alt Hacker NewsAMD's utter incompetence when it comes to the software side of things is truly, truly baffling to me. It's not like you need a mountain of developers, a team or two on the right project would do wonders for their market share.
For example: Implement the CUDA. CUDA's won, hands down, that toothpaste is solidly outside the tube. Luckily, to the outside observer CUDA is just an API, and API's aren't copyrightable. Literally nothing is stopping AMD from hiring a relatively small team of developers to make AMD GPUs CUDA-compatible.
Replies
likewise. i'm bewildered throughout the years.
my suspicion is that it is the company culture: the hardware engineers are the real engineers. software is a triviality left for the lesser minds. the consequence is they mess up every product... everything they do needs software.
HIP tries to be like this, almost API compatible with CUDA such that you just need to do find and replace. I think they even had a script to do this for you.
But the issue remains that the actual support and debugging tools remain so atrocious that it doesn't help to combat the CUDA monopoly. They've further burned a lot of trust by never really delivering on their promises to do better unless you're a customer large enough to get personalized attention from their engineers.
This ends up being a double whammy because not only are you pushing away smaller businesses, you're also pushing away single developers that go on to influence purchasing/development decisions.
AMD's entire software development strategy is insane. OpenCL was doing reasonably well, and then AMD have just fully dropped support for some reason. For the - albeit not huge - but actual cross platform API that people were using to develop for their GPUs. For a while, a few cross platform tools had OpenCL backends, but nobody has been able to get AMD to fix any of the damn bugs. In my testing, bug latency for even the most trivial but important bugfixes is often 4+ years, which is utterly mad. Some parts of their compute stack is so broken its clear that nobody has ever used it. There are exploitable privilege escalation vulnerabilities caused by threaded race conditions that are wontfix
They could support OpenCL 3.0. Nvidia do. AMD just chooses not to, even though they're the ones that desperately needs to support it most
Instead, we got ROCm which has been a disaster from start to end. It barely supports windows or consumer GPUs, for some reason. Its a buggy mess, for some reason. HIP/ROCm has worse performance than OpenCL, because they downgraded their compiler and stopped extracting read/write information on variables leading to a massive loss of parallelism and utilisation on their GPUs.. for some reason. Why? What are they doing? How is this so rubbish?
Literally ALL of this is WONTFIX, and I don't have a clue why. I've filed bugs, was part of their vanguard supporter program, have tried to reach out to AMD people to (gently) explain why good support is important. Or even just figure out what technology they're even intending to support for GPU development. Is ROCm deprecated? What should we be using on windows for GPU compute on consumer hardware AMD? For the love of god amd I want to make you money
As of 2026, the best cross platform cross vendor API for doing GPU compute is.. drumroll.. OpenCL 1.2. Vulkan is getting there, but its still missing a bunch of stuff. And this is literally AMDs direct fault at this point