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

Dissecting the Apple M1 GPU, the end

rosenzweig.io

Post reply on HN
Print view View original

231–240 of 243 posts

Re: Dissecting the Apple M1 GPU, the end

#231

Post by Incipient »

MangoToupe wrote: post #219
tlamponi wrote: post #147

Earlier quoted context omitted.

There was a time when people said that about AMD. Don't get me wrong, Intel's outlook is IMO currently indeed rather bleak, but I would not completely write it off just yet.

AMD is equally fucked. Building off of IP-locked architectures is just a graveyard. Even apple will hit a wall one day.

There are a myriad of companies that have thrived in "IP locked" environments, a host that have failed too. Equally there are heaps that have thrived and failed in "IP open" environments.

I think at best you could say it's more challenging or perhaps risky being a bit restricted with IP, but I'd call it miles away from a "graveyard".

You can hardly call Intel/amd/qualcomm etc all struggling due to the architectures being locked down.

Look at powerpc/Isa. It's (entirely?) open and hasn't really done any better than x86.

Fundamentally you're going to be tied to backwards compatibility to some extent. You're limited to evolution, not revolution. And I don't think x86 had failed to evolve? (eg avx10 is very new)

Re: Dissecting the Apple M1 GPU, the end

#232

Post by abnercoimbre »

mrcwinn wrote: post #41

Huge respect for her. I consider myself a talented software engineer. I can’t get anywhere close to this accomplishment.

Genuine question: is she a once-in-a-generation prodigy? We forget this class of people indeed exists. As fellow professionals are we inspired or deeply ashamed of ourselves?

Re: Dissecting the Apple M1 GPU, the end

#235

Post by amiga386 »

ActorNightly wrote: post #10

Im glad she stepped away from Asahi linux. Its absolutely great from a techincal perspective and the progress that team has made, but talented people like her shouldn't be trying to reverse engineer software/hardware from shitty anti-consumer company that can make the entire project work in a heartbeat by publishing documentation, in lieu of building better stuff from the ground up.

Ultimately we didn't need Jon Lech Johansen's work (and Derek Fawcus's, and others') on cracking DVD DRM or cracking Apple's FairPlay DRM, as there have always been alternatives. But their efforts did push the fight against DRM in our favour, and who knows what the world would be like if we had done nothing?

Creating things is a gamble, as mass adoption is almost never by technical merits, but by marketing. So you could make open documented everything but still end up with nobody benefiting from that openness, because a competitor (whether open or not) wipes you out. You saw this happen even in the era where electronic devices were expected to come with full schematics -- there were winners and losers even then.

But, if something has become widespread and well adopted, and it's not open, that's a problem. It absolutely should be opened up and documented. Especially if it's not because the money-grubbing creators of the something are deliberately hiding how it works and locking down control in order to extract more money from everyone else's pockets. The sooner you put an end to that, and the more often you fight against that, the sooner society itself becomes more efficient and fairer for everyone.

Re: Dissecting the Apple M1 GPU, the end

#237

Post by monocasa »

pjmlp wrote: post #165
monocasa wrote: post #38

Earlier quoted context omitted.

Good luck to her. That's one of the pieces of Intel I think will survive its slow motion implosion.

Hardly, Phoronix has a few reports from Linux driver folks being layed off at Intel.

Everywhere in Intel will be subject to layoffs.

My point was that the graphics division itself will still be around, as integrated mobile SoCs are basically the only revenue stream Intel still has a good handle on. That requires a graphics core, and all of the other usable options are either not for sale to Intel, have burned Intel in the past, or are owned by Arm.

Re: Dissecting the Apple M1 GPU, the end

#238

Post by nemomarx »

6244859432 wrote: post #234
berbec wrote: post #233

Earlier quoted context omitted.

Just random internet bigotry - Alyssa Anne Rosenzweig is transgender. [1] 1: https://web.archive.org/web/20250520182445/https://rosenzwei...

[flagged]

When someone tells you they prefer going by a nickname, do you also complain that they're forcing you to ignore the evidence on their paperwork and refuse?

Re: Dissecting the Apple M1 GPU, the end

#239

Post by AlexeyBelov »

6244859432 wrote: post #234
berbec wrote: post #233

Earlier quoted context omitted.

Just random internet bigotry - Alyssa Anne Rosenzweig is transgender. [1] 1: https://web.archive.org/web/20250520182445/https://rosenzwei...

[flagged]

The evidence of my eyes is two anonymous troll accounts that were just created for this thread. HN mods should really do something about it, it's way too easy to sign up and ban evade.

Re: Dissecting the Apple M1 GPU, the end

#240

Post by contrarian1234 »

ac29 wrote: post #51
contrarian1234 wrote: post #46

I never understood this project. Maybe I'm missing something, but the timescale is such that by the time they're done the product isn't even being sold anymore At least with Panfrost it made more sense bc it still being used M1 chip laptops can only be bought second hand at this point

> M1 chip laptops can only be bought second hand at this point New M1 Macbook Airs are still available at Walmart (maybe elsewhere). But even if not, who cares? People are still writing code for computers that haven't been sold since the 1980s.

Yeah, I get if it's just a for-fun kind of thing. I didn't mean to knock it. I guess generally people want their work to end up being used. The time-line is such that that seems semi-impossible.

As the other replies show, you can still buy this machine, but it sounds like it likely won't be for too much longer

The developers involved must be acutely aware of it. Maybe they have some sense that the work will easily update to current M chips. Or maybe they don't really care about that. It was just an interesting exercise and they move on

Post reply on HN