News for hackers. Stuff that matters.

A comment on Nvidia announces its most powerful AI chip as it seeks to become a platform co.. Parent comment

Re:Nvidia announces its most powerful AI chip as it seeks to... by wmf on Monday March 18, 2024 @11:39PM (#39752724)
It would be kind of genius for Nvidia to "open" the CUDA APIs (which have already been unofficially reverse engineered anyway) but not the code. Maybe they'd also officially support HIP and SYCL. Maybe they could open SXM after all competitors have already committed to OAM. They'd create the appearance of opening up while giving up very little.
[ Parent | on HN ]
Re:Nvidia announces its most powerful AI chip as it seeks to... by sitkack on Tuesday March 19, 2024 @01:58AM (#39753309)
By "Opening Up" they cement their leadership position. AI frameworks are already targeting CL, SPIR-V, etc. The low level details will fade and so will Nvidias api dominance.

The MI300 smokes the H100 yet here we are.

[ Parent | on HN ]
Re:Nvidia announces its most powerful AI chip as it seeks to... by incrudible on Tuesday March 19, 2024 @04:47AM (#39753911)
Just because they are a target doesn’t mean things just work. Historically, AMD hardware for GPGPU becomes obsolete well before the software landscape catches up. I am not going to risk my time and money finding out whether history repeats itself, just for a few potential FLOPS per dollar.
[ Parent | on HN ]
Re:Nvidia announces its most powerful AI chip as it seeks to... by sitkack on Tuesday March 19, 2024 @10:27AM (#39756063)
Don't disagree, but it is is nuts how AMD is leaving billions on the table by not finishing the project by writing the software.
[ Parent | on HN ]
Re:Nvidia announces its most powerful AI chip as it seeks to... by incrudible on Wednesday March 20, 2024 @01:51PM (#39769936)
Even if they finished it yesterday, it would take years to convince everyone that this time it will be worth investing in AMD, at which point the whole datacenter AI hype may already be over.
[ Parent | on HN ]
Re:Nvidia announces its most powerful AI chip as it seeks to... by sitkack on Thursday March 21, 2024 @12:34AM (#39774913)
The AI hype will not end. AMD just has to support transformers. OSS is already building it, at which point you can finetune an open weights model on an MI300. Nvidia is in the same place that Cisco and Sun were during the dot.com boom. People are just buying the mainstream think that works. It doesn't take years, it takes an accountant.
[ Parent | on HN ]