The Ryzen 7 5800X3D is a truly special CPU from AMD, a tribute to the long-lived AM4 platform whose gaming performance dwarfs other Ryzen 5000 processors due to its 3D V-Cache design. Now its first two successors have arrived, trying to capture that same magic, the Ryzen 9 7950X3D and 7900X3D – with the Ryzen 7 7800X3D set to debut in April.
All three should be standout performers, but can they take the “best gaming crown” from the Intel 13900K? With more powerful sockets, faster DDR5 RAM, and more advanced manufacturing, will a larger L3 cache be as transformative for the Ryzen 7000 as it was for the Ryzen 5000? To find out, we’ve been testing the flagship Ryzen 9 7950X3D, which offers 16 Zen 4 cores and 3D V-Cache for £699/$699 – the same MSRP as the original 7950X.
Before we get into content creation and gaming benchmarks, it’s worth outlining what makes the 7950X3D’s hardware so fascinating. Ryzen CPUs have used chiplet designs from the start, with a single chiplet (“CCD” in AMD nomenclature) for the low- and mid-range parts, two chiplets for the high-end part, and continuing with the 7950X3D.The biggest difference here is the 7950X3D motion asymmetrical Design where one chiplet received a 3D V-Cache upgrade in exchange for a slight loss in maximum frequency, while the other chiplet retained the smaller cache size and frequency of the 7950X it was based on.
This asymmetrical design may sound odd, and it does have its drawbacks, but there are some clear advantages here too. First, a CPU with one 3D V-Cache chiplet is cheaper to produce than a CPU with two chiplets, thereby lowering the price. Another advantage is that programs that benefit from a larger cache, such as games, can run primarily on these cores, while tasks that don’t benefit can take advantage of the higher frequencies available on non-V-Cache cores.
To ensure games and applications are assigned to the correct cores, AMD makes a number of additions to its chipset software. If a Dual CCD 7000X3D CPU is detected while installing new chipset drivers, you will get background services, background processes, new entries in Device Manager, and more. (Conversely, if you install the same chipset driver version without the X3D CPU, and then change to another one, none of these extra features will be activated; you’ll need to reinstall the chipset driver. Ask me how I know! )
This is pretty clever stuff, AMD uses the Windows Game Bar to detect when a game is active and “parks” the frequency cores, in the case of the 7950X3D 16-31, to ensure that Windows prefers to use the V-Cache with game cores . This means you can use the Game Bar override to tell Windows that any application is a game to run it on the high-cache core, and there are also registry keys to disable this behavior for specific games. Well, I’m talking about specific games, but there’s only one title out there right now – League of Legends. Finally, you can also modify this behavior in the BIOS by manually selecting Auto, Cache, or Frequency cores as the first choice for everything – a feature we tested on page 6 to see how much of an advantage 3D V-Cache really has.
You can see in the table below where the three Ryzen 7000X3D CPUs stand against their peers. As a reminder, each of these Zen 4 designs benefited from a range of improvements over their Zen 3 counterparts, citing a 13% increase in instructions per clock (IPC) as well as more internal improvements such as an improved execution engine and Better branch predictors. There’s also DDR5 and PCIe 5.0 support, with 5nm CCD and 6nm I/O chips, and a new AM5 socket that unlocks extra power and performance — albeit requiring a new motherboard, new RAM, and possibly new cooling.
| CPU design | Promote | according to | L3 cache | thermal design plan | Suggested retail price | |
|---|---|---|---|---|---|---|
| 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 | 170W | $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 | 170W | $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/TBD |
| Ryzen 7 7700X | Zen 4 8C/16T | 5.4GHz | 4.5GHz | 32MB | 105W | $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 | 105W | $299/£319 |
| Ryzen 5 7600 | Zen 4 6C/12T | 5.1GHz | 3.8GHz | 32MB | 65W | $229/£249 |
For our testing, we used the same basic setup as in our Ryzen 7600 review, just with the latest AMD chipset drivers installed. That means our graphics card requires an ASRock X670E Taichi motherboard, a G.Skill Trident Z5 Neo DDR5-6000 CL30 RAM, and an RTX 3090 Strix OC from Asus. Cooling is provided by a 240mm Alphacool Eisbaer Aurora AiO, which is compatible with the new AM5 socket.
For storage, we use three PCIe 4.0 NVMe SSDs to hold all our games – a 4TB Kingston KC3000, a 1TB PNY XLR8 CS3140, and a 1TB Crucial P5 Plus. Our rig came with a 1000W Corsair RM1000x power supply.
To maintain compatibility with past benchmarks, we are using a version of Windows 11 prior to 22H2, as it provides the best performance for Ryzen 7000 processors. We also used the latest BIOS, version 1.15.SMU215, and the latest version of AMD chipset drivers on the ASRock X670E Taichi motherboard while conducting our tests.
To test other platforms, we use the ASUS ROG Crosshair 8 Hero for the Ryzen 5000 test, the ASUS ROG Maximus Z590 Hero for the 11th Gen Intel test, the ASUS ROG Z690 Maximus Hero for the 12th Gen test, and the Gigabyte Z790 Aorus Master 13th Gen Tested; all of these are high-end boards for their respective platforms. DDR4 motherboards use G.Skill 3600MT/s CL16 memory, which is the best choice for DDR4, while DDR5 motherboards use DDR5-6000 CL30, as mentioned earlier.
Before we get into the gaming benchmarks that make up pages two through six, let’s set up some quick content creation benchmark scenarios: Cinebench R20 3D rendering and Handbrake video transcoding.
Not surprisingly given that Cinebench and Handbrake didn’t benefit from the extra cache we see slightly lower results here compared to the higher clocked 7950X 3D which was about 10% faster in Handbrake export and about 7.5% faster in Cinebench % Multi-Core Test In the single-core test, we expected the results to be closer since the 7950X3D can use its higher frequency cores, and indeed only the 7950X had a 1% gap in this test.
This also puts the 7950X3D behind the 13900K in multi-core workloads, but the top Intel CPU only performed about 10% better in Handbrake and about 13% better in Cinebench – so the gap isn’t that big. The 7950X3D is still significantly faster than any previous generation competitor, and by a wide margin — for example, 36% faster than the 5950X in Cinebench, and 32% faster than the 12900K.
| Expo R20 1T | CBRE R20 Metric Ton | Harvard h.264 | High-quality HEVC | HEVC power usage | |
|---|---|---|---|---|---|
| Ryzen 9 7950X3D | 788 | 13807 | 95.73fps | 40.70fps | 232W |
| Ryzen 9 7950X | 798 | 14837 | 105.15 frames per second | 45.10fps | 368W |
| Ryzen 9 7900X | 791 | 11324 | 79.38 frames per second | 33.77 frames per second | 288W |
| Ryzen 7 7700X | 768 | 7894 | 56.69 frames per second | 25.95 frames per second | 266W |
| Ryzen 5 7600X | 750 | 6063 | 44.35 frames per second | 20.28 frames per second | 236W |
| Ryzen 5 7600 | 706 | 5632 | 41.09 frames per second | 18.72fps | 196W |
| Ryzen 9 5950X | 637 | 10165 | 70.28 frames per second | 30.14 frames per second | 237W |
| Ryzen 7 5800X3D | 546 | 5746 | 42.71fps | 19.10fps | 221W |
| Ryzen 7 5800X | 596 | 6118 | 44.18 frames per second | 19.50fps | 229W |
| Ryzen 5 5600X | 601 | 4502 | 31.75 frames per second | 14.43 frames per second | 160W |
| Core i9 13900K | 873 | 15570 | 104.67 frames per second | 41.20fps | 473W |
| Core i5 13600K | 767 | 9267 | 62.37 frames per second | 26.44 frames per second | 254W |
| Core i9 12900K | 760 | 10416 | 70.82 frames per second | 29.26 frames per second | 373W |
| Core i7 12700K | 729 | 8683 | 57.64 frames per second | 25.67 frames per second | 318W |
| Core i5 12600K | 716 | 6598 | 44.27 frames per second | 19.99 frames per second | 223W |
| Core i5 12400F | 652 | 4736 | 31.77 frames per second | 14.70fps | 190W |
| Core i9 11900K | 588 | 5902 | 41.01 frames/second | 18.46 frames per second | 321W |
| Core i5 11600K | 541 | 4086 | 29.00 frames per second | 13.12 frames per second | 250W |
Power consumption here is very good, and in the same HEVC encoding workload (368W), the Ryzen 9 7950X3D draws significantly less power from the wall (232W) than the 7950X, partially making up for its drop in performance. It’s also significantly better than the Intel Core i9 13900K (473W). We’d like to see less power draw for non-AVX workloads such as gaming, but for content creation power consumption is definitely relevant to the current economic climate – it speaks to the efficiency of AMD’s latest processors more generally.
Now, let’s get into the fun stuff – games. We’ve tested a range of titles, so choose your favorite from the links below, or simply hit the next page button to continue the journey.
