Isn't the "goodness" of CUDA really down to its mass adoption -- and therefore its community and network effects -- not strictly its technical attributes?
If I recall, there are various "GPU programming" and "AI" efforts that have existed for AMD GPUs, but none of them have had the same success in large part because they're simply non-"standard?"
I don't think it's just adoption and network effects, though that is part of the equation. The other big (bigger?) piece is that the CUDA landscape is very complete, with libraries and examples for many different kinds of use cases, and they are well documented and easy to get started with. Ctrl+F this page for "ecosystem" and you'll find another comment that explains it better than I can.
I thought OpenCL was supposed to be the "standard"? From the Wikipedia page, it's largely vendor neutral and not that much younger than CUDA (initial release Aug 2009 vs Feb 2007). Maybe some more knowledgeable people can comment why it seems to have been outcompeted by the proprietary CUDA?
The question was why don't they have anything as good as CUDA, not why don't they adopt CUDA itself.
Isn't the "goodness" of CUDA really down to its mass adoption -- and therefore its community and network effects -- not strictly its technical attributes?
If I recall, there are various "GPU programming" and "AI" efforts that have existed for AMD GPUs, but none of them have had the same success in large part because they're simply non-"standard?"
I don't think it's just adoption and network effects, though that is part of the equation. The other big (bigger?) piece is that the CUDA landscape is very complete, with libraries and examples for many different kinds of use cases, and they are well documented and easy to get started with. Ctrl+F this page for "ecosystem" and you'll find another comment that explains it better than I can.
I thought OpenCL was supposed to be the "standard"? From the Wikipedia page, it's largely vendor neutral and not that much younger than CUDA (initial release Aug 2009 vs Feb 2007). Maybe some more knowledgeable people can comment why it seems to have been outcompeted by the proprietary CUDA?
2 replies →