ARM developments coming to a Windows laptop near you

[pdxindy]

So, while Apple has a good strategy for consumer computers, that is not necessarily a win for ARM in general. Qualcomm is not going to unseat Intel/AMD in the near future. Moreover, backwards compatibility is still going to be huge for a while. Long enough, IMO, to see Intel/AMD continue to evolve.

ARM is not some silver bullet. Execution is where it's at and Apple, for all of their gains, are still not on top in terms of overall performance. It has nothing to do with ARM and everything to do with the specific compromises that they're making in their designs.

It is hard to imagine how Qualcomm succeeds when they have no control over software. Apple was/is able to force the Mac world to develop for ARM. Why would any company on the PC side spend the time to develop for Windows ARM when the vast majority of computers are Intel?

For me, overall performance is becoming less significant. There is a point at which lots of computers are fast enough to handle what I need and then it doesn't matter in any practical sense which one is faster.

My new M1-Pro MBP is fast enough that it effectively fulfills my needs without getting in the way. So at this point, it doesn't much matter if it were faster. What I do care about is that it stays cool and quiet while I am using it... which it does. I was working on a small project in Bitwig this morning and the laptop was cool to the touch, not even warm.

[ghettosynth]

So, while Apple has a good strategy for consumer computers, that is not necessarily a win for ARM in general. Qualcomm is not going to unseat Intel/AMD in the near future. Moreover, backwards compatibility is still going to be huge for a while. Long enough, IMO, to see Intel/AMD continue to evolve.

ARM is not some silver bullet. Execution is where it's at and Apple, for all of their gains, are still not on top in terms of overall performance. It has nothing to do with ARM and everything to do with the specific compromises that they're making in their designs.

It is hard to imagine how Qualcomm succeeds when they have no control over software. Apple was/is able to force the Mac world to develop for ARM. Why would any company on the PC side spend the time to develop for Windows ARM when the vast majority of computers are Intel?

Well, to be fair, the vast majority of desktop/laptop computers are x86. The vast majority of "computers" of course are mobile devices and ARM is dominant.

But, yes, it's a huge mountain to climb on the windows side and you only need to look at the "lazy M1 devs" thread to see that it's not an overnight process and that if it wasn't for Apple's "our way or the highway" on the Mac side you simply wouldn't see any of that development.

Windows x86 emulation will continue to lag for some time. Of course ten years is a long time and we would absolutely expect to see that sort of thing improve. We should also see it become easier to develop for different architectures. However, my point was that this isn't intrinsic to ARM or even RISC in general. It's more about market forces. Microsoft isn't going to get a license to make x86 chips so they have to make something else, e.g., RISC V, or ARM, or something.

Microsoft wants to be Apple and they have achieved some of this success on the gaming console side. They want you to buy their hardware and develop for their closed ecosystems. I'm not sure that they have the ability to execute as well. In the past, I think that Google has been more successful with Chromebooks.

https://www.geekwire.com/2021/chromeboo ... ket-share/

For me, overall performance is becoming less significant. There is a point at which lots of computers are fast enough to handle what I need and then it doesn't matter in any practical sense which one is faster.

My new M1-Pro MBP is fast enough that it effectively fulfills my needs without getting in the way. So at this point, it doesn't much matter if it were faster. What I do care about is that it stays cool and quiet while I am using it... which it does. I was working on a small project in Bitwig this morning and the laptop was cool to the touch, not even warm.

Agreed. However, I was only addressing the argument that the ARM ISA is the secret sauce, it is not. Apple has chosen a decent set of tradeoffs for consumer computers and the M1 chip is really interesting. However, Apple does not make "ARM designs" they make Apple designs that execute ARM instructions. That said, many people, e.g., Gamers, still care very much about the top end, and even if they don't make use of it, it drives sales.

Your argument also applies to x86 machines. My AMD Lenovo that I now use for work stays cool and quiet and runs all day on batteries. I don't need the max performance there either. Here I mean performance in terms of battery life. I also value things other than top end CPU performance for different applications and my Lenovo is a better work choice than a MB Air. It was a lot cheaper, runs software that won't run well on an M1 Air, allows me to quickly shift my work machine to a different machine by moving the SSD, and is easily sufficiently powerful, cool, quiet, and power friendly to get the job done.

As cool, pun intended, as I think the M1 chip is, I think that the biggest impact will be how it shakes up the industry. It's not going to rocket Apple sales to some huge percentage of the market, but it is, and in fact, already has, forced strategic shifts among competitors.

[machinesworking]

Again, the main driving force behind recent chip improvements all come down to smaller, less power hungry chips at the same performance level, it’s why i7 and i9 won out over Xeon even on desktops for the most part. So if/when it comes out that Arm is the way forward in terms of power VS performance we will see a change in architecture. Apple are right now doing this, and we’re seeing Microsoft and Qualcom reinvested in this because frankly of the results they’re seeing from Apple.

It is a common and outdated view that ISA is the reason for the difference, it's not. It has to do with implementations, research focus, and priorities. Apple has been heavily invested in power friendly mobile chips and it's easier to scale those up to a low power desktop than it is to scale down a high power desktop.

However, it's a misconception to think that it's because ARM, and that x86 is intrinsically doomed as an ISA. There has only been one chip in the last two decades that directly executes x86, the original Atom. In terms of RISC vs CISC, RISC has won, all modern x86 chips operate a RISC core and the x86 decoder is not the reason for significant increased power consumption.

So, while Apple has a good strategy for consumer computers, that is not necessarily a win for ARM in general. Qualcomm is not going to unseat Intel/AMD in the near future. Moreover, backwards compatibility is still going to be huge for a while. Long enough, IMO, to see Intel/AMD continue to evolve.

ARM is not some silver bullet. Execution is where it's at and Apple, for all of their gains, are still not on top in terms of overall performance. It has nothing to do with ARM and everything to do with the specific compromises that they're making in their designs.

https://www.extremetech.com/mobile/3120 ... based-macs

Yeah I'm well aware that x86 is basically RISC at this point, the article you post is from before the release of Apples chips into the market, and has some gems in terms of limited research on the part of the writer, this is one:

At the same time, however, Apple isn’t shifting to ARM in a year, the way it did with x86 chips. Instead, Apple hopes to be done within two years. One way to read this decision is to see it as a reflection of Apple’s long-term focus on mobile.

Yeah, that's just wrong, the switch from PPC to x86 was announced with a two year transition period that Apple completed in a year and a half.
There is no qualitative difference in approach to the transition, this writer is reaching for straws. I appreciate the straightforward talk about RISC vs CISC, but given that, his assessment of the situation is not anything like complete, or even that relevant two years later.

But I digress, the fact is people are IMO not looking at this from the perspective of pure technology, but from a subjective angle. There are obvious advantages that have been blatantly on display since Apple Silicon came out, but instead everyone is concentrating on the obvious Apple agenda of it, tiny boxes and limited but good performance. It's pretty obvious that the same concepts can be used at higher wattages to get better performance out of new designs from AMD, Intel, Qualcom etc.
It's weird to me that people can't see this, and like most "vs" type discussions on things like tech that people invest their personalities into, it ends up being like kicking water uphill to discuss it.

[BONES]

It is hard to imagine how Qualcomm succeeds when they have no control over software.

They succeed the same way Wintel did, by working in partnership with Microsoft.

Apple was/is able to force the Mac world to develop for ARM. Why would any company on the PC side spend the time to develop for Windows ARM when the vast majority of computers are Intel?

Well, if ARM really was the future, then it would be a no-brainer for any dev, wouldn't it? And we'd have already seen lots of development for Windows on ARM. The almost total disinterest from software vendors tends to show that they don't see ARM as the future at all and they are surely in a better position to judge than you or I.

For me, overall performance is becoming less significant. There is a point at which lots of computers are fast enough to handle what I need and then it doesn't matter in any practical sense which one is faster.

Yep. I had my 6-core i7 gaming machine and it was way more power than I needed, so now I'm back on a good ol' mobile i7 with four cores and having no problems at all.

I was working on a small project in Bitwig this morning and the laptop was cool to the touch, not even warm.

I wouldn't have a clue how hot or cold my laptop gets or how often the fan kicks in. I know how super-quiet the fan is because sometimes it's on at the end of a session but I have no idea at what point it might have kicked in. People seem to worry about the weirdest things.

Microsoft wants to be Apple and they have achieved some of this success on the gaming console side.

I don't think they do. I think they have, at some point, realised how shithouse they are with consumers and they'd prefer to be IBM, with no retail business to deal with at all. They make most of their money from enterprise solutions, which is why Windows has been pretty much free for many years now.