Windows laptop to rival M1 Macbook Pro

[Jac459]

Anyway, the M1 never has been the fastes CPU. Only people who never used anything else than Macs think so.

Only you think they think that... Lol.

[Jac459]

Seems we will finally have something we didn't have since 2020 thanks to the new snapdragon...

Great news. I don't like being pinned into the Apple ecosystem.
One of the selling points for Windows has been backwards compatibility; how is this affected by switching to ARM architecture?

Unfortunately it will be exactly the same than MacOS....

[chk071]

Anyway, the M1 never has been the fastes CPU. Only people who never used anything else than Macs think so.

Only you think they think that... Lol.

Nope. I read it lots of times here. I wouldn't say it if I hadn't.

[machinesworking]

Seems we will finally have something we didn't have since 2020 thanks to the new snapdragon...

Great news. I don't like being pinned into the Apple ecosystem.
One of the selling points for Windows has been backwards compatibility; how is this affected by switching to ARM architecture?

Unfortunately it will be exactly the same than MacOS....

So far it's been much worse, Microsoft has never rolled out anything like Rosetta for their Arm chips, so pretty much nothing we would want to use works on Arm Windows.

It's almost always that a manufacturer can ramp up a chip faster once there's a benchmark to work against, in this case Apple Silicon, but the massive issue is no impetus for DAW developers, let alone plugin developers to code for Windows Arm. The ecosystem just doesn't provide a unified monetary incentive to develop something like Rosetta, so adoption is just not happening.

I'll call it right now, Arm Windows is not going to be a thing for niche communities like audio, DAWs, plugins, etc. for at least another 8+ years, and that's if this thing comes out tomorrow with even more impressive specs.

[Jac459]

Seems we will finally have something we didn't have since 2020 thanks to the new snapdragon...

Great news. I don't like being pinned into the Apple ecosystem.
One of the selling points for Windows has been backwards compatibility; how is this affected by switching to ARM architecture?

Unfortunately it will be exactly the same than MacOS....

So far it's been much worse, Microsoft has never rolled out anything like Rosetta for their Arm chips, so pretty much nothing we would want to use works on Arm Windows.

It's almost always that a manufacturer can ramp up a chip faster once there's a benchmark to work against, in this case Apple Silicon, but the massive issue is no impetus for DAW developers, let alone plugin developers to code for Windows Arm. The ecosystem just doesn't provide a unified monetary incentive to develop something like Rosetta, so adoption is just not happening.

I'll call it right now, Arm Windows is not going to be a thing for niche communities like audio, DAWs, plugins, etc. for at least another 8+ years, and that's if this thing comes out tomorrow with even more impressive specs.

I think ultimately Microsoft will improve its conversion layer at Rosetta level.

But one thing I see as a problem in the transition is that Apple was benefiting from being the only offer for RISC processor. So basically you wanted battery and thermal efficiency, no choice. You need to accept Apple ecosystem and Rosetta.

Tomorrow when Microsoft will do the same transition, people will have the choice to be in MacOS which has achieved this transition. Or have the same issues MacOS was having 2 years ago.

And I don't think Microsoft can be bold enough to drop x86.... It can't abandon Intel and its friends... So we will have 3 platforms at the same time.

Sure you always have the die hard windows or Mac fans...
But for the pragmatics laptop users, I think you are right, windows choice is not for exactly tomorrow.

[machinesworking]

I think ultimately Microsoft will improve its conversion layer at Rosetta level.

But one thing I see as a problem in the transition is that Apple was benefiting from being the only offer for RISC processor. So basically you wanted battery and thermal efficiency, no choice. You need to accept Apple ecosystem and Rosetta.

Tomorrow when Microsoft will do the same transition, people will have the choice to be in MacOS which has achieved this transition. Or have the same issues MacOS was having 2 years ago.

And I don't think Microsoft can be bold enough to drop x86.... It can't abandon Intel and its friends... So we will have 3 platforms at the same time.

Sure you always have the die hard windows or Mac fans...
But for the pragmatics laptop users, I think you are right, windows choice is not for exactly tomorrow.

The problem is 100 fold for them though, first off they've never done it. Remember Rosetta is actually Rosetta 2, the first was PPC to Intel. They also have a far less unified ecosystem, most of Windows machines are not made by Microsoft. The only way they will be able to pull it off is if they somehow make a Universal Binary like environment for developers, get most of them on board and provide a translation layer like Rosetta, which they've never done in house.

So all this has to pass by investors and board members, which is why we just don't see the same sort of thing from Microsoft. They will glacially move to Arm probably, but only with serious risk taking, and they're a mostly risk averse company.

[PAK]

I think ultimately Microsoft will improve its conversion layer at Rosetta level.

The problem is 100 fold for them though, first off they've never done it.

They have - just not to the same degree, because they don't have the same control over the ecosystem. It wasn't an accident that Apple killed 32 bit support in Catalina - right before they launched Apple Silicon the next year. The problem, for Microsoft, is some users still expect them to carry all the 32-bit stuff, and earlier, over to ARM too. In the meantime Windows 11 supports Arm64EC. It allows things to be compiled, for x64, and still run natively on ARM. The "EC" stands for "emulation compatible".

All of this isn't just about software though - it requires Microsoft become involved with chip designers. They already did this, in cooperation with Qualcomm, to underwhelming results based around the Snap Dragon. The reason it might be "different this time" is the new Qualcomm chip is really Nuvia's - the start up they bought formed by key Apple Silicon engineers. We'll see what the actual performance is, but it seems it's going to be competitive with Apple Silicon at least.

Either way, Microsoft doesn't just have regular consumers to consider in all of this. They have something much more fundamental to their bottom line - their corporate / data centre Azure clients. They're looking at things like power bills and questioning whether they should stick with Microsoft. So, really, they're under pressure more from that side than the consumer side. Though pressure is also going to heat up on the consumer side because "some analysts" are saying Apple are going to come after the wider Windows laptop / Chromebook market, in 2024, with a "low cost" sub $700 Apple laptop.

So I don't think we'll be waiting anywhere near 8 years to see some major impacts with all of this stuff, and things will likely become clearer with Windows 12. I don't think Microsoft can afford to wait for Intel possibly not to deliver with replacements for their (now stagnating) Core i series.

[uOpt]

Rosetta 2 is really impressive from a software engineering standpoint. Replicating it for Windows ARM might be easier said than done even for Microsoft.

[VariKusBrainZ]

You might be surprised about emulation on windows arm

[uOpt]

You might be surprised about emulation on windows arm

Emulation as in for a virtual machine that interprets at runtime is easier, in fact you can use qemu for that.

A static translation before runtime, but with optional runtime translation like Rosetta 2 does is a different matter. What Apple pulled off there is really impressive.

[koalaboy]

Rosetta 2 works for some scenarios. Anyone using docker (or virtual machines) with x86 images on an arm mac, despite it now having rosetta 2 capabities, knows how terrible the performance still is.

It's also not just the software - all hardware drivers will also have to be re-written, and this could take far longer as many companies don't really maintain multiple OSes for those.

Anyone thinking it would be 'easy', just take a loot at how complex is still is to run Windows audio software on Linux - the same architecture and just a different software layer. Sure, Microsoft would be looking to sort it out, but they barely support other x86 platforms as it is, beyond their own minimal software needs.

[Jac459]

I think ultimately Microsoft will improve its conversion layer at Rosetta level.

The problem is 100 fold for them though, first off they've never done it.

They have - just not to the same degree, because they don't have the same control over the ecosystem. It wasn't an accident that Apple killed 32 bit support in Catalina - right before they launched Apple Silicon the next year. The problem, for Microsoft, is some users still expect them to carry all the 32-bit stuff, and earlier, over to ARM too. In the meantime Windows 11 supports Arm64EC. It allows things to be compiled, for x64, and still run natively on ARM. The "EC" stands for "emulation compatible".

All of this isn't just about software though - it requires Microsoft become involved with chip designers. They already did this, in cooperation with Qualcomm, to underwhelming results based around the Snap Dragon. The reason it might be "different this time" is the new Qualcomm chip is really Nuvia's - the start up they bought formed by key Apple Silicon engineers. We'll see what the actual performance is, but it seems it's going to be competitive with Apple Silicon at least.

Either way, Microsoft doesn't just have regular consumers to consider in all of this. They have something much more fundamental to their bottom line - their corporate / data centre Azure clients. They're looking at things like power bills and questioning whether they should stick with Microsoft. So, really, they're under pressure more from that side than the consumer side. Though pressure is also going to heat up on the consumer side because "some analysts" are saying Apple are going to come after the wider Windows laptop / Chromebook market, in 2024, with a "low cost" sub $700 Apple laptop.

So I don't think we'll be waiting anywhere near 8 years to see some major impacts with all of this stuff, and things will likely become clearer with Windows 12. I don't think Microsoft can afford to wait for Intel possibly not to deliver with replacements for their (now stagnating) Core i series.

I agree with your view...

And on top of that, for big corporates, $ bill is not the only bill to be considered anymore, "ESG" bill is really becoming a strong thing, it is actually the third thing the most discussed by big IT departments arround the world (first thing being "whats on the menu today" and second being "copilot/chatGPT").

It is a desperate effort to reduce CO2 consumption. As a result, if one of the big hardware player (HP, Dell, IBM) manages to provide a new gen of server on ARM, it will sell like donuts next to a police station.

[Jac459]

Rosetta 2 works for some scenarios. Anyone using docker (or virtual machines) with x86 images on an arm mac, despite it now having rosetta 2 capabities, knows how terrible the performance still is.

It's also not just the software - all hardware drivers will also have to be re-written, and this could take far longer as many companies don't really maintain multiple OSes for those.

Anyone thinking it would be 'easy', just take a loot at how complex is still is to run Windows audio software on Linux - the same architecture and just a different software layer. Sure, Microsoft would be looking to sort it out, but they barely support other x86 platforms as it is, beyond their own minimal software needs.

I think nobody here is saying that it would be easy
Other than that I agree on what you said.

[jamcat]

There is no Windows laptop to rival the M3 Max MacBook Pro.