Hacker News Reader

Comment by delusional | Hacker News Reader

The AI is the least interesting part of this announcement. Microsoft is giving ARM another try with a special branding that's supposed to guarantee some level of performance and quality. That's possibly huge news.

View on HN

I still won't believe it until reviewers get hands on. In October 2022 I wanted an efficient laptop for web development but didn't want to grab a mac, so I bought a Lenovo Thinkpad X13s with the Snapdragon 8cx Gen 3, which Qualcomm made similar promises about and it lived up to exactly 0% of them. The main draw, the battery life was worse than my old Blade 14 that ran Fedora, windows on ARM and their current emulation continues to suck, every so often core services would just not work. The Linux support is only just barely getting there, Ubuntu and Armbian have custom images, where a lot of stuff still doesn't work. The camera will never work because of proprietary blobs and the battery life is way worse than windows, no suspend, audio barely works, etc. And those aren't just problems of Lenovo or that thinkpad it stems from the platform. I ended up buying a macbook a few weeks ago and it was 2 years late, should have got it in the first place.

It's also notable that Qualcomm is officially upstreaming kernel support for the Snapdragon Elite platform that Microsoft is pushing, so those systems may actually not suck at running Linux.

In the past a few Snapdragon 8 Gen x chip dev kits have had Linux support if I recall correctly. I'd love to have built a device out of them but they seem quite expensive unfortunately for consumers ($800-1000, often from grey market sources). It's nonetheless good to see Linux support.

Snapdragon 8cx Gen 3 chip laptops are only just getting there. Last time I tried to daily it maybe 6 months ago, they didn't even have hardware rendering on the GPU, everything was rendered on the CPU. Ubuntu and Armbian have custom images for the x13s, no idea about the Volterra but it's much worse than the windows on ARM experience for that laptop and that's saying something.

That's disappointing :(

Yeah, “ARM-based Linux laptop with 22h battery life” is much more interesting to me than “Windows 11 ARM-based AI PC”. If the TPU can eventually be utilized by open models under Linux that’s just a cherry on top.

Personally agree. But let's be honest: the big shift to ARM in desktop will only happen with Windows.

Acorn tried very hard, but without a critical mass of software ported to the platform it was nearly impossible - Apple was the only other platform that survived.

Now, with open source, it’s easier to open up a place for other architectures. Android and iOS took over a lot of space traditionally reserved to Windows and Microsoft is not oblivious to that fact.

Lenovo claimed 28h on their last arm laptop, real world was more like 10 for light tasks and 2 to 4 for heavier work like development

Laptop battery life estimates outside of Apple are downright criminal. I'm still bitter over the time I bought a fully loaded Sony Vaio which had an advertised battery life of 11 hours. When in actuality, I only got less than three hours of battery life and complained, they explained to me that they meant it gets 11 hours of standby mode with the lid closed. I only made the mistake of buying a non-Apple laptop one more time with a fully loaded XPS which could barely manage 4 hours of battery life. I honestly have no idea how anyone can put up with non-Apple laptops. It's pure garbage out there.

2-4 hours sounds wrong. Were you mining bitcoin in the background?

"Battery life on the ThinkPad X13s will depend on your workload ... and while you might not get close to the 28 hours that Lenovo touts in its own testing, our testing with PCMark 10 got us an amazing 15 and a half hours..."

I was working on a Rails project and had a YouTube video playing in the background, its a 49whr battery and the CPU can easily burst above 15w and usually sustains around 7w plus backlight and all the other components of a laptop, a sustained power draw of 15w while doing development work isn't unusual.

Yes, and Lenovo is releasing a new Qualcomm-powered ThinkPad, which are known to be a Linux-friendly laptops.

> The ThinkPad T14s Gen 6 is, of course, business-focused. It will have the same Snapdragon chip, storage capacity, and webcam but will support up to 64GB of memory and one of three 14-inch display options: an IPS with up to 400 nits of brightness; an IPS touch display; or an OLED that covers 100 percent of the DCI-P3 color gamut, also with 400 nits of brightness.

>Lenovo expects the Yoga Slim 7x 14 Gen 9 to start at $1,199 and the ThinkPad T14s Gen 6 to start at $1,699. Both will be available in June.

Source: https://www.theverge.com/2024/5/20/24160819/lenovo-qualcomm-...

They might be "known" (recent experience may vary. mine sure did) but if you can't get support for the pre-installed Linux, you're just asking for pain.

Hmm. The first thing I do when I buy any new machine is to format the hard drive so I can install my own OS on it. I don't really trust any manufacturer to get that right.

So, if it's so hard to get Linux running that you need to have it preinstalled for you, then it's not really a good Linux machine in my view.

It's not as much "How do I install Linux?" as it is "The manufacturer supports these Linux images but my hardware isn't working right with it so I can call them and they're supposed to have an answer as to making it actually work". Otherwise you're just as good to buy any random laptop and try to make sure everything is supported yourself (not a horrible option, just not the premise of these kinds of laptops).

I had decent luck with Dell (though it was an n=1 interaction so I'm not sure how it indicates overall) ~5 years back on this where there was some issue with the dual GPU nature of the 7730 where on this model you could actually completely bypass the iGPU (it wouldn't even show up as a PCIe device anymore) for the main screen but it was causing some sort of display desync after a few minutes on Linux but not Windows. Loaded up the official image, reproduced, opened a ticket, they sent a firmware patch, it worked.

That and that the manufacturer has worked to ensure that at least some version of Linux works on it well, i.e. has done the systems integration work. Otherwise it can be a death of a thousand paper cuts, where things kind of mostly work, sorta, occasionally.

I usually take it as, they installed Linux on it and support it. I will use my own install after formatting a drive (or carrying one over from the previous machine) but it’s more like a seal of approval that Linux works.

And if the company are good stewards, they will upstream any drivers/kernel modules for that hardware too.

Lenovo has been selling a Qualcomm laptop, the Thinkpad x13s, for several years already with questionable at best Linux support so I wouldn't expect the new ones to be much better.

But that's only one device. This is an entire generation of laptops all running the exact same chip.

It isn't so much about the chip but that there are almost no standards for how things should work when compared to a PC. Almost every PC boots the same. Almost no ARM device boots like another.

There has only ever been 1 or 2 ARM SoCs for Windows for the last 10 years. It didn't make Linux support on them easy.

> [Lenovo pricing]

Also available @ 75% off in 12mos on eBay because the used Lenovo market is a beautiful, beautiful thing.

I just picked up an unused T15 Gen2 11thGen 16Gb/256Gb for $390. In Warr until late 2026.

Bought a thin client today w/ similar specs for $350.

I thought Microsoft requires UEFI + secure boot for Windows, and with no disabling option in the firmware setup, for Windows Arm PC? Or maybe it was "only" Microsoft Secure Boot and you can actually use a Linux distro? If this is the latter case, can you build and run the kernel you want or not?

You can use secure boot on select distros. Coincidentally, they're the better ones as well (Debian, Fedora, openSUSE, etc...)

You should also be able to enroll your own keys to boot any distro under secure boot

Yeah but then you need to resign every update, if the distro doesn't do it for you.

Secure boot can now be disabled on ARM.

Exactly. They're now aggressively competing with Apple with supposedly better performance and better battery life. That's huge. Apple was years ahead with the introduction of the M1 but it appears competition has finally caught up. All of this has to be proven in real life though, but so far all the newly announced ARM devices [0] look like impressive MacBook competitors on their own.

[0] So far it seems this are the devices that have been announced.

Microsoft: Surface Laptop 7, Surface Pro 11

Dell: XPS 13, Inspiron 14 and 14 Plus, Latitude 5455, Latitude 7455

HP: OmniBook X and Elitebook X

Lenovo: Yoga Slim 7 and ThinkPad 14s Gen 6.

Samsung: Galaxy Book4 Edge

Acer: Swift 14

AIUI the snapdragon x elite they’re based on benches between the M3 and M3 pro for CPU.

I am excited on having an ARM on a PC with Linux support but I never see Windows as an OS optimized for batteries beyond if they use ARM or x86. The Apple advantage continues to be a complete control of the device from hardware to the operating system, while the bloatware of Microsoft Windows makes an arbitrary use of resources.

This is not to say that Microsoft Windows is not an advanced OS, the problem is that it is not laser focus optimized.

I guess that's true but you can't make a leap like Apple did with their transition to ARM every few years. So it's good to see the competition catch up. And I'm perfectly happy with Windows laptops trailing Apple by a close margin instead of a 4 year gap.

I'd bet a tenner that they've caught up with where apple was 3 years ago rather than 1 year ago, at best.

They do some sleight of hand when doing the comparisons during the release today.

The Surface Laptop is physically a competitor to the actively cooled MacBook Pro (in fact it’s thicker).

Their performance and battery metrics are against the slimmer and passively cooled MacBook Air.

Their performance comparison is between the M3 and the Snapdragon X Elite when the M3 has throttled (their wording is sustained performance)

Their battery comparison is between the M3 and the Snapdragon X Plus.

That they interchange both freely is a strong tell that their devices don’t compete on all fronts like your comment suggests.

The only area where I think the snapdragon X will compete is price, because they’re claiming it’s $200 lower than a comparable M3 MacBook Air (but don’t disclose the exact comparison).

Still, a very strong showing from Qualcomm/Nuvia. However the sleight of hand leaves a bad taste in my mouth.

Unfortunately, even though I wish I could get an M4 (or equivalent) ARM SoC in a PC, it is very unlikely that these chips are going to live up to the hype. Qualcomm has a bad track record of overstating and cherry-picking benchmarks.

All of their claims so far have been both impressive, and very squishy. Microsoft's own messaging around performance and battery life for their Surface devices was also unusually squishy. And I am a huge Surface Pro fan that would love nothing more than a fanless AND fast ARM Surface Pro.

I live in San Diego, and know countless Qualcomm employees, none of them give a shit about anything other than modems really. The rest of the SoC is just something to push 5G. They care as much about CPUs/SoCs as Intel did about 5G modems.

> Microsoft is giving ARM another try with a special branding that's supposed to guarantee some level of performance and quality.

I hope so. I've been a happy Windows for Arm user (via Parallels on Apple Silicon) for a year+ and it's been good. Based on that, I think drivers are going to be the biggest PITA for ARM-based PC users for the first couple years — for example, Google Drive doesn't work for that reason.

I tried Windows on UTM (based on qemu, I believe) and the graphics were choppy. I attributed that to the lack of graphics acceleration. Is it also the same on Parallels?

Parallels has excellent graphics acceleration.

Yep. Both Parallels and VMware have good graphics acceleration, but Parallels is better. I too have been running a Windows ARM VM for work for a year or so.

> I think drivers are going to be the biggest PITA for ARM-based PC users for the first couple years — for example, Google Drive doesn't work for that reason.

Google Drive does ship with Arm64 drivers, and patching the platform check out of the installer gets them installed just fine (40 84 f6 74 08 -> 40 84 f6 90 90).

No idea why they're blocking the install.

'AI' is frequently the least interesting part of any announcement.

Not quite. Hidden at the very end of Microsoft's blognouncement[1] is this tidbit (emphasis mine):

> We look forward to expanding through deep partnerships with Intel and AMD, starting with Lunar Lake and Strix. We will bring new Copilot+ PC experiences at a later date.

So it's less Microsoft pivoting to and giving ARM a try again but rather testing the waters and distributing the risks by introducing ARM into a line of laptops and tablets that will still be fundamentally x86. Arguably, the only reason ARM is first to store shelves is because Qualcomm released this generation first before Intel and AMD.

This isn't as significant as Apple throwing Intel out to pasture and converting to ARM wholesale, not yet anyway.

[1]: https://blogs.microsoft.com/blog/2024/05/20/introducing-copi...

apple never wanted to use intel hardware, they were forced to by motorola/IBM et al; whoever was selling them the PowerPC chips told them to pound sand because the xbox and playstation needed way more PPC chips than apple. Apple made a business decision to switch to intel, which caused a bit of a to-do in the community at the time. That apple switched off intel at their earliest possible chance - that which took time to "design" their own ARM cpu - doesn't really mean anything, in my opinion.

I wonder how many people remember all of the hardware platforms that NT 3.51 and NT 4 ran on (Sparc, etc)

IBM didn't want to design a power-efficient PowerPC chip. And then they switched from Intel once Intel stopped being able to design new power-efficient chips.

Well, that's not what the story was at the time. "You don't order enough" was the reasoning.

Further, nearly contemporaneously after apple M chips debuted intel released chips with P and E cores. Furthermore, Intel made lots of chips that have TDP under 10W, even 5W - multicore, even. I still use them to run HA VMs for emergency communications internet gateways.

"power-efficient" is a weird thing to claim, anyhow. What does it mean? PPC were much faster per socket than maybe even server class chips by intel, if you wanted power efficiency you could run them slower and get whatever FLOPS/J intel could give.

I am very sure i remember Apple not having a choice.

> "power-efficient" is a weird thing to claim, anyhow. What does it mean?

Cycles per Joule, or cycles per second per watt. While staying at the laptop-level available power envelope.

> PPC were much faster per socket than maybe even server class chips by intel

The absolutely top POWER chips were at the time _somewhat_ superior to Intel (and AMD), but not by much. And that superiority was achieved by raw strength, the POWER chips had a very large die with tons of additional cache, and ran at higher frequencies (i.e. more power dissipation).

However, laptop-class chips were absolutely underpowered. Intel ran circles around them. The same was true for consoles, PS3 had a puny underpowered CPU (with multiple co-processors that were supposed to make parallel tasks easier), and XBox360 was barely better than then current top desktop CPUs.

> The relationship between Apple and IBM has been rocky at times. Apple openly criticized IBM for chip delivery problems, though Big Blue said it fixed the issue. More recent concerns, which helped spur the Intel deal, included tension between Apple's desire for a wide variety of PowerPC processors and IBM's concerns about the profitability of a low-volume business, according to one source familiar with the partnership.

>IBM loses cachet with the end of the Apple partnership, but it can take consolation in that it's designing and manufacturing the Power family processors for future gaming consoles from Microsoft, Sony and Ninendo, said Clay Ryder, a Sageza Group analyst. "I would think in the sheer volume, all the stuff they're doing with the game consoles would be bigger. But anytime you lose a high-profile customer, that hurts in ways that are not quantifiable but that still hurt," Ryder said.

furthermore, if you look past the "Intel is just too slow for the power envelope" that you stated:

>The "bad quality assurance of Skylake" was responsible for Apple finally making the decision to ditch Intel and focus on its own ARM-based processors for high-performance machines. That's the claim made by outspoken former Intel principal engineer, François Piednoël. "For me this is the inflection point," says Piednoël. "This is where the Apple guys who were always contemplating to switch, they went and looked at it and said: 'Well, we've probably got to do it.' Basically the bad quality assurance of Skylake is responsible for them to actually go away from the platform."

So both of the claims about cycles/J being the reason are just sound bite ahistoria.

>The PowerXCell 8i powered super computers also dominated all of the top 6 "greenest" systems in the Green500 list, with highest MFLOPS/Watt ratio supercomputers in the world.

that's 2008 - when apple originally announced they'd complete the move to intel.

PS3 underpowered?

>According to Folding@Home in a statement on its website; “Using the Cell processor of the PS3, we should be able to do more folding than what one could do on a PC. Also, since the PS3 has a powerful GPU, the PS3 client will offer real time visualization for the first time.

Also if you compare like for like, the CPUs IBM put out in 2005 with 3.2ghz had a TDP of 75W. Intel CPUs in 2005 at 3.2GHZ used 130W TDP. both used 90nm processes.

You're fractally wrong. I could actually keep going with every claim you made and put paragraphs in. I'm sorry.

> furthermore, if you look past the "Intel is just too slow for the power envelope" that you stated:

Well, yes. Intel was not able to deliver fast mobile CPUs on time. So Apple decided to give up and do it on their own.

> PS3 underpowered?

Yes. It was terribly underpowered. I worked with it back then, and it was slower than a 1GHz Pentium for practical tasks. The CPU in PS3 was clocked at around 3GHz, but it had full in-order execution. So any code with branches just died.

Sony's way around it was SPUs - special coprocessors, that were even more underpowered with little local memory. But there were 8 of them, and the common data bus was pretty fast.

It worked great for graphics and for something computation-heavy like Folding@Home. Kinda like modern GPUs. But it sucked for general-purpose code.

> Also if you compare like for like, the CPUs IBM put out in 2005 with 3.2ghz had a TDP of 75W. Intel CPUs in 2005 at 3.2GHZ used 130W TDP. both used 90nm processes.

Now compare their performance for actual tasks. You'd be surprised.

Cell architecture bombed. As a result, Sony switched to a regular architecture for PS3.

Microsoft has no courage. They have to keep catering to every possible audience, so they’re not willing to pull the plug on x86, which means ARM will always play second fiddle.

Microsoft and Windows (and by extension x86) achieved their desktop market dominance by respecting that most people want backwards compatibility.

Everything that has tried to go or is going against that tide either failed (eg: Itanium, Windows RT) or never had market share to lose in the first place (eg: MacOS, Linux in the consumer space).

Microsoft would be stupid to be "courageous" and drop backwards compatibility, that would even trump Apple's courage abandoning the headphone jack. It also makes business sense to keep your eggs in multiple baskets, assuming those baskets are each commercially viable.

> Microsoft has no courage

Sadly the headphone jack is definitely dead.

> They have to keep catering to every possible audience

That's not Microsoft's job at this point, they heavily invest and push the envelop where makers don't want to take the risk, but from there it's for Lenovo, Dell, HP, Asus etc. to decide what they want to market and which chip to push. The same way some of them put weight behind AMD while other went full Intel 100% of the time.

I don't think the Intel or AMD chips will be (literally) as cool. AMD mini-PCs are _close_ to what an M1 can do, but still too power-hungry and slow.

Think it’s worth mentioning that their Qualcomm exclusive windows laptop deal ends soon and this should allow AMD and NVIDIA to ramp up (arm) windows laptop cpus soon within the next few years.

Microsoft has been consistently trying to give ARM a try since the surface RT. Consumers are not going to bite. marginal power saving is not meaningful.

The first iteration of Windows ARM didn't have any x86 emulation layer, so that one was doomed from the start. The second iteration did, but it initially couldn't run 64bit apps and the performance was poor. They do have 64bit support now and it sounds like the emulation performance has come a long way.

Here is my question though, comparing how this works on Mac.

Will Windows have the opposite? ARM running on x86?

I continue to wonder how Microsoft expects to work long term. Are they expecting that every developer is just going to keep x86 and ARM based app perpetually or users be stuck always using that emulation layer if they are running ARM?

Microsoft won't be able to 100% transition to ARM like Mac did. At some point all Intel Mac's will be old enough to no longer get the latest version of Mac and for developers to stop targeting and they drop Intel support.

I just don't see many developers bothering with an ARM native Windows version when doing so means they have to support both or risk annoying customers later.

I think the official line from Microsoft would be that most software should be using .NET anyway, and in that case the same binary should Just Work on either architecture. In reality there is still a lot of native software though, so who knows how that will play out. Games in particular will always be native.

If it's like their previous ARM Windows attempt, existing native software won't work in any event because the entire platform is locked down.

Windows 11 ARM isn't locked down at all. I run it on a daily basis.

You can run any legacy Win32 .EXE on it that you want? I didn't know that. Good to hear if true.

It's actually kinda annoying once I started paying attention, as many software vendors just detect "Windows" and give you a x86/x64 installer, even when the company offers a ARM64 build that would presumably be faster or be more energy efficient. I installed a bunch of stuff that were Intel binaries without even knowing that I wasn't running native. But I haven't noticed any performance issues, and yeah everything just works.

In 2018 that lockdown situation morphed into "S Mode" which you can turn off in the control panel. The only trick is that you can't turn it back on. It's just that the ecosystem isn't there, both in terms of developers and performant devices.

Hopefully today's announcement is a turning point for that but atm windows on ARM is about on the same tier as a pre-carplay infotainment system.

Does Microsoft even push or care about .net anymore? They seemed to move on after UWP and now that seems to be forgotten in focus of more web apps.

Microsoft care about .Net. It runs the Corporate world like Java

You have to understand that Windows comes from a separate division than .NET and they have no overlap. Microsoft isn't a cohesive company. .NET comes from the developer division (DevDiv) and UWP comes from the Windows division (now Server & Cloud). The Windows folks always hated .NET and the developer division has been lukewarm about UWP.

The Microsoft panel of this comic sums it up nicely: https://bonkersworld.net/organizational-charts

Not really, plenty of Windows workloads are still about C++, in-proc COM and user space drivers, do require C++.

Not that WinUI really matters after all the mess WinRT/UWP went through, but it is basically C++ COM/WinRT with .NET bindings.

I think the idea is to all apps and developers gradually transition and develop with ARM support - after all even the mobile devices will be running on ARM sooner or later so future apps, games will be developed with ARM in mind anyway. x86 apps will be supported - with some paid support for example.

But it all depends on the market share of ARM at one point. But you can run DOS apps still so with emulation layer - and the increasing performance of ARM - one way or another old apps will be able to run on ARM. For those who will need to those.

Unlike Mac, Microsoft just can't drop past generations and call it a day.

> But it all depends on the market share of ARM at one point.

Right thats kinda my point, unless I have missed it I have yet to see any real talk about ARM on custom built machines and I doubt gamers are going to give that up anytime soon.

Apple was able to force the transition to happen. I highly doubt Microsoft is going to risk actually dropping x86 from Windows on any reasonable timescale and there has to be something for ARM to x86.

Unlike when Apple announced that all of Mac was transitioning, there isn't a reason for a developer to think that anytime soon they can drop x86, so why complicate what they have now by adding ARM?

> Right thats kinda my point, unless I have missed it I have yet to see any real talk about ARM on custom built machines and I doubt gamers are going to give that up anytime soon.

A lot of gaming these days is running on mobile phones and portable PCs - and now laptops - will highly likely leverage ARM sooner or later. Add to that some eGPU with Nvidia cards and you get a monster.

Intel is in a deep trouble.

>Unlike when Apple announced that all of Mac was transitioning, there isn't a reason for a developer to think that anytime soon they can drop x86, so why complicate what they have now by adding ARM?

ARM is the future as there is a desire to have long battery life and performance increase. Microsoft right now does have x86 emulation layer and app support right now is much better already than it was before (in RT era where it did not even have the emulator).

Devs are developing apps across all the devices and ARM based Mac is already requires you to develop ARM compatible apps.

>I have yet to see any real talk about ARM on custom built machines and I doubt gamers are going to give that up anytime soon.

The vast majority of gamers game on smartphones and tablets with ARM processors.

Some of the biggest gaming hits recently have also been cross-architecture and cross-platform, namely Genshin Impact and Honkai: Star Rail. Native ARM and x86 releases, runs on Windows, Android, and iOS. There are also gaming hits like Fate/Grand Order that don't have an x86/Windows release at all due to not even considering desktops/laptops.

The future is already here.

> The vast majority of gamers game on smartphones and tablets with ARM processors.

Those are clearly not the gamers I am talking about. The gamers I am referring too are not switching to playing on mobile phones. If they are switching to handheld devices they are going with x86 devices like the Steam Deck.

There is a massive market out there of games that do not support those platforms. That are only just now scratching the surface with games like Death Stranding releasing on iPhone and Mac.

Except for Nintendo the 2 main AAA consoles are x86 based, and I have seen no rumors of that changing.

So great, there are large mobile games but lets not pretend that there is not a huge market that the future is not already here for and shows very little signs of actually changing anytime soon.

https://steamcharts.com/ that is what I am talking about. Which unless I am mistaken the only one of those in the top list that actually runs on mobile is PUBG.

> There are also gaming hits like Fate/Grand Order that don't have an x86/Windows release at all due to not even considering desktops/laptops.

That is nothing new, Pokemon GO came out in 2016. That isnt a sign that gaming is changing but that gaming is expanding to include new types of players. But the "hardcore" AAA gaming market still very much exists, and is firmly on x86 right now.

> Which unless I am mistaken the only one of those in the top list that actually runs on mobile is PUBG.

Even in that case it's "kind of but not really". PUBG Mobile is a distinct game from regular PUBG, they have similar core gameplay but they are developed independently of each other.

Good to know, thank you. I figured they went the Fortnite route and it was the same game.

But I don't play PUBG, so my main point stands. None of the top steam games support ARM.

Fortnite is the outlier there, being the exact same game across every platform. COD Mobile and Apex Mobile are/were also officially sanctioned clones of the original game, similar to PUBG Mobile.

Porting a game from x86 Windows to ARM Windows may take some effort, but for most games, nowhere near as much as porting to a different operating system. There just isn’t that much assembly code or even SIMD intrinsic use in your average game. And thanks to Microsoft’s Arm64EC ABI, the conversion from x86 to ARM can be done piecemeal. If, say, the game depends on some proprietary third-party library that isn’t willing to offer an ARM version, that library can be run in emulation while the rest of the game is compiled natively for ARM.

The AAA game world is very conservative, so I can’t guarantee that PC game developers will port their codebases to ARM. It really depends on the size of the audience and how well the x86 emulator works as a substitute. Even if ARM takes over on Windows laptops, I’m not sure laptops are enough, when laptop users are already accustomed to not being able to run AAA games well.

But if the audience gets large enough, it’s hard to believe that developers won’t try recompiling. It’s just not the same level of effort as a port to Mac or Linux.

> The AAA game world is very conservative, so I can’t guarantee that PC game developers will port their codebases to ARM.

Unreal, Unity, CryEngine and Godot all support ARM, so - testing and third-party binary libraries aside - there shouldn't be any reason to not have an ARM port.

Indeed – though many AAA games use in-house engines.

>Those are clearly not the gamers I am talking about.

You specified gamers, you should have explicitly specified PC gamers if they are who you referred to.

Note that PC gamers are, as much as they deny it, a minority of out of all gamers as a whole. The vast majority of gamers play on mobile or consoles, and of those mobile far outnumbers consoles too.

Consoles can also switch processor architectures with the changing forces of the wind, they don't have to support backwards compatibility unlike x86 and Windows. If Windows ends up becoming more ARM dominant than x86, consoles will likely follow suit to make subsequent Windows ports (and then also mobile ports?) easier.

Going on a tangent, I find it very annoying that PC gamers despite being the minority somehow want to claim gamers aren't gamers. PC Master Race is a meme, not reality.

>Which unless I am mistaken the only one of those in the top list that actually runs on mobile is PUBG.

Stardew Valley at #10 also has mobile ports.[1][2]

>But the "hardcore" AAA gaming market still very much exists, and is firmly on x86 right now.

The games I cited are AAA games, FSVO AAA; they are developed and/or published by big, established studios and/or publishers. Frankly, I find the AAA moniker worthless these days, but I digress.

[1]: https://apps.apple.com/us/app/stardew-valley/id1406710800

[2]: https://play.google.com/store/apps/details?id=com.chucklefis...

> mobile far outnumbers consoles too

That didn't sound correct to me, and I found an article [0] that says the numbers are pretty similar.

I play some games on PC+PS5 and some on mobile, so I'd probably count as both a mobile and "legacy" gamer, but if I had to choose one gaming market to immediately disappear from the face of the earth, it would be mobile gaming for me, absolutely.

[0] https://business.yougov.com/content/48108-console-vs-mobile-...

The graph is misleading because they group PC and consoles together against mobile. That implies mobile would slaughter both of those segments individually.

It's also missing some very important markets like Japan and South Korea, presumably included in "48 markets multi-market average" but not explicitly shown individually. Makes one wonder, eh? :V

> Right thats kinda my point, unless I have missed it I have yet to see any real talk about ARM on custom built machines

You can buy an ATX motherboard with an Ampere ARM server chip on it, I wonder if those will be able to boot Windows...

https://www.newegg.com/asrock-rack-altrad8ud-1l2t-q64-22-amp...

Gaming is barely aa blip in the overall market these days, and they're pretty much the only people who want X86 at this point.

> I just don't see many developers bothering with an ARM native Windows version when doing so means they have to support both or risk annoying customers later.

The market dictates what developers do. If Windows on ARM is the new shiny and it hits the three key laptop parameters of no fan noise, long battery life, cool case, then people will buy it and developers will build for it.

Also this time the cpu is actually fast. The previous ones were netbook slow.

I'm a Windows ARM user (Surface Pro X). For me the benefits (fanless, battery not running down randomly in your backpack, phone charger compatibility, integrated LTE, 16G RAM in that envelope), are worthwhile.

No one cares for power saving. Turn it into higher performance at same power usage and people will bite. Of course it has to actually be a real upgrade like the Apple Silicon chips were.

>No one cares for power saving.

I bought a MacBook Pro just to have 20 hours battery life, even if I don't like macOS too much.

Hm true, it does matter a great deal on laptops.

People care about fan noise.

Said like someone who's never actually owned an efficient machine, frankly. Not having a jet turbine under your desk is kinda life changing, actually.

Unfortunately announcement was very thin on actual details. For example is the new surface fanless? How's the battery etc?

As a fan (heh) of fanless PCs, it's remarkably difficult to filter based on that info.

An interview with Satya and the WSJ indicates it is actively cooled. Joanna asks him whether you’ll be able to hear the fans and he says they’re quiet versus missing.

Really? The Recall feature is pretty interesting and potentially quite useful. I see their AI implementation as V 1.0 of bigger things to come.

The huge news is indeed that this time vendors support Microsoft's ARM efforts.

Does anyone know if these chips will have SVE/SME?

To answer my own question, it appears the Oryon CPUs do not according to this LLVM patch: https://github.com/llvm/llvm-project/blob/8aebe46d7fdd15f02a...

There's a pretty good interpretation of it here: https://chipsandcheese.com/2024/05/15/qualcomms-oryon-llvm-p...

> Microsoft ... to guarantee some level of performance and quality.That's possibly huge news.

Only when you lived on a remote island, with no access to news. /s

Reply on news.ycombinator.com