Dell Considering ARM-Based Smartbooks

wonkavader sends us this quote from an article in PCWorld: "In an effort to expand its Linux offerings, Dell is researching new netbook-type devices and will soon offer netbook Linux OS upgrades, a company official said on Wednesday. The company is researching the possibility of offering new Linux-based mobile devices called smartbooks, said Todd Finch, senior product marketing manager for Linux clients, at the OpenSourceWorld conference in San Francisco. The company will also upgrade its Ubuntu Linux OS for netbooks to the latest version in the next few weeks ... Smartbooks with Arm chips have inherent advantages over x86 chips like Atom, such as lower power consumption and longer battery life, according to Finch. The chips are also becoming more powerful, as indicated by the growing number of applications on smartphones, he said. 'I think it's natural and reasonable for us to begin looking at them as they begin scaling their processors up.'"

Re:ARM vs x86

[Anonymous Coward]

Linux. Arm-based netbook already out by Always Innovating:
http://www.alwaysinnovating.com/home/index.htm

Most people don't use netbooks for more than email/browsing. This is great for them.

Re:ARM vs x86

[operator_error]

but it also has a huge disadvantage - it does not run x86 programs

Not necessarily a problem at all. If the user chooses Ubuntu, then synaptec, ( or apt-get, aptitude, etc.) will install an application successfully with something that works, transparently.

Re:ARM vs x86

[kamatsu]

Debian has a complete ARM distribution including all of those things you describe. It wouldn't be hard for Ubuntu to shift their distribution efforts to ARM. In fact, it's just changing a few lines in a shell script.

Re:ARM vs x86

[pantherace]

They shouldn't, but a lot of programmers have gotten used to tricks which work on x86.

Trust me, I've used (at various times) linux/alpha,sparc(64),arm,x86,x86-64,powerpc windows/alpha,x86,x86-64 solaris/x86,x86-64,sparc openvms/alpha.

The most consistent of those are the various Linux distributions, most mainline software has been whacked enough that it works. Though even there, sometimes people use those tricks, or they make assumptions about sizes, Netscape was a problem on alphas, on both Windows and Linux, because it assumed 32-bits on integers and a few other things, when they originally ported it... segfaults. Openoffice still may not compile on alphas, or other 64-bit systems (sparc64s as I recall used to run star/openoffice in 32-bit)

Currently, binaries I can think of are flash, nvidia, ati (Accelerated OpenGL was a pain in the past on many Linux systems), and that's mostly it that I've used for years, aside from some commercial games.

Re:ARM vs x86

[Lennie]

Ubuntu will have an ARM-architecture for their new release: Karmic Koala, scheduled for release in October 2009

Re:Stockholm syndrome

[Lennie]

First of all, these devices have some limited capabilities, that means their is a certain set/type of programs you'd expect them to run, specifically mostly a browser, an e-mail program, some light Office work maybe.

And pretty much all applications in Debian (and soon Ubuntu) are able to run on ARM/Linux. Only other thing you might want is Flash on these devices to possible watch some video's in webpages.

And their has been an ARM-build of Flash for years (look at Nokia N810 for example).

Re:ARM vs x86

[Tubal-Cain]

At the bottom of Mathcad's Wikipedia page [wikipedia.org] you'll find 9 open source options.

Re:This just in...

[pslam]

All instruction sets are like that. RISC or not makes no difference.

The key difference with ARM is it's almost beautiful to look at: it's (mostly) orthogonal, has a regular but very powerful syntax, and it's easy to see the data dependencies. I'd say hand-coding ARM assembly is easier than any other processor I've done it on (and that's lots).

I agree with his friend, and would go further to say anyone who's serious about programming should learn at least one assembly syntax to know what's going on under the hood, and ARM is the best to try.

Re:This just in...

[pslam]

Smaller batteries also means smaller devices, a fact that many manufacturers have been taking advantage of for a long time. For example, almost every mobile phone out there is ARM powered.

Re:ARM vs x86

[pslam]

And Debian has been distributing an ARM version for over a decade, now. It worries me how that fact is missing from reporting.

Re:A Big Up Yours

[shutdown -p now]

If ARM-based netbooks become popular, you will see an ARM port of Win7 in a few months, with a thorough porting guide for applications, tools to check for potential problems, etc (most cross-architecture quirks were already ironed out when x64 and especially Itanium support were introduced).

People kinda miss the fact that most applications are just a recompile away from a different architecture, so long as OS is the same - and not just FOSS code. Yes, you cannot do the recompilation/porting yourself, so there is some disadvantage, but you can be sure that, if there's market, all products that are still being actively developed will be ported.

Re:This just in...

[TheRaven64]

Describing ARM assembly as RISC is barely true. It's RISC in the sense that it's largely orthogonal (i.e. there is only one [sensible] way to do anything) but it's a very rich instruction set. For example, every ARM instruction can be predicated and execute conditionally on the value of a condition register. There is no dedicated shift instruction, just a barrel shifter which can be applied to the result of any instruction, meaning that constant multiplies which can be implemented by a single add-shift combination can be a single add instruction on ARM. Oh, and you definitely don't know your history. The ARM ISA was designed at Acorn as a chip to replace the 6502 in their computers. It was intended to be programmed in assembly language initially (the first OS was and so were many programs) with a BASIC interpreter as the high-level language option. Compilers came several years later.

Re:ARM vs x86

[TheRaven64]

Adobe produces a Flash plugin for ARM Linux, but they do not distribute it to end users. I'm not sure exactly how they do distribute it, but I think they license it (including source code) to SoC manufacturers, who then tailor it towards their CPU and ship it to OEMs.

Re:ARM vs x86

[TheRaven64]

Both ARM and x86 are:

• Little endian (actually ARM is bi-endian, but is usually run in little-endian mode).
• ILP32.

x86 permits (but discourages because they are slow) unaligned loads and stores. ARM does not, but a compiler can work around this with a (slow) sequence of add shift and mask instructions (the shifts are free on ARM) so this isn't really a problem. Porting from x86 to ARM is largely just a matter of making sure you don't use any inline asm or CPU-specific intrinsics.

Of the three examples you list as binary-only, two are drivers so are irrelevant (they will either be ported or OEMs will use different hardware). One is Flash, and Adobe ships an ARM Linux plugin to OEMs already.

This is what I'm excited for

[Anonymous Coward]

http://www.alwaysinnovating.com/touchbook/
Always Innovating Touch Book.

ARM based tablet/netbook with (supposedly) 15h of battery life. The keyboard is detachable, and the tablet itself can run for 5h. The second battery is in the keyboard compartment. This is something that I've had my eye on for a while. And their already working with Ubuntu for a compatible release.

Re:Uh-huh.

[electrosoccertux]

I just bought an Acer Aspire One AO751h.

It has 6-8h (wlan/no wlan, depending on brightness). it's got an Atom z520 and the GMA500 graphics chip (very low power) which has PowerVR which can accelerate 1080p videos. There's some driver issues ATM (jumping through hoops with special settings in KMPlayer) but h264, AVC1 play great. VC.1 plays as well, but I have use DXVA checker to get it to play without dropping frames-- showing that the capability is in the hardware, just needs some driver work.

It weighs 3lbs, has a 11.6" screen (1366x768) and a full-sized keyboard. It's the perfect size for a netbook; the 10.1" screens don't have enough vertical viewing resolution and you end up scrolling up/down all the time in Excel spreadsheets and Firefox/Chrome, especially if you roll with the taskbar on the bottom like most people. 768vert is 28% more viewing area vertically compared with the 10.1" models.

With Win7 on it and 2GB RAM, it flies; I love it.
There's really no need to wait.

Re:Can you scale an x86 processor down?

[faragon]

Yes, they do OoOE, but not with the insane amount of register renaming of the OoOE-x86/OoOE-PowerPC ones, nor with the same alternate execution depth. The ARM Cortex OoOE is a very power-wised balanced OoOE, however, and is just my opinion, completely unnecesary (you could put 3 in-order-execution cores instead of the 2 out-of-order-execution ones).

Re:Can you scale an x86 processor down?

[Hurricane78]

Only that their printer *will* actually print. With CUPS I found it to be less hassle to make a printer work on Linux than on Windows.

Tell me one thing that they would want to do, that is not limited because of performance.
Hardware on Linux: Works.
Browsing, music, movies, e-mail, chatting, instant messaging, etc: All works nicely, and out of the box.

If you think otherwise, that you haven't used any recent Ubuntu or similar distribution.

Additionally, the distribution will of course be adapted to the laptop, by Dell.