ARM-Powered Linux Laptops Unveiled At Computex

Charbax writes "At Computex in Taipei on June 2-6th, several companies unveiled ARM-powered laptops that are cheaper ($99 to $199), last much longer on a regular 3-cell battery (8-15 hours) and can still add cool new features such as a built-in HDMI 720p or 1080p output, 3D acceleration, connected standby and more. The ARM Linux laptops shown as working prototypes at Computex will run Ubuntu 9.10 (optimized for ARM), Google Android, Xandros OS for ARM, or some Red Flag Linux type of OS. In this video, the Director of Mobile Computing at ARM, is giving us all the latest details on the status for the support of full Flash (with all actionscripts), the optimizations of the web browser (accelerating rendering/scrolling using the GPU/DSP), the stuff that Google is working on to adapt Android 2.0 Donut release for laptop screens and interfaces and more. At Computex I also filmed an interview with the Nvidia team working on Tegra laptops, the Qualcomm people working on Snapdragon devices and the Freescale people doing their awesomely thin ARM laptops in cooperation with manufacturers such as Pegatron as well."

Price?

[siloko]

Nowhere in the article does it mention $99. The quote is "Some of the ARM-based systems will sell for as little as $199." Now $199 is pretty cheap but that is a starting price and will unlikely be the mean let alone allowing for $99 units. The summary is misleading.

Here, we obey the laws of physics

[Kupfernigk]

readable outdoors in daylight

implies either a monochrome e-ink display or something with enough backlighting to overcome skylight - which is where your battery life is disappearing to. Even LED lit displays are not going to give you what you want. OLED may one day get there, but is two technical breakthroughs short.

Even at 100% conversion rates - which are not likely to be attainable - I doubt you would get 20+ hours from a 3 cell battery on a 10 inch screen. A very rough calculation shows that you would need about 2W just to light a daylight readable display at 100% conversion. A 3 cell battery is around 22WH. That means that it could run the display alone for only around 11 hours.

Re:Will they run Linux?

[Erikderzweite]

They will be able to run Windows CE (which is a good thing for Linux though :) ).

Re:Who's gonna sell these?

[Anonymous Coward]

Since Microsoft does not even have an operating system that runs on these machines, if Microsoft were to do as you said, it would be the very definition of an Anti-Trust lawsuit, in the case of using one's business in one sector to influence another. As much as they can get away with that in the States, it'll never fly in the EU. And as the market has shown us, Linux adoption is much higher there anyways, so it would be a massive victory for any company in the EU selling these machines.

So, who's going to sell them? Chinese/Taiwanese/Japanese OEM companies. Who's going to buy them? EU markets. Who's going to advertise them heavily? Absolutely nobody (since that's mainly where Microsoft and Intel come in and stomps companies, which should be illegal but somehow flies).

Re:Here, we obey the laws of physics

[TheRaven64]

implies either a monochrome e-ink display or something with enough backlighting to overcome skylight - which is where your battery life is disappearing to.

Or a transflective TFT. They're fairly common in small devices, relying on a backlight in low-light conditions but being reflective (front-lit) in bright sunlight. Because they don't use the backlight in direct sunlight, the battery lasts longer when they are front-lit.

Re:Price?

[adam1101]

And many people that don't care about "using their old programs or data" because they never owned a computer before.

They will care about being able to use existing Chinese apps and games, which are pretty much all Windows-only. I don't know if you've actually been in China, but Windows is even more entrenched there than it is in the West.

For them Linux is perfect (they won't have to pirate MS Office.)

For them Windows is much better, because all the Chinese software that everybody around them is already using will work, and they don't give a hoot about piracy. In fact, lots of them don't even have a concept of "software piracy". Software is just something you copy from someone else, or buy from the street vendor for a dollar.

Re:$99 huh

[blind biker]

The Eee PC and similar netbooks don't have a CD/DVD drive, either, and yet they sold millions. I don't think people are quite as interested in "that cd you just bought from walmart".

Re:Will they run Linux? RTFA dude, they do!

[Tubal-Cain]

Get Wine on there ASAP and you're away for a good number of Win apps too...

Correct me if I'm wrong, but I get the impression that Wine doesn't translate across architectures. Any windows apps you hope to run on Wine would need to be compiled for Wine from source. So all ARM/Wine apps will either:

• Be compiled for a Windows ARM port. But switching to ARM would be a good opportunity for Microsoft to make a clean break of backwards compatibility, perhaps with an entirely new OS line. Perhaps one called 'Chairs'?
• Be an open source app compiled in ARM. Seeing as most open source apps are already ported to Linux, using Wine on ARM would have novelty value only.

Re:ARM floodgates blown?

[drinkypoo]

HTC has been making phones with Qualcomm chips for a while, I just got myself a refurb HTC Fuze when I sold my soul to AT&T (I live in GSM land, and they own it all here now, literally.) It's also the Sprint-sold HTC Touch Pro. 528 Mhz Qualcomm chip, VGA display, respectable 3d acceleration, halfway decent touch. "A $500 value" free with a two year torture session. In real-world reviewed testing they manage five days of standby time and you get maybe six to eight hours of use... on a 1350mAh battery! (You can get power from any old USB connection with the included dongle, which also gives you headphone and headset connections.) That's a prior-generation version of this same idea, using am ARMv6 core (which runs ARMv4 binaries quite nicely, thankyouverymuch.)

I'm no Windows Mobile fanboy, the phone gets chunky here and there. There ARE some hacks you can make (I used "Advanced Config", which should work across all Raphael devices) to dramatically increase the responsiveness (caching mostly) and you can find a list on xda-developers raphael forums. Touch Flo 3D is no iPhone interface, and you get dropped to the Windows interface on a regular basis, but that's far less odious than it used to be and besides, it's possible to run Android on Touch Pro already. I would never have got this thing if I thought I'd be stuck with WinCE forever. Best acronym ever.

Re:Will they run Linux? RTFA dude, they do!

[wisty]

You could run it in an X86 emulator, but the number of tuples you would burn sort of defeats the purpose of using ARM.

Re:Will they run Linux?

[Anonymous Coward]

I think the main reason is that here in Europe the mobile phone providers have been actively pushing mobile broadband and offer plan and netbook combo deals. Furthermore, it is a lot easier to get a broadband SIM card here in Europe and by far cheaper. Lastly, I would say these seem quite popular with the public-transport commuter crowd.

Re:Will they run Linux?

[dunkelfalke]

I know you were joking, but a Smart is surprisingly spacious.

Re:No one can stop the x86 train...

[TheRaven64]

VIA ultra-low voltage chips

Oh, this made me laugh. The OMAP3 used in these laptops is considered power-hungry in ARM circles. It draws 250mW when using the ARM core (complete with FPU and vector unit), the DSP, the OpenGL 2 ES GPU, the 512 MB of flash and 256MB of RAM and the other integrated components in the package. In contrast, the best 'low power' x86 chips use 2W for just the CPU and need more power for the GPU and supporting chipset. When you factor everything in, the best x86 solutions need over an order of magnitude more power for the same level of performance. Even the Geode has an embarrassingly high power consumption (close to 7W for a complete system, excluding display), and it doesn't even come close to the performance of a 250mW ARM system.

Re:Great. More prototypes.

[JohnBailey]

These things have been hyped on trade shows for over a year now. Call me when they actually have something a consumer can buy in a store.

Does this count.. http://www.maplin.co.uk/Module.aspx?ModuleNo=261613 [maplin.co.uk]

Why phone vendors don't want to play with Redmond

[symbolset]

It's an interesting story. [theregister.co.uk]

Re:Price?

[Anonymous Coward]

The $99 price is officially announced by Nvidia at http://techvideoblog.com/computex/nvidia-tegra-overview-by-michael-rayfield-general-manager-of-the-mobile-business-unit/

Check out that video, I should have linked to that instead of the geeky video with the Engineer dude at Nvidia.

Re:Will they run Linux?

[Theolojin]

Well, they're not going to run Windows any time soon. Good opportunity here. I hope the application availability is going to be good - as a Nokia Tablet user I've been running a variant of Linux on a ARM processor for some time now and I can't wait to get my hands on a ARM netbook.

For all the fun poked at Debian for having such lengthy release cycles, I, for one, am glad that ARM is a fully-supported architecture. (Part of the reason for the long release cycles is each supported platform must be ready for the release.) I could have essentially the same setup on my x86-based laptop as I could on an ARM-based laptop. In other words, application availability really isn't an issue when it comes to the ARM platform.

Re:RiscOS

[david.given]

The new owners of (one of the branches of) RISC OS has actually released it as shared source, available here [riscosopen.org] if you're interested.

I'll agree that the overall design is very elegant, and the module system is nice, if rather primitive (relocatable code with no fixups! Aaaah!). But we know more about operating systems now, and a lot of Acorn's design decisions are no longer valid: for example, you actually get better performance from a preempting kernel than a cooperative one (because you can do work while other tasks are blocked on I/O).

Some of the misfeatures I found include:

• OS_Heap is vastly more complex than it needs to be because it contains lots of code that scans through your stack looking to see whether you're calling OS_Heap reentrantly. Ew. Why? Because on RISC OS you're allowed to allocate memory inside interrupt handlers. Ew ew!
• Virtual location 0x00000000 has to contain a mapped page. Why? Because there is one user accessible byte of kernel private data there, which most of Acorn's modules insist on referencing. This means you can't detect null pointer dereferences. (And the byte is specific to Acorn's RISC OS kernel. R2 doesn't use it, but it's still got to be there.)
• There are about four different heaps available to modules --- all of which are used for much the same task, and all of which have different APIs. (Historical reasons.)
• Two of the standard modules --- FileCore and MessageTrans, I think --- are mutually dependent. FileCore's startup calls MessageTrans, which calls back to FileCore. It only ever worked at all because both modules were preloaded in ROM. If, like R2, you build your module list dynamically at boot time you're in a world of pain.
• A number of the APIs require you to pass in an address to a buffer, where the top 8 bits of the address contain flag bits. The kernel is at perfect liberty to hand you a buffer at an address greater than 0x00ffffff, and indeed, sometimes does.

That's off the top of my head, and that's omitting some of the plain style issues, such as APIs where the user has to pop a word from the stack before returning to invoke certain behaviour; the merry mixing of supervisor-mode and user-mode code (usually in the same code path. Did you know that supervisor-mode instructions on ARM are not guaranteed to be the same from one architecture to the next?); the bizarre maze of callbacks, service handlers, vectors, system control handlers, and environment handlers by which you get information back from the kernel...

Although, as you say, it did all work very well, although I did find my share of programs that would hang/crash/put the computer into some weird state. And it was impressive what it could do; this was all before the 386 had really hit the market.

But I don't think it's at all useful in today's world --- the big killer is that not having threads or processes or Posix it can't run any modern software. So no Chrome, although I note that some hero has managed to get Firefox 2.0.0.12 working on it.

Obligatory flame

[Nursie]

I stopped reading after the printer lie.

Linus supports (other than Lexmark, I hear, but I've never even seen one of their printers here in the UK) more printers than windows these days, and without the need to download 10s of megabytes of extra driver and add-on crap from the manufacturer.

You're an idiot.

Re:Will they run Linux?

[xaxa]

Possibly your wife, when the Hummer rolls over and crushes her...

Re:Will they run Linux?

[bhtooefr]

The ARMs aren't THAT slow. If it's usable on an Atom, it'll probably be usable on a modern ARM.

And, here's the kicker... the ARMs have hardware DSPs that the Atoms don't, making them potentially much faster in certain situations, if software takes advantage of the DSP. (Granted, that does require modification.)

One thing that doesn't require modification to the code, it just requires the JVM to be modified, is Java acceleration - many ARM processors, including the ones under discussion, have Jazelle support, which means that they natively support a subset of Java bytecode. Seeing as "complete development environments" were mentioned... and Eclipse is written in Java...

Re:Obligatory flame

[spandex_panda]

I take up your challenge:

Printer: http://search.dse.com.au/nav/cat2/computersandofficeequipment_printersscannersandfax/cat1/computers-and-office-equipment/0?sessionid=4a2c65840edb2330273fc0a87e0106aa [dse.com.au]

http://ubuntuforums.org/showthread.php?t=953292&page=2

Solution, go to canon website download .debs in a tar. Install and done.

I am actually too lazy to try DVB cards or wireless cards, but I have found the DVB cards work out of the box on Linux usually. My wireless card has given me problems but there are open source drivers which can be easily downloaded and compiled.

Case in point though. I have a USB webcam where the windows drivers have been lost, it works out of the box on ubuntu, you can not find the drivers anywhere for Windows though!

Re:Obligatory flame

[Randle_Revar]

all-in-one printer:
      Wal-Mart: HP Deskjet F2210 - HP works, of course
      Best Buy: Epson - Factory-Refurbished Stylus NX300 - works (print and scan)
      Staples: HP Refurbished Photosmart C6280 - HP works, of course

USB WiFi:
      Wal-Mart: CNet CWD-854 - works
      Best Buy: Dynex Wireless-G DX-BUSB - does not work
      Staples: Netgear WG111 - works

USB TV Tuner:
      Wal-Mart: Hauppauge WinTV HVR-850 USB - works
      Best Buy: Pinnacle PCTV HD mini Stick USB - does not work
      Staples: only tv tuner device was Diamond All-in-Wonder Radeon HD 3650, which I think does not work

6 out of 9 (or 6 out of 8 since a radeon isn't a usb device)? Not too bad, especially since two of the categories are dealing with USB crapware.