Huawei is sending samples of its new AI chip meant to rival Nvidia (www.scmp.com)
from yogthos@lemmy.ml to technology@lemmy.ml on 29 Sep 2024 19:51
https://lemmy.ml/post/20849010

archive.ph/Y7MVs

#technology

threaded - newest

gwilikers@lemmy.ml on 30 Sep 2024 07:52 next collapse

Am I missing something here? Did Nvidia leave a void?

msage@programming.dev on 30 Sep 2024 11:17 next collapse

Oh yes, companies knock themselves over trying to gobble up any AI chip in existence.

yogthos@lemmy.ml on 30 Sep 2024 12:08 collapse

The US put export restrictions on Nvidia in China, so they can only sell older chips there.

utopiah@lemmy.ml on 30 Sep 2024 12:32 collapse

As per usual, in order to understand what it means we need to see :

  • performance benchmark (A100 level? H100? B100? GB200 setups?)
  • energy consumption (A100 performance level and H100 lower watt? the other way around?)
  • networking scalability (how many cards cards can be interconnected for distributed compute? NVLink equivalents?)
  • software stack (e.g can it run CUDA and if not what alternatives can be used?)
  • yield (how many die are usable, i.e. can it be commercially viable or is it R&D still?)
  • price (which regardless of possible subsidies would come from yield)
  • volume (how many cards can actually be bought, also dependent on yield)

Still interesting to read after announcements, as per usual, and especially who will actually manufacture them at scale (SMIC? TSMC?).