD is fabricating the Zen 4 compute dies (CCDs) on TSMC's 5 nanometer production process and the IO die is made on TSMC 6 nanometer. Just these numbers alone would suggest that Zen 4 is extremely energy efficient, but it seems clocking Ryzen 7000 at these high frequencies and TDPs has cost AMD quite some efficiency. We've upgraded our power measurement pipeline, and can now measure "chip-only" power consumption (as opposed to "whole system" before), and we record samples much faster, with full integration of the results. This lets us pair up power measurements with individual benchmark runs and perform more complex analyses, check out page 23. Compared to the Ryzen 9 5950X, the 7950X has somewhat better single-threaded efficiency, but ends up 30% less efficient in multi-threaded workloads, with much higher overall performance though; gaming efficiency is roughly similar. If you prefer greater efficiency, you can always dial down the clocks and voltages slightly, and greatly reduce the power consumption with minimal impact on performance. AMD's Precision Boost offers plenty of dials to achieve that, loving it. Compared to the 12900K, the 7950X is much more efficient in multi-threaded workloads, by around 50%, but single-threaded, which of course is an edge-case for high-core CPUs, is much better on Intel. In gaming, both are roughly similar, achieving Frames per Watt scores, that are within a few percent of each other.

What will be a big source of discussion for Zen 4 is the extremely high temperatures. Using our high-end Noctua air cooler, there was a constant struggle trying to keep the CPU from its 95°C thermal limit - at which point it will start lowering boost clocks. Even adding all-in-one watercooling wasn't able to make a significant difference. AMD is very clear in their cooler guidance for 7950X and 7900X, mentioning "240-280 mm liquid" as the recommended cooling configuration. The only AIO that showed meaningful gains over air was the Arctic Liquid Freezer II, which offers an "offset mounting" configuration that moves the center of the cold plate away from the center of the CPU to sit right on top of the compute dies on Zen 4. This gained us a couple of degrees improvement, but still not enough to consistently stay below 95°C. I had to undervolt the processor by 0.05 V for that. In cases where this made any noteworthy performance difference, due to less throttling, I added a gold bar to the application charts. We're talking low single digit percentages here, so no big deal, and it confirms that you can run 7950X with good air cooling and not lose a lot of performance. For gaming, temperatures are a complete non-issue, we reached well below 80°C here. AMD was also clear that 95°C is the new 65°C: "TJMax is the max safe operating temperature – not the absolute max temperature. In the Ryzen 7000 Series, the processor is designed to run at TJMax 24/7 without risk of damage or deterioration. At 95 degrees it is not running hot, rather it will intentionally go to this temperature as much as possible under load because the power management system knows that this is the ideal way to squeeze the most performance out of the chip without damaging it."

I'm just surprised that temperatures are so high. Maybe it's the smaller die size of the compute dies, paired with high clocks and voltage that leads to these temperatures, or some compromises were made to achieve compatibility with AM4 coolers, or maybe both. I'm sure in the coming months we'll learn more about the cooling challenges, how to overcome them, and cooler vendors will probably release solutions optimized for Socket AM5. Right now you need to be aware that cooling Zen 4 is much more difficult than Zen 3, and that you should really stop worrying about 95°C temperatures. AMD is very clear in their messaging that these temps are normal and expected.

During testing I didn't encounter any major bugs or issues; the whole AM5 / X670 platform works very well considering how many new features it brings; there's one big gotcha though and that's startup duration. When powering on for the first time after a processor install, your system will spend at least a minute with memory training at POST code 15 before the BIOS screen appears. When I first booted up my Zen 4 sample I assumed it was hung and kept resetting/clearing CMOS. After the first boot, the super long startup times improve, but even with everything setup, you'll stare at a blank screen for 30 seconds. To clarify: after a clean system shutdown, without loss of power, when you press the power button you're still looking at a black screen for 30 seconds, before the BIOS logo appears. I find that an incredibly long time, especially when you're not watching the POST code display that tells you something is happening. AMD and the motherboard manufacturers say they are working on improving this—they must. I'm having doubts that your parents would accept such an experience as an "upgrade," considering their previous computer showed something on-screen within seconds after pressing the power button.
Update Sep 29: I just tested boot times using the newest ASUS 0703 Beta BIOS, which comes with AGESA ComboAM5PI 1.0.0.3 Patch A. No noticeable improvement in memory training times. It takes 38 seconds from pressing the power button (after a clean Windows shutdown), until there ASUS BIOS POST screen shots. After that, the usual BIOS POST stuff happens and Windows still start, which takes another 20 seconds or so.

Just like previous AMD Ryzen processors, all AMD Zen 4 CPUs come with unlocked multipliers, which makes overclocking much easier. Our highest manual all-core OC turned out to be 5.1 GHz only, held back by the fact that it's impossible to keep the fully loaded CPU under 100°C, even with a large AIO. This means that you're limited in how much voltage you can feed the beast, and you thus end up with much lower clocks. You could decide to not run Prime95 and test stability with lighter loads only, but that's not what I consider a stable OC. I ran all our benchmarks at 5.1 GHz All-Core OC, and this config ends up consistently slower than the stock configuration due to the high boost clocks on the R9 7950X. Especially in lightly threaded applications, there's obviously a big difference between 5.1 GHz All-Core OC and 5.75 GHz automagic boost with 1-2 threads active. So, unless you have exceptional cooling or are super lucky with the silicon lottery it makes no sense to go for an all-core OC, not even for rendering workloads. The better way to OC Ryzen is using PBO "Auto Overclocking", paired with Curve Optimizer, same as Zen 3. Spending just a few minutes on that will yield you a few percent in OC performance—not a lot, but the times of huge OC gains are a thing of the past—CPU manufacturers are getting better and better at giving you the best performance out of the box.

What is surprising on my Ryzen 9 7950X is that at stock, the clock speeds on all cores in the second CCD are considerably lower than in the first CCD. I'm seeing the same behavior with the 7900X, which is dual-CCD, too. The clock difference is clearly noticeable across all thread counts, from single-threaded to 32 threads active, and not related to CPPC2 "best cores". The delta is between 100 and 300 MHz, which is much more than I would have expected from random silicon variations. If the Windows scheduler puts workloads on the second CCD, then these will run a few percent slower than had they been scheduled on the first CCD. I asked AMD about this behavior and whether it's expected, they haven't responded yet, I'll update this review when they do.

As mentioned before, Zen 4 now comes with integrated graphics. These "just work"—if no discrete graphics card is installed, plug the monitor cable in the motherboard, boom, everything works. Windows Update will install the right driver, or you can grab the official AMD Radeon drivers. Overall IGP performance is outstanding, and plenty for everything except serious gaming. Some lighter 3D apps work perfectly fine, too, and get hardware-acceleration, just like all video decode and encode workloads, for video conferencing as an example. While AMD is very clear that the integrated graphics are not for gaming, I still ran our IGP test suite and have to say I'm impressed—the Zen 4 IGP roughly matches the IGP of the Core i9-12900K, which uses Intel's latest Xe architecture, that they've spent a ton of die area on. A real, discrete graphics card is still much faster, even the most entry-level Radeon RX 6400 offers four times (!) the FPS. For all other typical consumer activities, these integrated graphics are awesome and they'll be a huge selling factor for cost-optimized or compact office systems, a market where Intel has traditionally dominated, because discrete graphics cards weren't required.

AMD wants $700 for the Ryzen 9 7950X, which is a lot of money, but you get an awesome CPU for that. It can handle the toughest workloads, as long as they scale well across multiple cores. Less parallelizable workloads run very fast, too, thanks to the highest boost clock in the Zen 4 range, but the differences will be smaller. The Ryzen 9 5950X is currently $550, even at that price point it makes sense to prefer the 7950X—its price/performance ratio is slightly better, and you get a better upgrade path. On average, Intel's Core i9-12900K does offer slightly better value, due to its $585 pricing, despite its lower performance. If your apps scale well, then you should still prefer the 7950X, because time is money. If you're just running Photoshop and MS Office, then a more midrange CPU like Ryzen 7 or Core i7 will let you save some money that could go to a faster/bigger SSD, or more memory, for example. Overall, the platform cost of Ryzen 7000 is also very high. The new motherboards are extremely expensive, and there's only X670/X670E available for now, the more affordable B650 boards will be released in October. Given the positioning and its price point, platform cost will make up a relatively smaller percentage of the total cost with the Ryzen 9 7950X, unlike the 7600X, for example, where the motherboard costs more than the processor. Intel's new 13th gen CPUs are also getting announced this week. It seems like the processor market will become even more interesting in the coming months, and I'd expect that prices will go down, leading up to 2023.