AMD Ryzen to leave Intel 12th, 13th, 14th Gen in the dust in benchmarks due to AVX-512

Intel debuted AVX-512 instructions in its mainstream Core lineup with the 11th Gen Rocket Lake processors. Later with 12th Gen (Alder Lake) and newer chips (13th Gen Raptor Lake) though, Intel adopted a hybrid CPU design approach called "Big-Bigger" and simultaneously dropped AVX-512. The reason behind this was probably the high power consumption AVX-512 brought on, even though there were clear performance benefits when it could be taken advantage of. However, Intel Core engineers likely didn't feel like the extra performance trade-off was worth it.

Rival AMD however, introduced AVX-512 much later, with Zen 4 or Ryzen 7000. And it looks like the move would be paying off for AMD soon. Popular synthetic CPU benchmarks like y-cruncher and Geekbench are expanding support for AVX-512 in their tests.

y-cruncher developer expresses disappointment in Intel's decision to remove the instructions as they feel there are several consequent drawbacks, adding that it is difficult to optimize for:

I've been asked a number of times about why I haven't done any optimizations for recent Intel processors. The latest Intel processor which y-cruncher has optimizations for is Tiger Lake which is 2 generations behind the latest (Raptor Lake). And because Raptor Lake lacks AVX512, it can only run a binary going all the way back to Skylake client (circa 2015).

[...]

Removing AVX512 is a huge step back in more ways than just the instruction width. It also removes all the other (non-width) functionality exclusive to AVX512 such as masking, all-to-all permutes, and increased register count. From a developer perspective, this very discouraging since most of the algorithms I've been working on since 2016 have been heavily influenced by (if not outright designed for) AVX512.

The lack of AVX512 is likely why Tiger Lake and Rocket Lake outperform Alder Lake in single-threaded benchmarks where memory bandwidth and core count are not a factor.

With its latest version 0.8.x, y-cruncher notes some significant performance boosts on Ryzen 7000 when computing Pi. It says AVX-512 can provide anywhere between 23-31% improvement compared to the previous version.

Meanwhile, Geekbench has now expanded support for AVX-512 with FP16 or float16 instructions. This means AMD Ryzen 7000 processors will in fact perform better than Intel 12th Gen and newer CPUs in certain Geekbench tests as well. Primate Labs, the makers of Geekbench says AVX512-FP16 will be included in "several image processing functions".

Aside from that, the new Geekbench 6.1 also addresses thermal throttling on Samsung Galaxy S23 and similar devices.

Edit: As reader DougQuaid pointed out, AMD's Zen 4 does not support AVX-512 - FP16. This has now been corrected. The following AVX-512 instructions are supported by Zen 4 according to y-cruncher's Alexander Yee:

• AVX512-F

• AVX512-CD

• AVX512-VL

• AVX512-BW

• AVX512-DQ

• AVX512-IFMA

• AVX512-VBMI

• AVX512-VNNI

• AVX512-BF16

• AVX512-VPOPCNTDQ

• AVX512-VBMI2

• AVX512-VPCLMULQDQ

• AVX512-BITALG

• AVX512-GFNI

• AVX512-VAES

   

AMD Ryzen to leave Intel 12th, 13th, 14th Gen in the dust in benchmarks due to AVX-512