Wednesday, September 22nd 2021
Intel Core i9-12900K "Alder Lake" Beats Ryzen Threadripper 2990WX at Cinebench R23 nT
An alleged Intel Core i9-12900K "Alder Lake-S" sample is shown beating the 32-core AMD Ryzen Threadripper 2990WX HEDT processor at AMD's favorite benchmark, Cinebench R23, in its multi-threaded (nT) test. At this point it's not known whether the i9-12900K is overclocked, but the CPU-Z instance in the screenshot reads 5.30 GHz, which could very well be the processor's stock Thermal Velocity Boost frequency. The sample scored upward of 30000 points, putting it above the Threadripper 2990WX reference score in Cinebench.
The 2990WX is based on the "Zen+" microarchitecture, and released in 2018, but is a 32-core/64-thread chip that should have ripped through this rendering workload. The i9-12900K, on the other hand, has eight "Golden Cove" performance cores that have HyperThreading, in addition to 8 "Gracemont" efficiency cores that lack HTT. This benchmark was run on Windows 10, which lacks awareness of the Intel Thread Director, a hardware component that optimizes utilization of the two kinds of CPU cores. Windows 11 is known to feature better awareness of hybrid core architectures. The i9-12900K sample is possibly installed on a Gigabyte Z690 AORUS Ultra motherboard, and has 32 GB of DDR5-5200 memory (two modules, logically four 40-bit channels).
This would be a giant-slaying act by Alder Lake, as its motley crew of 8+8 cores is able to overcome an enormous CPU core-count deficit compared to the Threadripper 2990WX. To its credit, the 2990WX is a 3-year old processor based on a core with a much lower IPC than "Golden Cove." It also features a sub-optimal 2+2 channel DDR4 memory layout that AMD later corrected with the centralized memory controller on the IOD, with the Threadripper 3000 series.
Source:
REHWK (Twitter)
The 2990WX is based on the "Zen+" microarchitecture, and released in 2018, but is a 32-core/64-thread chip that should have ripped through this rendering workload. The i9-12900K, on the other hand, has eight "Golden Cove" performance cores that have HyperThreading, in addition to 8 "Gracemont" efficiency cores that lack HTT. This benchmark was run on Windows 10, which lacks awareness of the Intel Thread Director, a hardware component that optimizes utilization of the two kinds of CPU cores. Windows 11 is known to feature better awareness of hybrid core architectures. The i9-12900K sample is possibly installed on a Gigabyte Z690 AORUS Ultra motherboard, and has 32 GB of DDR5-5200 memory (two modules, logically four 40-bit channels).
96 Comments on Intel Core i9-12900K "Alder Lake" Beats Ryzen Threadripper 2990WX at Cinebench R23 nT
but, weirdo that I am, I am most pleased with CPU-Z recognising it as 8 + 8 cores and does not just say 16 cores / 24 threads, now if only task manager could show the same distinction.....come on microsoft!
Intel is doing it right imo , get the most out of what you have and perfect it , instead of spending more money on rnd to develop new processes, once you hit the wall , then you move on ....
That being said , i still think amd has a competitive edge , I just wish they could work on 7nm for a longer period and get the most out of it , they won’t have to spend more on rnd and tsmc will have more room to provide for other chip manufacturers.....
or 8+8/32 Threads as stated here:
cdn.wccftech.com/wp-content/uploads/2021/09/Intel-Core-i9-12900K-Alder-Lake-Desktop-CPU-5-GHz-Benchmark.png
And 5950x at 4700 already beats 2990WX:
www.notebookcheck.net/16-core-Ryzen-9-5950X-overclocked-to-4-7-GHz-on-all-cores-1-12V-outperforms-32-core-Threadripper-2990WX-in-Cinebench-R23.505662.0.html
So nothing really to get excited about. It is just Intel being Intel again ..
Intel will have an answer in a couple years because these processors are going up against AMD's 3D-V Cache. And don't stand a chance. Not yesteryears hardware.
However, I will wait for DDR5 to mature before getting anything running on DDR5.
Not going from high-end DDR4 to entry level DDR5 thats for sure. Latency looks horrible. Can't wait to see more benches.
Probably won't upgrade before AMD has AM5 out and Intel releases chips on their 3 or 4 node anyway. By then, I will decide if it's going to be AMD or Intel. It's not that I lack CPU power anyway :D
40% faster CPU score in Ashes and 2000 more points in Cinebench R23.
And this is 8C/16T + 8 Efficiency Cores vs 16C/32T.
AMD better wake up again. However, they lost the node advantage. AMDs good run might be over now (time to focus on value/perf again + GPU market - Nvidia sits at 83% dGPU marketshare now and Intel dGPUs are incoming, aiming for AMDs biggest segment; Low to Mid-end)
I would not be surprised if AMD has less than 10% dGPU marketshare in 2023..
The images clearly show this is on a PCIe5 board. Its like apples and oranges, comparing PCie5 to PCIe4 that Threadripper is still on. Just not enough information to make an accurate comparison, same old Intel marketing spin BS.
30549 vs 30054 is too narrow of a margin, LOL a 3-year old cpu vs a brand spanking new engineers sample.
The big cores (Golden Cove) will support AVX-512.Yet another "useless" benchmark ;)
Let's wait for proper benchmarks of real world tests.