AMD's Ryzen 9 5950X Shatters PassMark Records

Ryzen Desktop Processor

Ryzen Desktop Processor (Image credit: AMD)

If the Ryzen 5 5600X left you impressed, the Ryzen 9 5950X will blow you out of your chair. The 16-core monster has catapulted its way to the top of the mainstream processor chart in both single-and multi-thread performance.

The Ryzen 9 5950X scores are no longer available, but German publication ComputerBase managed to grab screenshots of the Zen 3 flagship's results before they were erased. Given the circumstances, it's uncertain if the Ryzen 9 5950X was overclocked or whether it was paired with memory that surpass the official supported DDR4-3200 specification. While we wait on the full review, we recommend you take the PassMark scores with a pinch of salt.

While AMD has been injecting more cores in mainstream chips for a while now, the chipmaker's offerings aren't quite up to par with Intel's parts when it comes to single-thread performance. If these PassMark numbers are accurate, it would appear that Zen 3 has finally tipped the scales in AMD's favor.

AMD Ryzen 5 5950X Benchmarks

Swipe to scroll horizontally
ProcessorPassMark Single-Thread ScorePassMark Multi-Thread Score
Ryzen 9 5950X3,69345,563
Ryzen 5 5600X3,49522,824
Core i9-10900K3,17624,261
Ryzen 9 3950X2,74739,277
Core i9-10980XEN/A34,138

The Ryzen 9 5950X is reportedly up to 34.4% faster than the Ryzen 9 3950X in single-thread performance and up to 16% in multi-thread performance. For reference, the Ryzen 9 5950X comes with a 3.4 GHz base clock and 4.9 GHz boost clock, while the Ryzen 9 3950X has a 3.5GHz base clock and 4.7 GHz. It was to expected that the Ryzen 9 5950X would be superior chip.

In comparison to its Intel rival, the Ryzen 9 5950X seemingly delivers up to 16.3% higher single-thread performance than the Core i9-10900K. Now, you have to remember that the Core i9-10900K features a 3.7 GHz base clock and a whopping 5.3 GHz boost clock. We're not underestimating Zen 3, but it's a bit hard to swallow that the AMD chip with a 400 MHz lower boost clock would outperform the Core i9-10900K. For now, we'll have to trust PassMark's metrics until we get the chip in our lab for thorough testing.

Possessing substantially more cores, the Ryzen 9 5950X's multi-thread performance doesn't raise any eyebrows. The 16-core processor purportedly offers up to 87.8% higher multi-thread performance than the Core i9-10900K. Intel doesn't offer more than 10 cores on its mainstream processors so the Core i9-10980XE, which is a HEDT (high-end desktop) SKU, will have to be the point of comparison for multi-thread performance. Despite being at a two-core disadvantage, the Ryzen 9 5950X's multi-thread performance is apparently faster than the Core i9-10980XE by up to 33.5%.

If Ryzen 5000 (Vermeer) can deliver, Intel could be in big trouble since it's Comet Lake-S army will likely not be able to fend of Zen 3. It looks as though even Intel's upcoming Rocket Lake processors might not be enough.

Zhiye Liu
RAM Reviewer and News Editor

Zhiye Liu is a Freelance News Writer at Tom’s Hardware US. Although he loves everything that’s hardware, he has a soft spot for CPUs, GPUs, and RAM.

  • tummybunny
    As someone who's been around a while it's hard not to feel that both AMD and Intel deserve this outcome.
  • Schlachtwolf
    Amazing to think that the upgrade for me from my 2700x to say a 5900x will give me an FPS boost similar to a GPU upgrade.... mad....(and a new GPU will be coming too when the market settles in....... 2025 LOL)
  • sizzling
    Schlachtwolf said:
    Amazing to think that the upgrade for me from my 2700x to say a 5900x will give me an FPS boost similar to a GPU upgrade.... mad....(and a new GPU will be coming too when the market settles in....... 2025 LOL)
    Only if you are entirely cpu bound with the 2700x.
  • nofanneeded
    I wonder how Threadripper 5000 would perform ... Bravo AMD !!!
  • Schlachtwolf
    The 2700x has been great to me but a RTX 3080 or (likely) Radeon 6900xt would certainly push it to and probably beyond it's limits. In RDR2 I have played on my system and on an Intel 10900k with an RTX 2080 and both with 3600mhz Corsair RAM.... you notice a big difference in the FPS on higher settings, so the 2700x is hitting it's ceiling on new GPU's.
  • VforV
    We're not underestimating Zen 3, but it's a bit hard to swallow that the AMD chip with a 400 MHz boost clock would outperform the Core i9-10900K.
    *400 MHz lower boost clock you mean.

    Also, "a bit hard to swallow"? There are dozens of leaked benchmarks by now shopwing the same scenario, yet it's still hard to swallow for some people.... pffft. Well prepeare to "swallow" for real in 2 weeks time when AMD kicks Intel's a***...
  • Psyadin
    Would've thought TH would know about IPC, but apparently they dont, frequency is only one part of the equation, it would be like measuring your property size by length and no width.
    In synthetical tests where more each cycle is more or less maxed out, the equation is frequency x ipc, to my knowledge Intel and AMD doesnt give an exact number, and it would be hard to verify even if they did, but we know for certain that AMD's is well over Intel's since at least Zen 2.
    So the 400MHz difference doesnt matter that much, and assuming AMD's IPC is 10% or more over Intels it will far outweigh that difference... at least in synthetic benchmarks.
  • RodroX
    Well this is no surprise, it was kinda what you would expect after seen the Ryzen 5 5600X result last week..

    I know this is only a leak, and only one test, but the most amazing part for me is that the Ryzen 5 5600X (6 cores / 12 threads) appears to have a multicore result of ~22800 points, when the Core i9 10900K have 24200 point with its 10 cores and 20 threads, that for me is kinda mind blowing.
  • semitope
    Till actual benchmarks show up, these passmark results with 1 CPU of unknown settings should be ignored. Everything else you'll compare it to would have thousands of results averaged
  • Jesse_20
    Does no one proofread their work these days? So many grammatical errors.
    Quality of the site keeps diminishing.