Technology & Positioning
With the Ryzen 9000 Series, AMD is launching their new Zen 5 microarchitecture, which brings improved IPC and better energy efficiency, while staying compatible with all Socket AM5 motherboards. For their recent processor releases, Intel has been betting on a mix of P-Cores and E-Cores, AMD's Ryzen 9000 processors are still pure P-Core designs, which simplifies scheduling. Last week, we posted reviews for the Ryzen 7 9700X and the Ryzen 5 9600X, today we have for you the Ryzen 9 9900X (this review) and Ryzen 9 9950X.

While the IO die is exactly the same as on Zen 4, the new Zen 5 "Granite Ridge" CPUs introduce a new CCD design, which is still eight cores per die, with 32 MB L3 cache, but fabricated on TSMC's 4 nanometer production process. An interesting change is the switch from 256-bit to a full 512-bit data path for AVX512, more on that later, the frontend and L1 and L2 caches have been improved, too. Clock speeds are roughly similar to previous gen, the 9900X runs at a maximum boost frequency of 5.6 GHz (same as 7900X), base clocks are a bit lower though, 4.4 GHz vs 4.7 GHz. While last week's processors both came with a lowered TDP of 65 W, the 9950X retains the same 170 W TDP as its predecessor, and the 9900X runs at 120 W, compared to 170 W on the 7900X. At $500, the strongest competitors for the 9900X are Intel's 14700K ($380) and 14900K ($555), AMD 7800X3D ($370), 7950X ($525), 7950X3D ($530), and of course 7900X ($360)—this is a long list, because the 9900X's price point is so high.

Application Performance
Just a few weeks ago, we introduced our new 2024 CPU Test Suite, which runs the newest versions of our apps and includes new workloads, especially for AI-related tasks, which are becoming more and more important every day. Averaged over these 49 tests, the Ryzen 9 9900X achieves a 4.4% performance uplift over the Ryzen 9 7900X, which is much less than expected. Depending on the application, there's some pretty big differences in terms of scaling, we even saw some negative gains gen-over-gen. The biggest winners are some heavy commercial applications, like V-Ray rendering, NoSQL database and AI workloads, which see double-digit improvements. These numbers suggest that it might make sense for enterprises to absorb the higher hardware cost, because the increase in productivity will make the money back in no time. On the other hand, for the vast majority of users and even creators, the differences are barely noticeable. When we removed the power limit and enabled PBO auto overclocking we saw some performance improvements, but I wouldn't say these processors are "power constrained." AMD's choice of a TDP of 120 W is very reasonable, also in terms of cooling requirements, and it's easy enough to adjust the limits anyway. There are several strong competitors for 9900X. My favorite is the Core i7-14700K, which is much cheaper, because it's one tier below, but still offers comparable performance to the 9900X, at just $380. If you prefer AMD, then maybe consider the 7950X, which costs just $20 more than 9900X and will give you a 16c/32t monster that crushes all compute heavy workloads.

AMD's new Zen 5 architecture is setting new standards in single-threaded performance. For example, Zen 5 has broken records in MP3 encoding, achieving speeds nearly 10% faster than the 6 GHz 14900K—AMD's IPC gains are truly remarkable. As mentioned before, AMD has implemented a full 512-bit data path for AVX512, which should help, especially in commercial applications that are optimized to make use of this modern instruction set. For consumers, it's mostly irrelevant though. Using AVX512-VNNI instructions in our Stockfish Chess test, the 9900X shows a decent 10.5% performance improvement, but it runs into the power limit. When we removed that limit, and added some PBO overclocking, that grew into a 23% gain, which is quite nice, but comes with serious power consumption of 280 W for the CPU alone—these new AVX512 units do consume more power than before.

Gaming Performance
AMD's Zen 4 Ryzen processors are formidable gaming machines, especially the X3D series, led by the 7800X3D. After their Computex announcement, to manage expectations, AMD confirmed that their new Zen 5 processors will not be able to beat the gaming performance of the 7800X3D. Our gaming tests show that AMD has definitely improved gaming performance over Zen 4—the 9700X beats almost any previous AMD processor—only the 7800X3D and 7950X3D are faster. At 1080p, the 9900X is 2% faster than the 7900X, but is not fast enough to beat last generation's X3D, especially the 7800X3D enjoys a solid 9% lead. Intel's high-end Raptor Lake processors deliver slightly higher FPS, too, by 2-5%, depending on the model. Interestingly, there is very little difference in gaming performance between all Zen 5 models, they all sit in a narrow band that's just 2% wide. While the 9900X is a fantastic gaming processor, just like all the other Zen 5 models, I still find it hard to recommend for gaming, because of the price point. You will have the same gaming experience if you buy a 9600X for $280, and save $220, which could go to a faster graphics card, to definitely improve both the minimum and average FPS. Another great choice for gaming is the 7800X3D, which is a bit more expensive, but offers notably higher performance and gaming efficiency at the same time. Intel's Raptor Lake CPUs (13700K and up) are also a bit faster, at considerably lower pricing.

Just two days ago, AMD notified us that "Ryzen 9 9900X and 9950X have Windows Game Mode core parking optimizations installed by the AMD PPM Provisioning File Driver," and that "Windows game mode must be enabled," "Xbox Game Bar must be enabled and up to date in the Microsoft Store," "the legacy Control Panel Power Options must be set to the default 'Balanced' scheme," and that "sometimes Windows does not apply the correct provisioning after the CPU installed has changed. You can try uninstalling then re-installing the AMD Chipset Driver as a workaround, but a fresh install of Windows is ideal." That's a lot of hoops to jump through for a dual CCD processor. Yes, we tested both 9900X and 9950X with those optimizations enabled. Previously this was required only for the dual CCD X3D models, so that games could be pushed onto the cores with 3DV cache. On Zen 5, AMD is using the same mechanism to improve game thread allocation, probably to put them on the cores with the highest default clocks. The difference should be pretty small though. I doubt it's more than a few percent and I probably would rather get rid of Game Bar instead.

Integrated Graphics
While the RDNA 2 integrated graphics are not a main selling point of the Ryzen 9900X, we've still tested it. Performance for non-gaming tasks is good, you'll be able to run all productivity, office, web browsing and media playback apps easily. From a technical perspective, this is the same iGPU as on Ryzen 7000 processors, because AMD is using the same IO die, where the iGPU lives. In gaming, the performance is identical to that of the Ryzen 7000 iGPU. AMD has always made it clear that the iGPU is not for serious gaming, yet it's still faster than what Intel sells in their Raptor Lake processors, and it works with all games, while the Intel iGPU failed in 3 out of 9 game tests.

Power Consumption
At stock, our Ryzen 9900X press sample provided by AMD wasn't able to run at DDR5-6000 stable and crashed with a bluescreen within minutes. The solution was to increase the memory controller's voltage (which is tied to "SoC Voltage") from the default of 1.2 V to 1.25 V. Now everything was fully stable, but power consumption is increased by a small bit of course (not more than 3 W). All power results in this review are based on SoC voltage of 1.25 V, there is no other way to run this processor at DDR5-6000. I looked further into the instability, and it turns out that the processor is 100% stable at DDR5-5800. AMD's official product spec is DDR5-5600, so DDR-6000 is technically overclocking, which isn't guaranteed, so legally the processor is not defective.

AMD's Ryzen 9 9900X uses a bit less power than its predecessor, also because the TDP has been lowered from 170 W to 120 W. The differences aren't huge though. On average, in our mix of 49 applications, the 9900X consumed 113 W while the 7900X used 123 W. In demanding workloads this can grow to over 40 W. Gaming power is slightly improved, too, but only by a few %, and the X3D processors still rock here. While power compared to the Intel 14900K is much better, the differences against 14700K and 13700K are slim enough to not matter much. With our 2024 Test Suite we've added full system idle power consumption measurements, here the 9900X doesn't do so well, just like the older Zen 4 processors. You should expect around 20 W more than a similar Intel system. The underlying reason seems to be the tiled die design, especially the IO die, which can't fully power down in idle. Interestingly, the die of the second CCD on the 9900X barely makes a difference in idle power, which suggests it has good support for clock gating.

Cooling Requirements
Given the 120 W TDP, AMD recommends pairing the 9900X with a liquid cooling solution. We've tested it on an air-cooled Noctua D15, and temperatures are fine, which suggests that weaker air coolers should be able to handle the 9900X, too, as long as you run it at stock settings. At first glance, temperatures might appear much better (71°C vs 92°C), but a large part of that is due to much lower power consumption during the Blender test, which we use for temperature measurements (173 W vs 200 W). During their press briefings AMD mentioned a 15% / 7°C improvement in thermals due to optimized floor planning and better placement of the thermal sensor. They also confirmed that no significant changes were made to the IHS and the TIM. If you want to dive into overclocking, then you better bring serious cooling. Once unleashed, power consumption and heat output climbs quickly, and you'll often run into the 95°C thermal limit—even with watercooling. The problem is not the cooling itself, but the heat transfer through the IHS, which is a compromise that AMD chose to retain cooler compatibility with Socket AM4.

Platform
We've seen very long boot times from the bigger AM5 processors in the past, and it seems that this is something that AMD can't really fix conclusively. While we found memory training times with the monolithic Phoenix 2 (Ryzen 8500G) much improved, that doesn't use a separate IO die, Zen 5 has only small improvements. After installation (or from a power disconnected/CMOS clear, with MCE disabled), the time spent on memory training is almost three minutes—feels like an eternity. However, once that is complete, powering the machine on is very fast and takes only 12 seconds to the Windows startup screen, which is just as fast as on Intel systems. Good job, AMD for improving that! As expected, you get full support for PCIe 5.0 x16 for graphics and 5.0 x4 for the M.2 SSD on all high-end motherboards. I also really like that AMD continues to support Socket AM5. They actually confirmed that future CPUs will support the socket, too, until 2027 at least! Intel will be switching sockets again this year, with the release of Arrow Lake.

AMD is announcing new 800-Series chipsets with Ryzen 9000, but the motherboards aren't ready yet. These are expected to come out later this summer, with the biggest innovation being USB4 and Wi-Fi 7. All features, including overclocking, Curve Shaper, etc., are fully supported on existing AM5 motherboards (after a BIOS update).

Overclocking
All Zen 5 processors have their multiplier unlocked for easy overclocking. Using a classic all-core OC approach I got 5.3 GHz stable in all workloads. Actually I could even boot at 5.5 GHz and run some light tests at 5.4 GHz—better than all my other Zen 5 samples—looks like I got lucky with the silicon lottery. As mentioned before, my CPU can't handle DDR5-6000 at stock, so maybe this would have been a high-end EPYC, but fell through the memory clock requirement and thus ended up as Ryzen. While temperatures were a big problem during OC with the 9950X, they weren't a problem with the 9900X. Actually, the dual CCD design helps spread out the heat over a larger area, which makes the processor a bit easier to cool. I still couldn't get all cores stable in the heaviest tests, so I ended up running 5.3 GHz, which is pretty good. Still, the PBO method of overclocking yields much better results, because the processor can boost high for low-threaded workloads and runs at around 5.3 GHz with all cores fully loaded, too. The "PBO Maximum" run is with the power limits removed, PBO Scalar set to x10, Clock to +200 and Curve Optimizer at -10.

Despite the excellent stability and performance of Zen 5, the overclocking experience with Ryzen Master is lacking. Whenever I used it for PBO/overclocking and the system crashed, POST would freeze at code 1A. Resetting didn't solve the problem, requiring a CMOS clear and a three-minute wait for memory training before testing could continue. I'm confident AMD will eventually address this issue, but it's surprising that it wasn't noticed prior to release.

Pricing & Alternatives
According to AMD, the Ryzen 9 9900X will sell for $500, which is much too high, and doesn't make much sense given the current competitive landscape. The 7900X is currently $360, AMD wants you to pay $140 for a few percent in extra performance. The previous-gen 16c/32t flagship 7950X sells for $525 right now—paying $25 extra for an extra four cores is a no-brainer if you need application performance. For gamers, the 7800X3D ($370) is the best CPU they could buy, and the 7950X3D ($530) isn't that much more expensive than the 9700X. Intel also has plenty of options like the 14900K for $555 and the 14700K for $380. Last but not least, the Zen 5 9600X will give you the same gaming performance as 9900X for just $280. Don't get me wrong, 9900X is a fantastic processor that's strong in both gaming and applications, but it's priced way too high. If it was much closer to $400, or even just $430, things would be different, but at $500 it will be a tough sell. Considering that the 7900X suffered the same fate, and AMD made it sellable with heavy discounts, this is probably the future of the 9900X, too, so don't give up all hope.

Upcoming Releases
Now that all the (expected) AMD Zen 5 processors are released, we have clarity regarding AMD's lineup. It seems likely that AMD will release a 9950X3D and 9900X3D later this year, which will offer improved gaming performance, likely beating the Zen 4 X3D models, but AMD won't sell those cheaply. It's also highly probable there will be a 9800XD that will be priced a bit lower than the 9900X, so it might be worth waiting for. On the Intel side, we'll see Arrow Lake later this year, which comes with a new architecture without Hyper-Threading, which should provide an additional boost to single-threaded workloads and gaming, gains for highly multithreaded applications could be limited. Still, especially in the context of the 9900X I'd be curious to know what Arrow Lake brings to the table and how it will affect AMD's pricing strategy.