ARMdevices.net

@Charbax yes Microsoft could follow Apple’s lead and build a similar tablet on CE or Windows Phone 7, and have developers create CE and Web based applications for it….Mostly, it would be a rival to the iPad with similar battery life attributes.

@Charbax:

In my opinion, Windows on ARM is not interesting if it isn't backwards compatible with the ten thousands of existing Windows-applications.
Therefore, what I thought was the most plausible scenario, is Microsoft is developing a way to run legacy x86 applications on ARM. To make a good x86-emulator, they need access to low-level details of the ARM instruction set.

Also, they were working on the Courier tablet, but it seems they were dependent on nVidia's Tegra2 for it. However, there have been doubts about nVidia, and it might be they don't want to be dependent on 3d parties (such as Intel, TI, Freescale etc.)
See: http://www.semiaccurate.com/2010/04/14/samsung-…

That's an awesome speculation.

Yup it would make sense for Microsoft to realize x86 emulation on ARM and maybe in a way utilize their Azure cloud computing system, deliver apps on demand, through the cloud, using some kinds of x86 app remote virtualization.

This way, amazingly, Microsoft may not even require people run Windows as the OS, they would sell you powerful apps and online tools on demand. They would then support x86 virtualization and emulation on Linux even and eventually go for the whole web apps and cloud computing thing.

Do their ARM as more emulative of x86 as possible, what else?

Sorry Charbax, but I don't agree with you. I don't think cheap ARM laptops are going to gain a lot of ground anytime soon. Especially since netbooks are doing so well. With Intels new dual core Atom and I think SSD's will start to be cheaper soon, Microsoft can just keep putting their OS on cheap netbooks that can already run their OS and already run all the .exe's that have been made and not have to make a whole new OS.
They may try a windows phone tablet. Ipad rival seems good to me but really I'm confused as to what Microsoft is going to do with the ARM license.

Emulating a CISC using a RISC is NEVER a good idea, takes too many instructions. So say it takes on average 5 RISC instructions to emulate a CISC instruction, your 2GHZ arm processor is now effectively running at 400Mhz. THere is zero advantage in this. Its now MUCH slower than a 1Ghz Atom, and is taking just as much power.

Where are these devices, that's what we want to know. Microsoft have caught on at last, but we are all still waiting for Marvell's Armada 510 and Freescale's i.MX535 PC-class System-on-chip based devices to hit the market.

Marvell showed the tiny, beautiful fan-less e-box off in January, with Full Ubuntu-arm kernel support for all those lovely on-chip interfaces, the DSP for video decode and the graphics core, and that was 4 months after the chip started sampling. We are now nearly a year on, and still these Microsoft-eaters are nowhere to be seen.

At the price points talked about ($100-$150 dollars) these machines would really shift, but no manufacturer seems to want to produce the designs.

We want to buy them. Where ARE THEY!!!!

I forgot to use the “sarcasm” tag in my post. On the other hand, Microsoft was never shy to kill any good idea that it could grab.

[...] are going to put Windows 7 on ARM to defend against Linux in the sub-netbook segment. Certainly a useful move – but do you need [...]

WHAT are you talking about ….?
ARM doesn't emulate anything – it has it's own instruction set (not related to X86). Why are you talking about emulation?
ARM is NOT Transmeta which did emulate X86 in RISC.
I think you need to check out what the ARM architecture is all about – its certainly not about trying to emulate X86 or anything else for that matter!

The answer to this question isn't software — there's zero reason for Microsoft to license ARM for any of the things we know that they do. There's also zero reason for Microsoft to license ARM for any use which is well-served by the very diverse ARM ecosystem that already exists.

No, this is about hardware, and its about something that existing solutions aren't suited for (and, by extension, a unique enough use that current ARM license holders aren't keen on addressing). The most likely candidate is Zune or a more-traditional portable gaming device to compete with the PSP replacement and upcoming 3DS from Nintendo. Next up would be some other skunk-works device — something like Courier, maybe, but even that would be quite well served by OTS solutions, or by any of the custom chip integrators.

Some less-likely candidates would be an ISA extension to make x86 emulation more efficient, and/or Jazelle-like instructions to execute managed code more efficiently. Least likely of all, but still a possibility, would be that it has something to do with the next Xbox — ARM isn't known for being a high-powered architecture, but tossing 16-32 ARM cores on a die and hitting 2Ghz in current thermal/power constraints wouldn't be terribly difficult. An ARM core might serve as the ISA for an SPE-like DSP core (a'la Cell), or opposite, might serve as traffic director/general integer compute core for many bespoke SPE-alikes (essentially serving the same role as the PPC core in the Cell, which, on its own is only about as powerful as an Intel Atom CPU, so an ARM core could certainly serve that role).

Yeah Microsoft always looking for monopoly in new product and try to capture global market