Hacker News Reader

Comment by georgehotz | Hacker News Reader

I have had multiple replies from Lisa Su on both Twitter and by e-mail. It doesn't help.

AMD is structurally incapable of fixing these issues. They don't even have a 7900XTX in CI. Crashing the firmware is so trivial there's no way anyone fuzzed anything.

They debug by the application, adding mitigations at many layers of the stack to make it work. While this strategy is fine for the 20 mainstream games that come out per year, unless you root cause issues and have a good CI, you'll never build a stable GPU for general compute.

View on HN

They don't even have a 7900XTX in CI.

You read stories in the glory days of Microsoft where they would bend over backwards to ensure backwards compatibility [0]. Going from that to what you say is depressing. It’s actually hard for me to accept your statement because such a failure would not only require QA/the PM to be totally ignorant of current testing best practices, but also require HR to have catastrophically misidentified both those people and the people who supervise them.

Keep up the great work.

[0]https://devblogs.microsoft.com/oldnewthing/20191119-00/?p=10...

The shocking thing about AMD in the last 5 years isn't that they got Ryzen to work, it's that they didn't screw it up within two releases. There are no other technology companies in the world who have a will to fail so ingrained in their corporate DNA as does AMD.

Yikes. For context, the 7900XTX and 7900XT are the only consumer GPUs that AMD officially support running AI workloads on, and the XTX is the one they recommend out of the two. Other cards in the same generation and any from previous generations are officially not supported or tested and may not work. So it's not like they have a particularly large amount of hardware that needs to be included in their testing environment.

Genuine question for you then, why are you investing your time reversing their firmware if it sounds like you think they're failing/will fail? Why not just "adopt" the golden child/golden standard NVIDIA now for your usecase, reverse that (if needed) and move on? Genuinely curious what you gain from trying to make AMD work/investing in AMD on your end here. Is it that you see a value proposition through the "muck" of crashy-kernels that I'm missing?

This is my weekend hobby, I'm learning about GPUs. Besides, I'd like to see competition to NVIDIA.

From a business side, tiny corp is selling boxes with both AMD and NVIDIA, we'll let the user choose how much they value money vs pain.

What are your thoughts on using Intel GPUs for deep learning?

Did you ever consider using NVIDIA? I thought the machine you were planning to sell originally had RTX 6000s.

Also, isn’t the real problem with the 7900 XTX is the lack of interconnect? I really appreciate the long list of stuff you are agitating AMD about. Infinity Fabric and blower fans would be great too. Another POV is none of that is ever going to happen, and everyone will have to wait for a Chinese GPU manufacturer to create real competition. I can’t recall a single time in my life a non-Chinese company - electronics or otherwise - delivered the “same” features for a lower price.

Thoughts on IREE[0]'s ability to close this gap?

[0] iree.dev

> Why not just "adopt" the golden child/golden standard NVIDIA now for your usecase, reverse that (if needed) and move on?

Oh oh, I will answer you with the business side of the thing: because Nvidia sells exactly the same thing as Tinycorp. Also, Nvidia has shown that they will crush you if they get a whiff that you are doing something they don't like, for example, speaking with a competitor. With a business partner that gets you by the balls like that, it is not sound business strategy not to depend on them.

Maybe they are not allowed to compete...

Reply on news.ycombinator.com