zlacker

[parent] [thread] 6 comments
1. ianlev+(OP)[view] [source] 2024-02-12 17:52:49
And indeed more than succeed, they invented x86_64.
replies(2): >>stcred+Y2 >>sangno+X6
2. stcred+Y2[view] [source] 2024-02-12 18:07:43
>>ianlev+(OP)
And indeed more than succeed, they invented x86_64.

If AMD invented the analogous to x86_64 for CUDA, this would increase competition and progress in AI by some huge fraction.

replies(2): >>pjmlp+Md >>jvande+lZ
3. sangno+X6[view] [source] 2024-02-12 18:27:00
>>ianlev+(OP)
x86_64's win was helped by Intel's Itanium misstep. AMD can't bank on Nvidia making a mistake, and Nvidia seems content with incremental changes to CUDA, contrasted with Intel's 32-bit to 64-bit transition. It is highly unlikely that AMD can find and exploit a similar chink in the amor against CUDA.
replies(1): >>LamaOf+uc
◧◩
4. LamaOf+uc[view] [source] [discussion] 2024-02-12 18:55:45
>>sangno+X6
If they're content with incremental changes to CUDA then it doesn't cost much to keep updated compatibility and do it as quickly as any users actually adopt changes.
◧◩
5. pjmlp+Md[view] [source] [discussion] 2024-02-12 19:03:55
>>stcred+Y2
Only works if NVidia misteps and creates the Itanium version of CUDA.
replies(1): >>stcred+Vf
◧◩◪
6. stcred+Vf[view] [source] [discussion] 2024-02-12 19:15:25
>>pjmlp+Md
You don't think someone would welcome the option to have more hardware buying options, even if the "Itanium version" didn't happen?
◧◩
7. jvande+lZ[view] [source] [discussion] 2024-02-12 23:05:21
>>stcred+Y2
Why "if only". Intel had been around forever when AMD showed up. CUDA isn't unassailable
[go to top]