No, this is not the reason. The reason is that Apple didn't half-ass anything, and they executed with competence. The Apple M1 is so good, it outperforms most everything else natively in single-thread, and every other previous x86-based MBP when using Rosetta, all the while offering superior battery life, and its not even that expensive because those are the entry-level models.

Microsoft never had anything remotely like this. Their hardware is ok, but they forgot one important detail, that Apple has taken to heart like nobody else:

"People who are really serious about software should make their own hardware. (Alan Kay)"

Why would anyone want to switch to ARM on Windows if it wasn't a lot better than x86? What if their WOW x86 emulation was really slow?

For Microsoft, "own hardware" means plugging cots parts together. The best they can achieve is a "premium look & feel", but they can't do anything game-changing. For Apple, it means owning and designing the most important chips themselves. Because they are controlling the whole hardware & software stack, they can do things nobody else can. This is the difference in understanding design as "how it looks" vs "how it works".

Apple has a custom MSR in their M1 implementation to toggle Total Store Ordering, which enables them to run either the native, more relaxed ARM or the more strict x86 memory model. They probably use other tricks as well, but there is a reason Rosetta is as fast as it is.

Never half-ass two things, whole-ass one thing. - Ron Swanson

While funny, the moral I perceived is "Apple whole-asses everything thanks to vast invested resources", which is unfortunately not as actionable as the Ron quote

Apple was routinely criticized for years for not doing more things. Why don’t they have more models? Why doesn’t their software have infrequently-used features X and Y? Why aren’t there a million levers and dials so I can customize my machine?

It’s because they were whole-assing one thing.

I haven’t been keeping track, but I think it’s true that the breadth of their product line is bigger or nearly as big as it has ever been, and certainly much bigger than 10 years ago. They got there organically over the last 20 years, instead of developing shovelware.

spit my coffee out...

Do you happen to have a source on the TSO switch ? It sounds fascinating.

Exhibit A: https://www.realworldtech.com/forum/?threadid=193883&curpost...

Exhibit B: https://github.com/saagarjha/TSOEnabler

Almost every software offering under the sun runs natively on Windows, so it doesn't exactly make sense to give MacOS a pat on the back for that particular achievement with WoW.

I am all for innovation and I welcome any chip that runs cooler while offering better performance with the apps I use. But so far, 90% of the praise I've seen of M1 either comes from benchmarking software that doesn't reflect my real world usage, or directly from Apple's marketing.

It's a bit early to draw conclusions about how much of a smash hit M1 is going to be. We don't know much about its performance, aside from a few benchmarks here and there.

Eight years after Microsoft's first attempt to move to ARM with Windows RT, you still can't run legacy 64 bit Win32 software on a Surface Pro X. There is absolutely no excuse for that.

Just as there was no excuse for the Surface RT not having a touch native version of Office despite them selling it as a tablet, or the Surface Pro X not having ARM native versions of ALL of Microsoft's software.

Where's the ARM native version of Visual Studio?

How can you expect other vendors to take Windows on ARM seriously if Microsoft doesn't?

Are we really sure ARM is where the advantage here is with Apple? Not the 5 nm process?

Let's be real, it's just on par with the Zen 3 in single threaded perf which is on a 7nm+ process. When Zen 4 comes it might be moot.

AMD's entire market cap is practically on par with Apple's cash on hand (so very different companies in terms of size/capability) yet they are equal in performance single threaded? Not only that, in practical terms the AMD chip is very desirable in many situations because it has twice the cores.

AMD has beat Intel despite being at a process node disadvantage in the past.

We do have an entire shopping list of ways in which Apple's ARM core implementation leads vs. x86 in their five watt iPhone chip.

https://www.anandtech.com/show/16226/apple-silicon-m1-a14-de...

Fair point. Without moving the goal posts, I should have clarified that I was referring to Windows on x86, and have never owned a Windows ARM device. What I was trying to say is that, if you want compatibility with the broadest swath of software, you don't generally buy a Mac. This is not as true as it used to be, but it's still true.

I don't have a Surface Pro X and would never buy one, but it's an example of how ARM isn't necessarily the answer to the user's problem, because the user may well use his device for more than running Microsoft Office and browsing the internet.

A glimmer of hope from https://docs.microsoft.com/en-us/surface/surface-pro-arm-app... on Oct 1, 2020:

With 64-bit emulation coming soon in Preview via the Windows Insider program, you'll be able to run 64-bit (x64) apps on Surface Pro X.

64 bit support should be coming soon [1]: With 64-bit emulation coming soon in Preview via the Windows Insider program, you'll be able to run 64-bit (x64) apps on Surface Pro X.

> Where's the ARM native version of Visual Studio?

Well, there is VSCode... [2]

[1] https://docs.microsoft.com/en-us/surface/surface-pro-arm-app...

[2] https://www.zdnet.com/article/microsoft-new-vs-code-update-i...

> We don't know much about its performance, aside from a few benchmarks here and there.

I don't know what to say to that, except that we must be living in different universes.

I haven't used an M1 device. Frankly, outside of Apple no real-world user (in my universe, can't speak for yours) has long-term experience with it.

Do you always believe everything that a marketing department tells you?

Does all of your real-world experience with consumer products tend to line up with benchmarking software and short-term reviews based on a test drive?

To the contrary, I placed an order for an Intel i7 MBP13 (32GB RAM, 4TB SSD) around the same date as the Apple Silicon presentation took place. I've read every review I can find, seen the benchmarking runs, and -- in light of the fact that my new device won't ship until mid December -- I could still cancel/return my order, buy an M1-powered MBP13, and save some cash.

That is to say, I have skin in the game.

Say what you will about my igneous domicile, but buying a new-to-market processor architecture is not something that I am willing to do just because some guy at The Verge or Tom's Hardware or AnandTech made some bar graphs or felt it necessary to treat hype as fact. I'm not saying they're dishonest or anything, just that the information is thin because the typical user isn't an instance of Geekbench.

My advice to them: Use the device for a couple of weeks, install all of your work software on it, put it through its paces, and then tell me about what you observed. That's the kind of review that would satisfy my curiosity. I do a lot of financial stuff on my machine, and the software (or software-based services) in that world is often proprietary, poorly developed, and/or expensive. Compatibility and performance of those programs are more important to me than battery life while browsing YouTube or render time in Final Cut Pro.

I take your point there - maybe I was a little short with no reason.

I'm in a similar position, I bought a 16 i9 at basically the same time. Certainly not adopting the first generation of a new processor architecture on a machine I use for work.

However, render time in FCP and number of channels in Logic Pro cannot be understated if they're really this good. It's crazy to see a sudden uplift in what's capable there. A lot of what I've read even indicates that general system use (windows opening /closing, scrolling, system animations) are noticeably faster on the M1.

Plus, I'm happy to take Apples word as they were totally right with the A12z in the iPad Pro. That thing outperforms most windows laptop in creative workflows.

It’s probably more complicated than that: Microsoft’s customers have traditionally been very different from Apple’s, and they valued perpetual backwards compatibility over Apple’s approach.

Basically Apple is selling a cozy computing environment that some people really like, whereas the operating system division of Microsoft is selling a compatibility layer between a bazillion of long tail niche applications and every single piece of hardware in existence that isn't either significantly smaller or larger than a PC (phones, supercomputers) or made by Apple. If Microsoft had an LTS version of Windows 2000 that was 64 bit, a lot of people would be willing to pay a premium. Microsoft thinks they are selling another cozy computing environment, but generally people don't really want more OS level bells and whistles.

Off topic ramble: unfortunately, Microsoft to have a bit of an acquired blindness regarding what users actually want (LTS Windows, innovate in Office or other installable software if you like) because paid software updates has so long been their culmination of visible success. Had they somehow switched to subscriptions by the time they moved to 32 bit Windows would be an entirely different beast today.

Microsoft is trying to have it both ways; turns out that's not what people really want (and when it's all about people they just fail -- Windows Phone anyone?), but because business clients (which generally doesn't mean the actual business end-users but some other department) tend to buy whatever looks good on paper.

It all just keeps going on with no real improvements. This disconnect between people and products has many forms, be it an IT project that is over budget and doesn't deliver the right stuff (often comes down to business people making impossible promises and requirements that were never properly gathered and keep changing).

Getting something that is good for the user instead of getting something that is short-term good for the business would be a good start.

>(and when it's all about people they just fail -- Windows Phone anyone?)

Everyone I knew who had a windows phone seemed to love it. If MS were taking a run at it today it would likely succeed. The nail in the coffin was Google refusing to write a Youtube app for Windows Phone and THEN blocking the app that MS wrote themselves.

I'm still shocked the EU didn't tear them apart on that move. It's about as anti-competitive as it gets. Honestly I'm still sad they gave up, we could've used a third player, even if they were going to always be a minority player.

There were hundreds of nails in the Windows Phone coffin. The straight incompatibility between 7 and 8 was a gigantic fuck you to every early adopter, dev or otherwise.

I still consider the Windows Phone (8 and 10, not 7) the best Phone OS I have ever used, and the Nokia Lumia hardware was great. The problem, as you mentioned, was that third party app support was dismal and the few apps that did exist where on the whole of a far lower quality than the Android or iOS equivalent.

> The nail in the coffin was Google refusing to write a Youtube app for Windows Phone and THEN blocking the app that MS wrote themselves.

How the hell were they not sued for that?? That is anti-competitive as hell.

Google is still doing shit like that e.g. by actively hindering Picture-in-Picture on iOS for years.

Google was de facto running the Federal Trade Commission during most of the Obama administration. The lead commissioner was a paid shill who wrote "academic" articles about how Google shouldn't face antitrust scrutiny while a law professor.

I had one and I loved it as well.

> The nail in the coffin was Google refusing to write a Youtube app for Windows Phone and THEN blocking the app that MS wrote themselves. There were at least 2 great third party youtube apps and youtube worked well in the browser as well. Including background play support which seems not to work on other platforms easily.

It works very well on Android, but it is a feature exclusive to YouTube Premium.

It also works great with Newpipe. They now even have their own F-droid repository so that you can get the latest updates asap when YouTube makes a breaking and they fix it.

It works very well on Firefox on Android, out of the box, with uBlock to get rid of commercials too. I shudder to think that one day Mozilla will tire of supporting FF on Android and I'll have to start using a YT app again...

Of course there were people that loved it, just like there were people that loved BlackBerry phones, people that loved PocketPC and people that loved the foot-x-ray-machine at the shoe store when that was a thing. I wasn't trying to say that it was bad or universally hated, and even if it was you'd find a niche that likes it ;-)

It's just that it wasn't loved by the masses, just like Zune wasn't loved by the masses, Windows S-mode isn't loved by the masses, and pretty much any Windows-on-ARM effort isn't loved by the masses.

Regarding YouTube: odd that they wouldn't make an app, you'd think that even with trident support they could at least put a web view in an app or something. You can't force a company to maken an app for another company, and with the current media/rights laws you can't make one yourself either, so not much of a lawsuit there.

Indeed, when I had a chance to play around with one I thought it was a pretty decent take for a first try.

I had no idea Google played this dirty trick, if that’s how it went they’d indeed deserve a thorough spanking for it.

I suspect Microsoft being Microsoft they must’ve made other blunders to murder the platform so badly though

I'd say the bigger nail in the coffin was, ironically, Microsoft's level of security/sandboxing.

tl;dr - Snapchat and a lot of other 'killer apps' never were written for the platform, and at least some of that was because the apps on Android/iOS did things Microsoft wouldn't let you (I think tracking when the user took a screenshot was one)

Snapchat refused to make a Windows Phone app. Microsoft even offered developers and resources to help, and they still refused.

I wonder how would it work if they ripped out all the crufty apps and UI configuration tools and replaced them with simple (dumbed down) controls like Gnome or Apple do.

All the more advanced knobs made available via Powershell, also like Gnome or Apple. Indeed any professionally managed configuration isn’t maintained with point’n’click panels anyway...

It's not as easy it sounds. It took a whole generation to decouple printing subsystem from UI libraries.

Also power users tweak a lot of little advanced settings via these point'n'click panels anyway. Even tuning a desktop PC needs a lot of tinkering down the road.

Windows is a very powerful platform as is but, it's not well suited to everything due to its architecture and design choices. This situation is equally valid for all operating systems out there.

Maybe it’s time for them to split out NT and a home/office-light edition again

They do, with Server 2019 for corporate servers. But your point is that no one is running Server 201x as their desktop OS, unlike the NT 4 .1 days, which then was rebranded into windows 2000, or NT 5

There's also the LTSC (Long-Term Servicing Channel) build of Windows 10[1], which doesn't have any of the consumer bits and bobs, and other than security fixes isn't designed to have any feature updates post deployment.

---

[1] https://techcommunity.microsoft.com/t5/windows-it-pro-blog/l...

LTSC is fantastic in that it ditches all the cruft. I wish they had a consumer-friendly license for it, though. Microsoft's vision for consumer windows 10 is no longer an unobtrusive platform that doesn't get in the way of the user, much less force unwanted features on them.

I wouldn't say _no one_, but it's definitely not common: https://www.windowsworkstation.com/

I definitely remember running windows 2008r2 on my home desktop because at the the time that was all that was available in the "your educational institution neither gives .edu emails or has arranged anything with microsoft" tier of dreamspark.

During my stint at MS many years ago, we were running 2008r2 as a Workstation OS and it was pretty good.

And the fact that qualcomm seems to be lagging behind apple's SoC.

Microsoft should split windows into two distinct versions.

Microsoft Windows Enterprise - focus on long term compatibility etc (essentially what windows is now)

Microsoft Windows X (seeing as they like using that letter) - which is way friendlier to breaking changes and advancing the OS.

I can't see any way for Windows to actually improve long term without doing this.

That "X" would basically be what's running on the ARM surface and arguably the whole line of precursors back to CE could be described like that. They end up wildly unpopular because they have all the bells and whistles people begrudgingly accept to get the one feature they want (continuity) but lack this one that they accept the others for. Continuity isn't just for enterprise. Non-technical users abhor having to re-learn anything or give up the procedures they struggled so hard with adopting (there are probably still some who feel uneasy if they haven't run defrag for more than two months, despite of having switched to NTFS more than a decade ago) and gamers and the like have lots of niche hardware and software solutions that easily puts the unsupportedness of vintage enterprise software to shame. Continuity is a key feature.

What I think could work, and would eventually lead to the same situation as "Enterprise/X" would be the reverse: a "regular" Windows aiming to eventually build a Microsoft equivalent of the Apple walled garden and a "continuity subscription" that's more conservative than what Windows is now (trivial example: opt into Cortana instead of opt out, barely)

PS: I understand that you are talking about low level technological differences and I seem to be hung up with UX, but I believe that separating groups on the high level would be the only viable way to eventually follow different paths on the low level.

The worshiping of windows compatibility /continuity (which I do consider to be impressive from an engineering standpoint) is astounding to me. It leads to what we have now - which, to me at least, is a frankly unusable operating system. A hodge-podge of 20 years worth of patterns, menus and systems. Some menus in windows feel they're in a completely different operating system.

This is why they should have 'Enterprise' and 'X' to me - they dont even have to look at different, but it frees one windows kernel team to do whatever they want and not have to worry about supporting loads of legacy systems. They'll naturally deviate.

It's actually kind of ironic to me that the best feature they've released in years (WSL) is essentially jamming another operating system into the mix. Like there's not enough going on there already!

I'm reading so much about how good Proton is that I'm considering dropping windows even for my gaming machine.

I consider the wildly mixed UIs not a symptom of compatibility but a symptom of the succession of hopeless attempts to be more than that. The incomplete new GUI generations are surely not caused by compatibility burdens affecting the kernel team.

No but they're a symptom of the 'all work on the same product' situation - if they were split, the people wanting to update the UI etc would work on the new version alongside the kernel team for the new version.

Microsoft's main problem is marketing - to be more specific, a naming problem.

Windows everywhere

They doomed Windows phone with the name WINDOWS.

Their windows on ARM, Win RT push was also handicapped and doomed the moment they named it Windows...

Explanation.

Windows means:

1. BSOD, instability, viruses,

2. Expectations of ability to run billions of existing windows software

The first thing that popped in anyone's mind about Windows Phone was 1 above.

For ARM, many frustrated customers returned their store-bought Windows ARM devices after they got home and could not run normal vlc, chrome, existing games...

Microsoft having realized this, has spent the last 4 years+ working on the ability to run classic windows apps on windows for ARM. They placed their bet on faster chips - Apple got there first.

If Microsoft names their ARM operating system something else, initial marketing dollars would be initially higher. But it would be worth it.

To developers, nothing will change, it's the same UWP, same APIs.

Imagine xbox was called "Windows Gaming". It too would be dead.

> Imagine xbox was called "Windows Gaming".

isn't Windows gaming the richest market for games since a proper market for games exists?

Yes! Today is a good day to reread Phil Haack’s 2014 essay on the lengths MS goes to for backwards compatibility: https://haacked.com/archive/2014/05/27/backwards-compatibili...

Edit: Haack heavily cites Raymond Chen from 2003: https://devblogs.microsoft.com/oldnewthing/20031015-00/?p=42...

On the flip side, supporting all that backward compatibility has been a massive anchor on Microsoft's ability to move Windows forward from decisions made in the past.

Yeah Microsoft is a like King John signing the Magna Carta to keep the thrown. Apple is like Louis XIV.

I believe you mean throne.

Still, as developer on the JVM I find the concept of a "Game of throwns" exceptional.

Ahah, "Game of throwns". Can only imagine the disaster :D

Be sure to catch every episode.

And optionally rethrow.

Oops!

And of course Sinofsky from this summer.

https://medium.learningbyshipping.com/apples-relentless-stra...

Apple is also delivering some immediate obvious value to make for the migration pain, given the huge battery life increase and what look like substantial performance gains over the previous models.

Oh, sure, and Enterprise SKUs are going to be pinned into those holes forever, but Microsoft could've been more aggressively pushing their platform on the consumer side if they wanted to make any of the changes they've tried in the last ten years stick.

The view from the outside on every Windows platform initiative in the past year is dismal failure, even if the platform has incrementally benefitted a bit from each.

Most software is effectively still written on the same platform it would've been for Windows 7. That's a bad look for everyone building everything there since.

Even internally: Microsoft's own platforms largely failed to embrace UWP. If your own internal developers aren't buying in, why would anyone else?

If you liked apple because it ran your Mac apps, you already got off that train when they moved to Intel in the first place.

No, you got off that train during the 68K to PPC transition.

6502 to 68k

I actually did leave Mac at 68000 to 68020 transition. Color was interesting but too much money for a high school kid... Cue a shiny 386 for which you could by Turbo Pascal!!

Now that was a real disaster that made me leave the Apple ecosystem.

I think we'd see a serious boom in Windows on ARM if ARM vendors had a performance benefit to offer and Microsoft implemented a backwards compatibility layer as Apple has in this case. The difference here is in the offering. Microsoft was making a tablet offering where they're just not competitive with no ability to use x86 software while Apple here is offering much more - backwards compatible and all.

Until then there's just no reason to bother with ARM and vendors and users alike can see that.

Only recently has Microsoft announced a backwards compatibility layer, if we can see some hardware vendors step up to Apple levels of performance, there might be a serious shot now: https://www.extremetech.com/computing/315733-64-bit-x86-emul...

> Microsoft implemented a backwards compatibility layer as Apple has in this case

Windows-on-ARM already has 32-bit x86 emulation - and 64-bit x64 emulation is coming soon:

* https://docs.microsoft.com/en-us/windows/uwp/porting/apps-on...

* https://www.extremetech.com/computing/315733-64-bit-x86-emul...

I haven't looked too deeply, but it seems to use the same WoW (Windows-on-Windows) mechanism that's been present in Windows NT going back to at least Windows XP 64-bit Edition (no, not 2005's "Windows XP x64", but the original 2001 Windows XP for Intel Itanium IA-64).

> The WOW64 layer of Windows 10 allows x86 code to run on the ARM64 version of Windows 10. x86 emulation works by compiling blocks of x86 instructions into ARM64 instructions with optimizations to improve performance. A service caches these translated blocks of code to reduce the overhead of instruction translation and allow for optimization when the code runs again. The caches are produced for each module so that other apps can make use of them on first launch.

Looks like Windows dynamically translates code in DLLs/EXEs on a JIT/on-demand basis and caches each translated run-of-code.

Running x86 code via Rosetta on the MacBook Air is significantly faster than it is running on the previous generation. With the Surface X, that isn't remotely the case. It's a pretty major gap.

This is from the Verge's review of the Surface Pro X:

> The worst part is that even if an app installs, it doesn’t mean you’re going to have a great experience. Photoshop installs and opens just fine on the Surface Pro X, but the usability of it is terrible. https://www.theverge.com/2019/11/6/20950487/microsoft-surfac...

Maybe it's improved since then? Everything I've heard suggests there is a pretty big performance penalty running x86 code on Windows ARM. If there is a significant performance hit on the M1, the performance gains of the M1 are so good, it erases the difference.

The only way an emulation layer works is if the new platform is fast enough to erase that difference. When Apple moved from PowerPC to x86, it nearly did. With this transition, Apple knocked it out of the park. Microsoft? Not so much.

Rosetta seems to have a similar performance hit to Microsoft's solution (10-30% overhead compared to native, depending on the workload).

The difference is that Microsoft is locked into using Qualcomm chips, which just aren't very close to Apple's performance.

The impression I got was that it was a lot slower than a 30% performance hit, but perhaps the Qualcomm CPU is that much slower? If so, it really begs the question—Why bother making a tablet-laptop thing which performs so poorly?

The tablet-laptop form factor is great, and a lot of people simply don't need high performance but do need to run some specific line of business application from 2003, so including x86 emulation makes sense.

It's a bad cycle.

Lackluster performance means low adoption. Low adoption means developers have no incentive to support it. Lack of developer support means little or no software will be written for the platform so performance never improves. Repeat.

> Lackluster performance means low adoption

And yet Electron rules cross-platform desktop development, not Qt nor wxWidgets or openstep.

Developers are fine with worse performance - even lackluster performance - if a platform or framework offers some other compelling advantage, especially _developer performance_ (i.e. it saves money).

Electron saves money because creating one-off custom UI widgets can be done easily and cheaply with just HTML+CSS+JS. While there are many things that are literally impossible with HTML+CSS+JS at-present (such as "splitting" a rendered UI, like how iOS 5 did with its notification banners and home-screen folders), but besides certain flashy effects everything is doable quickly. Whereas building a custom UI in any other platform would take days or weeks just to get a prototype done if you couldn't hack-on additions to an existing "UI component".

So your line of reason would be improved by saying:

"Lackluster performance with no other benefits means low adoption", the rest then follows.

Because those performance stats are the usual level, Apples own chips just exceed the general trend by a huge amount. The hardware is just better than Qualcomms or any other widely available ARM cpus.

Basic example is how even the MS's very own Visual Studio is horrible slowbloat (won't even mention how ugly), compared to let'say IntelliJ editors on the same Windows that are noticeably faster.

Have you tried running VS in Software-mode without hardware acceleration of the UI/graphics?

Huge difference in performance though.

M1 seems to run things on Rosetta emulation in some cases faster than they ran natively because of how much uplift the hardware has.

Microsoft ARM devices have such lower performance than any emulation will run worse than natively, making it essentially unusable as a serious offering.

Both behaviours make perfect sense for the respective companies.

Apple is selling hardware (and now services), the software at most should break even. Apple gets obvious benefits from its approach - verticle integration, hw sales...

Microsoft sells software. MS doesn't care if the platform is x86 or ARM. Why would it? MS gets paid either way. As far as its concerned it would offer the option (to make Windows future-proof and run on more devices) and let the market decide.

Microsoft cares because if Mac laptops have twice the battery life of windows laptops, nobody will blame Intel

Won't they blame Dell/Toshiba, not Microsoft? In my experience, users think of the vendor as being entirely responsible for battery, even when the software may be using it inefficiently.

Users will see all their friends using Apple products getting much better battery life, and their friends not using Apple products getting much worse battery life.

Agreed. People blamed Apple for Intel’s foibles up until a few days ago.

It still doesn't make sense for MS to push ARM that way - Apple isn't offering its ARM processors to 3rd parties. The best non-Apple processors are obviously x86 right now.

My question is what is the play for Intel and AMD right now? Are they shitting their pants or is this something that was somewhat expected given the progression of Apple's chips and their announced plans? Do Intel/AMD understand how Apple managed to so outmaneuver them? Is the performance gain Apple has delivered an inherent advantage of ARM over x86, and if so, does that imply that we are going to be seeing a migration to ARM in the PC space in the coming decade?

I'm deeply fascinated by these questions. I'd love it if someone in the chip design industry who is knowledgeable about the situation would comment.

AMD is fine. Its processors are competitive (often superior) and it's gaining marketshare.

Intel has been in trouble for some time due to its own failure to downscale, a failure which has little to do with x86. They'd be losing to someone else if Apple/AMD hadn't been there.

Intel still has some time to recover: it has plenty of money, and Apple/AMD are unable right now to take full advantage of their position. Apple/AMD rely on a supplier (TSMC) which does not have the capacity the supply all or even most of the processor demand, so Intel gets quite a bit of marketshare almost by default.

Given that Apple only has about 10% of the world desktop market, and zero server presence, very unlikely.

Apple isn’t selling to third parties, but I bet Google wants to. Google must be putting pressure on Qualcomm, and maybe Nvidia, to make something competitive. If they try, within a few years I’m sure someone can.

Imagine 3 years from now there are a bunch of Chromebooks selling at Chromebook prices, but offering intel performance. That would start to hurt Microsoft.

From the way they have been selling in Europe, there is a very long path to go.

Chromebooks are only relevant in US school system.

Chromebooks are actually widely used in schools in the Nordics.

Here in Germany they just do timid apperances in some consumer stores and then are reduced in endless offerings until someone finally takes them away.

Very few people are willing to pay the same price as a regular laptop for what seems just a big browser OS.

Is it though, or is it lacking the software investment that Apple made.

There are a good amount of chipsets that are now on 5nm processing. But only Apple made a full desktop OS to go with it.

Having used linux on High end ARM phone chips, my 4 year old Intel laptop, also runnign linux, is far more performant than they are.

I think one of the other reasons is that Apple designed their own cpus. And put a lot of effort in. Microsoft just grabbed some off the shelf stuff for the surface. So it’s not even close in terms of performance.

Imagine if Microsoft said you could get a surface with arm processor, 16gb of ram, with 20+ hours of battery life and the same perf as an i5 processor. And it lived up to the hype. People would love it.

My 2c

Surfaces are very, very custom. Their x86 versions even are almost entirely bespoke: There are custom Surface drivers for nearly every component. Their hardware team is top notch, and I guarantee you hardware is not where Microsoft is falling flat.

The Surface Duo is another example: The hardware was ready for years. If it'd launched when it was, they'd have scooped the foldable trend super early. But the Windows on ARM platform was still such a joke, they gave up and put Android on it.

> Surfaces are very, very custom. Their x86 versions even are almost entirely bespoke: There are custom Surface drivers for nearly every component.

Yet almost all use Intel CPUs, and the performance of the Surface Go / Surface Go 2, is abysmal compared to the M1.

The Surface Pro X?

Single Core score ~800 vs ~1700 for the M1

Multi Core score ~3000 vs ~7500 for the M1

https://browser.geekbench.com/v5/cpu/compare/4799088?baselin...

It's a ARMv8 CPU... How is it 'very custom'

> Their hardware team is top notch

That keyboard... it makes the bad Apple ones look top shelf. I know removable keyboards are a different thing but I cringe remembering the clicks and trackpad.

Honestly, the typing experience for me of the Surface keyboard is perfectly fine... it's just that the form factor makes lap-use difficult. I didn't think it'd be a problem for me, but a year or two in, I felt regret there.

The Surface Book seems like a vast improvement in that regard... but it's extremely expensive, and relied on an overengineered latching mechanism. Dell made a really neat Book alternative, the Latitude 7285, which was really compelling, but by the time I was in the market, they'd abandoned that form factor already. :/

I don't know, I think the Surface pro keyboard is surprisingly good all things considered. I agree the trackpad doesn't feel great, but it is accurate and usable enough and the typing experience is certainly no worse than any other similarly sized laptop I've used and better than many. For a removable keyboard I think they've done a great job.

Microsoft can't even let go of x86 and make x86_64 mandatory. It could them years to recognize ARM as a valid target.

At it‘s most basic, Microsoft‘s makes billions by enabling enterprise customers to remain lazy.

If you don‘t force them they will never move an inch. They‘d rather pay insane amounts for support for decades old shitty software than to finally modernize.

It‘s easy money for MS. Though I‘m sure Windows devs hate it.

Microsoft was committed to supporting Windows on Itanium but Intel mucked it up in several areas - if they perhaps got AMD onboard (a free license to the architecture - which would have made AMD64 moot) and made the first variant faster than x86 and consumer priced/focussed.

I am afraid that X86 will live on forever more with just more wacko instruction extensions that nobody will use.

Apple own the whole vertical of their platform.

Microsoft don't: they are in Google's position with Android, that they have to work with the third party hardware vendors and their value-subtracted software and so on.

>they are in Google's position with Android

Not true, Google just doesn't have to care what CPU is in your Phone.

You can have Mips/Arm/x86/powerpc whatever you want.

Microsoft still has a severe UX disadvantage with Windows and failed to understand that simplicity over complexity always wins. Most Microsoft products from Ballmer era are trying to do too much. VS Code is the prime example that you can produce a product with few features, people are going to use it and if you don't commit the previous mistakes that poison the UX of Word and Windows you are going to succeed. 5 years ago if somebody told me that I am going to be using a Microsoft product more than Vim for programming and text editing I would have not believed.

Microsoft has the ability to innovate and produce amazing products it just they carry too much weight from the past and that holds back entire product lines.

Why Windows on ARM is doomed? It's there. Microsoft just doesn't have immediate need or reason to switch to it immediately, while Apple does.

I'd say it's doomed because not even Microsoft has bothered to create ARM native versions of all their software.

Apple went all in on ARM native versions of their stuff on day one.

Well consider this, when WOW came out it supported PowerPC and Intel so WOW has effectively lasted longer than Apple on Intel did

World of Warcraft came out over half a year before Apple announced the Intel transition.

Also Adobe release Photoshop beta for Mac/Windows ARM.

It is ironic that windows ARM was released long before, but Adobe support it until Mac ARM on Market.