ARM Stealthily Rising As a Low-End Contender

snydeq writes "InfoWorld's Neil McAllister examines how the ongoing rise of netbooks, decline of desktops, and the smartphone explosion are reconfiguring the processor market, putting Intel's Atom processor on a clear collision course with ARM. And here, on the low end of computing, Intel may have finally met its match. Thanks to a unique licensing model, ARM will ship an estimated 90 chips per second this year, and the catalog of OSes and apps available for ARM has been growing for decades, including several complete Linux distributions such as Google's Android OS and Chrome OS when it ships. 'One thing ARM doesn't have, however, is Windows,' McAllister writes, something that could ultimately stymie ARM's plans to compete on the low end of the netbook market. And yet Intel's bet on Windows and its x86 compatibility appeal among developers could backfire, McAllister writes. In the end, it's all about performance. Thus far, Intel has yet to demonstrate a model with power characteristics comparable to those of the current generation of ARM chips, which are fast proving their ability to handle high-performance applications."

Comments

Windows is fading into the background

[Kupfernigk]

It's possible that the desktop dominance of Windows will keep Arm out of the small computer market. But a lot will depend on developments with Oleds and e-ink. Currently the display is the power hog of all-in-one computers, which means that changing the cpu energy consumption makes relatively little difference. But once Oled and e-ink displays reduce the power consumption needed for the display, the cpu becomes more significant. As screen sizes on convergent devices fall - I personally suspect that the 5.5 to 7 inch diagonal screen will come to dominate in truly portable devices - the resulting limit on battery size will be the difference between an all-day device and one that cannot get through a working day. This is where the new generation of Linux distributions like Maemo and Android running on Arm will deliver a visible benefit, and the end user - who doesn't really care whether he has to run "word" or "floop" so long as the document opens correctly and edits - will be more interested in whether he can go from 7 a.m. to 7p.m without a charge.

I'm writing this on a netbook running Ubuntu Netbook Remix 9.10 and it just works (TM). It would work just as well on an Arm processor.

In the real world, I'm sure that Microsoft will be able to roll out Windows Mobile on Arm one microsecond after Dell tell them that their new 7 inch communications centre and ebook reader will have to run an OS supplied by Canonical.

Re:MAME on ARM in Debian

[0100010001010011]

My SheevaPlug arrives via Fedex in about 30 minutes :).

It's going to be like Christmas in a few hours. The Fedex box will be ripped apart strewn across the living room as will be the product packaging. I'll plug it straight into the wall and Ethernet, realize it doesn't do much. Break out the 8GB SD card and not sleep tonight.

Who cares about Windows?

[cheesybagel]

Even if there was a Windows port, if you cannot run the vast set of Windows applications a port is useless. You would be better off running a Linux distro since it effortlessly comes with most categories of apps people need, because said apps are open source and usually can be recompiled fairly easily. If most Windows applications were targeted at .NET by now I could see a point, but they are not.

Re:ARM/Linux in the Tesla Roadster

[shutdown -p now]

The article says that a port of Windows could be important to the future of ARM, and that Microsoft has no plans to do such a port. (Does anybody remember when Windows NT was supposed to be ported to DEC alpha, HP PA-RISC, and IBM PowerPC?).

Thing is, there may not be such public plans today, but don't think it would take all that long should MS change its mind. NT was designed from ground up to be portable; heck, its early builds ran on Alpha before they did on x86. And it wasn't "supposed" to be ported to MIPS and Alpha and PowerPC - it was ported to all those platforms, and successfully ran there, though that configuration was never popular, and so support was dropped in W2K.

In fact, a version of NT running on PowerPC still exists today - it's the nameless OS inside Xbox 360...

Software for a hypothetical Windows ARM port is a more interesting topic. Of course, you can be sure that most Microsoft software - most importantly, IE and Office - would be ported right away. For other stuff, it may not be as hard as it seems - it's not the 90s anymore, and you don't see many people hand-coding asm for performance, or using dirty architecture-specific tricks. Windows went through multi-architecture support pains when x64 and Itanium were introduced - and it was a lot of headache back then, because of all the bad code that assumed sizeof(void*)==sizeof(int) etc - so now the tools are there to handle a transition (C++ compiler will give warnings for nonportable constructs, for example), code for most products that are still being developed had been cleaned up, etc. It's still not quite just a recompile away, but it's close enough.

Which means that pretty much every application that is actively developed for Windows today, you'd probably see ported to ARM in short time should there be demand: Flash, Quicken, new game releases...