zlacker

[return to "AMD funded a drop-in CUDA implementation built on ROCm: It's now open-source"]
1. hd4+bg[view] [source] 2024-02-12 15:24:21
>>mfigui+(OP)
The interest in this thread tells me there are a lot of people who are not cool with the CUDA monopoly.
◧◩
2. smolde+fr[view] [source] 2024-02-12 16:10:11
>>hd4+bg
Those people should have spoken up when their hardware manufacturers abandoned OpenCL. The industry set itself 5-10 years behind by ignoring open GPGPU compute drivers while Nvidia slowly built their empire. Just look at how long it's taken to re-impliment a fraction of the CUDA featureset on a small handful of hardware.

CUDA shouldn't exist. We should have hardware manufacturers working together, using common APIs and standardizing instead of going for the throat. The further platforms drift apart, the more valuable Nvidia's vertical integration becomes.

◧◩◪
3. mnau+Et[view] [source] 2024-02-12 16:19:42
>>smolde+fr
Common API means being replaceable, fungible. There are no margins in that.
[go to top]