zlacker

[parent] [thread] 5 comments
1. RamRod+(OP)[view] [source] 2024-02-12 15:44:37
> and would also act as a disincentive for NVIDIA to make breaking changes to CUDA

I don't know about that. You could kinda argue the opposite. "We improved CUDA. Oh it stopped working for you on AMD hardware? Too bad. Buy Nvidia next time"

replies(3): >>mnau+r7 >>freeon+8b >>outsid+By
◧
2. mnau+r7[view] [source] 2024-02-12 16:14:58
>>RamRod+(OP)
Also known as OS/2: Redux strategy.
◧
3. freeon+8b[view] [source] 2024-02-12 16:30:28
>>RamRod+(OP)
Most CUDA applications do not target the newest CUDA version! Despite 12.1 being out, lots of code still targets 7 or 8 to support old NVIDIA cards. Similar support for AMD isn’t unthinkable (but a rewrite to rocm would be).
replies(1): >>lambda+7t1
◧
4. outsid+By[view] [source] 2024-02-12 18:15:48
>>RamRod+(OP)
NVIDIA is about ecosystem plays, they have no interest in sabotage or anti competition plays. Leave that to apple and google and their dumb app stores and mobile OSs.
replies(1): >>0x457+XE
◧◩
5. 0x457+XE[view] [source] [discussion] 2024-02-12 18:46:19
>>outsid+By
> NVIDIA is about ecosystem plays, they have no interest in sabotage or anti competition plays.

Are we talking about the same NVIDIA? The entire Nvidia GPU strategy for nvidia is - make a feature (or find existing one) that performs better on their cards - pay developers to use (and sometimes misuse) it extensively.

◧◩
6. lambda+7t1[view] [source] [discussion] 2024-02-12 23:03:56
>>freeon+8b
That's exactly the point I was making above.
[go to top]