[HN Gopher] Microbenchmarking Intel's Arc A770
       ___________________________________________________________________
        
       Microbenchmarking Intel's Arc A770
        
       Author : pantalaimon
       Score  : 85 points
       Date   : 2022-10-20 13:28 UTC (9 hours ago)
        
 (HTM) web link (chipsandcheese.com)
 (TXT) w3m dump (chipsandcheese.com)
        
       | timw4mail wrote:
       | I'm kind of disappointed all the tests were without ReBAR. I'm
       | curious as to what kind of difference that makes.
        
         | onli wrote:
         | You are probably aware, but ReBAR has a huge effect on these
         | card, so much that it is a requirement to get stable and
         | acceptable FPS in games. I would assume that means the effect
         | would be huge also in those smaller benchmarks.
         | 
         |  _edit:_ But was it really off in general, or only for the
         | PCI-e 3.0 part of the article?
        
           | wtallis wrote:
           | ReBAR only matters for transfers across the PCIe link, so for
           | most of the tests in the article it's completely irrelevant.
           | And even for the PCIe transfer tests, ReBAR should only
           | affect the portions of the tests that are accessing more GPU
           | memory than the default BAR size used for 32-bit
           | compatibility.
        
             | timw4mail wrote:
             | Considering how much the Arc series loves to share system
             | memory, I'm not so sure.
        
               | wtallis wrote:
               | Intel's having to find and fix every place their drivers
               | assumed that the CPU's RAM and the GPU's RAM were the
               | same pool of memory. As a result, it's completely
               | unsurprising that they're struggling with performance in
               | real applications and games, but it doesn't provide a
               | reason to expect microbenchmarks such as these to run
               | afoul of an irrelevant problem.
        
         | ErneX wrote:
         | These GPUs practically require ReBAR, performance without it
         | drops significantly.
        
         | clamchowder wrote:
         | Apologies for the confusion, the tests were run with ReBAR.
         | I've updated the article to reflect that
         | 
         | Shouldn't affect the conclusion for anything besides the PCIe
         | copy to/from GPU tests.
        
         | TomVDB wrote:
         | It would make zero difference, except, maybe, for the PCIe
         | bandwidth test. Because all other tests were stressed
         | interfaces that reside within the GPU silicon itself or between
         | the GPU and the DRAM.
         | 
         | Even the numbers of PCIe bandwidth test might not change much,
         | even it's trying to test medium size memory to memory block
         | transfers.
        
       | atlgator wrote:
       | I would love to get my hands on one. They seem to be sold out
       | everywhere.
        
         | MegaDeKay wrote:
         | If on Windows, you might want to check out the Gamers Nexus
         | reviews on YouTube: the Intel drivers are a dumpster fire and
         | build quality is... not great.
         | 
         | https://www.youtube.com/c/GamersNexus/videos
        
       | DRAGONERO wrote:
       | Is the benchmark suite available somewhere?
        
         | clamchowder wrote:
         | (Author here) See
         | https://github.com/clamchowder/Microbenchmarks/tree/master/G...
         | 
         | It's very much a work in progress, as noted in the article. And
         | some of the stuff that worked reasonably well on my cards, like
         | the instruction rate test when trying to measure throughput
         | across the entire card, went down the drain when run on Arc.
        
           | jra101 wrote:
           | Have you tried reducing the register count in your FP32 FMA
           | test by increasing the iteration count and reducing the
           | number of values computed per loop?
           | 
           | Instead of computing 8 independent values, compute one with
           | 8x more iterations:                   for (int i = 0; i <
           | count * 8; i++) {             v0 += acc * v0;          }
           | 
           | That plus inlining the iteration count so the compiler can
           | unroll the loop might help get closer to SOL.
        
             | clamchowder wrote:
             | The problem is loop overhead matters on AMD, because AMD's
             | compiler doesn't unroll the loop. Nvidia's does, so it
             | doesn't matter for them.
        
       ___________________________________________________________________
       (page generated 2022-10-20 23:01 UTC)