For years the sector has been considering Windows failure on ARM While watching with a touch of envy as Apple has effortlessly made its Macs with its own processors dominate the market, the debate has reignited with the arrival of the MacBook Neo, an affordable laptop for an Apple product, featuring an iPhone chip and performance that has left many speechless.
The question you see everywhere is something like: "How can a laptop that costs around 700 euros run so well with a mobile phone chip?" To be clear, Doubt comes late and is badIt was clear for over a decade that ARM processors would seriously challenge x86 chips in laptops, and Apple has been able to take advantage of that trend while Microsoft remains stuck in its own past.
From iPhones to the MacBook Neo: the logical path towards ARM in laptops
When John Gruber reviewed the iPhone 6S in 2015, he found that the Apple A9 performed comparablely to the 1,1 GHz Core M in that year's MacBook, a machine that cost over $1.300; that fact alone made it clear the brutal power of Apple's mobile chips compared to much more expensive laptop processors. It wasn't a technical curiosity; it was a very clear sign of where the market was headed.
From the beginning of the last decade, it was clear that the SoCs in iPhones and iPads matched or surpassed many x86 laptop processors; this made it inevitable that Apple, sooner or later, abandon Intel and make the definitive leap to ARMThe fact that the change came with the M1 in 2020 only confirmed something that the most attentive had been waiting for for years.
The M1 blew us all away with its blend of power and efficiency, but the MacBook Neo has had a different effect: it has democratized that idea in a much more affordable format, with an Apple A18 Pro chip straight from the iPhone. While it may be less powerful than the desktop or high-end M series, For the average user, the performance is more than sufficient. and the price is in a range where there are hardly any rivals as well-rounded.
What's striking is that Apple, a brand famous for not being cheap, has launched a laptop with this positioning: capable hardware, good battery life, a mobile ARM chip, and a very competitive price. This move, which many had long considered possible, has been interpreted by some as proof that The vision for Windows on ARM made sense…but it arrived at the wrong time and with the wrong execution.
In this context, Steven Sinofsky, former head of Windows and Windows Live, emerges. He has purchased a MacBook Neo and describes it as a paradigm shift. What's interesting is not only his enthusiasm for the Apple device, but also his direct comparison to what Microsoft attempted in 2012 with Surface RT and Windows RT, asserting that, ultimately, That was the MacBook Neo of its time.But the market wasn't ready.
Surface RT and Windows RT: Windows' first major stumble on ARM
When Microsoft launched the Surface RT, the idea on paper didn't sound so far-fetched: a lightweight device with an ARM processor (NVIDIA Tegra), good battery life, a relatively affordable price, and a Windows interface adapted to the new touchscreen format. In practice, That proposal clashed head-on with the reality of the Windows ecosystem and ended up becoming a textbook example of how not to make a platform transition.
The Surface RT featured a 10,6-inch screen, a size that seemed manageable on paper, but which in real-world use fell short for extended work sessions. The touchscreen was eye-catching and attractive, but time has shown that for most serious tasks, people still prefer a keyboard. mouse and keyboard as main toolsTouchscreens are fine as a complement, not as a complete replacement.
The real problem, however, wasn't so much the hardware as the software. Windows RT was, in essence, a Windows that looked like Windows, but that I couldn't run classic Windows applications based on x86. The entire Win32 universe that had made the platform so popular disappeared at a stroke in these computers.
Users were faced with a system that was unsuitable for running their usual programs: there was no compatibility with most games, productivity tools, utilities, or long-standing professional applications. Microsoft tried to compensate with an adapted version of Office and a few select titles, but it was clear that the catalog was meager. For anyone who had been using Windows for years, Windows RT was not "the Windows we've always known"And that weighed heavily on the purchase decision.
In the end, Surface RT and Windows RT didn't solve any specific problems: they didn't offer better performance than a conventional laptop, they didn't have a mature app ecosystem like iOS or Android, and to top it all off, They did not allow taking advantage of the gigantic catalog of Windows x86 softwareIn practice, the product was left in no man's land, a hybrid that did not stand out in anything and which, logically, was rejected by the market.
An operating system that looked like Windows… but wasn't
Many analysts and users criticized Windows RT from day one for being "Windows that isn't Windows." For years, the platform's great value had been the ability to use applications like Photoshop, AutoCAD, media players, office tools, messaging, games, and utilities of all kinds; suddenly, Microsoft intended for the public to accept a cut-down version where the vast majority of that software did not exist.
The situation was complicated because, in parallel, tablets with iOS and Android were growing at a breakneck pace, with app stores full of games, productivity apps, email clients, social networks, multimedia, and just about anything the average user could ask for. Compared to that, the Windows Store catalog for RT seemed meager. far behind in quantity and quality of the dominant mobile alternatives.
On the other hand, the Surface Pro range of Windows x86 tablets demonstrated that there was indeed a demand for hybrid devices that ran "full Windows." These devices, with Core i5 processors, offered complete compatibility with desktop applications, albeit at much higher prices; the implicit message was clear: If you wanted the authentic Windows experience, you needed x86and pay for it.
This dichotomy confused the average user: on one hand, there was a cheaper but limited Surface RT tablet, on the other, the more expensive but fully compatible Surface Pro. Instead of a clear evolution to ARM, Microsoft created a scenario where the traditional version of Windows remained the standard, while Windows RT seemed like a half-finished experiment..
The financial result was devastating. Less than a year after its launch, in July 2013, Microsoft had to acknowledge a loss of around $900 million due to the Surface RT and Windows RT fiasco; they were left with a significant stock of unsold devices, and the market had already practically written off the platform. In the eyes of many, Windows RT was doomed from the start..
The ecosystem factor: where Microsoft stumbled and Apple did its homework
Steven Sinofsky insists that the failure lay neither in the hardware nor in the overall concept, but in the management of the ecosystem. Microsoft presented Windows RT as a version of Windows for ARM focused on the new WinRT platform, more secure, with modern touch apps, and designed for mobility. The problem is that The company tried to push the ecosystem too quickly.without giving developers and users a comfortable bridge between the old Win32 world and the new application model.
Meanwhile, Apple pursued a completely different strategy. When it introduced Macs with Apple Silicon processors, it didn't do so as an optional alternative, but as the new standard: it stopped releasing new Macs with Intel and forced the transition of the entire range to ARMHowever, great care was taken to ensure that most users would barely notice the change in architecture.
To achieve this, Apple implemented a series of measures that Microsoft has never been able to replicate with such success: tools to easily recompile apps, clear incentives for developers to migrate their applications, and an emulation layer so polished that, for most, The non-native apps continued to function without any noticeable problems.The result was that Mac users didn't feel like they were entering experimental territory.
In contrast, during the Windows RT era, users encountered a device where the "new" ecosystem was practically nonexistent, and the "old" one was incompatible. This head-on clash with public expectations led to The Surface RT seems worse than any high-end Android tablet or iPadbecause they neither had a good touch ecosystem nor supported classic desktop applications.
Today, with the MacBook Neo, Apple has a clear advantage: its ARM ecosystem has been well-established since the M1, the tools are mature, and the average user understands that if they buy a Mac, their programs will work, the battery will last longer than on an average laptop, and the performance will be very solid for everyday use. In other words, The MacBook Neo solves the same old problems, but more efficiently.while Windows RT didn't solve anything that wasn't already better covered by other products.
Windows on ARM in 2024-2026: Copilot+ PC, Snapdragon and the same old ghosts
A decade after the Surface RT debacle, Microsoft returned with Windows on ARM, relying on Qualcomm chips and the narrative of AI-powered computers, the so-called Copilot+ PCs. On paper, these machines promised high autonomy, competitive performance, and integrated AI capabilities thanks to their NPUs, all in more efficient ARM architectures.
However, reports that have emerged indicate that the Snapdragon 8 for laptops is barely selling in the general market, and rumors suggest that Microsoft is... reconsidering its exclusive bet on ARM in its own Surface, studying a return to Intel and even AMD in some future models.
Adding to this climate of uncertainty is the news that Qualcomm has canceled its Windows development kit for Snapdragon, refunding existing orders. For many, this move is a sign that the bet on Windows for ARM has not taken off as expected and that There is not enough critical mass of developers and users as if to maintain the momentum solely through faith in the platform.
In 2024, Microsoft boasted that 87% of usage time on Windows on ARM was already being spent on native ARM64 applications or optimized experiences, thanks to Prism, its new emulation layer for improved performance in x86 and x64 applications. On paper, this figure sounded very promising and reflected a significant improvement over the Windows RT era, where virtually There was no native software ready for ARM.
Even so, market perception remains lukewarm. For the professional or business user, any doubts about compatibility far outweigh the potential advantages of performance or battery life. If a corporate VPN, a specific security solution, or a critical application doesn't work on ARM, The purchase decision falls apart completely.However eye-catching the marketing of integrated artificial intelligence may be.
The burden of legacy: why Microsoft is having such a hard time
One of the key factors explaining the relative failure of Windows on ARM is the weight of the past. There are millions of Windows computers in businesses and homes running applications that are between 5 and 40 years old, many of them custom-developed, without active maintenance or a cloud alternative. For these users, Backward compatibility is not a whim, it's an absolute necessity.
When an organization upgrades its PC fleet, it wants assurances that everything will continue to function exactly the same. With a Windows laptop running on Intel or AMD x86, those assurances are high; with ARM, there's always the question of whether emulation will be sufficient, whether there will be performance issues, subtle incompatibilities, or bugs that will only surface when it's too late. The perceived risk is enormous and, by definition, Companies hate technological risk they cannot control.
For someone to consider buying a Windows laptop with an ARM processor, there are three basic conditions that must be clearly met: that everything works the same as on an Intel or AMD PC, that the battery life is as good as advertised, and that the price is significantly lower than the x86 alternative. If even one of these points is lacking, The proposal makes no sense and the user sees no reason to complicate their life.
Currently, the ARM chips used in Windows don't outperform the best x86 options in overall performance, the battery life advantage isn't always dramatic, and the final price of the systems is usually similar or even higher. If we add to that the fact that Windows for ARM still suffers from occasional compatibility issues, the conclusion for many is simple: There is no clear incentive to change.
Microsoft, moreover, cannot do what Apple did: it cannot suddenly turn off the tap and declare that, from now on, everything will be ARM. With an installed base of billions of devices and an enterprise ecosystem tied to legacy software, any abrupt move could trigger a chain reaction of problems, migration to other platforms, or a massive freeze on updates. In that context, Forcing a complete transition to ARM seems almost impossible in the short term..
Apple versus Microsoft: two opposing strategies for the same challenge
Apple has won this battle, at least for now, because it has applied a radically different strategy. When migrating its Macs to ARM, the company opted for a relatively quick but carefully planned transition, supported by a mix of transparent emulation, powerful development tools, and crystal-clear messagingIf you want a Mac, it will have an ARM chip; there is no alternative.
The great advantage of this approach is that the user barely notices anything: they continue using their usual programs, which are updated to native ARM versions seamlessly, while simultaneously enjoying tangible improvements in battery life, noise, heat, and, in many cases, raw performance. The Mac mini with M1 was a perfect example: an inexpensive desktop within the Apple standard, very efficient, quiet, and surprisingly powerful for its price. It offered more than many expected from an entry-level Mac..
With the MacBook Neo, Apple has lowered the entry bar for the ARM ecosystem in laptops even further, offering an affordable machine—by Apple's standards—based on an iPhone chip, with good battery life and seamless integration with macOS. Users don't have to worry about architectures or complicated compatibilities: they buy a Mac and Everything works as expected from a Mac, without further complications.
Microsoft, on the other hand, has been erratic. First, it presented Windows RT as the natural successor to x86 for certain product ranges, but without a solid plan to populate the new ecosystem with apps. Then, with Windows 10 and 11, it maintained ARM as a kind of advanced alternative, interesting from a technical point of view, but never the platform's primary focus. This ambiguity means that developers and manufacturers haven't fully committed to making a strong investment through architecture.
Meanwhile, attempts to revive ARM with campaigns like Copilot+ PC are hampered by memories of the past (Surface RT) and lingering doubts about compatibility and real-world performance. In contrast to the coherence of Apple's ecosystem, Windows on ARM... It still looks like an ongoing experiment. for many potential buyers.
All of this leaves us with a curious scenario: Microsoft was one of the first major players to attempt to bring a complete desktop system to ARM with the Surface RT, but it failed; Apple arrived later, more calmly, preparing the ground for years with its iPhone and iPad chips, and when it made the leap to Mac, it did so with almost all its homework done. Today, the MacBook Neo is perceived as the successful materialization of an idea that was too far ahead of its time in 2012 in its time in the Windows world.
Ultimately, what all these attempts show is that simply changing the architecture isn't enough: the new product needs to clearly improve upon what already exists and solve real problems for the user, whether in performance, battery life, price, or ease of use. At this point, Windows on ARM remains stuck, trapped between a past it can't shake and a future that hasn't quite materialized, while Apple's ecosystem has managed to make its move to ARM seem perfectly normal.
