Asahi Linux Update Brings Experimental Apple M2 Support

[qarium]

Please write a novel, I love your style.

all you get with that style is censorship

[ddriver]

So having been proven wrong with your first, unfounded allegation, you immediately pivot to a new, unfounded allegation?

Words to lookup in the dictionary: "proven, unfounded, allegation", since you are clearly not using the properly.

My "unfounded allegation" is stating the fact that apple is in a position to make this as hard as they want to. They control the entire product chain, from chip design through device design all the way to the operating system and all vets software that it runs, and they also happen to be a trillion + dollar company with immense legal capacity, if they don't want a 3rd party os running on their hardware, there are scores of ways to achieve that in an instant.

Just because they haven't opted to block that for now doesn't mean they are not in the position to do so. In fact, they may not even opt to do that until the end of time, that doesn't mean that they are not in the position to do that, which is what my "allegation" was. Naive wishful thinking otherwise, absent any factual substantiation does in no way make facts into allegations, much less proving them false. Here's another word for you to lookup - hypocrisy, and next time you insinuate someone is making unfounded allegations, make sure your own "argument" doesn't consist entirely of precisely that. Projecting your flaws and poor practices onto others is not a road toward personal development... but still, feel free to take that as unfounded allegation as well.

There is plenty of precedent in favor of my view. They are the only vendor to strictly prohibit installing software from your own source. Most of their update releases plug holes that facilitate jailbreaking. The company has done a lot, and spent mountains of money to precisely obstruct their devices from running any software they haven't vetted. They make products riddled with "planned obsolescence", they make repairs as close to impossible as they can, they have a long long list of anti-user behavior and nothing to even suggest the opposite. This is not unexpected, atypical or illogical behavior.

If there's any unfounded and I dare say outright ridiculously false allegation here, it is that the currently existing possibility to boot "something else" exists to the common good of the end user. Apple has never done anything like that ever, and currently, I wouldn't say there are any factors that mandate this to change. Apple has nothing to compete with really, is not moving in wintel direction, wintel is and has been moving in ios direction. They are gonna close up windows and x86 for the simple reason they are salivating at the profits that walled garden ecosystems are generating. Yet those who appear to live under the rocks seem to not have noticed that transition.

I did not elaborate why apple is most likely to obstruct that for the simple reason I presumed that history so far has provided ample evidence those are not typical practices for that company. But since some geniuses appear to need proof that water is wet now, I can obviously provide factually substantiated argumentation for my "unfounded allegation". Do you have any argumentation in favor of your unfounded allegation about my "unfunded allegation"?

[Dukenukemx]

Is there any laptop vendor with similar performance per watt ? I want to try AMD/Intel laptop, but their battery life sucks after some months of usage and overheat a lot even with fan, most good Intel/AMD laptop are quite bulky too. M-series MacBook are one of the best Laptops I have ever seen, only thing that sucks is OSX which Asahi Linux resolves.

If you push the CPU on a M based laptop you will consume the same power as an AMD or Intel. Like for example compiling the Linux kernel. It's even worse when it comes to gaming as the GPU isn't as good as AMD's. The power efficiency of the Apple M chips comes from their fixed function hardware for video editing, and their single threaded power consumption is really low. Basically the M chips are great for video editing and watching videos along with web browsing.

[Dukenukemx]

They can put in as many cores as they want when they redo the floorplan of the chip. They already have a fast enough fabric. It'd be indeed difficult to find a RAM interface as wide as the one they currently have (it already outperforms server ram) but I'm sure they could put some more ram chips on an interposer or something. Maybe borrow IBM's OMI interconnect.

Actually the M series chips are built cheap, as they have unified ram. Sure it has a lot of bandwidth and low latency, but that's at the cost of not being able to upgrade the ram. The ram sits right next to the SoC, which gives it those benefits. Putting ram slots with the M chips will hurt it's performance massively. There's a reason why x86 chips have so much cache, and that's to deal with the ram latency. The further you move the ram from the CPU, the worse the latency gets. The Apple chips could never be used for serious sever work. And yes, CPU's care more about latency than bandwidth. GPU's care more about bandwidth than latency.

[Dukenukemx]

Credit where it's due, big BIG kudos to Apple for doing all of this.

You're thanking Apple for doing nothing?

[Dukenukemx]

Microsoft has a long history of not following standards either.

Microsoft owns PC as much Gabe Newell. Apple though totally owns you.

[ddriver]

If you push the CPU on a M based laptop you will consume the same power as an AMD or Intel. Like for example compiling the Linux kernel. It's even worse when it comes to gaming as the GPU isn't as good as AMD's. The power efficiency of the Apple M chips comes from their fixed function hardware for video editing, and their single threaded power consumption is really low. Basically the M chips are great for video editing and watching videos along with web browsing.

Most of the performance miracles on the M chips is due to asic acceleration, not general purpose compute as some perpetuators of uarch myths of awesomeness would have people believe. Running translated code, or running code that doesn't benefit by those accelerators reveals the actual performance characteristics of the cpu itself, rather than the "soc performance".

I always say it is very important to have benchmarks for both. Today we largely have "platform" performance tests, with many optimized routines running on different purpose specific hardware, which is just not a good indication of how well the actual cpu component is capable of on its own.

I think apple deliberately paid for that type of promotion to the tech "journalists" / corporate shills, for a number of reasons, the immediately obvious ones being "look at how glorious our cpu is" and "hide the true origin of improvements at soc level, provide misdirection to the competition". I remember Ian at AT mindlessly parroting their marketing slides, on subjects he has proven time and time again to be conceptually clueless. But the fact is there was no magic or awesomeness to the M1, it was a process node advantage, quite fast integrated memory, and a lot of non-cpu accelerators. This thing had 20 billion transistors, when an 8 core amd apu had about 10. It is evident when it is running general purpose code, and it is evident when it is running translated code that has no way to directly tap into accelerator apis. Any performance or efficiency advantages evaporate, solid proof it is all in the accelerators, and the actual cpu part is ... somewhere in the ballpark of zen 2, just on 5 nm.

x86 is getting there currently, but since it is n hardware vendors running a number of different os and software from scores of vendors, it will take longer to properly leverage that. Apple in comparison controls everything top to bottom, they have the guarantee what hardware baseline all software will run on, they can integrate asic functionality on system library level, into the app sdk, and the benefits are automatically propagated to 3rd party software with little to no need to manually optimize anything.

[hamishmb]

You're thanking Apple for doing nothing?

No, I'm thanking Apple for designing power efficient hardware, and fixing bugs in their firmware that only affect Linux - see the posts that came before mine.

[ldesnogu]

Most of the performance miracles on the M chips is due to asic acceleration, not general purpose compute as some perpetuators of uarch myths of awesomeness would have people believe. Running translated code, or running code that doesn't benefit by those accelerators reveals the actual performance characteristics of the cpu itself, rather than the "soc performance".

Yes and llvm provides "ASIC" acceleration for SPEC which helps Apple chips. You won the stupidity award of the month. And that's not the first time you achieved that.

[hamishmb]

if you go deeper in the so called supply chain fiasco of the chip crisis you will find that there was never a crisis in the supply chain....

there are only 2 companies who was the bottleneck ASML (EUV lithography) and Ajinomoto (BF-Substrat (Ajinomoto-Build-up-Film))...

for Ajinomoto its super simple to just build more factories for more Substrat

and ASML did not even need to build more factories they just developed High-NA EUV,... High-NA means they can just do the EUV lighting job much much faster...

https://semiengineering.com/gearing-up-for-high-na-euv/

this means there was never a problem in the supply chain...

what really happened: monero crypto coin did eat all the CPUs and ethereum crypto coin did eat the gpus.

and eat all the cpu/gpus means they did consume all no matter how much you produce if you produce the double they consumed the double if you consume 100 times more they consume all the 100 times more.

I know someone who works in electronics manufacturing and I can tell you that it's about far more than GPUs. Also, how would people buying GPUs cause a chip shortage?