Hacker Newsnew | past | comments | ask | show | jobs | submitlogin
Available Today: Windows Dev Kit 2023 a.k.a. Project Volterra (windows.com)
370 points by mpalme on Oct 24, 2022 | hide | past | favorite | 542 comments


This is Microsoft's second go at a developer-focused ARM kit. You may recall the first was called the Snapdragon Developer Kit [1] and was a miserable machine. It appears Microsoft has already pulled it from the Microsoft Store [2].

There is no other supported desktop/CI/etc. solution for Windows on ARM. (Mac mini with Parallels/VMware is a popular unsupported alternative. It may still be your fastest.)

Seems like a truly decent machine (ordered one) but I'm not sure if there's remaining appetite for Windows target-specific development. After years and years of development platform missteps and PC usage pattern changes, I fear web (electron, etc.) has become good enough for Windows users.

[1] https://developer.qualcomm.com/hardware/windows-on-snapdrago...

[2] https://www.microsoft.com/en-us/d/ecs-liva-mini-box-qc710-de...


Snapdragon Developer Kit was built with a "cost above all" mandate, $219 target pricing but it just wasn't powerful enough.

> There is no other supported desktop/CI/etc. solution for Windows on ARM. (Mac mini with Parallels/VMware is a popular unsupported alternative. It may still be your fastest.)

Azure cloud instances are also an option.

> but I'm not sure if there's remaining appetite for Windows target-specific development

I think that there is. If there isn't, Windows has bigger problems...


I feel Microsoft is in a similar position to it’s windows phones in regards to ARM.

Additionally they don’t have a system like Rosetta 2 to help with translation, a LOT of windows apps are x32/64 based, and there has to be an incentive for devs to make the switch, Apple faced this issue when moving from PPC, they used I believe Carbon to help, and over time took carbon off and forced devs to update to apples current architecture at the time.

Electron unfortunately does seem like the potential way forward for apps, not just because of Windows and their handling of ARM, but it’s also potentially cheaper to develop against as its multi platform, 1Password is a prime example of this, and outside of the hardcore audience like us, they won’t notice the increased resource use of electron.

Microsoft has proven however electron can be good, see Visual Studio Code, plus with the advent of chrome os gaining traction, it’s likely we may head into a web first world, outside a niche subset of users.



> and there has to be an incentive for devs to make the switch

Why make the devs switch? It would be much more in the spirit of Microsoft (opposed to the common spirit of Apple and GNU/Linux) to encourage to develop parallelly for ARM, too.

The switch from 16 to 32 bit Windows was "support running 16 bit applications still for a long time". Similarly for 32 to 64 bit Windows (32 bit applications are still perfectly supported). This is in contrast to how how it is typically done on GNU/Linux (distributions typically make it hard to support 32 and 64 bit at the same time) or what Apple typically does (only support the old architecture only for a short time and nudge developers heavily to do the switch).


> Additionally they don’t have a system like Rosetta 2 to help with translation

They do. I'm not sure it covers x86-64 yet, but it does x86.


x86_64 is covered since Windows 11


Electron is on the way out on Windows as they move to WebView2. https://developer.microsoft.com/en-us/microsoft-edge/webview...


I like the idea of WebView2, but it seems to be abandonware. They don't fix the most obvious bugs you run into immediately when trying it out. https://github.com/MicrosoftEdge/WebView2Feedback/issues/229...


It's not abandonware... it's a core component to all of Microsofts desktop apps starting with the new Teams and Outlook.


> but it just wasn't powerful enough.

4 GB of RAM and eMMC would never be comfortable for Windows development. I have a very small laptop like that and I'd never think of using it to do .NET development on it. OTOH, for Python-backed web apps and API-based services, it's quite sufficient.


That devkit wasn't intended to be a development but a testing device.

But then they made a big mistake: putting Windows Home on it instead of Pro... so that it didn't have RDP...


> Azure cloud instances are also an option.

I couldn’t find ARM with windows, can you help?


It's in Public Preview, I believe. I was able to create one relatively easily. https://imgur.com/a/bHXcNTz


That doesn't show desktop Windows on an ARM VM.


It is desktop Windows afaik. Server Windows is not quite publicly available yet (on arm64).


I am working on the Windows on Arm project at Linaro, enabling various OSS programs for WoA (currently working in Blender), there is some appetite :)

Feel free to ask questions!


How long do you think it will be until many things are ported to arm?


Define 'many'! Several large projects that may well be prerequisites/dependencies (ie, Python, CMake, .NET) are available natively, so most projects should be able to be ported with relative ease.


I remember when that came out. Shocking it's already replaced.

Anyone who is still interested can buy one on eBay straight from Microsoft's eBay Outlet (which surprisingly exists):

https://www.ebay.com/itm/255784903190


> There is no other supported desktop/CI/etc. solution for Windows on ARM.

What about Surface Pro X? Not desktop enough?

https://learn.microsoft.com/en-us/surface/surface-pro-arm-ap...


>> but I'm not sure if there's remaining appetite for Windows target-specific development

I mean.....video games are still huge?


In general yes, however not sure windows on arm is a target platform for game devs?


I am hearing great things about the Lenovo Thinkpad X13s (both performance and battery-wise), and hope more ARM models will follow.


> I fear web (electron, etc.) has become good enough for Windows users.

You seem to be talking without regard to the actual facts. Plenty of programming languages offer first class Windows support, including:

Go, Rust, Zig, D, Nim, Python, PHP, Ruby, and others.


The most interesting part to me was the quote from Denuvo, presumably testing out pluton:

  “The Volterra devices are neat and powerful, ideal for us to test our market-leading anti-piracy and anti-cheat game security technologies. They’re also very quiet, and they just work out of the box.” – Reinhard Blaukovitsch, Managing Director of Denuvo by Irdeto


Yay, more walled gardens that will break for legitimate customers while pirated version will continue to work forever.

https://news.ycombinator.com/item?id=29149555


> “Performance with Volterra has been refreshingly fast. We have run Actipro’s x86-based product installers, and they all worked. Our WPF and WinForms controls all work well with no changes, even the ‘API-heavy’ ones, as does all test collateral.” – Boyd Patterson, Senior Software Engineer, Actipro Software

I just thought the same, but not WPF native on arm?, but an emulation layer?, still looks very promising.


WPF on .NET 6 is native-on-ARM64 now, apparently:

* https://developer.arm.com/documentation/102341/0400/WPF-and-...

* https://github.com/dotnet/wpf/issues/1817

* https://devblogs.microsoft.com/dotnet/net-july-2021/#windows...

> Windows Forms and WPF are now supported on Windows Arm64. We added support for Windows Arm64 with .NET 6 Preview 1 and have since backported it to .NET 5 (with 5.0.8). Windows Forms and WPF work the same on Arm64 as they do on x64. You could already use x86 emulation to run Windows Forms and WPF apps on Arm64, however, there was a performance cost to doing so. With this release, you can your apps natively with full performance.


Cool, thanks you for the links. I somehow missed that.


I wonder how many years until every DRM vendor can enforce pluton as a baseline. 5 more?


I just read about Pluton and honestly I'm not even shocked anymore. I'm glad I could enjoy free computing while it lasted.


In the name of securitay!


I'm stoked to see this as it's one more platform that can run OpenBSD 7.2. From the release announcement last week (https://www.openbsd.org/72.html):

    New/extended platforms:

    - Added support for Ampere Altra
    - Added support for Apple M2
    - Added support for Lenovo ThinkPad
      x13s and other machines using
      the Qualcomm Snapdragon 8cx Gen 3
      (SC8280XP) SoC.


I want to be on OpenBSD so bad but support is not the same as full support. There are still issues like battery life, GPU, etc. AFAIK everything still happens on the CPU, which increases battery draw.


Microsoft's Apple envy continues - additionally hobbled by Qualcomm's indistinguished/2nd rate hardware and disinterested developers.


Well, if they had copied Apple, they would have offered a $499 developer test kit that you could use for a year to port your app(s) to ARM and then return for a $200 credit.

Then you wouldn't be stuck paying way more for much less capable hardware.


$500 credit.


That's right. They did bump it up to a full refund.


I’d like to be able to run Windows on mac M series.


Or full-featured Linux, for that matter.


Arm64 vms seem to be working great. And pretty soon Asahi will get to the point of being completely usable as a daily driver


Why would I buy a Mac Mini for Linux when devices like this exist? The Qualcomm chip is officially supported and much cheaper than buying the Mac. Seems to be much less hassle and better value on the Microsoft side of things.


> Qualcomm

The main reason so many Android devices are throwaway junk is Qualcomm. Don't rely on them for anything that you still want to keep running for more than a couple of years.


We used a Qualcomm module for wifi/bluetooth connectivity on an IoT device and the bluetooth just straight up didn't work for a handful of Android devices. We had to wait like a year for a mitigation, since it was closed source firmware. That was after like a 3 month process of actually find someone who would hear our issue.


Where does it say Linux is officially supported?


The chip is having mainline Linux support merged with 5.20, according to Phoronix: https://www.phoronix.com/news/Linux-5.20-SoCs-8cx-Gen3-Arm


I've tried it. It seems like even with a 6.0 Kernel, additional DTBs are required to boot.

https://twitter.com/alexellisuk/status/1585562359254421504?s...


Sure, doesn't mean that everything works though. The Thinkpad x13s has the same CPU and apparently is quite tricky to get booted. Once booted audio, sleep, bluetooth and the like has been problematic. Granted that's more of an issue with a laptop than a SFF like the DevKit.

Things do seem to be improving. Releasing the Thinkpad x13s has brought attention to the 8cx gen 3 from the Linux community.


I'd wait to hear if this board has Linux support. It might have a locked bootloader.


Specs seem to list UEFI Secure Boot, which has been unlockable on all of Microsoft's Surface devices and previous ARM outings. Shipping this with a locked bootloader would be one hell of a breach from tradition (and probably undermine it's usefulness as a dev box anyways).

Regardless, the internal SOC is the same as the new ARM Thinkpad which also shipped with an unlocked bootloader. Pretty much everything suggests that this will ship unlocked.


> Specs seem to list UEFI Secure Boot, which has been unlockable on all of Microsoft's Surface devices and previous ARM outings

That doesn't quite seem accurate.

For one example: https://www.theregister.com/2016/07/15/windows_fix_closes_rt...


For all ARM64 Windows devices, UEFI Secure Boot is end-user configurable. And yes that includes Surface Pro X, Pro 9 arm and this devkit.


$699 for a Mini, $599 for this. The 32GB ram may be nice, but I haven't felt ram pressure on my M1 16GB even running Windows VM on Parallels once, with VSCode, Teams, ~20 tabs of Firefox, DBeaver, Outlook, and Kitty open at once.

Would like to see perf numbers. Could be an interesting box, maybe a nice home server.


that 699 mini also has a lot smaller ssd which is pretty important.


Yeah I am seriously considering getting one of these. 32gb of ram for 600 dollars plus its a decent specked arm machine with plenty of ports


I wish macOS just was Linux.


What more *nix do you want on macOS than is already there?

I can understand from the lockdown perspective (i.e. just give me a Vulkan driver on macOS) but as a polished DTE with a unix terminal I find it pretty decent.


ZFS or any other filesystems.

Package manager that controls the entire system, not just add ons.

Choice of desktop environments, no silly $3 apps to enable simple things like edge snapping. Options for tile based window manager, etc.

Decent cut/paste (no control-v and control-c instead of sweep and click to paste).

SystemTap for performance analysis.

Seamless use of containers with the same environment as the host OS.

Ability to create binaries that don't require the approval of some Apple run web service to run.


That being said BetterTouchTools and Keyboard Maestro are far above what you can possibly find in any linux software.

MacOS desktop by itself is pretty limited, but once you had these two softwares, the desktop experience become pretty nice, far above anything you find in any completed linux desktop.

But it is it. It's only a 'desktop experience'... Coupled with deep integration with ipad and iphone.

If you want more of a *nix integration, Brew is pretty nice and do more of less the job. As for container, dockers work seamlessly on the command line and integrate well with the file system, even if containers still run atop of a virtual machine.


>> What more *nix do you want on macOS than is already there?

Linux package manager (not brew)

Linux kernel with cgroups


MacOS is great and all but no Ctrl + Alt + T for terminal, no deal.


There must be a way to set up a hotkey for this right?


Not at the moment, I tried everything.

My 2 most important shortcuts: Ctrl+Alt+T for Terminal and Ctrl+Alt+C for VSCode are extremely difficult to achieve on MacOS.

This is why I prefer Linux, it has keyboard shortcuts that are more consistent and freedom that's important to me.


Try this: https://github.com/koekeishiya/skhd

If it’s customizable, global, keyboard shortcuts you’re after, skhd should do the job.


Thanks for the suggestion, I'll give it a try.


Bettertouchtools and keyboard maestro.


Asahi Linux is making decent progress.


How does battery life compare to macOS?


Have you tried Parallels Desktop?


If only MS was a larger company with more market push to get QComm to put effort in it or find another supplier... /s

Or maybe, in some twisted logic they want to sabotage their own ARM products in favor of Intel


The difference here is that the M1 in the M1 mac mini is faster than Apple's phone SoC (the A16). The Qualcomm 8cx gen 3 (what a terrible name) in this machine is NOT faster than even Qualcomm's phone SoC (the 8 gen 1), let alone Apple's phone SoC.

It's absurd to be selling a desktop PC that's weaker than a phone.


It's a chicken or the egg dilemma. Non-Apple ARM is too slow, so nobody wants Windows ARM, and nobody wants Windows ARM, so nobody is seriously working on consumer-targeted ARM. It will probably fail because Microsoft is generally bad at managing and marketing their hardware projects and coordinating with manufacturers and retail partners. Also, yeah, the pricing and performance is abysmal. But there is some sense to trying to lay the groundwork for future ARM SoCs running Windows.

Right now, the biggest driver for Windows ARM adoption is, ironically, Apple's M1 and onwards because of people running it as a guest OS through Parallels.

Also, frankly, I'm okay with Microsoft failing at projects like this. We don't need another Apple-like presence on the market.


The solution to chicken-or-egg problems is known: you spend money to just overpower the problem. But MS and Qualcomm aren't doing that; they're half-assing their ARM hardware to save a little money.


Does Qualcomm actually need to whole-ass a solution though? They're making gobs of money on mobile SoCs, any additional market for these chips is just gravy. Microsoft is the party that stands to really gain from a successful x86 alternative here but they don't seem like they're willing to pony up Apple or Google money to design their own chips yet.


Qualcomm is fully capable of building an SoC with eight X2 or X3 cores, for example, if MS is willing to pay for it. I think it's on MS that they didn't set higher performance goals for Qualcomm. Nvidia can also design good ARM chips (see Orin) but MS went and got married to Qualcomm (never do this!) so they can't use them.


Qualcomm's value proposition isn't the chip itself. It's the package, that includes the baseband.


Yup.

Compare this to Apple's $499 ARM Dev test kit that you could return for a $200 credit.

If Microsoft is serious about ARM, they need a very low barrier of entry for those willing to port their software.


Apple's ace card is that they bought PA Semi and have a competent team to pull off what Qualcomm apparently can't.


Well, the Nuvia acquisition seems to indicate that at least Qualcomm wants to. might be some generations before we see their designs.


Lol, Qualcomm is Oracle of hardware world. Buying a company is a no way an indicator that they want to do anything. They haven't delivered a good SoC since Apple released A7.


The Snapdragon 870 just to name one.


To name what? Subpar SoC that looses in benchmarks compared to older A14? If it weren't for patents Qualcomm holding, there would be so many good SoCs on a market.


Except for the fact ARM is suing them to prevent using any of the Nuvia tech in their chips, since the license terms with ARM didn't convey with the purchase.

There is a good chance they get on the verge of shipping and find they can't actually sell any of their new chips...

https://www.axios.com/2022/09/06/arm-qualcomm-nuvia-chip-gia...


...and for some reason I conspire Apple is behind this


>The solution to chicken-or-egg problems is known: you spend money to just overpower the problem.

Ah yes, "just spend more money" has solved so many problems over the years. Can anyone even name a single time a lot of money was spent and the product failed? Victory is practically guaranteed!


>I'm okay with Microsoft failing at projects like this. We don't need another Apple-like presence on the market.

Personally I’m not ok with being locked out of potential computing power by platform so I can’t get excited as MS struggles to keep up.

It’s either MS keeps up or the future ahead is dark.


> It’s either MS keeps up or the future ahead is dark.

You say that as if Apple will become some sort of PC-market hegemon, driving consumers and professionals to use its proprietary ARM hardware and OS despite having traditional Windows- or Linux-based workflows, simply because Apple's ARM is just so much faster.

Here's the third option: nobody cares what Apple is doing over in its corner of the PC market; the PC market remains an x86 market; and it continues to be driven by the needs of corporate buyers buying 1000+-part orders of PCs to outfit entire (non-IT!) businesses with; where those businesses don't care about having the fastest computer, but simply need "a" computer, with support and parts their internal IT department can swap out when needed; where the biggest factor driving purchases is TCO; and where TCO is driven down by commoditization and competition, not by vertical integration.


This still misses a large risk for Microsoft. Business software is moving to web apps. The orders of 1000+ PCs you've described will be soon replaced by BYO personal devices (who's owners prefer vertical integration like Apple for ease of consumer use) and Chromebook esque devices which are even cheaper than traditional PCs for a business.

Microsoft cannot keep doing what they are doing.


I wish Apple's proprietary computing platform keep niche but continue building good products. However iPhone become top share on some market and perhaps is it much higher share on enterprise?


PCs and "sealed consumer information-consumption appliances" (phones, tablets) are different market verticals — they will never converge, because the buyers of them have different needs that result in mutually-exclusive design constraints.

Even in Japan, where "nobody" owns a home PC, offices are still full of PCs; nobody is being handed a Macbook to do their work from, let alone a phone or tablet.

In my opinion, it's perfectly fine for Apple to win the "sealed appliance" space — since ~everybody else (your Samsungs, your Xiaomis, even Google) is just trying to do the same proprietary vertical-integration play that Apple is doing in that space; they're just worse at it than Apple is.

In contrast, Apple will never win the PC space, since their whole market strategy — "give people something different that forces them into our ecosystem" — is (and always has been) anathema to how boring, non-IT businesses want to use computers. Even companies whose workflows depend entirely on macOS "killer apps" (like desktop publishes back in the 1980s), who begrudgingly buy Macs to fulfill those needs, constantly lobby their app-developer ISVs to go multiplatform, so that they can toss out the Macs and revert to doing the same boring PC-centered IT that every other company around them is doing.

(And do note that Apple isn't trying to win the PC space. They were trying, historically — maybe up until the year 2001 or so. But ever since the iPod, and then the iPhone, Apple's strategy has moved to treating the Mac as a halo product category — a nice-to-have for those already in the Apple ecosystem for other reasons — rather than as something that's going to usurp the PC in its place one day. It's why, around that time, they killed XServe and Airport routers: they saw no further benefit in attempting to achieve corporate ubiquity, rather than complementarity.)


>> Also, frankly, I'm okay with Microsoft failing at projects like this. We don't need another Apple-like presence on the market.

Don't we?


I’ve gotta agree with you there. I love my Apple gear, but I hear and understand the criticisms against them. The above argument sounds to me like “you think one monopolist is bad? Wait until there are 2 of them!”


It will be an oligopoly then.


But wouldn't that be trending the right direction? Seems like a case of letting the perfect be the enemy of the good.


No, in the non-Apple world, Microsoft doesn't drive the computing industry. It's market-driven coordination between a multitude of manufacturers and retailers where no one instance has control over the others. Apple is a vertically integrated monopoly that effectively rules all its partners (from apps to hardware) with an iron fist and can push its decisions on users (and partners) unilaterally.

So no, I don't want Microsoft to become another Apple.

From my point of view, Microsoft failing to do what Apple does is a good thing because it means there's some semblance of a working, competitive (if imperfect) market. They can't force other companies to dance to their tune.


> so nobody is seriously working on consumer-targeted ARM.

Consumer targeted arm chips power most phones and tablets. Many of them are more than a handful for i5 and i3 class Intel chips and draw a lot less power. When you are talking about a $599 price point, you aren't talking about top-shelf Intel, anyway.

> But there is some sense to trying to lay the groundwork for future ARM SoCs running Windows

Now that ARM has grown up to be a viable alternative to Intel, this makes a ton of sense.


i3/i5 12 and 13 gen?

Outside m1/m2 there isn't any chip that can be compared to i5 intel 13 gen especially for PC space [that is in the same price bracket]. (maybe some arm chips can get close to i3 12100f but you can get it for 90 usd and get good single core performance so im not sure if in this price point is even any arm alternative)


what's the incentive for a chip manufacturer to put out a better ARM chip (eg comparable to the Apple M1)? I don't know that world well, so my best guess is that the margins go to the OS manufacturers, so unless you have guaranteed commitment from MSFT for a given volume, and some sharing of the margin, it's too risky to invest in the R&D to make a better ARM chip. The market for x86 chips is large and known, so for someone like Intel, it makes more sense to invest in the i3/i5 than in a new line of ARM. But that's all speculation - would love to hear the perspective of someone who understands the industry better.


> what's the incentive for a chip manufacturer to put out a better ARM chip (eg comparable to the Apple M1)?

Intel & AMD’s consumer and/or server market share.

I can’t imagine going back to an Intel after an M1. The battery life is better. It’s dead silent. It doesn’t get warm. It’s like a totally different kind of object.

I know PC people (those who want Windows) aren’t interested in an M1 Mac. That’s fine.

But I see PC laptop reviews with 4 or 6 or maybe 8 hours of battery life. They get hot but the fan isn’t “too loud”. And I know the performance isn’t the same.

And I just wish reviewers would call it out. They’re not on the same level. I’m sure fanboys would complain about the comparison in every review, but why shouldn’t Windows users have something much better? It’s been proven possible. Hold AMD/Intel/Qualcomm to the fire more.

If Qualcomm could get a chip with reasonable performance at a reasonable price that just doesn’t get hot and waste all its battery playing space heater, I bet they could really get a hold on the laptop market.


Proving your point, Amazon designed graviton, an ARM processor for their data centers.


Data centers know power/heat is everything. I’m not surprised they’re leading the way.

But as a consumer you’re stuck. You should have a machine with the performance of any normal/good laptop with way better thermals and battery life.

I’m not sure what the Uber-high end laptop would look like but surely it would do better than today.

The scale needs adjusting. What counts as “too hot”, “too short battery life”, “minimum performance.”

It all should have changed. But it didn’t. The industry acts like Macs are magic and therefor incomparable. “Of course that Boeing goes faster, it’s a jet engine plane. You can’t compare that to our cars.”

They’re both computers. It can be done. So why are Intel/AMD/Qualcomm getting off the hook to such a degree?

I just don’t understand it. It’s almost like Stockholm Syndrome or something. “Intel is nice to us, who are we to complain?”


>The industry acts like Macs are magic and therefor incomparable.

The cheapest M1 is $1300 in Canada. MacBook Pros are well over $2000. These prices are far beyond the budget of the majority of computer users. They are also beyond the requirements.

>Of course that Boeing goes faster, it’s a jet engine plane. You can’t compare that to our cars.

No, it's like comparing a BMW to a Honda. You spend a lot less money and then are surprised it doesn't perform as well?


I’m only trying to talk about just the chips.

Yes, the MBA is more expensive than most people would want to pay.

But it’s also not comparable to those laptops. It has a very high quality Retina display, the all-metal unibody construction, and Apple’s infamous profit margins.

There is no reason to think putting a similarly efficient processor in an otherwise standard PC laptop would raise the price much. Heck, Intel loves there margins too.

I don’t think the chip is too expensive. It’s all the other choices Apple makes that push the price up so much.


Apple has done pretty amazing things with the m1. IMO the most unique part is scaling memory bandwidth.

The vast majority of PCs are running 128 bit wide memory, with workstation CPUs like the threadripper (and pro) being the exception, but a VERY small fraction of the market.

The M1 has 128 bit wide 67GB/sec peak (that you'll never see) bandwidth, like most PCs. Upgrade to the Pro and you get 200GB/sec. Max will take you to 400GB/sec, and Ultra takes you to 800GB/sec.

On the Intel (i3, i5, i7, i9) or AMD (ryzen r3, r5, r7, r9) you get ... the same memory bandwidth. Check the 8 core vs 16 core scaling numbers and for most benchmarks you'll see poor scaling. Sure you can increase GPU performance by adding GPUs, which reduces (but not removes) the need for extra memory bandwidth. Sadly iGPUs (outside the XboxX and PS5) largely stink and are only good enough for non-GPU intensive workloads. Apple on the other hand does scale GPU performance, granted not to the levels that AMD and Nvidia do.

So why can't anyone in the PC space do more memory bandwidth and a decent iGPU, especially when for years the GPUs were in short supply and had exorbitant prices. I think it does come down to OS support, volume (which could be problematic if current GPU customers avoid you), and potentially reducing profits for AMD (who would have sold an expensive external GPU). Not to mention that fast/wide ram requires soldering chips on board or increasing size/cost with large banks of ram. Even servers with 8 memory channels (minimum 8 dimms) only get you to the M1 pro level (1/2 of the m1 max and 1/4th of the m1 ultra).

Apple can say we have X% of the market today, and all new customers will be on our new platform with 2 years, so the driver, OS, iGPU, memory bandwidth, etc will be amortized over substantial volumes. Additionally Apple gets a larger fraction of the revenue, since they aren't paying Nvidia or AMD for a GPU. Who is going to push a MBP or Apple studio competitor that could ship the same volumes?


Long term there is a huge market for server-oriented CPUs that can compete with Intel Xeon, which currently has a near monopoly in the server market. Note that a lot of companies are already working on this: Ampere (ARM), Amazon (Graviton3, etc.), and likely Nvidia, Rivos (RISC-V), etc.

The best of market ARM designs don't really compete head-to-head with Xeon right now, but there are still a ton of server applications where they make already make sense. As a simple example companies like Google and Facebook have hundreds of thousands of servers that are doing things like running memcached or running some application like D/GFS where the server is mostly just doing a lot of I/O and doesn't necessarily need really beefy single-threaded CPU performance.

Longer term obviously if there are ARM or RISC-V CPUs that can compete head-to-head with Xeon in terms of features and single threaded performance then that opens up pretty much the entire enterprise/server market.


Performance is irrelevant depending on metrics.

If my new zCPU chip is 60% as fast as a Xeon at your task, that’s a problem.

If it can do it at 40% of the Xeon’s power, things get interesting.

I could use twice as many zCPUs, be 20% faster, and use 20% less power. That also means less cooling capacity in my DC.

Some tasks will always need the absolute best single threaded performance. But a lot don’t. And the Xeon’s power requirements leave a large opening we’re starting to see other companies poke at with things like Graviton.


> . But there is some sense to trying to lay the groundwork for future ARM SoCs running Windows.

They've been doing exactly that since Windows 8, if not earlier. Perhaps the Year of Windows on ARM is somewhere around the corner from the Year of the Linux Desktop

I think we are on the ... third? ... attention cycle for this? Because they were trying pretty hard when W10 dropped too


The year of Linux-on-Windows-on-ARM Desktop.


> Microsoft is generally bad at managing and marketing their hardware projects and coordinating with manufacturers and retail partners

X-Box would beg to differ since 2002.


XBox is a great example of horrible mismanagement that cost Microsoft many billions of dollars, was almost abandoned, and went through a very rocky path to get to where it is today which is still frankly not that great of a position (it lags behind Switch and Playstation):

https://www.shacknews.com/article/121384/last-one-at-the-tab...

Don Mattrick almost tanked that entire product and it survives today because Phil Spencer miraculously managed to turn it around after all of the previous leadership was forced out of the company.


>XBox is a great example of horrible mismanagement that cost Microsoft many billions of dollars

Guess what, so was the first NES, and the first Play Station, the first Game Boy, etc. for their respective companies.

Breaking into a new market, with a new product, in uncharted waters, with no prior experience, with no support from clueless executives who don't believe in the new product looking for any reason to stop you from burning cash, endless turf wars such large and expensive projects create, makes it is hard, brutally hard, for any company to succed on the first try.

>and went through a very rocky path

The Xbox 360 sold 85 million units, one of the best selling consoles of all time.


The first NES and Playstation were MASSIVE successes for their respective companies, as was the Gameboy. In fact, the NES is credited with single handedly putting an end to the video game crash of 1983. The Gameboy sold out in a matter of weeks and Nintendo managed to sell every single Gameboy that it produced for the course of its first two years.

Here's an article about how massive of a success the original NES and Gameboy were and how it revived Nintendo as a company:

https://www.polygon.com/2019/4/19/18295061/game-boy-history-...

I have no idea where you got the idea that any of those products were mismanaged or cost those companies enormous amounts of cash and in fact those specific examples are among the most successful product launches in video game history.

>The Xbox 360 sold 85 million units, one of the best selling consoles of all time.

The XBox 360 is the best selling console from Microsoft and ranks 9th among all consoles behind the Playstation, Playstation 2, Playstation 3, Playstation 4, Switch, Gameboy, and Wii.

Read into that what you will.


>The first NES and Playstation were MASSIVE successes for their respective companies, as was the Gameboy.

I never said they weren't successful, I said they were also mismanaged during development like you said about the xbox, because management at Nintendo did not believe in the product.

>I have no idea where you got the idea that any of those products were mismanaged

Documentaries and war stories on youtube rabit holes.

>in fact those specific examples are among the most successful product launches in video game history

Today it's easy to say that with hindsight, but before they were launched, during their development, many in the company did not believe in those projects would succeed at all, leading to many internal fights and turf wars.

Also, Nintendo has a number of fuckups that bombed as well. Anyone remember the Virtualboy? Or the Wii-U? Gamecube also didn't sell too good.

The only thing that moves Nintendo merch is their exclusive IP (Zelda, Mario, Pokemon, etc), as their HW products are mediocre at best both in technical capabilities and in quality.


Can you please link to a Youtube video or article indicating that Nintendo management thought the Gameboy or NES would be a failure or that Sony thought that the Playstation would be a failure.

I am looking at some quick sources that I can find, and it looks like the complete opposite, that the management at Nintendo was very eager to develop a home video game console based on the success of their arcade games. They believed in the NES so much that when Atari bailed on its partnership agreement with them (due in no small part to the video game crash of 83), they went ahead and decided to do it alone.

Here is an article that was posted to HN awhile back that does a very deep dive into the development of the NES. It's an excerpt from the book "Console Wars" and it does not paint a picture at all like the one you're suggesting:

http://grantland.com/features/the-rise-of-nintendo-video-gam...

A relevant quote is:

"Yamauchi wanted Nintendo to aggressively get into the videogame business, which was really two separate businesses: home consoles and coin-operated arcade games. He saw the potential in these industries and took the necessary steps for Nintendo to enter both."

As an FYI, Yamauchi was the President of Nintendo.


I'm not sure what point you're making here. As the OP said, you took some anecdata about the Xbox's issues and speak about it like it was a disaster. They had problems, as did Nintendo and Sega (as talked about in Console Wars). It certainly wasn't all roses. Sega stopped building systems altogether!

The Xbox 360/One outsold the NES/SNES. That's amazing.


I'm not the one making a point in this thread, I'm refuting the laughably false point that the other person made that the NES, Gameboy, and Playstation were mismanaged and ended up costing their respective companies significant sums of money.

>As the OP said, you took some anecdata about the Xbox's issues and speak about it like it was a disaster.

What anecdata are you talking about? I linked to a source on the subject and not just any source, it's an interview with the head of XBox, Phil Spencer. I have also followed up every single one of my posts with a reputable source.

OP claims that they also have sources that suggest that Sony and Nintendo did mismanage the launch of their respective consoles in a manner that cost them significant amounts of money, so I'd like to review those sources.

The fact that he was actively replying to me until I requested some evidence to back his claims is a strong indication that those sources never existed.


>The fact that he was actively replying to me until I requested some evidence to back his claims is a strong indication that those sources never existed.

I stopped replying to you because I saw you have a bone to pick and I have a life to live rather than trying to win an online argument with you.

The sources are on youtube, feel free to use the search function and watch them at your own convenience and make your own opinion. Good day sir.


What a sad excuse of a reply.

Just have the decency to admit you have no source, maybe you misheard something, instead of doubling down on your ignorance. It's unbelievably pathetic otherwise.


> The Xbox 360 sold 85 million units, one of the best selling consoles of all time.

Define the best? It is 10th in a list of consoles that sold at least one million units. It sold less than console release later and went through even rockier path - PlayStation 3.


Surface products as well.


In a way, it's not much different than when they were porting Windows NT to Dec ALPHA. (But I suspect ARM has legs, where Alpha didn't)


Didn’t NT actually start on the Alpha and end up with x86 becoming a co-platform (and later main platform) during development?


Close. The original NT targeted an experimental Intel cpu called i860 (RISC based) nicknamed “N10”. NT was, according to Wikipedia, a play on words for the target cpu architecture.

From Wikipedia: “One of the original NT developers, Mark Lucovsky, states that the name was taken from the original target processor—the Intel i860, code-named N10 ("N-Ten")”.

NT was always supposed to be multi architecture (MIPS, Alpha, IA-32). I’m not sure what happened to i860.

Link: https://en.wikipedia.org/wiki/Windows_NT


That’s right. I forgot a third (failed) processor. Thanks.


Exactly. Apple have the advantage of being able to tell consumers that if they want a new Mac, then they get no choice. They'll have to buy an M1 or M2. Despite of this Apple knew that M-series computers needed to be better than the previous Intel lineup, and noticeable so.

An ARM based Windows computer isn't a bad idea, I just question if Microsoft is able to deliver on it. Picking the same manufacturer, who repeatably failed to deliver usable ARM processors for desktop and laptops seem like a obvious mistake. This isn't their first attempt either, so why would I trust that this won't fail, like the last time? Apple had done this before an architecture transition before, Microsoft haven't, and I doubt they have the will to ensure that it will succeed. They are too tied up in the x86 world, too busy with Azure and they don't have the attention of the consumer market.

In terms of price, it's really close to the Mac mini. Factor in performance, then this thing is a bad deal.

The form factor is right for many uses, but I don't get who the potential buyers are.


>> An ARM based Windows computer isn't a bad idea, I just question if Microsoft is able to deliver on it.

Microsoft has already executed and brutally failed with ARM in the form of the disastrous Surface RT/Windows RT.

But, and more importantly.

Two people with unlimited resources are running a race car race.

One starts a year or two before the other. Even with unlimited resources; the other racer/team has unlimited resources, too.

So let’s say they can both reach a maximum of the speed of sound.

Apple’s already been going the speed of sound for a couple years now; they also have the advantage that their vehicle in the race has software and hardware that are married.

Microsoft is not only nowhere near the speed of sound, but even though they also have unlimited resources; they are severely hampered by separate hardware and software, with the exclusion of their surface tablets, whose previous incarnation of this race car model crashed and burned on the race track.

This isn’t a race where MS can or will catch up. They’re already years behind. Not that I encourage anyone to use an OS with built-in ads anyway. Just use Linux at that point.


I think the primary reason that windows rt failed totally is no software can run on it at all. Not only you can't run x86 software. You can't even download random executable and run it. It is basically killed by ms itself. It is always a mystery to me that why ms would expect user to buy a device that nearly run nothing.

It looks like they want to address this. But I wonder if they will succeed this time.


>It is always a mystery to me that why ms would expect user to buy a device that nearly run nothing.

Because at the end of the day, Windows RT was a creature born of greed. They saw dollar signs- Apple's 30% App Store cut- and as such wanted a machine that forced you to buy software only from them. There was no technical reason that normal software couldn't run on Windows RT, given that MS themselves did it with Office.

So confident were they that this would work that they threw the tablet features onto Windows 8 proper, relegating the reason people buy computers to a secondary function- after all, paying MS for the privilege of developing software was going to be the New Way forward. Besides, don't you want security?


IIRC, Office RT was made as preinstalled Win32 app, not made on Windows Runtime.


If MS showed up with a backward compatibility layer as good as Apple’s Rosetta, they’d instantly be a strong competitor. Maybe not for gaming systems or high-end workstations — at least not immediately — but in the huge space of people who want a battery sipping laptop with access to a vast amount of software they’re already using.



I don't think MS needs to push this if they want to succeed, they need to start making a consumer friendly machine that Windows runs fantastically on.

For an "acceptable" laptop, the price point is already pretty close to $1000, and before I would have a tentative recommendation of MacBooks/Macbook Airs because of the learning curve of MacOS. With M1/M2 and how much better it is than anything else on the consumer market, I openly recommend it to anyone in the market for a new consumer machine. Gaming isn't even that much of an issue anymore, so for casual players it's pretty fine.

I was discussing this with a colleague last night, but the M1/M2 chips and complimentary hardware let Apple do some amazing stuff out of the box without adjustment that Windows simply has no answer for. The integration of the complimentary hardware with the M1/M2 chips is so strong that I stumbled onto features I completely missed announcements on, and it legitimately "wow'd" me.

- Live Text caught me off-guard while drag/drop-ing an image to a chat app. I couldn't stop testing its limits and reading the dev docs

- I took surprise calls from really crowded + noisy places and was in disbelief that my call partners couldn't hear anything but my voice in crystal clear quality

- I ran games and software that just weren't possible on Intel Macs through Rosetta at pretty fine FPS/quality without incident

- I didn't need to change a single program from my workflow

Microsoft can likely do the same but they need to put the legwork in to make it happen. Personally I understand they have no interest in this and it makes sense -- they want you on Azure with your server workloads and this keeps the lights on at Microsoft, and as best I know the consumer market (not considering gaming) still favors Windows. But I guess that's why projects like this confuse me a lot since it must be a pretty substantial RND and manufacturing cost, neverminding advertising, but Microsoft doesn't seem to have their heart in it.

It's not about backwards compatibility - consumers don't need to keep Windows 3.0 apps running, not a statistically significant portion anyways, they just need modern apps to run fast and well, long battery life on portable devices, quiet machines, and that's it, but seems that this just isn't something Microsoft is interested in taking over.

I really can't think of Windows features in decades that "wow" so much as you just know what you get with Windows regardless of the version in terms of basic features; what worked on Windows XP probably works on Windows 11, but even that is starting to erode in a slow and painful way. There are quite a few programs on Windows I get the impression that Microsoft just doesn't want me to be running, but things like the Windows Store, Windows' implementation of security for unsigned apps, etc, these all feel like Microsoft isn't confident enough to fully invest into these new features or to drop them in order to advance.

Microsoft definitely has the talent and cash reserves to pursue a strong consumer laptop to compete with Apple; for whatever reason, they don't seem to have the interest though for consumer devices. Probably the simplest reason is the server market is theirs and this is plenty of money, but I just can't get why they continue with such forays then.

Edit: just elaborated on price point for consumer laptops and recommending machines.


> Gaming isn't even that much of an issue anymore, so for casual players it's pretty fine.

I don't expect "casual" gamers setup Wine+Proton+MoltenVK. It's better to say that let's play game on PlayStation. If you mean really "casual", they play on iPhone.


Cloud gaming is an option as well.


Ah that's good point for who with great network connection.


8cx gen 3 is definitely faster multicore and about the same singlecore. The 8cx chips also have a wider memory bus and pretty massive GPU in comparison the the mobile parts. Looks like also far better cooling than a phone or thin and light laptop here as well.


Yeah I don't know what the parent is talking about Gen 3 is lightyears ahead of Gen 1 of this line of chips. Passmark has it at 4k for Gen 1 and 11k for Gen 3.


If you're referring to my comment, I'm not comparing the 8cx gen 3 to 8cx gen1, I'm comparing 8cx gen 3 (this desktop/laptop chip) to 8 gen 1 (QCOMM's flagship phone SoC).

As I said in my comment, QCOMM's naming sucks. cx is their desktop/laptop line. They have no designation for their phone line, so the confusion isn't surprising.


Prefer 32GB than a better CPU.

By the way When the fuck did 32GB become an entry requirement, I am so saddened by the crappy software and stacks that treats memory like an infinite resource


Agree. Further I nowadays see crappy software developers instead of being apologetic or at least modest, claim some kind of moral high ground along the lines of "This software wouldn't even exist if not for our accomplishments, so be thankful to us"

I'd hope some one have already or will write thesis on correlation between Rise of Javascript stack and narcissism in software industry.


The "beggars can't be choosers" attitude probably correlates with a lot of software becoming $0.


And just for comparison, the cheapest Mac mini with 32GB RAM is $1699. I can forgive a slower CPU in a machine that's about one third the price.


presumably they want it to be able to run Teams


Yeah, Where would computer industry be if physical collaboration with open office setup and online collaboration with teams/slack does not take place?


GP was a dig at Microsoft Teams being utterly bad software that not only uses absurd amounts of memory for a chat app but also hogs all kinds of resources while still being generally laggy on high-end machines.


The only dev environment that I know of where 32GB is an entry requirement is Professional (i.e. non-trivial, non-hobby/learner) Android development. You certainly do not need it to develop Windows applications.


Being old enough to remember when 32K was generous space, memory is an infinite resource.


What Andy giveth, Bill taketh away. Blame Microsoft for that.


I think the point of this machine is to give Developers a test bed for apps they port to ARM Windows.

However, given that it took Microsoft more than a decade to decide to port their own Visual Studio to ARM Windows, I'm not sure why they think third parties are chomping at the bit.

If Microsoft wants to copy Apple, they need to copy the decision to immediately port all their first party software.


> the point of this machine is to give Developers

At 700€ price tag what a gift! It's 3-400€ too expensive for a very dispensable toy, especially given that the managed stack (.net) of Microsoft development tools can be tested on Raspberry Pi or M1, which are both very popular with developers.


Yes, I've mentioned elsewhere that they really should have gone all in on copying Apple and not only make the Dev kit cheaper, but issue a credit on future hardware purchases if the Dev returns the Dev kit when they get done porting their software.

It's not like you want to keep something this underwhelming forever at that price point.


Why would developers want to waste time porting to ARM Windows though?


A Mac mini with otherwise comparable specs (32GB RAM, 512GB SSD) is nearly three times as expensive as this Microsoft machine. I don't think it's fair to expect a CPU that beats the M1 or phones that cost more than $599.


I'll be curious to see benchmarks but it would be funny if Windows for ARM running in Parallels on an M1 MacBook Air ends up being a better dev machine than an official Windows for ARM dev box produced by Microsoft.


on Vmware fusion tech preview...can't say I've pushed its performance, but it is really quite smooth so far...


I hadn’t thought of that - but holy shit; I’d put money on that being the case. How embarrassing.


It is a dev kit, for preview, I'm guessing them going with cheaper, available, to get software builts out that perform on shit hardware, is better than developer on better than consumer grade equipment only to have the applications choke when consumers have it.

I admit I'm disappointed in the showing, and I think that Microsoft not loosing out on this market is important to them. I'd be willing to be bet 2nd gen of this will likely be produced by a 3rd party vendor.


The Apple dev kit was about two thirds the speed of an Intel Mac Mini. Microsoft or Qualcomm hasn’t shown us what their production PC Arm chip will look like.


Production ARM Surfaces have been out for a while; they have lower specs than this dev kit.


I assume this dev kit is leading up to an announcement of significantly more performant PC hardware. If I’m mistaken, that’s disappointing.


M1 has 4 high performance cores called firestorm, and 4 energy efficient cores called icestorm. Same cores that are in the A14 which is in the Iphone 12.

Since then the A15 came out with some efficiency and performance improvements, it's in the M2 IPad and M2 MBA and presumably several future apple products.

The A16 has some efficiency and perfomance tweaks and is what's in the recently released iPhone 14 Pro and Pro Max.

So sure, the M1 in the M1 mac mini might have more power, cooling, and cores than the iPhone 14 pro, but the cores are actually slower at the same clock. Sure a phone will hit thermal limits sooner than a SFF PC.


Will this Windows Arm only get three years of updates, just like a Qualcomm phone?


Sure, Apple's got the superior ARM silicon. But an alternative take is that this SoC is plenty fast for non-poweruser desktop use (which should be the majority of users).

You don't need top shelf performance for browsing the web, checking your emails or writing some docs.


If people didn't want top-shelf performance why would they be using a desktop, instead of a phone, tablet, or laptop? Isn't performance the whole point of desktop?


Or screen size.


> You don't need top shelf performance for browsing the web, checking your emails or writing some docs

DIDN’T

Until electron


It really depends if they did their homework related to thermal management and ventilation. Some phones have relatively fast cpus but they throttle so quickly because of temp that power can only be used in short bursts.


This Qualcom chip is very similar to the Microsoft SQ3 that is available in the new Surface Pro 9.


Firstly its a dev kit, not a retail product.

But most people only need a browser and maybe a few simple products anyway. A phone SoC is probably enough.


Let's not forget that Microsoft did not put in the work to optimize for Arm like Apple did.


At least according to this related blog post[0] (and submission[1]), they've put in work, albeit not as thorough and effective as Rosetta 2.

> To boost performance, we have added vendor-specific optimizations so your apps run well on a variety of Arm hardware. We have several runtime improvements to targeting server throughput (RPS) and latency.

Seems largely focused on .NET 7 though[2].

[0] https://blogs.windows.com/windowsdeveloper/2022/10/24/availa...

[1] https://news.ycombinator.com/item?id=33319535

[2] https://devblogs.microsoft.com/dotnet/arm64-performance-impr...


Windows 11 can run AMD64 applications on ARM64 through a JIT, similar to how Rosetta operates [0]. I don't know if anyone has tried to compare them in benchmarks though.

[0] - https://blogs.windows.com/windows-insider/2020/12/10/introdu...


Arm64EC is also looks good


I'm surprised at the price for a 32GB RAM device. I'm tempted to get one just to double the ram of my desktop, despite probably being slower than my i7-3770. It might be a good platform to work on Scalable Vector Extensions (SVE) code without needing to rent a Graviton from AWS.

Anandtech about the cpu: https://www.anandtech.com/show/17127/qualcomms-8cx-gen-3-for...

Some info about the ARMv8.2-A Architecture (2017): https://community.arm.com/arm-community-blogs/b/architecture...


Funny, I have an i7-3770K as my gaming rig and to this day still play Warzone regularly on it. Such a workhorse for being 10 years old.


It appears the consensus is HN is hesitant about the Windows on ARM push. I really want this to succeed.

If this gains traction then we should start seeing support for windows laptops on ARM. As a dev who prefers windows but has a m1 MacBook b/c of the battery life I really hope this works out.


Yeah, I'm sitting here shocked at everyone's hate. I took a major leap and bought a Thinkpad X13s, with the same 8cx Gen 3 CPU this thing has.

I am beyond impressed at it. Windows 11 on it - while, yes, it's Windows 11 with its own concerns and issues - runs flawlessly. I've been able to run my traditional x86 and x64 workloads - even things like arm64 Tailscale without issues, and I get amazing battery life.

I am 100% for this thing. It's value for money in the high-performance ARM Windows rig market. It benches similarly to a ~11th gen i5 and runs the part.


There's a lot of the "it has less geekbenches than the M1 so it must be a total piece of crap" going on. I have a far less performant 8cx Gen 1 laptop with similar experience to you. It's no workstation but general usage and browsing work as well as you'd expect a modern laptop. With great battery life and total silence.

This thing is basically the most performant ARM soc on the market with a lot of memory, fast storage and good connectivity. Not bad at all for the price. Not to mention there's a good chance this might be the first cost-effective ARM Linux workstation.


> There's a lot of the "it has less geekbenches than the M1 so it must be a total piece of crap" going on.

It's not that it benches worse than an M1, but rather that it doesn't do well against comparable x86 laptops. The whole reason M-series sells as well as it does is because people don't feel like they're giving up generations of power increases by buying them.

Low power ARM laptops have been a thing for a long time in the form of Chromebooks, and adding Windows to the equation alone isn't enough to get most people to take notice. For Windows on ARM to work, the hardware it's running on has to be at least as powerful as mainstream x86 laptops.


I love windows and wanted to give it a try, but the lack of OLED option is bothering me.

Can you comment about the battery life and the "traditional x86" workloads? Does it use a just-in-time compilation? (I haven't looked at the details yet)

Can you dual boot to a Linux arm64 distribution?


> Does it use a just-in-time compilation?

It's a caching JIT that'll squirrel away the translations to use across program invocations, and even shared between different programs in the case of .dlls.

https://learn.microsoft.com/en-us/windows/arm/apps-on-arm-x8...


> I love windows and wanted to give it a try, but the lack of OLED option is bothering me.

My Dell XPS15 has an OLED, fwiw. It's a couple years old now but the latest models still have it (if you buy a 4K screen, anyway).


Thanks for the suggestion, but I don't buy dells anymore, too many problems with the BIOS.

I'll stick to Lenovo for a while!


There are Windows ARM laptops already, and their existence does a lot to justify that hesitancy.


Of course Qualcomm being chronically incapable of making fast CPUs doesn't help, but partly it's just a chicken-egg problem of lack of ARM-optimized Windows software caused by the lack of ARM Windows users.

Having a decent devkit may help bring more native ARM software, which may make these underpowered machines struggle less with speed and compatibility.


Genuinely asking because I don't know. Is there anything that's better about ARM or is the Apple/TSMC being better than Intel, AMD, Qualcomm, Samsung, et. al.?


Its less about the instruction set and more about the specific implementation choices and market they're competing in. Apple has been doing ARM SoC design for a long time and was an early 64-bit adopter, buys tons of bleeding-edge capacity from fabs, and is willing to design in more expensive features like tons of onboard cache that they can make up on margins in other ways because Apple sells complete devices, while Qualcomm etc are mostly in the business of selling chips. So while there are fundamental differences between x86 and ARM, the bigger differences are more "Apple vs other ARM SoC designers" than something fundamentally different about ARM that Apple takes advantage of.


1 - The instruction set is more interesting because it is smaller.

2 - Arm also implements weak ordering which is type of memory model. It allows instructions to be separated into groups based on whether they affect other instructions. This allows these groups to skip some waiting lines. (Magic). X86 has strong ordering, which can make it slower in specific scenarios.


3 - The instruction set is made of fixed length instructions

TBH there are various mostly-Apple/TSMC-"exclusive" tricks too, for why their chips are better than the others:

A - On Apple Silicon, pages are larger, but not too large

B - there are various accelerators leveraged by libraries provided in the OS (or the provided toolchains, etc.)

C - Apple got to use the best TSMC process years before the competition.

D - TSMC is ahead (I'm curious of what Zen4 will give on laptop, btw)

So it's mixed. The ARM ISA probably plays a small role in the perf of Apple Silicon vs. x86 chips, but is probably not the main cause of the perf gap.


If the company is selling just the chip, it's in their best interests to hold the die size down as much as possible because doing so directly effects their profit margin.

If the company is selling the whole device, including the chip, they can afford to throw more transistors and die area at performance with a much more minimal impact on their bottom line.


TSMC tech helps a lot, but is not exclusive to Apple. A relatively modern and sensible 64-bit ARM architecture helps too, but on top of that Apple seems to be really good at designing the chips.

OTOH Qualcomm is constantly lagging behind Apple, promising that next year they will be almost as fast as a previous-gen Apple chip. The most generous explanation I've heard is that Qualcomm is specializing in quantity over quality, and focuses on making CPUs cheap, rather than fast.


Actually, TSMC reserves the lowest node size to Apple (I don't recall for how long their deal will last). AMD's processors are about equivalent despite being on a larger node - so Apple actually didn't do a great job, but a lower node size is a lower node. That's good enough for users who (correctly) don't care about the why but just want to use their computer.


Previous laptops and current surface devices do support ARM, but IMO nothing MSFT has put out has been serious. (Looking at you Surface Pro X). This targeted solution correctly identifies the main problem with Windows on ARM. Lack of developer support. Outside of internal MSFT dev support no one has taken Windows/ARM support seriously. I hope these dev kits represent a wider market push that ultimately leads to better Windows laptops on ARM. (Ones that can close to gap with the current MacBook lineups).


The new Surface Pro 7 ARM looks serious — it does seem to be the first serious ARM entry, though.

Lack of Windows ARM devices for dev really has hamstrung Hololens 2 development — it’s also Windows ARM and there are very few libraries with Windows ARM compatibility. I altered many libraries myself, despite that the documentation was nearly nonexistent.


I’m out of the Windows ecosystem, except one machine at work.

I’d love Windows on ARM to work. I love my M1 and would like something comparable.

But Apple had the M1 and the ARM PCs seem to be based on iPhone 7 level chips (perceptually).

“You can have Intel or ARM. No one buys ARM so there is little software. Windows was ruined on ARM for a long time. Your ARM laptop will be way slower, but it will be cheap because the only ones you can buy are ultra-low spec with terrible components like eMMC storage.”

There is absolutely no compelling reason to buy one for any reason and they’re not fixing it.

The first MacBook Air was a horrible computer in many ways, but it really excelled in one that mattered to people.

These don’t have that one thing that matters. They’re just ‘meh’ computers that don’t run much. MS can’t fix it (without their own chip) and Qualcomm seemingly wont either.


> Your ARM laptop will be way slower, but it will be cheap because the only ones you can buy are ultra-low spec with terrible components like eMMC storage.

It's not clear what you're quoting from. But it's out of date, now. The last generation (or two?) of windows ARM laptops certainly have NVMe storage. And they're not generally cheap (in cost nor components).

> MS can’t fix it (without their own chip) and Qualcomm seemingly wont either.

Well, launching these laptops without a native Chrome build was a bit of a blunder IMO. That was ~2-3 generations ago and while Google might not ship Chrome for windows-arm64 (last I checked), you can download native binaries of Chromium.

IIRC MS does somehow co-brand w/Qualcomm for the SoCs used on the ARM Surface. But I kinda doubt Microsoft will start designing their own.


MS has been credibly rumored to be designing their own ARM processors for both Azure and the Surface lineup: https://www.theverge.com/2020/12/18/22189450/microsoft-arm-p...


> But I kinda doubt Microsoft will start designing their own.

There were job postings a few months back which would indicate you’re wrong, but who knows if those positions were filled, or if the project is still alive internally. We’ll see, I guess.


I have seen a lot of comments/discussions about battery life of laptops but my opinion is that laptops are really bad ergonomically for any kind of work that goes beyond few minutes. I always recommend people to attach the laptop to an external monitor, keyboard and mouse. It seems that people do use laptops for doing work on the go and that too for long hours.


I work remote, if I want to do anything computer related outside of work then it basically *has* to be on a laptop. I just can’t commit to spending more time at my desk for personal projects / etc. An efficient/fast laptop that I can use wherever (coffee shop / GFs house / Parents) is a game changer for me. Every Windows laptop I’ve owned has required to me be tethered to a wall socket after 1-2 hours of use. These MacBooks literally last an entire weekend of moderate - heavy use. It’s truly wild and is an experience that I crave for a windows laptop.


I definitely prefer having a dual monitor setup with a proper keyboard and pointing device at my desk, but my 16" MBP is quite serviceable when I want to sit elsewhere in my house for a change or take my work with me.

Now for 12"/13"/14" laptops, yes I agree that extended dev work isn't the most practical, but then again I spend most of my day with IDEs and simulators — it's probably a different story for someone who lives in a terminal or spends most of their time writing prose.


Once bitten, twice shy?

MS released Windows on ARM with Windows 8. It them abandoned it and the users.


Constant length instructions.

Chip design that's easier to put things like memory into an SOC.

Windows on ARM can only be good for computing.


Yeah I want arm windows exclusively so I can install a pirated version on arm bootcamp


I want to install Linux on one of these and use it as a low power consumption home server.


I have a N6005 box from AliExpress ($180 bare, about $300 with 32Gb RAM and a nice 512Gb M.2 nvme) that runs a bunch of workloads (Proxmox VE on the metal, opnSense in a VM doing all my routing, plus 4 LXCs with 20+ Dockers for various services).

Just looked over the energy report and it's averaging 0.18kWh daily (so the average power draw is 7.5W) - that's with a 3.5in SATA HDD. Better power use than my Raspberry Pi 4 with an external USB HDD!

If low power consumption is your main motivator, I'd say you can do a lot better with one of those instead.


Same here - with 32GB of RAM I could easily replace my main home server (which runs NixOS, so I can easily just rebuild the system for aarch64). I wonder how quickly people will get Linux running on these...


The M1 can run Asahi Linux, which I imagine is going to outperform this. https://www.androidauthority.com/snapdragon-8cx-gen-3-intel-...


Exactly, this is priced very similarly to the Mac Mini with M1 (soon to be updated to use an M2) and can already run Linux fairly well.


I wouldn't say the pricing is similar. This retails at $599 with 32GB of RAM. The Mac mini is $1,199 with 16GB of RAM.

edit: Was off on pricing of the Mac mini by $100.


Mac Mini has no option for 32GB of RAM, and with 512GB of SSD storage is $899. How is that similar in price to $599?


There are two nines. Pretty similar, I'd say.


Performing a substring comparison on integers isn't a useful operation.


It looks like a macmini


It tries really hard. The previous one looked like a Roku.


I was going to say you should just use a raspberry pi, but you can't get those any more anyway. :)


And RPIs are not very good regarding power consumption, neither idle power nor efficiency.


I don't know why you would buy one, even if it were available, unless power-efficiency was a crucial component for a server. Can you imagine what you could get, on eBay, in a small form factor, for $600?

I made off with a i3-8100T (about 3/4s as powerful, 35W TDP), with 16GB of RAM, and a 256GB SSD in an ultra-small-form-factor case for less than $150. For $600, it makes absolutely no sense against anything used. Let's say it drew 50W vs 20W total use at 10 cents per KWh. It would take 5 years to save $130, if you ran it 24/7/365. A $450 used Intel small-form-factor system would run circles around it in performance (especially after any emulation / code conversion) and you'd break even.


Imagine for a second there are other countries in the world.

Now let's say in Europe, just one of them was called "The United Kingdom", where after a recent (temporary) energy price cap, electricity prices rose to _only_ 36p/kWh (41 cents US at current rates), and further rise are expected, and the cap had an end, bringing us potentially to double the current uni rate, then do the maths again and see why it might be a "crucial" component for many.

Running old, cheap hardware with high power usage has been impractical here and many other parts of the world for quite some time and that was before recent disastrous rises.


Problem is, we are not talking about 10 year old hardware. I find it sadly more likely that hardware from 5 years ago will consume less than hardware from a couple months ago. I myself have an x86 atom where the entire system can idle at <2W, which is no easy feat unless you start reusing phone hardware...


More like 30-50c

https://www.eex.com/en/market-data/power/futures#%7B%22snipp...

Anything power hungry gets really expensive quickly for home use, these days.


Fair point. But idle power consumption will be like 7W vs 15W.


Would you mind sending a link to the i3-8100T box you bought?


Snapdragon chips have pretty good Linux support. And the GPU driver is in a good state with Freedreno for desktop use. So probably a decent chance of it being functional shortly.


From what MS says you cannot do it out of the box.


You'll need to wait a bit until a new device tree comes for these devices. Given that support for the ThinkPad X13s with the same SoC is coming along... going to happen pretty soon.


For others similarly interested, it turns out "X13s" is not "multiple X13 models" but rather "X13s" is the model number: https://www.lenovo.com/us/en/p/laptops/thinkpad/thinkpadx/th... and is currently listed at USD$995 for "Gen 1" and all the way up to USD$1570 for the high end


Which things currently work and don't with X13s?


Secure boot is going to be forced on.


which Linux can run on perfectly fine.


Depends if it will enforce Windows' CA only, or if it will trust UEFI 3rd party CA too. See also "Secured-core PCs".


Despite much FUD spread in Linux nerd circles online, Linux can run just fine on it.


Do you have a citation for this? I'm marginally interested in this device, but only if I am certain I can run Linux on it.


I have a Surface RT which disagrees.


In case it was not immediately obvious, this discussion is not about the Surface RT, but about the Windows Dev Kit 2023. It ships with a different Windows version, on a different Arm architecture, with different policies regarding locking down.


And? "In case it was not immediately obvious" by me specifically pointing an MS ARM device, MS's "policies" when shipping ARM devices have been much more locked down than x86. Around 50% of ARM devices released by MS so far have been completely locked down, incapable of running anything except for some version of Windows. Not just "FUD by nerd circles".


All ARM64 Windows devices have been "bootloader-unlocked". That has been the case for the last 5 years. Windows RT and Windows Phone devices are literally irrelevant to this discussion regardless of your anecdote.


In the past 5 years, how many ARM devices has MS released, other than the Surface Pro X ?

My "anecdote" is actually the majority of MS's history so far. You just can't claim it's scaremongering when they have released so many locked down ARM devices it's hard to remember all of them, and definitely easier to just remember the few "unlocked" devices they have ever released (just the X?).

And also "obviously" we should mention that even in the so-called unlocked Surfaces you are still forced to entirely disable Secure Boot in order to run Linux or _anything else_, with the consequences that implies. For example, it is dubious you will be able to load future versions of Windows, since those will require SB to be on (W11 already officially does, even if it doesn't seem to enforce it -- yet). A properly unlocked device would allow you to load your own signatures while keeping SB enabled.


> In the past 5 years, how many ARM devices has MS released, other than the Surface Pro X ?

Two Surface Pro X and one Surface Pro 9. I intentionally said "all ARM64 Windows devices", without restricting it to Microsoft ones, no matter how many there have been: it remains the current platform-wide policy, regardless of the FUD.

The rest of your comment is either strawmanning or shifting the discussion further away from where it started and I don't find it worth responding to.


> Two Surface Pro X and one Surface Pro 9.

Two devices compared to dozens of locked ones including the original Surface as well as the 2nd RT.

> I intentionally said "all ARM64 Windows devices", without restricting it to Microsoft ones

And how are non Microsoft devices relevant to a discussion of a Microsoft device? You made the same point before...

> The rest of your comment is either strawmanning or shifting the discussion further away

In case you didn't notice, all your 3 replies so far have been filled to the brim with _accusations_ of FUDing, irrelevance, strawmanning and whatever without even mentioning what is irrelevant or why. Everything I say is apparently an irrelevant anecdote, even thought it is actually the historical trend of MS ARM released devices, and the anecdote if anything is the last device only. Why are your arguments any better exactly?


Very interesting to see that it has a built-in NPU. Such co-processors are increasingly being included in recent high-end mobile phone platforms, used mostly to accelerate image processing algorithms.

Seems like this is a sign of things to come in Windows platforms too, since Microsoft are including an NPU in this devkit. The PC of the future may well end up being CPU+GPU+NPU as standard, much as it is CPU+GPU today.


I’m almost willing to bet this is early look at what a ‘native’ Microsoft Teams Room device is. Currently Microsoft relies on a partner ecosystem for compute in these, but given the push into more features built around real-time image processing, dealing with multiple cameras, multiple audio channels and other latency sensitive workloads this seems like a perfect platform to distribute that.


Honestly, if Apple were to care, they could undo the course of computing history where at one point Windows became the monopoly.

Launch a 500$ category laptop, say 3 gens old. That would soon be M1. Won't be very profitable, but a direct competitor to the typical crappy Windows low-end laptop.

Yet it's so much better in performance, security, battery life, total lifespan, etc. Double down on making it idiot-proof. Optimize the onboarding experience for ex-Windows users, including prominent placement of alternatives for popular W32 software. Make sure Office works well, and so on.

I would admit that it would take many years to make a dent, but there's no rush to it. Windows seems a sitting duck. Nobody, including Microsoft, seems to care about it.

Edit: oh yes, forgot about gaming.


> I would admit that it would take many years to make a dent, but there's no rush to it. Windows seems a sitting duck. Nobody, including Microsoft, seems to care about it.

Microsoft are too busy making money on Azure and M365 subscriptions for the hundreds of thousands of companies heavily invested in Microsoft apps. Windows is just one platform to get users to those apps. Microsoft are pushing Intune/MEM so organisations can push Microsoft apps to all devices including Apple ones. They don't care about the OS because it's not important.


I think the build quality of the thing might be more of the cost than the difference in power. Could be wrong, though.


They could make a m1 SE in late 23 with a plastic case as old MacBooks, same ports of m1 air, slightly less battery and 1080p screen at 500$ and it will sell like crazy.


Why would Apple jeopardize its brand (premium products) when they are already printing enough money?


Their consumer base is shrinking even as it becomes more profitable. I doubt they have the nerve to cannibalize their high-end branding and profits, but if current trends continue then they're in trouble in the long run.


I think the only problem with this is that that Apple permanently switched to higher resolution screens, text looks awful at 1080p since Apple dropped font anti aliasing. High DPI screens are a requirement.


Jeff Atwood 5 years ago: "the Qualcomm hardware people told Google internally they were 5 years behind Apple on hardware. This was about 1.5 years ago, but I believe that's about right." ( https://twitter.com/codinghorror/status/1050538041796788225)

And benchmarks continue to show: https://twitter.com/slightlylate/status/1584350796233117696


Someone had a battery life test for phones. The iPhones beat the android phones in nearly every test by hours. And they only had 60-70% of the raw battery capacity.

And you know they have faster processors too.

It’s so weird to me. The A series is kind of “that doesn’t count” because the competition can’t get close. The M series wowed people and then we all went back to normal AMD/Intel stuff like nothing happened.

If I want a new PC laptop, it just won’t compare because no reasonable part is available for the heat/battery life I could have gotten.


Who went back to AMD/Intel stuff? Any Apple user is far away from that, and comparing the performance, heat, noise, and battery differences between my Intel and Apple Silicon computers, I have no desire to regularly use an Intel machine until they make some stratospheric leap.

(I still need it for some x86 vagrant boxes I have though!)


I did. I moved over to linux mint for at-the-desk programming (though I still use a macbook while traveling). My linux desktop idles at 0% CPU, which is something macos hasn't been able to deliver for 5+ years at this point.

Current generation Intel / AMD desktop CPUs have similar performance to apple's M1/M2 chips. They just consume a lot more power. (Though apparently applying a slight underclock makes a huge difference.)

I'm not sure how recent intel / AMD laptop performance compares, but a PC with an x86 chip is still a great choice.


> They just consume a lot more power

This is the thing I’m stuck on. Computers are all roughly the same speed, for the point of this discussion.

But there is no contest in power. If you’re in the Windows world you have no good choices. Fast and hot to very slow and cool.

Where is the “pretty fast and nicely cool” choice? Someone proved it can exist.

But it’s not there and people don’t seem to be very upset about that. And I’m surprised.


Not upset but resigned I guess. Making do for now - typing this on my Macbook but my main desktop is Windows as is my work Laptop.

Could Windows have better battery life? Probably only by dropping backward compatibility and moving to a Linux kernel or something radical.


A lot of stuff on MacOS ran faster on an M1 with Rosetta 2 than it ran on the Intel chips Apple was selling the year before.

For something not highly performance sensitive like office suite or line of business stuff you’d probably be fine with a half-decent ARM chip.

So much of what we do these days is a web browser (ported) or a secret web browser (Electron/etc). I work in Java, that can be ported if it’s not already. I bet C#/CLR already are.

It’s not perfect. It’s probably not ready for 3D rendering or 4k video editing.

But a HUGE percentage of people could probably go ARM tomorrow if good chips were out there. MS has their own “Rosetta” already for running x86(-64?) on ARM. Is it slow?

From my (very weak) understanding I have the impression the ARM stuff available is really cheap/slow. Not unlike the performance of the first netbooks compared to real laptops. Things that just aren’t powerful enough.

Good for MS for making a dev machine. But a few good general purpose laptops with great battery life would probably move the needle a ton, even if only in the business market.


Yep. I'm curious if microsoft ever takes the reverse engineering work from Asahi linux and uses it to make a native Windows for Apple Silicon. Or just work with Apple like they did for Bootcamp and port the real Apple drivers to the windows driver model.

Apparently there's nothing stopping M1 macs from dual-booting other operating systems. As strange as it sounds, Apple laptops might make the best windows laptops you can buy today, so long as you don't need a powerful GPU. It would be like a reverse hackintosh.


The entertaining thing here is that Apple’s walled garden is intended to keep people in, but here it’s keeping them out. The M1 is a technical marvel, but if you’re not already tied to Apple, it’s nowhere near good enough to warrant leaping that wall.


But isn't the walled garden in thsi case part of the appeal? i.e. running windows on it bare metal would not be as performant because of all the tighly coupled hardware-software integration apple do?


The tighly coupled hardware-software integration is mostly a myth. Linux is super-fast on Apple Silicon and Windows or Android would probably also benefit.


"Apple should adopt Dell's business model, it will work briliantly" is a thing that Apple have heard for a long time.

When they tried in the 90s they went broke and had to be bailed out by Microsoft.


I meant the rest of the industry. They said “wow” and then went back to business as usual. They had talking points about how they’ll improve and performance (as always) and efficiency got talked up more.

But when will any of that show up? Hardware takes time but I just don’t see reviewers demanding it. They’ll be happy if it shows up but they’re not calling out “this isn’t good enough” and I just don’t get it. How can they use an M1 and go back to a hot low battery life Intel laptop in the same price class and say “this is fine?”


It's surprising that efficiency is barely mentioned still, in all the latest AMD / Intel reviews, maybe it's even a point that early reviewers have to avoid or something?

To me, the performance per watt is increasingly important not only as we start approaching branch circuit limits (1200-1500W max in the US), but also as (in many places) energy prices continue to rise.

Both idle power consumption (e-state, core sleeping, etc.) and max power consumption (with a few measures of work per watt) should be highlighted in any serious review, probably on par with whatever other benchmarks (like Cinebench, Blender, etc.).


Hardware Unboxed is the only reviewer that I see that discusses efficiency consistently when reviewing parts (be they CPUs or GPUs)


What can you do about it. Outside of Apple hardware running macOS you just can't get the same. We can demand all we want but no one can offer what is available from Apple yet. I'm certainly not surprised Qualcomm/Intel/AMD want to add more fire to their feet by talking up their competition.


But you have to demand it. People have to make clear to the vendors that this matters.

As long people don’t do that everything will continue as normal. It needs to be very clear you’re not doing good enough.

As long as reviewers keep ignoring the elephant in the room in their text and scores, how are you ever going to get what I’m guessing just about everyone wants? They may not know it (us Mac users were getting annoyed at heat and noise but didn’t know how much).

But I’m convinced if someone built it, users would come out of the woodwork. So many people doing “office” work that doesn’t require high power machines would benefit. Qualcomm just doesn’t seem to be trying very hard in the computer segment.


> Qualcomm just doesn’t seem to be trying very hard in the computer segment.

They're functionally a monopoly. They don't have to.


They’re a monopoly in phones. They’re no-one of consequence in computers. That’s where I want them to try.


Thanks for clarifying, now I get it! And fully agreed - once you use an M1 that's cold to the touch even while doing some heavy lifting, it's hard to imagine using an Intel machine that gets warm just running Zoom and a few browser tabs.


For VR and games, Apple doesn’t compare to PCs for now until Apple Reality is released. Apple tolerates video games now because of the revenue, but they still don’t like them. The M2 is the most amazing Apple upgrade that I’ve ever had, but it’s an ecosystem that doesn’t garner enough support for non-mobile, non-iOS video games.


It is astonishing to me that people are out there buying Intel PC laptops when the M1 MBA can be had for <$1000


It depends what is important to you.

Firstly I am mostly plugged in, so battery life is secondary.

Secondly I am a heavy Excel user, and it just isn't as good on Apple

Thirdly I use PowerBI which isn't available for Apple

Lastly I have hybrid Active Directory, and can deploy SSO trivially to my remote Windows users, which I can't on Mac. During Covid I could mail someone a PC and get them signed on to AD resources in minutes. My Windows users are first class citizens, easier to support, and happier with their IT.


I’m not advocating everyone switch to Mac. I’m sorry if I gave that impression.

I’m advocating Windows/Linux users should have the choice of a machine roughly as good. Doesn’t have to 1:1, but clearly there is a lot of space for “better” past what everyone is selling today.

And I don’t see that improving much after 2 years and I find it kind of sad.

Why should you have to switch OSes and everything to have a fast computer with good battery life that’s quiet and cool?

You shouldn’t.


Are you (MBCook) and sylens the same person? You are replying as if you are.


Sorry. You’re right. Heh.

No we’re not. I thought it was a reply to my comment higher above in the hierarchy.

Sorry about that.


THIS! Apple:- fair warning an inside Bollywood joke.

Apple:- “Mere paas M1/M2/MAX,A16, Hain “ and looks at MSFT and asks “Tere paas kya Hain”?

MSFT:- Shoulder shrugs and says “Mere Paas Excel Hain”.

Apple has nothing on Excel. And Numbers languishes Excel by a few generations. Esp after the boatload of updates that Excel APIs got last year. I guess it’s all about priorities and what’s important for that business at that time.


That wasn’t what I was trying to say.

I know there is Windows only software. I know there is software that’s better on Windows.

I don’t care if people switch to Mac.

They should all have the choice of a chip as good (or reasonably close) to the M1.

But they don’t. And it doesn’t look like that’s changing. And I find that sad.


Okay! Thanks for the clarification!


The base MBA comes with only 8GB RAM and 256GB SSD and has very limited upgrade option. Also Apple charges crazy prices for more RAM and storage.

For $1000 one can buy a Windows laptop with 32GB RAM and 1TB SDD, that can cheaply be upgraded to 64GB memory and 10+TB SSD (2x SO-DIMM slots + 2x M2 PCIe slots).


I’m not arguing people should switch to an M1 Mac.

I’m trying to say I think people should have the option of something like it on Windows, not just the hot/power hungry stuff out there today.


Um. Cheaply to 10+ TB SSD? Where are you shopping?! I want to go there.


Samsung 870 QVO 8TB is under $700.


> It is astonishing to me that people are out there buying Intel PC laptops when the M1 MBA can be had for <$1000

Can the M1 MBA run Windows?

EDIT: Or can it run GNU/Linux natively (see https://news.ycombinator.com/item?id=33320765)?


Apple Silicon Macs can run Windows and Linux in a VM. Asahi Linux (native) is making decent progress.


To caveat, Linux in a VM is decent enough for non GUI programs, Windows might have harder limitations on anything that wants GPU power or direct peripherics access, and Asahi Linux is a research project at this point.

We’re far far from the dual booting days, running a mac VM on a Windows machine is probably the holy grail of running every OS on a single machine at this point in time.


I understand the sentiment but a dead ssd won't permanently brick a large portion of x86 laptops.


More to the point: I can still access the data from the SSD even if every other component on the laptop was dead. Heck, I could pop the SSD into another laptop and continue working where I left off.


Comments like this astonish me, a little. First that Apple is so good at advertising. Well done! Second, that a Macbook Air is the only form factor anyone could conceivably need? Or that macOS covers all use cases?

My (largely useless in this context) personal anecdote is that I spent $1029 for a gaming laptop. AMD Ryzen 7 4800H (8-cores, 16 threads), 16GB, 1TB, RTX 2060 6GB. Gets about 4-5 hours of battery doing normal stuff. Plugged in for gaming. (Yes, it runs all of my Steam collection!) It's very fast for everything I use it for. (I mostly work on my desktop, but I've used it for various development, etc. - Visual Studio Community, SQL Server Management Studio, VS Code, WSL/WSL2 and all Linux-based development I've done, photo editing, etc., of course it can run web browsers and be used to check email!)

Why wouldn't I buy a laptop that costs this much and can do so many things well? Because it doesn't work away from plugs for a day? But I live and work at home. Near lots of plugs!


First, I'm not in Apple's corner by any means. Before I left this comment, I left another one on a different story critical of what they've been doing with macOS.

But I've tried a lot of laptops over the pandemic and nothing has come close to the Apple silicon ones in terms of responsiveness, power efficiency, and just plain reliability. And in terms of sheer battery life and weight, the MBA is king of the hill. Sure, I'm not gaming on it - but then again, I can't stand gaming on a laptop to begin with, as the heat and fan noise really bother me. A typical consumer picking out a laptop for college or for use around the house would be better solved with an M1 MBA than almost any Windows laptop unless they have an absolutely mission critical software need.

The MBA isn't for everyone - heck, it can only support 1 external monitor - but it 's the right fit for that middle part of the bell curve.


I guess that says something about the concessions some of us will do to avoid using MacOS.


An Intel laptop can be had for like $200


I have a macbook pro 16 (the last Intel one) and it has a claimed 10 hour battery life. I use it mostly for ms office (excel, PowerPoint, outlook) and have zoom/teams open in the background, as well as citirx for access to some Linux servers. I'm lucky to get an hour and a half. Fans running, etc. Window_server chwes up battery like no one's business. Nice hw is nice, but I'm very dissapointed at my first toe-dip into the mac world


I’ve got a similar machine. The last 15” Intel.

Honestly it’s kind of one of the worst. The 2010 was pretty great. If you ignore they keyboard issues (a BIG DEAL) they got better for a few years.

In the last few years it became very clear Apple just couldn’t do what they wanted as the Intel chips kept getting hotter and more power hungry.

I’m sorry you got that experience. It was and now is so much better. You kind of lucked into a low point.


That's interesting, thanks. I would have assumed Window_server would be a common issue cpu on all platforms, is that not the case? If so do you know if disabling 3d acceleration could help etc?


The M1 is a marvel, however there's a vast range of things it's not going to do anytime soon, e.g. CUDA, eGPU etc - nevermind games.

https://store.steampowered.com/macos

Had they not pushed Metal so hard over Vulkan - then things would be different. There's time yet for Apple to change course. (Seems Valve open sourced MoltenVK, so that's one barrier removed at least!)


This was the general sentiment when Apple was on PowerPC, and we were buying pricier mac laptops just because of the OS.

I don’t have actual experience on a windows machine right now, but if the linux subsystem is getting good enough, switching there could be worth the hardware and price penalty.


Well, $1000 upfront plus $100 a year for the right to develop apps.


$1000 is still more than I would guess about half of PCs.

It would surprise me if my mother, grandparents, sister, etc. have ever sent more than $800 on a computer.


The M-series is very directly comparable to, say, an AMD Zen 3 v-cache or a Zen 4 processor. In many scenarios the AMD is faster, and in others the M-series is faster.

The main advantage of the M-series if for laptops on battery. Desktops, gaming, and software development is still significantly better on Intel-compatible desktops. E.g.: I can run Windows 11 with WSL 2 and Docker. That lets me develop using multiple huge ecosystems of tools seamlessly. With Apple, you get immediately relegated to a tiny corner of "optimised for M1" software, and have to either emulate or use cloud-hosting for everything else.


Also cost. You won't be M1 price/performance ratio in a laptop.


That can't possibly be a useful benchmark. There are hundreds of Android models with different batteries and power characteristics. Was it just CPU power efficiency?


Apple's decision to keep the A15 in the lower-tier 14 models shows that they did not expect everyone else (Qualcomm) to stop innovating on performance.


This is very hard for me to believe, even when I see the test results. Everyone I know with an Apple phone is either over 60 or looking for a charger.


Is the chip in this PC anywhere on that chart? I assume it's most similar to the latest Galaxy's which are not far behind Mac. Which is pretty much fine as far as I can tell, I don't see many CPU-bound workstation activities for most folks. And if you are CPU bound, you don't buy a micro form factor.


This is pretty interesting. I'm just imaging a future with inexpensive Windows ARM machines, passively-cooled, and running in most of the offices of your country.

With most of the important Microsoft software already compiled to ARM, and with those kits available to developers to do compile theirs at a competitive price, I won't doubt that future could be possible.


There are huge amounts of existing x86 software in use, including Office add-ins, Explorer extensions, and COM dependencies in general, that won’t be recompiled to ARM. The only way ARM would take over is if compatibility with existing binaries is maintained (which is very difficult if not impossible for x86 DLLs within ARM-based applications), and if their execution remains sufficiently performant.


Microsoft takes years to develop their own version of rosetta after the fail of Windows rt. And I think you are already able to use it now, you can already run x86 on Windows on arm vm on mac(m1).

Just wondering the performance when running it on the dev kit they sell now.


You can run a self-contained x86 application and self-contained ARM applications, but you can’t integrate between ARM applications and x86 in-process components (DLLs etc.), and such integration is how a lot of things work in practice on Windows. Converting an x86 application to ARM means losing those integrations, and that is a major hurdle to adoption.



I saw the "Stackable" part and thought you could stack one on top of another to have a combined compute + memory available automatically. Now that would've been mind blowing


There's actually a small vendor that produced tiny systems that do literally stack like Lego, intended for cluster deployment in edge computing use cases like retail or restaurants. Stacking the units provides all necessary interconnects.

I can't remember their name for the life of me, but they demo'd it at VMWorld a few years back.

The economics versus boring pizzabox or compact blade systems probably never worked out in their favor, hence why I'm having so much trouble tracking them down again.

Edit: found them. I guess they're still alive. Hivecell: https://hivecell.com/


How is Snapdragon 8cx support on Linux? I recall seeing some basic dotfiles for the ARM Thinkpad when it hit shelves, but I haven't seen anything else. Is there a good chance it would run OOB on a recent kernel with this devkit?


Given https://www.phoronix.com/news/Linux-5.20-SoCs-8cx-Gen3-Arm, it should already work if you run something rolling release or similar that has reached 5.20. (Otherwise, waiting for 6.1 is probably simplest.)


If I click on the "You can purchase the dev-kit here" button, I get "We are sorry, the page you requested cannot be found".

Probably because the device is not available in my country, but then at least show a page explaining I can't purchase the device based on my IP or whatever. This was a really bad experience.


I connected to a UK server through my VPN and managed to put in an order :)


800$ in Canada, at that price point it’s impossible to justify buying it instead of an M1 Mac Mini for 100$ more as a general use machine, especially given the sorry state of Windows-ARM.

Given it’s a dev kit companies won’t care about cost, but I can’t see many being sold to independent/small developers, the excitement just isn’t there.


I initially had the same reaction but the 900$ Mac Mini has only 8GB of memory, 256GB of storage and can only drive two displays, vs this which has 32GB of memory, 512GB of storage and can drive three displays.


It makes sense. The Mac Mini is what you get if you want to make a Mac or iOS app, but you want to keep your Windows or Linux system for daily use.


There's a lot more use-cases though, of both systems.

If you want a "good" small computer and you already have a screen (or want to buy cheap ones) then these systems are fantastic.

Performance is completely fine for moderate-to-heavy workloads (assuming the heavy workloads are bursty) for the Mac mini, and hopefully this.

Both systems are what you would get if you didn't need a display or keyboard already, they're desktop replacements with a small footprint, and fantastic for the majority of computer workloads including a lot of development ones.


Are you sure? I don't want to go the anecdote route here but I know plenty of people who use Mac minis as daily driver desktop machines.


There's no conflict? It's a good daily driver for some, and a good second machine for others.


Yes, I’m pretty sure that the Mac Mini is good as a machine for developing Mac and iOS applications. I have myself used it for this purpose in the past, and it worked very well.


What's wrong with using a Mac Mini as a daily driver?


I think you may be responding to something I didn’t write :-)


I have a mac mini as a desktop in my home office and actual office, multiple screens, and almost all my windows are either browsers or shell windows to my Linux dev server. I've never developed a MacOS or iOS app.


I think I have to get one. To replace an obsolete HP Steam. I was gonna do the Mac Mini, but I don't think the M2 model is out yet. Just worry about Win ARM compatibility across remote desktop, citrix daas, etc ;)


> The Mac Mini is what you get if you want to make a Mac or iOS app

This is probably closer to the Mac Mini with M1 that they shipped to kick-start the Apple Silicon transition for desktop apps.

Because if I was a windows programmer for today's customers, I can't really build things on a "Windows on Arm" device like this.

Like Apple before, I hope this is just the first salvo against the Windows+Intel, before we all switch to Arm chips (including Intel fabs).


I imagine Dr. Su a couple of years ago whispered to a couple of lead technicians "you are going to start a little garage side project here at AMD; it will be the future"

AMD releases 32-core + ARM APUs



Maybe I am missing the point, but why wouldn't you rather buy a NUC instead of this slow box? I get it MS wants to be like Apple, meticulously copying anything that pops up there, but their brand is associated with different "experiences" and their main value lies in backwards compatibility and open hardware ecosystem.


The point of this is that it's an Windows-on-ARM devkit; you can't buy a ARM NUC that'll run Windows AFAIK (Intel doesn't make ARM machines), and you can't test ARM software natively on an x86-64 NUC.


I doubt this is going to feel slow in actual use. You've got 8 big arm cores with a lot of memory, fast storage, and active cooling.


I learned this lesson the hard way during the AMD Phenom era. Core count are not a good representation of performance because the 8 "big" cores could be blazing fast or be secretly powered by a hamster on a wheel. What is the actual benchmark performance on real applications you might use? Thats what matters at the end of the day.


There are already other machines with 8cx Gen3 CPUs on Geekbench if you want to compare: https://browser.geekbench.com/search?utf8=%E2%9C%93&q=8cx+ge... it's like a third slower than an M1 Max in single-core.


Having worked with Windows on ARM in the past, I _hope_ you're right. But my experience has been that a ton of code is still going through the x86 emulation layer, which IMO is woefully lacking in performance, particularly compared to Apple's Rosetta 2 (which is a magical marvel of engineering).


Rosetta feels fast because apple m cpus are about twice as fast as qualcom ones


Because Intel performance per watt sucks compared to the competition, mostly.


Because nucs are not arm-based.


meticulously? What is it that Microsoft has really copied in the last five years?


Like the UI changes in Windows 11? Surface hardware? Windows Store?


I don’t know if Windows on ARM has a future, but, IMHO, almost anything that encourages more happenings in the ARM ecosystem is a benefit—the whole world has gone mobile, and its longer battery life takes the cake over its more powerful x86/amd64 counterparts.


Does Linux work on it? No, not yet. It may need some custom firmware like you see in the instructions for the Surface.

Support for the Qualcomm chip did land in v6.0 of the Kernel - https://github.com/torvalds/linux/commit/b6a6535b339776d32fa...

I just had mine arrive today, which I wanted to enroll as an ARM64 build agent (with Firecracker) for self-actuated ( https://docs.actuated.dev ) - this does actually work with Asahi Linux and the Mac Mini M1, but that was double the price for half the RAM.

So far, it's not looking good, with Ubuntu Server 22.04 and 22.10 both crashing after the Grub menu. I'm sharing progress here:

https://twitter.com/alexellisuk/status/1585562359254421504?s...


It continues to suck how African countries are excluded from these product launches and from many products in general.

Microsoft has enough partners and regional offices because Office365 and Azure are the only worthy products for us in "shithole countries".

If Microsoft isn't intending on selling a lot of these devices, the cost of adding 1 or 2 African countries would be relatively small compared to the revenue they make from our regions.

I mean, expensive Macs came out, costing twice the shitty HP/Lenovo/Dell enterprise offerings with poor thermals and battery life. We bought them.

I would buy this device if it was for sale in my market, I see a benefit in testing my work on Windows ARM64.


>It continues to suck how African countries are excluded from these product launches

All South American, most Asian and most European countries are excluded. That sucks, too.


Do we have any sense of how much of this is driven by privacy regulation? Newer versions of Windows have been more aggressive about collecting user information, so this would not at all surprise me.


Given that they are launching in some of the EU countries, not much. That said, I do hope their data harvesting breaks some laws somewhere and they get fined handsomely. This trend is atrocious.


The EU has probably some of the strictest privacy regulations in the world. I don't really get why you would think that.


They exclude about 180 countries


Are there any actual stats of how profitable African countries are for Microsoft?


I'd have to look. My previous employer spends a life changing amount of money on Azure. MS has a lot of local partners. They have an office here which seems adequately staffed (might be mostly consultants). I'd say 90% of South African businesses use Office365 to done level. We have Xbox subscriptions. When the Xbox came out, people bought it.

On the surface, Microsoft should be making decent revenue. It won't compare with the revenue from the 8 markets they chose. However that's my point, if they include smaller markets strategically in some launches, it can benefit them in the future.


Excluded how you mean?


"The Windows Dev Kit 2023 is now available to developers in 8 countries: Australia, Canada, China, France, Germany, Japan, the United Kingdom and the United States."


Seems odd to leave India out while keeping China on the list next to France before Germany. Apple is going to manufacture the gadgets in India and the former Indochina region.

Will be interesting to see the memory capacity on the next new Mac Mini expected in one or two month.


"Microsoft tells Windows Central that the Windows Dev Kit 2023 is exclusive to Windows 11, with no official support for running other operating systems such as Linux or even Windows 10 on ARM. The product is designed for developers looking to optimize their apps for ARM on top of Windows 11."


That doesn't really mean anything. No laptop released by Microsoft or most of its OEMs officially supports Linux. I'd wager this one will be running some ARM distro within 5 minutes of the first developer getting their hands on it.


> No laptop released by Microsoft or most of its OEMs officially supports Linux.

I disagree that "Most OEMs don't officially support Linux". Lenovo supports Linux on 17 laptops released in 2022 alone [1]. Dell also has a similar effort, with Linux supported on the 2022 XPS 13 Plus [2] among others. HP has its DevOne[3]

But at the moment, I don't know of an modern, high performance ARM laptop with official Linux support.

The closest thing I know of is the Nvidia Jetson AGX Orin Dev kit by NVIDIA[4].

[1] https://support.lenovo.com/us/en/solutions/pd031426-linux-fo... [2] https://ubuntu.com/certified/laptops?q=&limit=20&release=22.... [3] https://hpdevone.com/ [4] https://www.nvidia.com/en-us/autonomous-machines/embedded-sy...


The word _official_ is italicized[1], and the words "the product is _designed for_" means there may be hope that this is possible.

1. https://www.windowscentral.com/software-apps/windows-11/proj...


UEFI Secure Boot can be disabled or put in a custom configuration on all arm64 Windows devices, including Windows Dev Kit 2023.


And people said they wouldn't help. This is intended for us!


Realistically you have to ask how many people is that now?

I have never seen an ARM windows machine in the wild.

In fact I barely even see any windows 11 machines.


Surface Pro X had some traction, but they share the chassis with Surface Pro 7 and 8 so you wouldn't be able to tell even if you passed by one.


No the 7 is different, but I believe the 8 and 9 look the same


Ah my bad, I meant 8 and 9.


The hardware looks well suited to a cluster to me especially at the price point and with the presumably low power usage. Smack three of these bad boys together and you’re got about 64gb of memory to work with and 32gb for redundancy for $1800. That’s not terrible considering SBCs only go up to about 16gb of memory and tinyminimicro boxes probably draw a bit more power and don’t cost less new…


The gen2 was scoring only half an M1 in performance. With the bloat of windows on top this is going to be horrible.


Despite being in Hong Kong I'm really excited to buy one if it is available in my region natively (so I don't have to pay in RMB which is very volatile recently). Most of you guys forgot that you can run WSL in ARM as well -- yes, the same WSLg is capable of running Android on Windows on ARM, natively and very much faithfully. This could mean a whole lot of things including a much-integrated testing and app development pipeline with a whooping 32G of LPDDR4X memory, that means it is finally possible to run a lot of dev software most likely such as java, dotnet and node stuff that can offer day-0 cross platform support. I'm thus very confused why Microsoft is marking Volterra to run AI workload despite the it should be richer on the other side


A lot of people commenting that MS should not have went with Qualcomm on this device, although I don't think this locks Windows into Qualcomm. More importantly, I'm super curious to see what Qualcomm announces in December in reference to the next generation chips being designed by the Nuvia team (some of the people that designed the M1). ARM is suing to prevent it, but that seems shortsighted, and I assume it will net out that the Nuvia design will launch, but may have to pay some additional license money to ARM.

FWIW the M1 is a little more than twice as fast on GeekBench, but also uses twice the power (TDP). So sort of similar speed score per watt, so possibly they can be within the same ballpark as Apple in the next couple years.


Haven’t bought a Windows machine in a few years. It seems like this would likely be useful as a general purpose development machine because there is probably no crapware installed. Do you think this is a plausible argument to buy one or are they now all infested no matter what?


I wouldn’t recommend Windows on arm if you just want a general purpose machine, it is still very niche and will need time to be well supported. Just get a desktop with a recent Intel or AMD and reinstall the OS if you want to ensure you have something clean.


Well, a general purpose development machine. I’d like to be able to run office and a go compiler, not games


I’m confused by your response, I did not mention gaming.


It might not get the longest-term support, but it seems like a half-decent deal. If you're comfortable turning it into a home server/compute module once it reaches EOL, I'd say go for it.


Yeah I like that choice to


* too


Buying from Microsoft themselves, or Signature Edition systems, is certainly a less bad option than typical windows manufacturers. You still get Candy Crush preinstalled though.


My experience with Windows 11 so far has been great.


I've written up my findings with trying to getting Linux working on the new Dev Kit - https://blog.alexellis.io/linux-on-microsoft-dev-kit-2023/


So, "everyone" is switching to ARM now?


Microsoft was there first, but handled it so poorly it got memory holed.

edit: I forgot about the A in ARM.


I think the Newton was the first major use of the architecture outside the Acorn Archimedes.

"...an advanced, low-power processor was needed for sophisticated graphics manipulation. He found Hermann Hauser, who had developed the Acorn RISC Machine that utilized what became known as the ARM architecture, and put together Advanced RISC Machines, now Arm Ltd."

https://en.wikipedia.org/wiki/Apple_Newton

Around this time, DEC also chose to implement their StrongARM, so that pushed into embedded.

https://en.wikipedia.org/wiki/StrongARM


It cost them an ARM and a leg.


Nah, Acorn was first.


Take a look at what Apple is doing today to find out what the industry is doing in a year or two.


Windows has been on ARM since 2012 https://en.wikipedia.org/wiki/Windows_RT


Windows CE was actually on it long before this.

The original launch of CE was on MIPS and Super-H, but ARM appears to have gained support with Windows CE version 2.2.0.

https://en.wikipedia.org/wiki/Windows_Embedded_Compact


Break the build tools for their cash cow about once a month, and forbid anyone from developing for it without spending thousands of dollars on new hardware?


Seems to be working out for them alright. I'd love to make fun of them for similar reasons, but look at market, and more importantly, mind, share...


Where would we be if we can't make fun of popular things?


Apple seems to be a lot more than 2 years ahead.

The M1 was November 2020. That’s two years ago, give or take a few weeks.

There is nothing close to the M1 available for a Windows ARM computer. There is nothing close from Intel/AMD if you just want an ultra-low power chip with very good performance.


This looks like a really good deal for folks using WSL2 and VS Code for development.


I kind of hoped this was a laptop instead of an all in one. I have a Lenovo x13s and I really like it. The windows on ARM experience with win 11 is almost surprisingly good compared to early arm iterations.


Ah yes, the Microsoft NUC.


I'm glad there's more serious competition in this space. And apparently I'm not alone in this opinion.

From 2018: https://news.ycombinator.com/item?id=17783924

> Oddly, what I'd really like to see is ARM enter the NUC space. Maybe I'm the only one, but I'd like be able to pay $200-400 for a small, low power usage, decently performant machine. The 8th generation Intel NUC are good, but 28W TDP and it'd be nice to get it much, much lower than that. I know these are a small fraction of the overall market but personally I think it'd be cool.

Once you want something more powerful than Raspberry Pi or a board based on a mobile SoC your options whittle down considerably. There are "mini/micro" PCs but they don't touch the lack of power consumption.


Well for what it's worth, Synology makes some ARM powered NASes. But I suppose those fall under the mobile SoC category.


I've always been tempted by them but the price per compute power never got to the point where I wanted to pull the trigger. I don't fault them at all for their design choices, they make complete sense, but I wanted the box to be more compute/memory heavy.


For me, Synology is the remaining niche where Intel is still competitive.

Whoops. Never mind; they have an AMD Ryzen model now: https://www.synology.com/en-us/company/news/article/DS1621_P...


Intel might take exception since it was supposed to be their thing for their ISA, but you know what? “Next Unit of Computing” is a lot more apt here, at least from a Redmond perspective.


You'd think the actual NUC would be the ultimate perfect WinTel machine—I mean, it's right there in the name!-but I can't even get the Windows installer to boot my NUC.


It could make for an interesting server for homelab if we can easily install Linux on it. Does anyone know if there is some kind of lock that prevents installation of another OS ?


I get it’s way harder for Windows but they failed the first time but Apple succeeded because they went all in for the M1. Windows ARM is doomed to fail with such fragmentation.


I'm so hyped about this.

I'm sure this won't be M1-levels of performance, but I'm glad to see Microsoft move towards ARM. I can totally see in a few years more manufacturers building ARM-based SoCs to run on laptops, desktops, etc.

Makes me happy I TA'd a course on ARM programming 10 years ago.I'm extremely bullish on ARM's future.


I have mixed feelings about how successful this will be. The 32GB of RAM is really cool for the price - but 8cx Gen 3 is still no M1. It's still significantly slower than an 11th Gen laptop Core i5. Probably not super fun to develop on, even if serviceable - but considering nobody has cared about Windows on ARM to this point, why would you spend $599 to suddenly care about it, when WoA has far less than 1% of Windows PC marketshare?

Because Microsoft says it's the future? Microsoft is the worst at these promises. That's what they said about Windows 8, then Windows RT, the Windows Phone, the Windows Phone 8 platform, Windows 10 Mobile, UWP in General, the Windows Store, the relaunched Windows Store, Windows on ARM years ago, Project Reunion with XAML islands, Windows 10 S, Windows 10 X, Desktop Converter Bridge, the iOS Converter Bridge... I suppose they kept their promises with DirectX and that kind of thing. Right now, developer apathy for Windows is nearly insurmountable, and has been for the last decade, and Microsoft's constant changing of directions does not instill confidence.

https://www.androidauthority.com/snapdragon-8cx-gen-3-intel-...


On the other hand, Microsoft had spot on bets on operating systems, Internet, 64-bit, cloud, gaming consoles, managed runtimes, programming languages, high-end consumer PCs, Linux integration, and open source (albeit late).

Yes, they might have dropped the ball on more than one thing. You're especially right about Windows app ecosystem today, but it's not like Microsoft is constantly failing. They're doing phenomenal job on many fronts. They're certainly not that easy to write off.


Windows Store and Desktop Bridge work though. You are right about everything else.


Yes... and no. The Windows Store works way better now than it used to, back in 2015. Now it's finally serviceable, but it's still loaded with junk that makes the iOS App Store look well-maintained. Discoverability is still poor (better than it used to be), and the number of people actually using it also remains low. So... it works, but it was hardly the future of app distribution on Windows.

As for the Desktop Converter, it's in the same boat. For the first few years, all it was, was a pile of PowerShell scripts. No GUI, mediocre documentation, run a pile of scripts to package your app for a Store almost nobody uses. Also the command to package the app requires Windows 10 Pro and, like, 30 command-line arguments that had to be perfect in order to work. Now it has a GUI, and more people use the Store than before, but the Store has abandoned the need to use it and now allows just directly downloading unpackaged Exes, rendering it mostly pointless.


Does it have a power brick larger than the NUC, like those things usually do?

Neither the Verge piccies nor the marketing movie are clear on this.


X2 - If its ARM / mobile chip I wouldn’t think that the power supply would be very big. Maybe it’s internal?


All that effort and the best ARM Windows development machine is still "a VM on an Apple Silicon Mac".


My understanding is that it's not possible to legally get a license for Windows in that case, as there's no SKU for it.

Has that changed? Otherwise it's nonviable at $DAYJOB,


What prevents you from using an existing not-hardware-bound license? Enterprise volume licenses should also work.


How much development experience targeting windows on arm do you have?


I’ve developed on WoA plenty and it’s my primary Windows dev platform, but targeting it? Lol. I spin up an Azure VM for testing.

Any business which might have made use of this to take advantage of things like ARM devices being cheaper/more available than x86 devices (you literally can’t find an x86 tablet worth using for less than $400) long since ported to Android. Microsoft sort of missed the boat with ARM. Every time I see those little android based terminals littered around small businesses I think “man this sure is a business Microsoft lost for no real reason”

I wish Microsoft would support the WoA use case people could actually take advantage of. Using windows in a Mac VM. That would be by far the most effective way to get people to develop for the platform, not this product.


Is this actually true? How is Visual Studio in parallels or crossover?


Technically, yes. If you look at the benchmarks, a VM of Windows on an M1 (base model) is significantly faster than both the SQ1 / 8cx Gen 1 and SQ2 / 8cx Gen 2 processors Microsoft included with the Surface Pro X (by like, 30%, it's not even close), though how it compares to 8cx Gen 3 / SQ3 is unknown. However, considering we're on M2 by now, and that we now have M1 Pro and M1 Ultra models as well... egh...


Having a functioning GPU with this box is likely to make a bigger difference than the benchmarks in general use.


32gb on an m2 is probably going to cost you around $3000. I don't think 32gb on m1 is even possible.


Sure - if you need all 32GB for development work, it is a great deal. But, with a processor slower than a midrange modern Intel laptop, good luck using it to its full extent without feeling quite slow. So slow you might almost prefer a 16GB M1 Mac mini with swap for the rest despite the VM.

Plus, if you read Microsoft documentation on Windows on ARM so far, Microsoft doesn't actually expect you to use an IDE on these machines - but rather run your code remotely on them. You'll be a lot happier with your IDE running on a more powerful machine. Of course, if you do that, the lack of a GPU on the M1 for a Windows on ARM VM becomes not really an issue.


16gb is the near the cutoff that I need to run all the app/containers/etc I need. If I don't have enough memory everything comes to an abrupt crawl.

When that happens it doesn't matter what CPU I have, things are nearly locked up.


But, with a processor slower than a midrange modern Intel laptop, good luck using it to its full extent without feeling quite slow

I have little recent Windows development experience, but I wonder if Windows on native Arm gets the same sort of latency/lagginess reduction that going from Intel to Arm on macOS does? Even if the raw processing power is less, I would be happy with the tradeoff if Windows felt even snappier.


VS .NET workload is advertised to be working already.


My mac studio with an Apple M1 Ultra has 64GB of unified memory.


M1 != M1 Ultra


Which probably cost you a bit more than $600.


Not great


So the big question... Can it run Linux?


Of course, but the real question is: can it run Windows?


Or: can it run unsigned code?


For a developer focused machine? Nearly certainly.


Or: will it blend?


I think Windows on ARM still faces the same problem it always has: there's little reason for most people to care and it creates a collective action problem.

With macOS on ARM, Apple said they'd stop making Intel machines and they launched a compelling new experience. If you were a developer, you knew you needed to get on board or be left behind. If you were a user, you probably wanted one of the shiny new M1 machines, but even if you didn't you were going to be moving to ARM the next time you upgraded a few years down the line. Users also knew that software would be ported to ARM because it was the only way forward for the Mac.

Microsoft isn't abandoning x86. So why should developers care about Windows on ARM? Why should users choose Windows on ARM when developer support is poor - Microsoft just got Visual Studio on ARM. Why should users choose Windows on ARM when the ARM processors being offered are way behind what Apple/Intel/AMD are offering for processors? With developers and users unenthusiastic about ARM, why should chip companies want to invest in laptop/desktop ARM chips? Why should a hardware company start making ARM machines that just incur losses for a few generations as they manufacture things users aren't interested in?

There are certainly reasons to care, but it's a lot harder to justify. Maybe Qualcomm thinks it can create a new laptop/desktop chip business to rival Intel/AMD. That's certainly a good incentive, but I'm sure they've had meetings where they've talked about Microsofts lukewarm support for ARM, how to get users to buy an inferior product in the meantime (eg. until they create better chip designs and until devs port things to ARM), how to get devs to port things to ARM despite little user demand, and how to get hardware companies to want their laptop/desktop chips despite all this.

I'm not saying that Windows on ARM won't happen. I think we're in a time when CPU-independence is a lot easier and there's a lot of money to be made. However, it won't happen nearly as quickly as Apple's transition - because people have a choice in the matter. Intel/AMD CPUs are likely to be significantly better (than non-Apple ARM CPUs) for years to come. If Intel is able to get back on track in terms of process, it'll be even harder for Qualcomm and others to match. And Microsoft is likely playing a harder game when it comes to translating x86 software. Apple mirrored Intel's memory guarantees in their ARM chips, but the ARM spec doesn't require that. That makes translating a lot easier/faster, but if Windows for ARM is going to work with weaker guarantees, that makes it a harder sell. With Intel and AMD doing decently well at the moment, there's less reason for users.

Again, I'm not saying it won't happen. Hardware manufacturers will like having additional suppliers (even if the new ARM machines are partly just to get some leverage with Intel/AMD). Microsoft will want to make sure that Windows doesn't suffer if Intel/AMD stumble in the future. It's just going to be a long slog convincing developers and users that it's worth their time and money. Over a long enough time frame, I think new apps will be ARM/x86 and users will be fine with them at the right price/performance point, but it's not going to be like Apple's transition where people were enthusiastic.


Who's the intended buyer for this? Someone intending to make Windows ARM hardware who wants a reference to test the drivers on? I find it hard to believe that app developers need it, because they can just use qemu.


I've spent ages trying to get Windows 10 Pro for ARM to run on my raspberry pi w/ 8GB. Seems to boot and then spend age doing sod all. Maybr some day it'll get better but until then I'm sticking with Linux. <3


They should give them away. I’d never spend $600 on any kind of windows product. Maybe we can install linux on it? But you’d have to be smoking crack not to buy a mac mini m1. It’s the best computer I ever used hands down


When Apple released its $500 ARM dev kit ahead of the M1, they offered a $200 credit in exchange for return. Something equivalent seems appropriate here. This isn’t meant to be a computer for mass consumption so shouldn’t be compared to the M1 Mini.


It seems to be priced relatively cheap already with 32GB Ram and a 512GB SSD. A Mac mini with only 16GB of memory already costs nearly twice as much and I really wish there were a 32 gb mini at all. I like my Mac mini, but I hit the memory limit from time to time.


Rumors claim a M2 mini in the next month, with a max of 24GB ram. Even more interesting is the rumor there will be a M2 pro mini which doubles the max ram and adds more CPU/GPU perf.


Can a mac mini m1 run Windows?


Virtualized, yes. For most purposes, an M1/M2 Mac is an excellent platform for developing stuff for Windows ARM64.


No, not x86 Windows. You can only virtualize ARM Windows, which the other answers mean.


Yes, in Parallels.



if you don't want to spend money on it, just dont buy one. It's ok for something to be made for people who aren't you.


I mean crack, or a specific use case that isn't Mac bound like some kinds of development.

But probably crack


Does windows on arm have emulation for x86 programs? I'd think something like what apple has done in the past with rosetta would be a necessity.

It doesn't have to be super fast, just work well enough.



https://learn.microsoft.com/en-us/windows/arm/apps-on-arm-x8...

Yes, but not for drivers (so device manufacturers have to write drivers specifically for ARM)


Yes, both 32-bit and 64-bit apps should work, as long as they don't rely on low-level x86 stuff.


Why isn't this capability highlighted more?


I don't know, perhaps because it's not that new. It's very nice to have though.

On the other hand this particular device is meant as a dev kit for making ARM apps (either new ones or porting traditional programs to ARM) - it would be a little bit ironic if x86 compatibility were to be highlighted here.


Why would anyone buy that before Windows users are actually using ARM?


Maybe the usecase for this 32GB pre-halloween box release is to run the dotnet stuff on an agnostic headless operating system at the edge fanning in FPGA sensor, actuator things. The next after ARM is RISCV. ARM has been done.


Maybe MS is giving these for free to important developers.


To run Linux on it, if it isn't locked out.


Minority opinion: As long as I can program it in assembler, and directly call GUI routines in the OS to build windows, etc., I don't care much what CPU its running on.


I have a feeling that Microsoft is eventually going to follow Apple’s playbook and completely exit the desktop OS business to focus on their own line of products.


Their current strategy is the opposite of what you stated: it is to ensure MS software and services run on every imaginable product, including those made by others, including competitors. Hence XBox pass, VS Code / Edge on Linux, Office 360 on iOS/Android, etc.


Are they actually becoming serious about ARM all of a sudden or is this just a fluke?

It speaks volumes that Visual Studio wasn't even available as an ARM version until now.


No, they have been for a while. You don’t port a whole proprietary os to another isa overnight.

And iirc windows rt already ran on arm.


First they to make the 64bit version of Visual Studio for x86-64. It seems the ARM port was pretty fast after that.


I wouldn't mind MS hardware if it were designed by the people who did the first Natural Keyboard or the Intellimouse…


Could this be the harbinger of a return to Windows Phone? The world sorely needs an alternative to the mobile OS duopoly.


Why?

(And let’s set aside the how they’d possibly be able to compete with the scale, market penetration, marketing spend, and mature app ecosystems of iOS/Android and Apple/Samsung.)

Privacy? Lack of advertising? Respect for the user’s choices? From the company that brought you Windows 11? Why does the world sorely need another closed-source operating system full of telemetry?


>Privacy? Lack of advertising? Respect for the user’s choices? From the company that brought you Windows 11? Why does the world sorely need another closed-source operating system full of telemetry?

Hardware. I want a Surface Phone.


I'd argue that Microsoft entering the mobile OS market is worse than having a duopoly


>> I'd argue that Microsoft entering the mobile OS market is worse than having a duopoly

Would please you elaborate your argument?

Personally, I think having more choices would be better. The Apple vs. Google duopoly is limiting for consumers and developers.

Would you rather have an expensive device that you barely control or a cheaper device that spies on you?

More choices and competition, please.


Not the OP but I've had this thought as well. Microsoft has an almost unassailable position on desktop even still. If they had a solid position in mobile they could probably expel Android/iOS from enterprise with the same bundling tactics they use to push out different software on Windows with their own (often but not always) inferior offerings. From there the consumer space would be weaker and enterprise may start to ignore iOS/Android altogether. iOS and Android may well be to entrenched at this point for this to be a realistic fear, but based on how aggressively Google reacted to Windows phone (The youtube app fiasco) I think they at least worry about it a great deal.

Ideally, we'd have 3 or more fairly evenly matched and interoperable OS choices on mobile and desktop but that doesn't seem likely to happen. Trapping the monopoly inside it's own castle may be the best we can get.

I feel similarly about people calling for Apple to open iOS up to different browser engines. Idealistically that is what I believe should happen, but realistically I think it would just result in Chrome being even more dominate. For the same reason I lament the death of IE and even the original Edge. I don't personally use IE or Safari but I benefited from them existing and having decent market share.

> Would you rather have an expensive device that you barely control or a cheaper device that spies on you?

It is unclear to me if modern Windows actually still spies on you any less than Google at this point. My feeling is if still does, it isn't by much.


>> Microsoft has an almost unassailable position on desktop even still. If they had a solid position in mobile they could probably expel Android/iOS from enterprise with the same bundling tactics they use to push out different software on Windows with their own (often but not always) inferior offerings.

>> It is unclear to me if modern Windows actually still spies on you any less than Google at this point. My feeling is if still does, it isn't by much.

Microsoft is in a good position to be a strong third contender in the mobile space, but that does not mean that they would be better in all aspects.

>> Ideally, we'd have 3 or more fairly evenly matched and interoperable OS choices on mobile and desktop but that doesn't seem likely to happen. Trapping the monopoly inside it's own castle may be the best we can get.

Yes. That is why I would like to see more choices with hopefully better treatment of consumers and developers. Right now consumers have limited choices and the mobile development experience is agonizingly painful. It seems like an opportunity for disruption, but the entrenched players are dug in deep and probably nearly impossible to dislodge.


So what exactly are “consumers” clamoring for that are not currently being delivered?


>> So what exactly are “consumers” clamoring for that are not currently being delivered?

Until other choices are available, people tend to accept the default or keep on doing what was done in the past.

There is a segment of consumers that would like choices beyond Apple and Google mobile operating systems:

https://www.theguardian.com/technology/askjack/2019/jul/04/c...

https://vivaldi.com/blog/technology/smartphones-5-alternativ...

https://www.pcmag.com/picks/break-away-from-android-ios-7-fr...

Personally, I would like to see more "convergence" devices that let the little computer I carry around with me be anything I want it to be: a programmable general purpose computer, a streaming media server, or whatever else I want.

There are some projects that offer such functionality, but most require expert knowledge to setup or are not very widely-adopted or not very mature:

https://maruos.com/

https://www.pine64.org/pinephone/

Walled gardens are not where innovation happens because the gardeners uproot whatever does not meet their vision.


You really don’t think you’re out of touch with what most users want?

> https://www.theguardian.com/technology/askjack/2019/jul/04/c...

Yes because using an operating system from the other 1 trillion dollar market cap company is going to be a better alternative. Meet the new boss…

> https://vivaldi.com/blog/technology/smartphones-5-alternativ...

And those alternatives are already out there and no one wants them in a first approximation to no one

> https://www.pcmag.com/picks/break-away-from-android-ios-7-fr...

Okay. So they are “out there to try”. Have the majority of users been clamoring for it?

> Personally, I would like to see more "convergence" devices that let the little computer I carry around with me be anything I want it to be: a programmable general purpose computer, a streaming media server, or whatever else I want.

And you are in the modernity and so much so that it wouldn’t be a profitable business. Do you think Microsoft is going to give you that?

> Walled gardens are not where innovation happens because the gardeners uproot whatever does not meet their vision.

Where are all of the “innovations” that the majority of people care about - or even enough to make a profitable business - on Android where you can sideload and have third party web browser engines?


>> You really don’t think you’re out of touch with what most users want?

I don't claim to speak for what most people want. I think having more options than iOS and Android could help promote more consumer-friendly choices.

>> Yes because using an operating system from the other 1 trillion dollar market cap company is going to be a better alternative. Meet the new boss...

It would be another choice. Yes, they have similar incentives, but more choices help to drive innovation and keep all players competitive.

>> And you are in the modernity and so much so that it wouldn’t be a profitable business. Do you think Microsoft is going to give you that?

No. Microsoft is a better position than many to be a third choice in smartphone platforms, but they have shown poor initiative in the mobile space. They could try again or it could be some other organization with sufficient know-how and daring. (Something disruptive like Tesla or Starlink perhaps?)

>> Have the majority of users been clamoring for it?

"If I had asked my customers what they wanted they would have said a faster horse." --Henry Ford

"Some people say give the customers what they want, but that's not my approach. Our job is to figure out what they're going to want before they do." --Steve Jobs

>> Where are all of the “innovations” that the majority of people care about - or even enough to make a profitable business - on Android where you can sideload and have third party web browser engines?

Android is innovative because it is more open than iOS. Even more innovation is possible given the right circumstances.


> I don't claim to speak for what most people want. I think having more options than iOS and Android could help promote more consumer-friendly choices.

Using Linux on the phone with the lack of integration, the poor interface etc is the opposite of “consumer friendly”.

Normal consumers are not asking for the ability to “program their phone and run media servers”.

> Android is innovative because it is more open than iOS. Even more innovation is possible given the right circumstances

“Open” is not an “innovation”.


>> Using Linux on the phone with the lack of integration, the poor interface etc is the opposite of “consumer friendly”.

Who said anything about using Linux on phones? I agree that a third smartphone platform would need to be user friendly. Whether based on Linux, OpenBSD, QNX, Symbian, or something is just a technical detail.

>> Normal consumers are not asking for the ability to “program their phone and run media servers”

No one asked for iPhone. They were quite happy with their Blackberry and Treo phones. My personal wants for a smartphone are not why having a third smartphone platform would help innovation and competition in the current stagnant duopoly.

>> “Open” is not an “innovation”.

Yes, but "Closed" sucks for everyone but the platform owners.

iOS developers have been suffering and Apple has little reason it fix the issues: https://www.wired.com/story/apples-app-store-review-fix-fail...

Android developers face similar troubles: http://www.fosspatents.com/2022/07/developer-class-action-se...

The current smartphone duopoly is just two competing monopolies with consumers and developers caught in the middle.

Some organizations are trying to get "Open" smartphone marketplaces and more choice and competition in the markets:

https://appfairness.org/

https://www.consilium.europa.eu/en/press/press-releases/2022...

Open standards and open markets encourage real competition and innovation.


> Yes, but "Closed" sucks for everyone but the platform owners.

If you ask 99%+ of phone users. They don’t care about a “closed” phone platform anymore than console owners care.

> iOS developers have been suffering and Apple has little reason it fix the issues: https://www.wired.com/story/apples-app-store-review-fix-fail...

And what percentage of iOS users that collectively making billions are “suffering”?

> Some organizations are trying to get "Open" smartphone marketplaces and more choice and competition in the markets:

And failing miserable. The market has spoken.

> Open standards and open markets encourage real competition and innovation.

And yet, it wasn’t “open standards” that brought any of the “innovations” to the market that users care about.

How many decades have Linux advocates been promising the “year of the Linux desktop”?


>> If you ask 99%+ of phone users. They don’t care about a “closed” phone platform anymore than console owners care.

No one cares until they are personally impacted.

>> And failing miserable. The market has spoken.

Markets shift with time and circumstances. Those on top will not be there forever.

>> And yet, it wasn’t “open standards” that brought any of the “innovations” to the market that users care about.

Most users are ignorant of the standards that they rely on. iOS and Android are built on POSIX standards and rely on numerous networking and telecommunication standards. The Internet and World Wide Web that people use their smartphones to access are built on standards. The "magical" experiences that Apple sells to users would not be possible without a veritable book of engineering standards:

https://www.ietf.org/standards/rfcs/

https://www.iso.org/home.html

https://www.ecma-international.org/

>> How many decades have Linux advocates been promising the “year of the Linux desktop”?

I am not sure why you keep pulling Linux into the discussion. Just because Android uses Linux does not mean that other smartphone platforms would use it.


>I'd argue that Microsoft entering the mobile OS market is worse than having a duopoly

We need something. I’ve lost all faith in the hardware direction of iPhone. The 14 Pro (let alone Pro Max) is an absurd monstrosity. And Google clearly has no interest in innovation beyond copying Apple.


> And Google clearly has no interest in innovation beyond copying Apple.

I disagree, I think both platforms have copied plenty from one another. I used to jailbreak my iOS devices to get similar functionality to Android. Hasn't been necessary for awhile, I feel like the platforms are near parity now, but claiming one is copying the other (with no reciprocity) seems farfetched.


GP is probably one of the "small phone people" to hazard a guess.


How do you mean? Im curious as to your thoughts


>How do you mean? Im curious as to your thoughts

They've lost any sense of maintaining a cohesive design, or keeping things sleek and convenient. Performance has plateaued to a level of diminishing returns, so the only way they can get people to buy a new phone every year since iPhone 7 is to say "hey we put a bigger camera on it".

Product ran free with that mandate, and now we have this abomination: https://hips.hearstapps.com/hmg-prod/images/trevor-raab-ipho...

I have this recurring fantasy of an alternate history timeline where Steve Jobs never died, and when an engineer brought him the first iPhone 7 prototype, he held it in his hand, flipped it over, felt the camera bump, and said "You're fired. Get rid of the bump". I just refuse to believe he would have allowed this to happen, and I refuse to believe that we can't have good cameras without bumps.


Great points


Why?


I was joking... Everyone Chill out


How far away am i where i can have a windows arm laptop to develop php applications on and superb battery life?


Is Windows a better option for PHP development? What other tools you use in your dev cycle that favor Windows?


I think Windows is the best platform for PHP development. I just need to install wamp and vs-code and then am good to go.

I can use whatever hardware i want without worrying about drivers and compatibility and just about every other advantage windows have.

I have not yet found an alternative for HeidiSQL thats free to use that works on other platforms, every other alternative costs money and or is not as good. HeidiSQL on wine is horrible.


Damn, is that my new Linux ARM home PC? hehe


So is it possible yet to install a Windows 11 copy on an ARM Mac in VMWare or such? Officially.

Lots of people have Macs already.


At this price, with that HW, they are either delusional, or they have to pay insane fees to Qualcomm, they should have built their own silicon

That's what decades are poor management gets you, you late to the party with expired food

Apple was smarter when they came up with their M1; with an aggressive pricing and excellent performance/watt


Fantastic! I can't wait to install Windows on my M1 MacBook!


game of thrones has nothing on that Microsoft video...


Well I'd actually buy one of these if I could lol


Well I'd actually buy one of these if I could lol


Who is still developing for Windows at all, much less Windows on ARM? It's mostly just games and a handful of professional apps, no?


Would this run WSL by any chance?


WSL already runs on Windows 10/11 ARM in non-S mode, so no reason to assume it won't.


This device uses UEFI or U-BOOT?


UEFI


Not to be confused with Voltera, a company in Canada that makes desktop PCB printers.


Sleek device...


Sorry Microsoft, you lost me with Windows 11. I'll wait until you release a worthy successor to Windows 10, just like when I skipped Vista.


Is there a list of things that are worse with 11? I had it pushed on one of my less frequently used devices and so far all that has been annoying has been the Explorer context menu which hides "less used" items (some of which I frequently need, of course) behind a "More" entry.


For me personally

- Taskbar cannot be pinned to the side on my widescreen monitors

- Items on taskbar cannot be un-grouped

- Cannot show text on taskbar

- News/weather widget is awful, full of clickbait news and tiny Weather widget, which is vastly inferior to having a live tile that opens to a full screen weather app

- Reduced start menu customization (live tiles / grid are replaced with folders that add an extra click)

Other than that I haven't used it enough to comment much more. I have it on my laptop which is mostly just used for gaming, and I can tolerate the taskbar and start menu regressions. But for me, most of using Windows is... using the taskbar and start menu. To take away most of their functionality seems like complete insanity!


This tool [1] can solve all of your Windows 11 problems. I simply refuse to use windows 11 without the full right click context menu. This open source program does that, and much, much more. Smaller task bar, grouping/ungrouping icons. So glad I found this.

[1] https://github.com/valinet/ExplorerPatcher


> - Items on taskbar cannot be un-grouped

Oh my god, so much this one. It's driving me insane. I keep hoping that the next round of Start menu "improvements" bring it back, but I keep getting disappointed.


- The awful new context menu in Explorer.


it's so terrible that they even have to include a "show the good context menu" button


Personally I find Windows 11 fine although the decision to make the start bar only icons instead of icon + text is a bit bizarre in today’s trend towards (ultra)wide monitors.


I despise the fact that there is no right-click/context menu on the task bar. I use that frequently.


I think there is a registry key for that? And the very latest Insider build brings back Task Manager on right click.

Explorer Patcher can fix it all but you shouldn’t be obliged to fight the OS to feel productive.


I'd also be curious to know. I've found Win11 to mostly be a skin on top of 10.

Maybe I'm biased because I'm on the Insider Program, and I get changes incrementally instead of as a big release in 2 years


They basically tried to remove everything that needs a 2nd mouse button. Plus many keyboard shortcuts are gone. It looks and feels like an iPad to me, not like an environment to be productive in.


Windows 10 (and perhaps 8) was the beginning of Windows as spyware and adware that exists to manipulate rather than serve the user (outside of past obscure edge cases like Windows's anti-piracy mechanisms and disabling debugging on audiodg.exe). Now we have forced Microsoft accounts at install time, gaslighting users for switching off Edge and sending your Edge browsing history to coupon clipping sites, telemetry in Windows and Visual Studio and every time you open a MSVC command prompt, Visual Studio phoning home your menu searched to the web for "cloud AI menu search results"...


I also skipped Windows Millennium and Windows 8.


Yeah ME was awful, too. I thought 8.1 was actually OK.


Lack of HDMI is disappointing.


HDMI connectors cost a lot of money to the HDMI licensing authority. This payment is very painful if you have a small quantity of shipments. DisplayPort is free. And as others have pointed out, you just need to buy a cable.


Wouldn't display over USB-C solve this as well?


It looks like this does support display over USB-C - the tech specs have a snippet saying that you should use DP over USB-C so you see the BIOS.


A <$10 cable solves this problem.


Too little too late. Apple silicon remains king of value. M1 Linux advances. I guess it's time to go Apple...


And here's the difference between Microsoft and Apple: When Apple switches to ARM, people believe them. When Microsoft switches to arm, people ignore them. Why? Because Microsoft only ever half-asses such changes (see the terrible SOC in this).


I think another considerable difference is that Windows global footprint is 10x that of MacOS, so Microsoft has to keep both backwards compatibility and OEM's production plans in mind.

There is nothing for people to believe or not believe. MS cannot cannot discontinue x86 overnight because their OS is used by a much larger proportion of the world.


> There is nothing for people to believe or not believe. MS cannot cannot discontinue x86 overnight because their OS is used by a much larger proportion of the world.

Apple managed a competent compatibility layer, albeit with some special sauce in the SOC to make it fast. Is that too much to ask from Microsoft?


Yes, maybe it is too much to ask.

Windows is a general purpose OS, which is why it dominates two enormous markets: business software and games software. Microsoft will usually err on the side of developers because of this. The two companies' philosophies will of course be different.

If MacOS had similar mindshare in those markets, Mac developers would probably ask Apple to avoid overnight changes like the discontinuation of x86 Macs.

MS still provides security updates for Windows 7 despite its EOL occurring nearly 3 years ago. This is because many organizations still run critical software that they cannot shift away from, for whatever reason. Apple doesn't have to do that because no hospital or airport is running their logistics on MacOS.

Even with all this baggage, Windows on ARM has been available in some form since 2012's Surface RT.


They have no better choice. Name me, today, a desktop class ARM SoC that isn't made by Apple and represents the median performance band of the class.

Fundamentally, ARM Holdings is what Antitrust legislation was supposed to break down. They own the "ARM" name and control who can license the ARM IP and most importantly, how.

Ampere, the folks behind a lot of ARM servers, are by contract barred from getting into the market of making ARM chips for phones, desktops, or otherwise. That's the form of their license: Server-grade 96-core behemoths running at 3+Ghz and with the thermal output of a small space heater.

ARM holdings sets all sorts of weird restrictions and forces market segmentation to make sure that nobody "Accidentally" makes something that they don't immediately approve of. Qualcomm is basically locked into making phone SoCs for all eternity until they renegotiate their license with ARM holdings. They're in a shit situation because they have competition all over the place (Allwinner, Rockchip, a legacy Intel series, NXP, and Samsung to name a few), letting ARMHoldings bully them into not making something that rocks the boat too hard.

Apple pulled a massive show of force in making ARM license them desktop grade chips. You see, Apple has been a license holder for ARM for a bit now (with the Ax series chips) and makes up, ballpark, 15% of worldwide phones and now >50% of US phones. Apple had already idly said "we could... you know, not use an integrated solution" when they fiddled with Intel's radio baseband.

For ARM to try and sue Apple for breach of contract for developing the Mx series of desktop class ARM processors and get away with it, they'd be putting their market share dominance in four different major markets at risk. Qualcomm can't do that.

So that leaves Microsoft, who does not want to get into the processor fabrication business and who is still reeling over the antitrust lawsuit 20 years ago (which, I'll point out, was mostly over a shared text mangling library, for what it's worth) out in the dust looking for options, and the option they get is "Whatever Qualcomm will ship them."


> Apple pulled a massive show of force in making ARM license them desktop grade chips. You see, Apple has been a license holder for ARM for a bit now (with the Ax series chips) and makes up, ballpark, 15% of worldwide phones and now >50% of US phones. Apple had already idly said "we could... you know, not use an integrated solution" when they fiddled with Intel's radio baseband.

> For ARM to try and sue Apple for breach of contract for developing the Mx series of desktop class ARM processors and get away with it, they'd be putting their market share dominance in four different major markets at risk.

As I understand, Apple has a special license with a lot more leeway than those held by other companies thanks to Apple having been one of ARM's founders[0], so they may not have had to do any negotiations at all since they had the rights from the get-go.

[0]: https://en.wikipedia.org/wiki/Arm_(company)#Founding


> Apple pulled a massive show of force in making ARM license them desktop grade chips

Afaik Apple has an Architecture License which means they can do anything they want. They were one of the companies who co-founded ARM.


Let me play my tiny violin for the Gigacorporation Microsoft that was so unfairly treated by ARM that they just had to take "Whatever Qualcomm will ship them". Poor Multi Billion Dollar, they never had a chance to compete on a fair playing ground.

Nah, this is just organizational incompetence. The same reason we got cortana, windows 8 or adds in the start bar.


> Apple pulled a massive show of force in making ARM license them desktop grade chips. You see, Apple has been a license holder for ARM for a bit now

apple was part of the original ARM joint venture and gave it its initial capital

it's been there since day 0


> grade 96-core behemoths running at 3+Ghz and with the thermal output of a small space heater.

I mean, have you seen some of the latest desktop-grade hardware? I have had space heaters with less heat output than a 4090 at full tilt.


The architecture license was acquired in 2008.

Is there any evidence that anyone has been refused a license to develop a desktop arm cpu?


Qualcomm is currently being sued by ARM for daring to make a core thats faster than the ARM cortex


Well when the M1 came out it was dramatically better than everything else available to Mac users outside some small use-cases (like Mac Pro + multi-GPU).

People would have mostly wanted it anyway.

That doesn’t seem to be the case with this hardware.


If the ethernet port doesn't support multigigabit Ethernet, that is a shame. WiFi 6 is great, but we need more development, deployment, and support of multigigabit ethernet for corporate and enterprise customers!


There’s little point to such a feature for the intended use case, which is just to run some VMs to test builds on. At $600 with 32gb of ram multi-gig is asking a little much.


Not it's not, even $150 SBCs like odroid-h3 can have 2x 2.5gbit ethernet ports.

For server usecase this is very unbalanced as far as connectivity goes. Say you want to use the modem or wifi for internet access. Modem gives you 5gbit/s and you'll get out to your network just 1gbit. Wasteful, and it needlessly limits the opportnities.


You can put a 2.5gbe adapter on the USB ports I guess.


That would likely be stretching the USB interface to the limit. Ethernet is full-duplex, USB is not.

Also this SoC doesn't even seem to have proper publicly available datasheet, and whatever marketing stuff qcom has on their website doesn't list USB at all, lol. So for all I care it can have just one host controller. Not interested in SoC with no datasheets, when it's not possible to answer basic questions about the SoC, like how many USB host controllers it has...



Hm. Thanks for clarification.




Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: