Intel Sapphire Rapids CPU with 20 cores and 40 threads shows up on Geekbench

Published: Jul 6th 2021, 07:01 GMT   Comments

Intel Sapphire Rapids with 20 cores tested

Despite the delay, the engineering samples of Sapphire Rapids server CPUs are already out in the wild. 

The processor that has been tested features 20 cores and 40 threads, in fact, there are two such CPUs running in a dual socket system. It is reportedly running at a 1.5 GHz base frequency but according to the JSON output from the Geekbench reported clock speed is 4.68-4.70 GHz. Whether that is a real clock speed or an error from Geekbench, we do not know, but it is worth noting that Sapphire Rapids will also launch as HEDT consumer CPU, so such frequencies are to be expected at some point:

Geekbench does not explicitly confirm these are Sapphire Rapids CPUs, however, the ‘identifier reveals the CPU family, which is Family 6 Model 143 Stepping 2. This model specifically is known as Sapphire Rapids.

The processor has been tested on the Intel reference platform codenamed VulcanCity. The system was equipped with 32GB of memory (presumably DDR5), but the Geekbench result does not confirm the frequency. More interestingly though, the CPU caches have been listed in this leak. The CPU has 75MB of L3 cache and 40MB of L2 cache.

Intel sapphire Rapids Specifications, Source: Geekbench

In its current state, this 20-core Sapphire Rapids CPU running in a dual-socket system is not very fast. It is easily outperformed by Ryzen 9 5950X which offers twice as high multi-core performance. The single-core performance is even worse, it is outclassed by most of the CPUs released in the last 5 years, including mobile CPUs. That said, the CPU is clearly not running at its full speed.

Intel sapphire Rapids Performance, Source: Geekbench

Intel recently announced that its Sapphire Rapids CPUs will enter volume production in the second quarter of next year. The manufacturer is also developing a version with HBM2e memory, which is expected to launch by the end of next year.

INTEL Xeon Scalable
VideoCardz2nd Gen Xeon3rd Gen Xeon4th Gen Xeon
ArchitectureCooper Lake Ice LakeSapphire Rapids
Release Year202020202022
Max Cores284056
Max TDP250W270W350W
Memory SupportDDR4DDR4DDR5
Max Mem. Channels688
Max Mem. Speed3200 (2DPC)3200 (2DPC)4800 (1DPC)
HBM Supportnonoup to 64GB HBM2e
Intel UPI linkUPI 1.0UPI 1.0UPI 2.0
UPI speeds10.4 GT/s11.2 GT/s16 GT/s
PCI ExpressPCIe 3.0, 40 lanesPCIe 4.0, 64 lanesPCIe 5/4, 80 lanes
Deep Learning InstructionAVX-512 (VNNI/INT8 & Bfloat16)AVX-512 (VNNI/INT8)AMX/TMUL (INT & BFloat16)
AVX-512 (VNNI/INT8 & Bfloat16)
Optane SupportBarlow PassBarlow passCrow pass
Compute Express LinknonoCLX 1.1 (4 x16 devices)
SKUsPlatinum, GoldAll except BronzeAll

Source: Geekbench via @BenchLeaks




Comment Policy
  1. Comments must be written in English and should not exceed 1000 characters.
  2. Comments deemed to be spam or solely promotional in nature will be deleted. Including a link to relevant content is permitted, but comments should be relevant to the post topic. Discussions about politics are not allowed on this website.
  3. Comments and usernames containing language or concepts that could be deemed offensive will be deleted.
  4. Comments complaining about the post subject or its source will be removed.
  5. A failure to comply with these rules will result in a warning and, in extreme cases, a ban. In addition, please note that comments that attack or harass an individual directly will result in a ban without warning.
  6. VideoCardz has never been sponsored by AMD, Intel, or NVIDIA. Users claiming otherwise will be banned.
  7. VideoCardz Moderating Team reserves the right to edit or delete any comments submitted to the site without notice.
  8. If you have any questions about the commenting policy, please let us know through the Contact Page.
Hide Comment Policy
Comments