That is contradicted by the headline. This easy confusion between CUDA (the API) and CUDA (the proprietary software package that is one implementation of it) illustrates the problem with CUDA.
ZLUDA seems to be an effort to fix that problem, but I don’t know what it’s chances of success might be.
By funding an API-compatible product, they are giving CUDA legitimacy as a common API. I can absolutely understand AMD not wanting a competitors invention and walled-off product to be anything resembling an industry standard.
They stopped funding the replacement, not CUDA.
That is contradicted by the headline. This easy confusion between CUDA (the API) and CUDA (the proprietary software package that is one implementation of it) illustrates the problem with CUDA.
ZLUDA seems to be an effort to fix that problem, but I don’t know what it’s chances of success might be.
It’s just a bad headline. They funded a CUDA replacement, then stopped funding it, as a result of which the project was released as open source.
By funding an API-compatible product, they are giving CUDA legitimacy as a common API. I can absolutely understand AMD not wanting a competitors invention and walled-off product to be anything resembling an industry standard.
It already has legitimacy. It’s their hardware that doesn’t, despite the decent raw flops and high memory.