Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

CUDA Moat Still Alive

semianalysis.com

Post reply on HN
Print view View original

171–176 of 176 posts

Re: CUDA Moat Still Alive

#171

Post by isthatafact »

Lerc wrote: post #35
Archit3ch wrote: post #27

Earlier quoted context omitted.

Coming up next: "We bought AMD stock on the open market and used it to compensate AMD engineers".

You joke, but it is almost a genuine investment opportunity here for a large player. Spend a billion on AMD shares, Spend another Billion on a out-of-house software team to solve the software solution to more than double the share price. Taking into account that there are players that already own billions in AMD shares, they could probably do that as well. On the other hand perhaps it would be better for them, as maj…

It does seem like a good idea. If the obvious major Nvidia customers see it as too risky (or we could speculate about other possible reasons why it has not happened yet), maybe some hedge fund who is struggling with where to put their cash could initiate and fund the project.

Re: CUDA Moat Still Alive

#172

Post by sangnoir »

mike_hearn wrote: post #166
sangnoir wrote: post #157

Earlier quoted context omitted.

> It's in the article The same article also states that AMD provided custom bug-fixes written by Principle Engineers to address bugs in a benchmark - this is software that will only become part of the public release in 2 quarters. I ask again, do you think AMD will not expedite non-public bug-fixes for hyperscalers? > You can't train with AMD, full stop, because their software stack is so buggy. Point 7 from the arti…

AI labs don't want to train models using a stack build some guy hacked up on his desktop last night that's been through no proper QA process. The cost of a training run that fails or results in a garbage model due to numerical errors are huge. Which is why, as they say clearly, nobody is training models on AMD. Only inference, at most. I'm not sure why you keep claiming they are training using private drivers. They c…

Now I i see how where talking past each other. I 100% agree that none of if the hyperscalers are currently (publicly) training on AMD silicon. I disagree with forward-looking statements like it "can't" happen, because I can guarantee you several of them are actively working on making it possible to train on AMD chips - that's just too juicy a target for a bonus packets all the way up to directors: "Our team saved the org $x0 million in TCO vy enabling training on MI300/MI400X in our new clusters"

Re: CUDA Moat Still Alive

#174

Post by mike_hearn »

sangnoir wrote: post #172
mike_hearn wrote: post #166

Earlier quoted context omitted.

AI labs don't want to train models using a stack build some guy hacked up on his desktop last night that's been through no proper QA process. The cost of a training run that fails or results in a garbage model due to numerical errors are huge. Which is why, as they say clearly, nobody is training models on AMD. Only inference, at most. I'm not sure why you keep claiming they are training using private drivers. They c…

Now I i see how where talking past each other. I 100% agree that none of if the hyperscalers are currently (publicly) training on AMD silicon. I disagree with forward-looking statements like it "can't" happen, because I can guarantee you several of them are actively working on making it possible to train on AMD chips - that's just too juicy a target for a bonus packets all the way up to directors: "Our team saved the…

Oh, I see. "Can't" means in the present tense in my previous sentence, it wasn't meant to be a definitive statement about the entire future.

Re: CUDA Moat Still Alive

#176

Post by eapriv »

ryao wrote: post #131
eapriv wrote: post #98

Earlier quoted context omitted.

> FFT based matrix multiplication, which is O((n^2)log(n)) What?

https://en.wikipedia.org/wiki/Schönhage–Strassen_algorithm I forgot the log(log(n)) factor. In any case, for matrix multiplications that people actually do, this algorithm runs slower than a well optimized O(n^3) matrix multiplication implementation because the constant factor in the Big O notation is orders of magnitude larger.

Schönhage-Strassen is not about matrix multiplication.
Post reply on HN