Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

Documentation for the AMD 7900XTX

github.com

Post reply on HN
Print view View original

71–80 of 154 posts

Re: Documentation for the AMD 7900XTX

#71

Post by bee_rider »

DiabloD3 wrote: post #40
ultra_nick wrote: post #31

Earlier quoted context omitted.

NVidia is making boatloads of money because their driver works and they have a software library called CUDA that accelerates neural networks. Nobody expects AMD to match them, but George thought they could at least write a GPU driver. If AMD can get that driver out, then George could provide a competitor to CUDA (for neutral networks only). They'd both make boatloads of money. However, AMD was less capable than expec…

Slight mistake in your description: CUDA is an out of date API that was replaced by Khronos's own official compute APIs. Khronos is a standards consortium that Nvidia is a founding member of. Although the marketing department at Nvidia still pushes for greenfield CUDA codebases, no new code should be written in it, and they should opt for open source international standards only. Khronos APIs are implemented by over…

That’s not a mistake, you just have a different preference.

Re: Documentation for the AMD 7900XTX

#72

Post by jiggawatts »

ein0p wrote: post #61
threeseed wrote: post #58

Earlier quoted context omitted.

> have him fix it to his liking Like when he strolled into Twitter, acted like he knew better than everyone, delivered nothing and then left. There is far more to achieving things in large companies than simply having great technical skills.

IDK, dude’s provably pretty capable as long as you don’t attempt to drown him in a bureaucratic swamp. Betcha he’d fix it pretty good.

> don’t attempt to drown him in a bureaucratic swamp

You weren't listening to the GP comment.

He's expected to succeed while being drowned in the swamp.

That's what most people mean and expect when they bring in an outsider.

"Fix our problems, but without changing anything to do with management."

Re: Documentation for the AMD 7900XTX

#73

Post by fragmede »

jiggawatts wrote: post #72
ein0p wrote: post #61

Earlier quoted context omitted.

IDK, dude’s provably pretty capable as long as you don’t attempt to drown him in a bureaucratic swamp. Betcha he’d fix it pretty good.

> don’t attempt to drown him in a bureaucratic swamp You weren't listening to the GP comment. He's expected to succeed while being drowned in the swamp . That's what most people mean and expect when they bring in an outsider. "Fix our problems, but without changing anything to do with management."

There's a difference between fixing technical problems and cultural problems. Hotz has proven to be a wizard at technical ones, while his skill with fixing cultural issues remains to be seen.

Re: Documentation for the AMD 7900XTX

#74

Post by rstat1 »

trws wrote: post #49
roenxi wrote: post #8

So for reference, this geohot is George Hotz who has a company Tiny Corp [0] in the space. Among a long list of things, he had a moment in the spotlight recently for a long rant [1] where he "gave up on AMD" because their drivers sucked more than he expected. The situation is reasonably complex - AMD have some programmers working on ROCm who seemed to be operating at standard fare when that is really not what AMD nee…

They do need to rework their development processes. It’s worth noting though that there are two (kinda three) different drivers with very different quality and different teams behind them. There’s the binary driver, on windows and the fglrx driver for Linux, that contains the full OpenGL stack and all the proprietary stuff. That’s been maintained and extended since the ATI days by a team primarily based in Markham. I…

I believe your info is a bit out-of-date.

FGLRX hasn't been maintained or actively developed for a quite a while and unless you're using super-duper-ancient HW no one actually uses it anymore.

AMDGPU is the the current Linux driver, combined with Mesa, and certainly is not 2D only.

Re: Documentation for the AMD 7900XTX

#75

Post by UberFly »

quantumwoke wrote: post #63
lannisterstark wrote: post #53

Earlier quoted context omitted.

>I'm cheering him on, even though his complaints were a little melodramatic. I've really soured on him when he literally went "I'm going to fix twitter search, hire me elon" and then went to public saying "Hey, want an internship? Fix it, and then MIT license your code(so he could use it lol), and maybe we can talk - oh btw I have no real authority to give anyone an internship." He might be a brilliant programmer and…

I soured on him a bit more when I watched some of his stream yesterday and he was ranting about diversity being a root of evil in Silicon Valley. Struck me that he hasn’t matured much since he was slinging mud at Sony and Apple.

He seems to be the type who doesn't care who's "soured on him".

Re: Documentation for the AMD 7900XTX

#76

Post by pxc »

trws wrote: post #49
roenxi wrote: post #8

So for reference, this geohot is George Hotz who has a company Tiny Corp [0] in the space. Among a long list of things, he had a moment in the spotlight recently for a long rant [1] where he "gave up on AMD" because their drivers sucked more than he expected. The situation is reasonably complex - AMD have some programmers working on ROCm who seemed to be operating at standard fare when that is really not what AMD nee…

They do need to rework their development processes. It’s worth noting though that there are two (kinda three) different drivers with very different quality and different teams behind them. There’s the binary driver, on windows and the fglrx driver for Linux, that contains the full OpenGL stack and all the proprietary stuff. That’s been maintained and extended since the ATI days by a team primarily based in Markham. I…

> The amdgpu driver is 2d only

Huh? People play 3D games on that driver. Or is the 3D part just Mesa in such cases?

Re: Documentation for the AMD 7900XTX

#77

Post by infogulch »

hughesjj wrote: post #66
lannisterstark wrote: post #53

Earlier quoted context omitted.

>I'm cheering him on, even though his complaints were a little melodramatic. I've really soured on him when he literally went "I'm going to fix twitter search, hire me elon" and then went to public saying "Hey, want an internship? Fix it, and then MIT license your code(so he could use it lol), and maybe we can talk - oh btw I have no real authority to give anyone an internship." He might be a brilliant programmer and…

Yeah, I was feeling similar vibes back when Lex Friedman interviewed him for some self driving stuff back in the day (...I've soured on lex too since he's not really an interviewer/journalist so much as a host/promoter and rarely gets to anything interesting in his talks imo). His work for jailbreaking/actually being able to own the devices we pay for back in the iPhone+PS3 days was and is still inspiring, but I thin…

This is a great take. We need more people that are willing and able to jump in and wreck new bullshit machines as they crop up in diverse industries. They can even keep their boastful grandstanding attitude as compensation.

Re: Documentation for the AMD 7900XTX

#78

Post by pezezin »

rstat1 wrote: post #74
trws wrote: post #49

Earlier quoted context omitted.

They do need to rework their development processes. It’s worth noting though that there are two (kinda three) different drivers with very different quality and different teams behind them. There’s the binary driver, on windows and the fglrx driver for Linux, that contains the full OpenGL stack and all the proprietary stuff. That’s been maintained and extended since the ATI days by a team primarily based in Markham. I…

I believe your info is a bit out-of-date. FGLRX hasn't been maintained or actively developed for a quite a while and unless you're using super-duper-ancient HW no one actually uses it anymore. AMDGPU is the the current Linux driver, combined with Mesa, and certainly is not 2D only.

Yup, FGLRX was deprecated from all major distros in 2016. Nobody has used it for ages.

Re: Documentation for the AMD 7900XTX

#79

Post by baq »

DiabloD3 wrote: post #40
ultra_nick wrote: post #31

Earlier quoted context omitted.

NVidia is making boatloads of money because their driver works and they have a software library called CUDA that accelerates neural networks. Nobody expects AMD to match them, but George thought they could at least write a GPU driver. If AMD can get that driver out, then George could provide a competitor to CUDA (for neutral networks only). They'd both make boatloads of money. However, AMD was less capable than expec…

Slight mistake in your description: CUDA is an out of date API that was replaced by Khronos's own official compute APIs. Khronos is a standards consortium that Nvidia is a founding member of. Although the marketing department at Nvidia still pushes for greenfield CUDA codebases, no new code should be written in it, and they should opt for open source international standards only. Khronos APIs are implemented by over…

CUDA is a vendor lock-in mechanism and they’ll happily endorse whatever Khronos says is standard… after they lose dominance.

It took 30 years for x86 and just barely so.

Re: Documentation for the AMD 7900XTX

#80

Post by KeplerBoy »

DiabloD3 wrote: post #40
ultra_nick wrote: post #31

Earlier quoted context omitted.

NVidia is making boatloads of money because their driver works and they have a software library called CUDA that accelerates neural networks. Nobody expects AMD to match them, but George thought they could at least write a GPU driver. If AMD can get that driver out, then George could provide a competitor to CUDA (for neutral networks only). They'd both make boatloads of money. However, AMD was less capable than expec…

Slight mistake in your description: CUDA is an out of date API that was replaced by Khronos's own official compute APIs. Khronos is a standards consortium that Nvidia is a founding member of. Although the marketing department at Nvidia still pushes for greenfield CUDA codebases, no new code should be written in it, and they should opt for open source international standards only. Khronos APIs are implemented by over…

Why do I have the feeling that CUDA will outlive whatever Khronos has proposed (do you refer to Sycl)?
Post reply on HN