- Windows on ARM has suffered a decade of setbacks due to compatibility issues, a lack of an app ecosystem, and the absence of clear advantages over x86.
- Surface RT and Windows RT were Microsoft's first major attempt with ARM, but they failed to better solve any real user problems.
- Apple has achieved a successful transition to ARM with its M chips and the MacBook Neo, thanks to total control of the ecosystem and highly polished emulation.
- While Microsoft grapples with the weight of the x86 legacy and modest sales in ARM, Windows remains strong in gaming, but faces increasing pressure from Linux and SteamOS.
For years we've heard that Windows on ARM was the future , the path to lighter, quieter laptops with incredible battery life. However, while Apple has flawlessly navigated its transition to ARM chips with the MacBook, Microsoft and its partners have been stumbling over the same obstacle for more than a decade. The contrast has become especially stark with the launch of the MacBook Neo, which has reignited the debate about why one company has achieved what the other has been unsuccessfully pursuing for so long.
The case of Windows on ARM is a perfect example of how a good idea can fail if the technical execution, the software ecosystem, and the user strategy don't align. Microsoft has been a pioneer in many areas, but in this particular case, it erred by arriving too early, unprepared, and without addressing people's real problems. Meanwhile, Apple waited, did its homework, and demonstrated that the same bet on ARM can be a resounding success when you control all the elements of the equation.
From the iPhone to the MacBook Neo: why ARM made sense years ago
The surprise many users feel at the performance of the "cheap" MacBook Neo says more about our memory than about the technology itself. Many people wonder how a laptop costing around 700 euros, based on an iPhone-derived chip like the Apple A18 Pro, can run so smoothly. For anyone who has closely followed the evolution of Apple processors, the question sounds almost absurd.
Back in 2015, analysts like John Gruber pointed out that the Apple A9 chip in the iPhone 6S was comparable in power to the chip in a MacBook of the time, a 1,1 GHz Intel Core M mounted in a laptop that cost around $1.300. In other words, more than a decade ago, Apple already had a chip in its mobile phone capable of competing with laptop processors, making it clear that Macs would eventually adopt ARM sooner or later.
The migration officially began with the M1 in 2020, a leap that left half the industry speechless with its combination of power, energy efficiency, and quiet operation. What the MacBook Neo has done now is take it a step further symbolically: it's an affordable laptop, without disguising the "iPhone" origins of its chip, yet it still offers more than enough performance for the average user. Apple has demonstrated that it can repurpose its expertise in mobile devices and iPads to lower the entry price to the Mac ecosystem without sacrificing the overall experience.
This move is especially ironic because, according to former Windows chief Steven Sinofsky, Microsoft attempted something very similar more than ten years ago with Surface RT and Windows RT. Their idea back then was essentially what Apple has now delivered with the MacBook Neo: a lightweight computer with an ARM chip, good battery life, and geared towards everyday tasks. Except, as we'll see, Microsoft ran headlong into reality.
Surface RT and Windows RT: Microsoft's first major ARM bet
When Microsoft launched the first Surface RT in 2012 , it did so with the ambition of paving the way for the future of lightweight computing. Conceptually, the idea wasn't far-fetched: an ARM device with an NVIDIA Tegra processor, a touch-enabled interface, good battery life, and an entry-level price of $499 ($599 with the Touch Cover keyboard), designed to cover the "basic computing" needs of studying, browsing, writing, and consuming content.
The problem was that, in practice, the Surface RT was a case of "wanting to but not being able to ." Those 10,6 inches weren't enough for comfortable work, the tablet format with an optional keyboard didn't quite work as a laptop replacement, and the touchscreen, however appealing it might have seemed at first, proved far less relevant for everyday use than Microsoft had hoped. The mouse and keyboard remained the primary input devices, and the overall ergonomics weren't entirely convincing.
But where the experiment truly failed was in the realm of software and compatibility . Windows RT didn't allow the running of classic Win32/x86 applications, relying instead on a new app model ( WinRT ) for which neither developers nor users were prepared. There was virtually no application ecosystem; the few available applications arrived in dribs and drabs, and many of the usual professional or enterprise tools simply didn't exist on that platform.
The result was devastating: Surface RT didn't solve any real problems better than traditional Windows x86 laptops. On the contrary, it was more limited in compatibility, didn't clearly excel in performance, and its tablet-with-a-keyboard proposition didn't offer compelling advantages for most users. The market reacted with indifference bordering on rejection, and in 2013 Microsoft had to swallow a charge of around $900 million for the Surface RT and Windows RT debacle.
According to Sinofsky himself, the problem wasn't so much the hardware or the premise of using ARM, but rather the desire to push the Windows ecosystem too quickly toward a new application model without providing a solid bridge to the enormous legacy of Win32 software. Windows RT was, in effect, a version of Windows crippled from a compatibility perspective, the exact opposite of what users and businesses were asking for.
Why Windows on ARM is still not taking off despite the advances
More than a decade after the Surface RT, the story of Windows on ARM continues to be fraught with setbacks . Microsoft has tried to relaunch the idea several times: from ultraportable laptops with Snapdragon chips to offerings like the Surface Pro X, and the recent wave of Copilot+ PCs that once again rely on ARM with Qualcomm's support.
At the hardware level, the situation has clearly improved. Qualcomm's current ARM chips for laptops are much more powerful and efficient than those early Tegra chips, and laptop designs are considerably more refined in terms of build quality, screens, keyboards, and battery life. On paper, many of these devices promise good battery life and adequate performance for general use, as well as incorporating AI accelerators for localized tasks.
The problem is that, even with these improvements, they don't manage to outperform their x86 Intel or AMD counterparts in the aspects that truly matter to the average user. They don't clearly outperform traditional laptops in terms of power, their energy efficiency is good but not revolutionary, and their prices tend to be on par with equally capable x86 machines. Add to this the fact that Windows for ARM still suffers from compatibility limitations, and the logical question for users is: why take the risk?
Microsoft has attempted to mitigate the situation with advancements such as Prism , the Windows 11 24H2 emulation layer that allows x86 and x64 applications to run with increasingly respectable performance, including support for AVX and AVX2 instructions. Furthermore, the Redmond giant boasted in 2024 that nearly 87% of usage time in certain configurations was already dedicated to applications with a native ARM64 version, a stark contrast to the Windows RT era.
Even so, everyday reality continues to reveal shortcomings. There are widely used tools that simply don't work, or don't work well, on ARM. Kaspersky's case is illustrative: neither its VPN nor its antivirus offers full compatibility with Windows on ARM, meaning that even in 2026, there will be security and networking software that leaves users who choose this architecture stranded. For the advanced home user and, above all, for businesses, these kinds of deficiencies are unacceptable.
The weight of legacy: companies, compatibility, and fear of change
The biggest obstacle for Windows on ARM is not so much technological as cultural and economic. The Windows ecosystem is burdened by decades of legacy applications , many of which remain critical for millions of business users.
For these organizations, the main requirement when upgrading equipment with tools like Microsoft Intune is obvious: that everything continues to function exactly as it did . When they buy a new laptop or desktop with Windows on Intel or AMD, they take it for granted that this legacy software will boot up without a hitch. It's a trust built over decades, and breaking it would entail costs in support, training, and operational risks that very few are willing to bear.
On ARM, that absolute guarantee is shaky. No matter how good the emulation is or how many patches are released, there's always the possibility that a key application will fail, perform worse, or simply not work at all. Companies hate risk as much as anyone hates an unwatchable reality show, so the incentive to launch onto a platform where perfect compatibility isn't guaranteed is practically nonexistent.
To put it simply, for someone to adopt a Windows laptop with an ARM processor, Microsoft would have to convince them that: all their software will run just as well as on x86 , battery life will be significantly better than on Intel or AMD, and the machine will be considerably cheaper than the traditional alternative. If any of these points fall short, the logical choice is to stick with what they know. And as of today, neither performance nor price clearly tips the scales in favor of ARM in the Windows world.
This puts Microsoft in a very different situation than Apple . While Apple completely controls its ecosystem and was able to nip the sale of Intel-based Macs in the bud, forcing a transition to ARM with extremely polished emulation, Microsoft has to maintain a universe of hundreds of millions of heterogeneous machines and programs. It can't simply announce that, from now on, all Windows PCs will be ARM-based, because half the planet would be unable to run its key tools.
The comparison with Apple: why the transition from macOS to ARM has worked
Apple has achieved with ARM something Microsoft has longed for for years: that most users barely notice the change in architecture. When Macs transitioned from Intel to M1 chips and successors, the translation layers and transition planning were so meticulously executed that for 90% of people, the Mac remained the same old Mac, only faster and with better battery life.
There are several keys to this success. On the one hand, Apple controls the hardware, the operating system, and a very significant part of the app ecosystem, allowing it to coordinate timelines and requirements without having to negotiate with thousands of manufacturers. On the other hand, it didn't offer an escape route : once the transition began, it stopped selling Macs with Intel processors and focused all its efforts on the ARM platform, reducing fragmentation and pushing developers and users in the same direction.
Furthermore, Apple framed the transition as a clear improvement: greater efficiency, longer battery life, and more power , while maintaining similar prices or, in some models, even lower prices than previous generations. The Mac mini with M1 was the perfect example of this approach: a compact, quiet, and surprisingly powerful desktop computer for its price range, demonstrating that a computer could be ultra-efficient without being an unattainable luxury.
In practice, ARM-based Macs solved the same problems as Intel-based Macs, but they did so better in almost every aspect. And, very importantly, the perceived compatibility for users remained extremely high thanks to the quality of the emulation and a well-orchestrated, gradual transition . When a user opened their usual applications, these worked as well as, or even better than, before, reducing friction and minimizing the perceived risk.
That's why the MacBook Neo, heir to that strategy and a direct descendant of the iPhone's chips, has been able to position itself as a kind of "Surface RT that actually works ." It's an affordable, aluminum laptop with great portability, long battery life, and the ability to run the same catalog of apps as any other contemporary Mac. Apple entered the realm of basic computing with its homework done, while Microsoft arrived with a chaotic mess of interfaces, app models, and half-baked architectures.
Windows, gaming, and the strategic shift towards other platforms
Although the core of Windows on ARM's problems lies in general compatibility and the professional environment, its impact is also felt in a key sector: PC gaming . Traditionally, if you wanted to play games on a computer, you chose Windows almost by default. Linux and macOS were relegated due to a smaller game library and more technical limitations, but this dynamic has been changing, partly due to Microsoft's actions and partly due to the efforts of other players.
Valve has been a major game-changer, seriously investing in SteamOS and Proton , enabling thousands of games designed for Windows to run on Linux with a level of compatibility unimaginable just a few years ago. The success of Steam Deck has showcased this progress, to the point that more and more gamers are considering switching to Linux, drawn by a more streamlined environment with fewer background processes and more predictable behavior.
Meanwhile, Microsoft is trying to maintain its dominance in PC gaming by improving Windows 11 and strengthening partnerships with hardware manufacturers. Instead of launching its own Xbox laptop, it has relied on partners like Asus and AMD for products like the ROG Xbox Ally, devices that run Windows 11 but boot into full-screen interfaces designed for controllers, attempting to combine the best of PC and console gaming.
On the technical front, Windows 11 has seen several improvements: DirectX Raytracing 1.2 , tools for precompiling shaders such as Advanced Shader Delivery, work on neural rendering techniques in preview mode, and, in the audio realm, support for LE Audio with lower latency. All of this, along with internal power adjustments, memory management, and CPU load reduction, improves the experience on laptops and portable consoles based on Ryzen APUs.
As part of this same strategy, Microsoft has promoted Windows on ARM as another way to expand the ecosystem's reach, allowing compatible games to be installed and run from the Xbox app on some ARM devices enrolled in the Insider program. The Prism emulator has also been improving the performance of x86 titles and its support for anti-cheat systems like Easy Anti-Cheat and BattlEye, trying to convince gamers that ARM can also be a viable platform for light gaming.
However, even with these advances, users still perceive an x86 PC running Windows as the safest option. The catalog is virtually universal, hardware and software compatibility is well-proven, and Windows' market share on platforms like Steam still hovers around 95%, compared to Linux, which, although growing, remains close to 3%. ARM, on the other hand, continues to be seen as an experiment in testing, not as the standard.
Signs of wear and tear: modest sales and changes of direction
Doubts about the commercial viability of Windows on ARM are not just theoretical. Reports of very weak sales for some Snapdragon-based laptops are starting to circulate widely in the market, to the point that some are even talking about "not selling a single unit" in certain segments. Recent rumors even suggest that Microsoft is more decisively reintroducing Intel (and perhaps AMD) configurations to the Surface line, thus reducing ARM's prominence.
Another striking symptom has been Qualcomm's cancellation of the Windows development kit for Snapdragon , with the corresponding refunds for orders. For a platform that desperately needs to strengthen its ecosystem of native ARM64 applications, the withdrawal of specific developer tools is a bad sign that adds to the feeling of a lack of commercial traction.
Despite all the talk about "AI PCs" or Copilot+ as the next big thing, the everyday reality is that these machines don't solve any problems better than a well-balanced x86 laptop can. Yes, they have AI accelerators and innovative features, but if users encounter incompatibilities, similar prices, and underwhelming performance, their appeal fades.
Meanwhile, the Windows ecosystem itself is facing pressure on other fronts, such as the forced transition from Windows 10 to Windows 11 , which has left many perfectly capable computers without official support due to hardware requirements like the TPM or certain processors. Between this pressure, the fragmentation of user experiences, and the still-nascent experiments with ARM, a sense of burnout is emerging, which other players, like Valve with Linux, are capitalizing on to offer alternatives.
All this context reinforces the idea that the failure of Windows on ARM is not a single, isolated mistake, but a sum of poorly planned decisions , rushed timing, and an enormous structural difficulty in moving a gigantic ecosystem to a new architecture without destroying the compatibility that has made it strong for decades.
Given this scenario, the conclusion is clear: Apple has been able to transform ARM into a tangible competitive advantage with products like the MacBook Neo, while Microsoft remains trapped between its x86 past, an ARM future that has yet to take hold, and a user base that, with very few exceptions, sees no compelling reason to risk switching. Until Windows on ARM offers an experience that unequivocally matches compatibility, significantly improves battery life, and translates into more attractive prices, it will continue to be perceived as a failed gamble that illustrates better than anything else the difficulties of reinventing a giant without breaking it from within.
Passionate writer about the world of bytes and technology in general. I love sharing my knowledge through writing, and that's what I'll do on this blog, show you all the most interesting things about gadgets, software, hardware, tech trends, and more. My goal is to help you navigate the digital world in a simple and entertaining way.