[HN Gopher] AMD Launches Instinct MI300X AI Accelerator, Up to 6...
       ___________________________________________________________________
        
       AMD Launches Instinct MI300X AI Accelerator, Up to 60% Faster Than
       Nvidia H100
        
       Author : fariszr
       Score  : 32 points
       Date   : 2023-12-06 19:24 UTC (3 hours ago)
        
 (HTM) web link (wccftech.com)
 (TXT) w3m dump (wccftech.com)
        
       | filterfiber wrote:
       | I really hope AMD gains more ground on the software side.
       | 
       | I know they are "technically supported" by onyx/pytorch etc. But
       | almost every project I've seen there's a bunch of asterisks, a
       | bunch of people needing different work arounds, and subpar
       | performance compared to the hardware.
       | 
       | Maybe it's because most people have their consumer cards? Nvidia
       | however doesn't have this issue, people on pascal are only just
       | now having compatibility issues with not supporting a high enough
       | cuda version.
       | 
       | The dream is they release a 4090-type competitor for a few grand,
       | it doesn't even need full 4090-level performance, just strap
       | 48gb+ of HBM and it would be a absolutely incredible deal for
       | individuals learning/researching -> who then use the full
       | enterprise stuff for full training/commercial inference.
        
         | rreichman wrote:
         | Their software is definitely improving and will get a whole lot
         | better in the next few years. It's worth many billions for
         | companies to exit the Nvidia lock-in.
        
           | fariszr wrote:
           | As long, as it doesn't support normal consumer hardware, AMD
           | won't get the support Nvidia GPUs enjoy from community
           | projects.
        
         | sbininit wrote:
         | I suspect that those who would otherwise be buying thousands of
         | H100s would simply put in the dev work to make the software
         | function well on AMD hardware
        
         | idonotknowwhy wrote:
         | I'm hoping for the same thing, but with Intel ARC. Intel gpus
         | are cheaper, but have even worse ML software for consumer
         | projects
        
       | rreichman wrote:
       | According to this and SemiAnalysis[1], the MI300X seems
       | impressive. Better than Nvidia's H100 and maybe a bit inferior to
       | the H200. Why is Nvidia worth $1 trillion more?
       | 
       | [1] - https://www.semianalysis.com/p/amd-mi300-performance-
       | faster-...
        
         | mattnewton wrote:
         | CUDA
        
           | rreichman wrote:
           | CUDA is becoming less important over time
        
         | tedunangst wrote:
         | $9B in profit vs $300M will have that effect.
        
           | rreichman wrote:
           | Sure but the future is not the past
        
             | riku_iki wrote:
             | AMD p/e is 900 vs 60 for NVDA, so investors already heavily
             | invested into AMD future.
        
       | Racing0461 wrote:
       | We need competition in the ai gpu space.
        
       ___________________________________________________________________
       (page generated 2023-12-06 23:01 UTC)