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.'"

Uh-huh.

[XanC]

And what reason do we have to believe this isn't a just negotiating tactic against Microsoft?

This just in...

[MrMage]

Lower power consumption leads to longer battery life.

In all seriousness though, I once had someone tell me as I was looking into programming in assembly that I should learn an ARM-Based syntax. It still hasn't paid off completely yet, but this is a step in the right direction.

MS will adapt. Eventually.

[Tubal-Cain]

MS might not be selling any ARM-compatible systems at the moment (embedded OSs aside), but I would bet they have experimental ARM builds of everything they've produced in the past 5 years.

Comment removed

[account_deleted]

Comment removed based on user account deletion

Re:MS will adapt. Eventually.

[Anonymous Coward]

The very applications that keep a lot of people running Windows instead of e.g. Linux also keep Windows firmly locked to x86.
Take away the third-party closed source applications/games, and suddenly Windows is looking pretty crappy even to your average consumer.
Apple handled this with emulation, but they were moving to a faster chip.

Re:Can you scale an x86 processor down?

[0123456]

Or does x86 inherently consume more power at the same performance level?

Difficult: ARM has traditionally had a very clean instruction set which eliminates a lot of the junk that an x86 requires in order to function, and it's much easier to take a chip designed for low power and increase the performance than to take a 100+W monster like an x86 and scale it down for low-power use. The modern 'x86', at least from Intel, is basically an x86 emulator wrapped around a RISC core.... the ARM effectively eliminates the emulator and just runs the RISC core.

If I remember correctly, the dual-core ARM chips I was working on a couple of years ago used about 1W of power to play 720p HD... an Atom has trouble doing that even with several times that power usage.

Re:This just in...

[amorsen]

The tradeoff supposedly was that RISC would give you less powerful instructions which were easier for the CPU to decode, but then it was expected that it would be more difficult for compilers or humans to write the instructions. It didn't turn out that way. E.g. x86 comes from the time of constrained 8-bit processors, and back then there were no wasteful niceties in the instruction encoding. Most of the worst nastiness is gone as of x86_64, and assemblers hide some of the rest from you.

If you want to program CISC, at least go for M68k/ColdFire, but you won't find many user-programmable devices with ColdFire anymore. ARM is everywhere. My personal favourite is SPARC, and SPARC machines are relatively easy to come by.

Test baloons?

[achten]

Do not know if it is due to the reporter or the strategy itself.
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.
It ends with
The company is also researching Google's Chrome for use in netbooks.
Makes netbooks-are-atom-and-smartbooks-are-ARM distinction.
However
Dell couldn't say whether it would ultimately offer a smartbook.
Maybe just floating of test baloons.

Re:Uh-huh.

[binarylarry]

Microsoft is planning to build "Microsoft PC" products that are Microsoft Software+Hardware.

You think Dell is just going to see back and watch that happen and not have a plan B?

Re:Ripe for adoption

[bemymonkey]

I don't think so. The whole point of a little netbook ("Oooooh, look at the cute little laptop!") is making your regular (lightweight) notebook apps portable. For the majority of consumers, that means they want to run exactly the same e-mail program, the same browser, the same IM program(s)... Realizing that they need to learn to use a completely different interface is going to be quite a shock. It was supposedly pretty much the same with the Ubuntu laptops certain manufacturers have been selling...

I don't really see the appeal in an ARM netbook. In fact, I wish my phone was x86 - the current standard Windows Mobile smartphone res of 800x480 is just fine for a stripped down XP or even Win7... hell, if they could just get the damned things up to 24 hours (or even 15 or so!) without a recharge, I'd be more than happy. If all you're interested in is ARM and long battery life, get a smartphone and a foldable bluetooth keyboard. Otherwise, get an x86-based netbook and be satisfied with the currently available 10 hours of battery life...

Re:Uh-huh.

[simula]

I have been raptly awaiting Pegatron's $200 arm netbook with an 8 hour runtime:
from January [engadget.com]
from July [ubergizmo.com]

If Dell is willing to ship what is practically the same device, then this competition can be nothing but good for everyone who wants one.

Re:This just in...

[Rockoon]

Lower power consumption leads to longer battery life.

I'm not sure that you understand how things work.

Lower power consumption leads to smaller cheaper batteries with the same capacity, a fact which manufacturers will surely take advantage of to increase profits. As long as competition exists which uses less efficient CPU's, thats the way its going to be.

Nice idea, but let's wait for what Apple is up to

[cheros]

Like it or loath it, Apple has seriously shaken up the mobile phone industry, and got away with something nobody else ever managed: taking a big slice of the carrier's cake on top.

If Appe brings out a sensible iTablet that actually works and is smart enough to work with the laser keyboard [vkb-support.com] (the Bluetooth version does proper HID support) I cannot see that fail, and it will probably nuke the market Dell is looking at.

The tablet in itself goes into markets at present taken by ebook stuff like the Kindle, and with a proper remote keyboard it hits the portable market - why take a whole system if it's that portable.

So I'd wait a bit - let's see what Apple is up to. I hope I'm right - it's about time for such a device.

Re:Uh-huh.

[doctormetal]

ehh? Microsoft has an OS that is suited for such devices. Either windows embedded CE or Windows Embedded nav ready.
Or even the CE derivative Windows mobile.

See http://msdn.microsoft.com/en-us/windowsembedded/dd630116.aspx [microsoft.com]

Re:This just in...

[lobiusmoop]

Agreed. I'm looking forward to getting something like the Gecko Edubook [laptopshop.co.uk] which can run on cheap AA batteries instead of an expensive custom Li-Ion battery.

Stockholm syndrome

[xororand]

It's interesting how some people are quick to declare portable ARM computers a failure because it won't run their favorite (proprietary) x86 programs.
That's the Stockholm Syndrome, only with software instead of human kidnappers.

Re:Can you scale an x86 processor down?

[Anonymous Coward]

ARM systems usually use co-processors for things like decoding video. A heavy use of co-processors lets the already low-power CPU be idle even more often.

Re:Can you scale an x86 processor down?

[faragon]

It is not the "x86 emulator", as it takes a tiny percent of the die, and 90-95% of instructions are decoded to one underlying RISC equivalent. Most power consumption is because of OoOE [wikipedia.org], huge pipelines, and huge caches. In my opinion OoOE processors are an aberration inteded to maximize serial code, by wasting 4 to 8x resources, as it is like having many processors executing future code paths "just in case" (misusage of instruction cache just to feed the OoOE jump prediction execution paths) while making a misuse of the system bus by loading data for instructions that will be discarded 1 of every 10 times (data cache misusage by fetching data for instructions that will be discarded in a major part). So in "advanced OoOE CPU" you're saturating the bus for computing worthless instructions. As example, in the area of a P4 CPU, you may had 8 to 16 MIPS or ARM in-order CPU cores, making much better usage of the shared cache, and with 4 to 8x more executed instructions/transistor, with efficient system bus usage.

Comment removed

[account_deleted]

Comment removed based on user account deletion

Re:MS will adapt. Eventually.

[bhtooefr]

What MS could do is change the executable format for Windows 8, to allow for fat binaries, and then make their compilers always compile to x86/Itanium/ARM fat binaries.

(Speaking of that, that's why MS was pushing .NET so hard, because Itanium was supposed to be the future, and MS didn't want to get left behind by CPU architecture changes - hence trying to move everything to interpreted bytecode.)

What's a 'smartbook'?

[TeknoHog]

I'm afraid it will be something weaker and smaller than current netbooks. A toy computer, compared to the real computers that run Windows on x86, like God intended.

The point is, why can't we have a regularly sized laptop with a sensible processor like ARM?

Re:A Big Up Yours

[TheRaven64]

Win64 is a bit weird, it's an IL32P64 architecture, which breaks a lot of C code aimed at x86 and full of assumptions like sizeof(long) >= sizeof(void*) (which is true on almost every other platform). ARM is a lot closer to IA32 than x86-64 is if you're coming from a C-like language. If you're not using inline assembly, it's probably easier to port from x86 to ARM than to x86-64. Int, long, and void* are all the same size, the alignment requirements are almost the same (a few things don't work on ARM, but they're things that are painfully slow on x86 anyway), and they're both little-endian.

I will buy arm architecture

[FudRucker]

in a laptop or netbook if it means longer battery life, I dont use laptops netbooks for CPU/GPU intensive things, mostly web surfing & email, IM, and occasional typing of documents in OpenOffice on Linux and since Linux already supports arm the switch to that architecture would be seamless...