Slashdot Mirror
ARM Attacks Intel's Netbook Stranglehold
Barence writes "British chip designer ARM is launching an outright attack on Intel with the launch of a 2GHz processor aimed at everything from netbooks to servers. ARM claims the 40nm Cortex A9 MPCore processor represents a shift in strategy for the company, which has until now concentrated on low-power processors for mobile devices. In the consumer market, ARM is pitching the Cortex A9 directly against Intel's Atom, claiming the processor offers five times the power while drawing comparable amounts of energy. 'It's head and shoulders above anything Intel can deliver today,' ARM VP of marketing Eric Schom claims. However, it has one major hurdle to overcome: it doesn't support Windows. 'We've had conversations with Microsoft and you can imagine what they entail,' says Schom."
Broken, first gen/beta ARM drivers for all my hardware!
moox. for a new generation.
I suppose Ubuntu Linux is just chopped liver.
C'mon people. Wake up! There are tons of operating systems out there. Some are even better than Windows! *gasp*
Linux already made MS drop their price, allwing cheap windows netbooks because of linux. It's not out of the question that a really compelling ARM netbook would scare them into ARM support. I would be surprised if they didn't have something similar to the x86 apple builds in the powerPC era. Of course windows is mainly valuable for its 3rd party software so people who buy these putative ARM/windows machines may be better off with linux anyway.
This program was made possible by a grant from the Ultra-Humanite, and viewers like you.
What is involved in porting code to a new chip? I've done some programming in my life, but it has mostly been limited to personal interest and school projects. I imagine it can't be as simple as just recompiling. So what does it take to port code?What are the hurdles? Assume (accurately) that I'm a total noob.
This won't end well. I have an ARM device (nokia n810) and it's great. But Wintel monopoly will kill this just like it did Sparc and IBM Power. I'm sure if it's as good as they claim it'll carve out a niche, but it won't directly compete in numbers or presence with intel CPUs.
Does ARM plan on integrated video along the lines of Nvidia and ION? http://www.nvidia.com/object/sff_ion.html
netbooks are a great place to quietly slip in non-windows OS's that meet customer needs. the mobile phone/smart phone market has shown that customers aren't slavishly devoted to Windows. they will buy what works.
This may be the first time I've ever heard it said that a processor doesn't support an OS... Usually it's the other way around.
Actually it's pretty much the other way around the fact that M$ doesn't yet support the processor is down to M$ not Arm. I suspect there are a couple of factors here firstly M$ PPC software supports older ARM processors & secondly it's just a case of re-tooling some of M$'s compilers to support the new processor and recompiling. However how it's going to support non .net/java windows software is another matter.
Build a Man a Fire, and He'll Be Warm for a Day. Set a Man on Fire, and He'll Be Warm for the Rest of His Life.
Clock for clock, the Cortex A8 is a bit faster than the Atom on most workloads (in about 10% of the power envelope). The A8, however, typically ships at about half the clock speed of an Atom (they go up to 1GHz, but 600MHz is the most common speed). The A9 is slightly faster than the A8 clock-for-clock, but goes to twice the clock speed and scales to four cores, so it's not a stretch to imagine that it's more than five times the speed of a single-core Atom. I've not seen any figures for the A9's power consumption yet though...
It's worth noting that ARM doesn't make chips, they are an IP-only company. ARM licenses designs to other companies who combine their cores with other stuff and ship them. One of the more high-profile Cortex A9 licensees is nVidia, who are using it in their Tegra line. Other existing ARM licensees, like Qualcomm, TI, Samsung and Freescale have already signed up for the A9 as well.
It's also worth noting that the A9 isn't really news. The designs have been available from ARM for a while now. I don't know of any shipping chips including A9 cores yet (being mass-produced, anyway; there are a few being sampled), but TI announced the OMAP4 series a little while ago which is based around the A9 and looks like a very nice chip for handheld machines.
I am TheRaven on Soylent News
People are willing to pay $15 more for XP (the cost of an XP Netbook license), but are they willing to pay $100 more for Windows (the difference between the cost of the announced ARM-based netbooks and a typical x86 model)?
I am TheRaven on Soylent News
A Microsoft refusal to support a really cool netbook technology would be a good opening for Linux.
That's like saying "Linux or even Ubuntu". :)
Microsoft used to have a laptop/netbook-friendly Windows CE version back in the late '90s, but dumped it in favor of the "Tablet PC" build of Windows NT around 2000-2001. It would be interesting to see them bring that back.
They still do, the problem is it's shit and it won't run any off-the-shelf applications. It's used in a number of industrial PDAs, particularly ruggedized, intrinsically-safe ones.
The way I see it, using CE on a laptop is far worse than Ubuntu because it looks like windows (95), behaves (mostly) like Windows, but won't run any Windows apps. In some ways it's the perfect combination - you get all the 'It-won't-run-Outlook/Oblivion/Photoshop' problems of Linux, all the 'It-won't-work-with-my-USB-doodad' problems of OpenBSD and all of the bugginess of Windows.
And unless it's CE6 (WM and most devices are still CE5), it will have that abysmal 32MB-per-application limit, so good luck porting any substantial win32 apps to it.
Much as I'd like a linux ARM netbook, I am a little worried that they don't seem to have 64-bit addressing in that architecture yet. It won't be so many years before it becomes a needed feature for a netbook too.
Actually, microsoft supported non intel before. Anyone remember the DEC Alpha chips? There was an NT flavor for that. It ran faster then the intel chips of the day.
It would not surprise me that in an microsoft lab there was windows for power PC, windows for ARM, windows for . It would be in microsoft's best interests to have them.
A) much simply needs recompiled, if it doesn't - with an app with the source - it's usually a bug.
B) No - wine is simply a conversion layer between the windows and linux calls - the windows program is never emulated.
C) No - again - not without emulation.
D) I think you can probably guess this one - but again no.
Emulation may be _lots_ slower than the host processor - slowdowns of ten times or more are not uncommon.
I'm guessing that means it doesn't (can't) do most of what an ATOM can do. No x86 support is kind of a dealbreaker.
Windows has pretty much a lock on the desktop, so the new chip won't have much market there. The desktop is also the declining market, so the new chip won't be missing that much. The big growth will remain in servers, where windows is optional at best, and netbooks/mobile devices where windows is a minority player. ARM may have made a rather astute decision to concede the dying segment to Wintel and make a big footprint in the markets that will continue to grow, and which also happen to do just fine without Windows. If they make sure to brilliantly showcase the not-windows OSs, ARM could come roaring back as a force to be reckoned with in consumer-level computing.
You are not all that bright. Some might even call you an idiot.
The ARM instruction set is not x86 compatible. End of story.
There are a lot of barriers to Windows adoption on the ARM processor that go beyond MS not really wanting it. If they really want to gain market share above and beyond cell phones and PDA's, ARM needs a strong partner to create a real, integrated, polished solution. And by solution I don't mean a device. They need to do something akin to the iPhone, in creating a nice device or set of devices with a consistent polished operating system and with an integrated ecosystem of solutions. The project is large in scope and they need a partner that preferably has an existing position to leverage, experience, money, and which is not beholden to Microsoft. A cell phone service company might be a viable partner or Canonical and someone, or RIM or Google or an appliance maker that has not entered the netbook market yet.
If they really want to sell netbooks with ARM processors in them they have to think big. They need to better than hope MS is scared. They need to commit to building a system that bypasses MS's core monopolies through vertical integration. This is no small task. They need the hardware, which has to be cheap and hit a sweet spot. They need an OS and applications. They need dev tools for applications and services. They need Web and network services integrated with the device. More than all those pieces which are out there, they need someone to put it all together in a nice package and usability test the whole user experience from buying to opening the box right up through using it for all the common tasks: Web surfing, E-mail, chat, word processing, potentially phone calls and videophone, playing games, playing music and video, and adding new applications. The problem with a lot attempts at this sort of thing is the assumption that someone else will take care of parts or that blaming someone else somehow makes a failure better.
As a product of British manufacture, is it safe to assume it will spend most of its lifetime at the computer repair shop?
That'll be news to the folk that have been using computers with ARM processors since the very early 1990s.
It depends on the distro. Debian has a complete ARM-port, Ubuntu was working on one last time I checked. Maemo is an ARM-only distro.
Nope.
Nope.
Not likely (assuming these are binary blobs). Flash video, avi/mpeg's and various other formats shouldn't be a problem though.
An ARM netbook wouldn't be someones only PC, just like current netbooks aren't. If it can do 90% of the things you're used to you're set.
This sig is intentionally left blank
Well, you wouldn't necessarily expect x86 support on a non x86 architecture, would you.
It need not, and should not, be a deal breaker though. Windows has run on other architectures in the past - Windows NT and its successors have variously run on PowerPC, Alpha and MIPS and Itanium.
Imagine a computer that does not run Windows. One that is not able to run Windows!
I want one. Now. (I assume that it runs a full Linux distro of course).
However, it has one major hurdle to overcome: it doesn't support Windows.
Fuck Windows. Seriously.
I've been unwillingly paying the Microsoft tax for TEN YEARS. All I ever do is wipe Windows and install Linux. If my new computer can't run Windows then... great!! Maybe I won't have to pay the tax.
I'd love a low-power, high-performance ARM notebook. I'd be happy with MIPS or Loongson (Chinese MIPS clone) as well. Debian already has a full-blown ARM port and I'll bet they could get it working on an ARM netbook in a day. Ubuntu would undoubtedly be soon-to-follow.
As a side benefit, having multiple widely-used architectures for desktop systems (x86 and ARM) would be a support nightmare for hardware companies that still keep their drivers proprietary and undocumented. Yeah, I'm looking at you, Broadcom and NVidia. This would just be another nail in the coffin for their obstructionist attitudes towards free/open-source operating systems.
My bicyles
It doesn't look like we'll have to wait too long to see these implementations in action either. Schorn reckons we'll be seeing ARM ecosystem products containing Cortex A9 designs in the first half of 2009 and then Osprey related silicon to appear later that year.
From the Hexus Article..
It sure would be nice to have an update to that linked article that was written a year ago. I've seen lots of info on ATOM since then, but not much on the A9 systems that should already be out.
What are we going to do tonight Brain?
32-bit only becomes a limitation on NetBooks when you start to get applications that can't fit comfortable in less than 4GB of RAM. This is not likely to be a problem for a few years. NetBooks may start getting more than 4GB of RAM in the next couple of years, but that doesn't require major changes, as long as the OS can address it and map it into processes' 32-bit address spaces (we still aren't getting many machine shipping with more RAM than a Pentium Pro with PAE can address).
I am TheRaven on Soylent News
If anyone ever starts a new CPU-related company, can you please call it LEG for the sake of "it cost an ARM and a LEG" jokes?
Thank you.
Most of the time, the only advantage x64 has over i386 is the larger register set. ARM had a larger register set to begin with. If you need 64-bit integers or > 4GB address space, ARM isn't an option, but many servers would be happy with 32-bit cpu (especially a low power, low heat one)
Do you even lift?
These aren't the 'roids you're looking for.
I run Debian on my ARM server (@ 500Mhz). It performs very well. Thanks for pretending!
Do you even lift?
These aren't the 'roids you're looking for.
Look through the Slashdot archives for the article containing benchmarks - I am too lazy to dig it out. It is a gross mistake to regard ARM as a RISC architecture. It is in the sense that the instruction set is orthogonal, but it is incredibly dense (much denser than x86). Almost every instruction can be predicated on one of the condition codes, which eliminates the need for a lot of branching (and, therefore, reduces the overhead from superscalar designs) and every instruction gets free use of a barrel shift on the result. Added to that, most ARM chips from the last decade support one or more of the Thumb instruction sets, which are 16-bit versions of the ARM instruction set, and most ABIs let you switch between these on a per-function basis, so you can compile functions that don't touch more than 64KB of RAM into thumb code and get even better cache usage.
You'd also be surprised at SIMD performance. The Cortex A8 and A9 support both Neon and VFP vector instruction sets. They are not so fast for double-precision vector floating point workloads, but on single-precision and integer SIMD loads they do reasonably well. For very FPU-intensive workloads you are generally better off using the DSP that comes with most ARM SoCs.
I am TheRaven on Soylent News
The only way the ARM manufacturers stand any chance is to either go for a completely new OS or jump onto the Linux bandwagon. Microsoft wont endanger their cooperation with Intel and AMD until ARM has secured a sufficient enough marketshare. This makes ARM and Microsoft a catch 22 happening. Any support will be superficial with lots and lots of fot dragging.
On the other hand Asus has shown just how successfull cheap small devices can be with Linux on them. If the ARM companies goes ahead full steam pushing devices with Linux Microsoft will be forced to jump aboard no matter what they really want to do. By then Microsoft wont be calling the shots and ARM will have a much better bargaining position.
I also think the ARM manufacturers should take a long hard look at the Wintel OEMs and think about their situation. Do they really want ot find themselves in a position where all their revenue is taken by a third party like Microsoft who doesnt contribute anything at all to the platform? Are they comfortable to be totally in the hands of a company that cant manage to turn out a new version of their OS in almost ten years?
HTTP/1.1 400
However, it has one major hurdle to overcome: it doesn't support Windows.
Fuck Windows. Seriously.
I've been unwillingly paying the Microsoft tax for TEN YEARS. All I ever do is wipe Windows and install Linux. If my new computer can't run Windows then... great!! Maybe I won't have to pay the tax.
On a serious note, why not get your computer built for you (or DIY if you can). I had mine built by a small local company (Intel core2 quad, 4Gig RAM and 250Gig hard drive so a decent spec) and it cost well under £300. It came 'empty' - no OS - so I could install Ubuntu with NO Windoze contamination. It works geat. It's never given me any trouble at all and it does everything I want, quickly and very well.
Smivs on the intertubes!
it doesn't support Windows.
That's not a bug, its a feature.
TFA mentions that Google Chrome OS should support ARM, so since we already see Google Phones with Android and Google Apps, I don't think it's overly optimistic to hope to see a "GoogleBook" or Google Tablet.
There are 75,000 apps for ARM iPhone OS X.
There are 10,000+ apps for ARM Android OS.
There are loads of apps for ARM Maemo.
There are loads of apps for ARM Symbian.
There are loads of apps for ARM Windows CE and derivatives.
There are loads of apps for ARM Linux and derivatives.
in the Windows world, you'll hear that the processor runs on Windows all over the place. They've been trained that Windows is the end all, be all, and center of the universe so the concept of "it runs on Windows" is their world. Talk about a CPU and _it_ runs on Windows is the norm. They really don't know how to think about it without Windows at the center or in a hierarchy of the hardware->OS->applications. They can't imagine a world without Windows. Combine that with software people and marketing people with no clue of hardware and you get "processor X doesn't run on Windows"
LoB
"Anyone who stands out in the middle of a road looks like roadkill to me." --Linus
The reason that DEC/MIPS/PPC NT failed had absolutely nothing to do with the lack of apps, but the lack of the value-proposition those companies provided to end-users. MIPS/ALPHA/POWER platforms were two-four times more expensive than commodity Intel platforms. No one forsaw the power-house that Intel was to become with the Pentium Pro/Pentium II, and that single factor alone killed those platforms.
The apps issue was only secondary. When NT 3.1 first came out, boatloads of apps were ported to MIPS/Alpha as well as Intel. BOATLOADS. I supported a network with all three (four, we had some pre-release IBM PowerPC's in 1997). But that cost differential is what killed any chance for a multi-platform windows.
This is a different case. A9 can be as cost efficient as Atom.
I've written networking kernel code for Linux, and never encountered any CPU specific requirements - it's all abstracted behind function calls.
The Internet's nature is peer to peer - 20050301_cs_profs.pdf
Doing so incurs a performance hit, quite a significant one, as well as using extra memory.... When DEC did it with the Alpha, current Alpha processors were hugely faster than any available x86 so even with the performance hit you got comparable or better performance than using a real x86 system. ARM processors are not as fast as current x86, and the performance would be so poor as to eliminate the benefits ARM has over Atom.
http://spamdecoy.net - free throwaway anonymous email - avoid spam!
With serious improvements within ffmpeg stack, w32codecs as mandatory package is already gone for some time. Most of newest netbook oriented distros (Moblin, Maemo, Ubuntu Netbook Remix) uses Gstreamer as multimedia engine, which has serious developers working for speeding up things for ARM platform. Also I bet ffmpeg guys already have been working on this.
user@ubuntubox:~$ stfu This server is going down for shutdown NOW!
1. Everything Ubuntu can offer is available for ARM: http://www.ubuntu.com/products/whatisubuntu/arm
4. W32Codecs is obsolete since FFMPEG does WMV & Quicktime. Real player is in Helix. All these are distributed by source and should work on ARM.
Science & open-source build trust from peer review. Learn systems you can trust.