A week ago we looked at the slightly lackluster AMD Ryzen 9600X and 9700X, which offered slightly better gaming performance in certain games, significantly more power for content creation, and slightly improved thermals. Now it’s time to look at the second half of AMD’s Zen 5 quartet, the £459/$499 Ryzen 9 9900X and £609/$649 Ryzen 9 9950X. These are powerful 12- and 16-core parts that should be of more interest to content creators than gamers, so are they more representative of Ryzen 9000? Can either of these surpass the top gaming performance of the 7800X3D?
Unfortunately, after four days of frantic benchmarking and troubleshooting, I’m not sure AMD succeeded on either of those counts. Our testing of the Ryzen 9900X and 9950X was marked by bafflingly poor gaming performance, including regressions compared to the 7900X and 7950X, and a few instances of genuine improvement, but not enough to make these CPUs worth recommending.
This review marks the first of at least two improvements to our benchmark suite compared to last week’s review, which itself marked the launch of a brand new automated benchmarking system from Digital Foundry and therefore includes a more limited selection of games than we’ve offered in the past.
First, in addition to 1080p, 1440p, and 4K data, we’ve added 720p data, which some people prefer for CPU benchmarks because it more effectively eliminates the impact of the GPU. Second, we’ve added two new games in new genres: Counter-Strike 2 for competitive shooters and Starfield for Bethesda-style RPGs. Both games present a serious challenge to modern CPUs, especially with high refresh rate screens now commonplace. Of course, we’ll continue to expand our testing in the future and look forward to some exciting possibilities.
Based on our 9600X and 9700X results from last week, gamble While performance isn’t improving at the same pace as previous Ryzen generations, we’re expecting at least a reasonable boost in content creation performance. This is thanks to an increase in IPC (instructions per clock, a metric of single-core performance), while clock speeds, power consumption, and thermals have remained flat or improved slightly. As with the 9600X and 9700X, the increase in single-core performance is a combination of architectural improvements (such as an improved branch predictor, better AVX-512 support, and a doubling of data bandwidth between the L2 and L1 caches) and a switch to a more efficient 4nm CCD.
With that in mind, the 9900X is rated at 120W TDP, down from the 7900X at 170W – though the 9950X is rated at the same 170W TDP as the 7950X it replaces. Improved thermal resistance also means temperatures should be lower at the same TDP – AMD claims 7°C lower temperatures at matched TDPs, which is likely to be more important for these high-core-count products, which are more often used for all-core workloads that maximize temperatures.
| CPU Design | Promote | according to | L3 Cache | Thermal Design Piezoelectric | Suggested retail price | |
|---|---|---|---|---|---|---|
| Ryzen 9 9950X | Zen 5 16C/32T | 5.7GHz | 4.3GHz | 64MB | 170 W | £609/$649 |
| Ryzen 9 9900X | Zen 5 12C/24T | 5.6GHz | 4.4GHz | 64MB | 120W | £459/$499 |
| Ryzen 7 9700X | Zen 5 8C/16T | 5.5GHz | 3.8GHz | 32MB | 65W | £339/$359 |
| Ryzen 5 9600X | Zen 5 6C/12T | 5.4GHz | 3.9GHz | 32MB | 65W | £269/$279 |
| Ryzen 9 7950X3D | Zen 4 16C/32T | 5.7GHz | 4.2GHz | 128MB | 120W | $699 / £699 |
| Ryzen 9 7950X | Zen 4 16C/32T | 5.7GHz | 4.5GHz | 64MB | 170 W | $699 / £739 |
| Ryzen 9 7900X3D | Zen 4 12C/24T | 5.6GHz | 4.4GHz | 128MB | 120W | $599 / £599 |
| Ryzen 9 7900X | Zen 4 12C/24T | 5.6GHz | 4.7GHz | 64MB | 170 W | $549 / £579 |
| Ryzen 9 7900 | Zen 4 12C/24T | 5.4GHz | 3.7GHz | 64MB | 65W | $429/£519 |
| Ryzen 7 7800X3D | Zen 4 8C/16T | 5.0GHz | 4.2GHz | 96MB | 120W | $449 / £375 |
| Ryzen 7 7700X | Zen 4 8C/16T | 5.4GHz | 4.5GHz | 32MB | 105 W | $399 / £419 |
| Ryzen 7 7700 | Zen 4 8C/16T | 5.3GHz | 3.8GHz | 32MB | 65W | $329 / £349 |
| Ryzen 5 7600X | Zen 4 6C/12T | 5.3GHz | 4.7GHz | 32MB | 105 W | $299 / £319 |
| Ryzen 5 7600 | Zen 4 6C/12T | 5.1GHz | 3.8GHz | 32MB | 65W | $229/£249 |
| Ryzen 5 7500F | Zen 4 6C/12T | 5.0GHz | 3.7GHz | 32MB | 65W | $200/£255 |
Per AMD’s recommendations, we used a similar physical setup to our previous Ryzen 7000 tests, including an ASRock X670E Taichi motherboard, G.Skill Trident Z5 Neo DDR5-6000 CL30 RAM, and an Eisbaer Aurora 240mm AiO. The biggest difference is that our RTX 3090 has been swapped out for an RTX 4090 Founders Edition, providing the extra graphics performance to push modern CPUs further.
While Ryzen 9000 was the star of the show, we also tested some earlier Ryzen 3000 and 5000 processors, as well as some Intel 14th Gen processors. Our Intel CPU testing was conducted on a Gigabyte Aorus Z790 Master using the same DDR5-6000 CL30 RAM, while the AM4 CPUs were tested on the classic Asus ROG Crosshair 8 Hero using the Trident Z Royal DDR4-3600 CL16 kit.
For storage, we used a 4TB Lexar NM790 PCIe 4.0 NVMe SSD and a 4TB Kingston KC3000. Our rig was equipped with a 1000W Corsair RM1000x power supply. Testing was performed on a fresh installation of Windows 11 with the latest Windows updates (23H2), chipset drivers (6.06.28.910), and BIOS revision (3.06 for ASRock AM5 motherboards).
| Cinebench | 2024 (1T) | 2024 (MT) | R20 (1T) | R20 (metric ton) |
|---|---|---|---|---|
| Ryzen 5 3600X | 77 | 578 | 485 | 3654 |
| Ryzen 7 5800X3D | 95 | 915 | 546 | 5746 |
| Ryzen 9 5900X | 98 | 1171 | 610 | 8393 |
| Ryzen 5 7600X | 114 | 845 | 744 | 5814 |
| Ryzen 7 7700X | 118 | 1127 | 758 | 7609 |
| Ryzen 7 7800X3D | 112 | 1074 | 688 | 6988 |
| Ryzen 9 7900X | 116 | 1605 | 776 | 11196 |
| Ryzen 9 7950X | 121 | 2004 | 784 | 14272 |
| Ryzen 5 9600X | 132 | 935 | 850 | 6358 |
| Ryzen 7 9700X | 130 | 1172 | 862 | 7851 |
| Ryzen 9 9900X | 135 | 1784 | 879 | 12617 |
| Ryzen 9 9950X | 138 | 2235 | 866 | 15850 |
| Intel Core i5 14600K | 120 | 1400 | 777 | 9420 |
| Intel Core i7 14700K | 127 | 1987 | 818 | 13614 |
| Intel Core i9 14900K | 133 | 2107 | 875 | 15297 |
Before we get into the gaming benchmarks, it’s worth spending some time looking at the results of the content creation tests. These test results provide a synthetic benchmark for gamers, indicating some of the performance levels we might expect in a best-case scenario, and of course, they’re also a useful indicator for anyone who actually intends to use a Ryzen 9000 CPU for 3D modeling or video production. These 12-core and 16-core parts tend to have impressive performance, and fortunately, this is also the case with Zen 5.
We hope to expand this in the future, but for now we have three workloads: Cinebench 2024 and R20 emulate the 3D modeling and animation program Cinema 4D, while Handbrake is actually just transcoding early DF Patreon videos to H264 and H265 formats – a common task for anyone working with video.
Let’s start by looking at the Cinebench 2024 results, which were recorded after a 10+ minute loop to ensure the CPUs couldn’t quickly get to a good score before thermal or power limits kicked in. Here we see single-core performance for both CPUs improving by around 15% over their predecessors, while multi-core performance gains were a bit more modest: both CPUs were up 11%. For comparison, Intel’s 14700K was 11% faster than the 9900X in this test, but the 9950X was around 6% faster than the 14900K.
Cinebench R20 is a traditional test for us and has been run before in a number of our CPU reviews, so we thought it was worth running in case you want to go back and compare results. We see similar margins here (11% improvement in multi-core performance for both Ryzen 9000 CPUs).
| Handbrake | H264 (frames per second) | HEVC (frames per second) |
|---|---|---|
| Ryzen 5 3600X | 26.66 | 10.80 |
| Ryzen 7 5800X3D | 42.00 | 18.71 |
| Ryzen 9 5900X | 57.59 | 23.83 |
| Ryzen 5 7600X | 41.29 | 18.31 |
| Ryzen 7 7700X | 53.27 | 23.65 |
| Ryzen 7 7800X3D | 49.63 | 21.54 |
| Ryzen 9 7900X | 78.35 | 32.59 |
| Ryzen 9 7950X | 98.58 | 41.68 |
| Ryzen 5 9600X | 42.51 | 19.77 |
| Ryzen 7 9700X | 51.80 | 23.79 |
| Ryzen 9 9900X | 82.96 | 35.33 |
| Ryzen 9 9950X | 103.25 | 44.97 |
| Intel Core i5 14600K | 59.42 | 25.39 |
| Intel Core i7 14700K | 80.26 | 31.07 |
| Intel Core i9 14900K | 85.06 | 35.08 |
Our final test is the Handbrake transcoding test, in which we convert an 822MB 4K video file using the H264 and H265 codecs with the Production Standard preset set to CRF 18. We’re using the latest version of Handbrake here, 1.8.1, which helpfully displays its stats in the app window rather than burying the average encoding frame rate in a log file.
These transcoding results show a much smaller advantage for the Ryzen 9000, something we also found with the 9600X and 9700X. The 9900X and 9950X both have around a 5% advantage on H264 and around an 8% advantage on H265 over their predecessors – this is likely due to improvements to the AVX architecture.
We didn’t get a chance to try out PBO and other performance enhancements this time around, but we did see a big boost with the 9700X at the expense of power efficiency, and I’d expect the 9900X and 9950X to be significantly faster if stock power and thermal budgets aren’t factored in.
With the content creation benchmarks out of the way, it was time for the gaming benchmarks. Unfortunately, this is where things started to go south.
