ARM Stealthily Rising As a Low-End Contender 285
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."
Windows missing ARM (Score:5, Insightful)
'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.
In my opinion, it's the opposite. One thing Windows doesn't have is ARM support (besides Windows CE). Manufacturers are already seeing the advantage of ARM, and the lack of Windows support isn't a deal breaker in every segment. I have a SheevaPlug which is an ARM device, and while most major Linux distributions have support for the architecture, Microsoft just has the one, and it isn't even a consideration for most users of the device.
Re:Fast is not always best (Score:4, Insightful)
I have a feeling the processor itself is not all that expensive in most "browse the web" computers. If ARM or some other processor is to make inroads it will have to be in the power department. A more efficient processor means a cheaper, lighter laptops with smaller batteries.
Stealthily?! (Score:4, Insightful)
Oh please!
It's not a stealth thing at all. The low power SoC market has always been ARMs. It's AMD (Geode... and then Intel's Atom) who decided to bring x86 to the low power market. If anything the article should focus on the troubles ARM is likely to face in the near future: unless RISC can continue to compete for price (aggressively), I doubt that adding more pipelines will make the general purpose platform developers happy - RISC bottlenecks will always be bottlenecks; x86 can simply gun for greater clock speed.
IMO Transmeta had it right: very long instruction words (which ultimately do 'everything'). Unfortunately it came 10 years too soon and no-one was ready because we didn't know "what" we wanted from a clock (or half clock etc if you're talking ARM...).
VLIW will be back soon enough, but I worry that it wont be the right place for ARM.
(nb: I am an ARM fanboy, having 'matured' in an ARM sponsored undergrad lab. it upsets me as much as anyone that ARM haven't tried to reinvent the wheel using the cash from their recent market dominance)
Matt
Re:Fast is not always best (Score:2, Insightful)
Re:Fast is not always best (Score:3, Insightful)
Well, that's what I'm looking for anyway...
Re:It's JVC's VHS-C versus Sony's Video8 again (Score:5, Insightful)
WIndowsMobile had years of time in the market before the iPhone and it had a lot more applications than the iPhone. The iPhone blew it out of the water in just a few short years.
Let me preface this with a disclaimer. I never really liked WinMo, and I can't wait till I can buy a cheap, fast ARM netbook to run linux on!
However, WinMo did not come about in the era of ubiquitous high-speed internet and wifi, large hosted storage and applications ('cloud' crap). I used to own an HP Jornada 320lx (precursor to netbook- a palmtop)
WinMo however sucked because of poor app compatibility. The portable versions of word and excel were pretty useless. Nobody uses these types of apps regularly on an iPhone. The iPhone is largely (but not solely) a toy used for music, video playback (youtube) and web browsing. When WinMo was relevant, processing power and internet availability were not up to it, and so the only people buying it were using it primarily for Calendaring, Portable Office, and the like, and it wasn't all that great at it, as I mentioned before. As such, the usability, simplicity and broad appeal of the iPhone is simply not there.
The scene is very different, it is hard to say just what will happen.
Re:It's JVC's VHS-C versus Sony's Video8 again (Score:3, Insightful)
Be careful about assuming causation here. It might have easily been that VHS-C sounded familiar to people who had VHS, and they went with what they knew. Video8 might have been just as successful if the names had simply been reversed.
Re:ARM/Linux in the Tesla Roadster (Score:3, Insightful)
Re:Windows CE and Windows Mobile (Score:4, Insightful)
It cannot run the same applications as windows, therefore it's not windows...
It is partially source compatible, but not enough to make any but the simplest of apps a direct compile... Linux/arm on the other hand, makes it possible to simply recompile the vast majority of applications so that they work (i have a sheevaplug running gentoo and i have done exactly that).
People buy windows because it runs the applications they have or are familiar with, the versions of windows which run on arm don't provide this.. Linux has a greater chance of running apps users will find familiar, since there are ports of things like firefox to arm.
Re:Windows CE and Windows Mobile (Score:5, Insightful)
Windows CE and Windows Mobile both support ARM.
There might not be "full-featured Windows" on ARM, but saying there's no Windows at all on ARM is just ignorance.
Except that with Linux, BSD, and even OS X, the code that runs on x86 is the same code that runs on ARM (and PowerPC).
With "Windows", the code that runs on x86 is not the same as runs on the embedded stuff: there's no "scaled down" version like the Unix-based systems. It's a completely separate OS. The only multi-platform stuff that Microsoft has is Windows for Itanium.
Just because the Microsoft marketing folks call it "Windows" CE or Mobile does not make it the same as the desktop / server OS. With the Unix-y systems, it is the code and OS (though perhaps cut down to the bare essentials). And that's what we're talking about here: taking the same code and simply doing a recompile. It's not going to happen with Vista or W7, but it can happen with other OSes (heck, even OpenSolaris is being ported as-is to ARM and PowerPC).
ARM == Hype (Score:5, Insightful)
Yes, ARM marketing (notoriously overoptimistic) says they will have a 2GHz A9 in 28nm, relatively high performance process.
But A9, in terms of efficiency, is not substantially better than where Atom will be. That shouldn't be surprising. They're both scalar architectures. They both have a little less than 15 useful registers. They both have similarly deep pipelines. They both rely on branch prediction for performance. Neither company has magic, it's not surprising that they're similar on the curve of performance / efficiency.
Put another way, your instruction encoding doesn't really buy you all that much.
Now ARM has some lower-end cores (ARM9, ARM11, Sparrow/CoretexA5) that are much more energy efficient than Atom. But they're also much lower performance.
But this is how ARM's marketing plays it out: we have super-efficient cores (ARM9)! We have higher-performance cores (Theoretically, A9)! You think that ARM cores are somehow both high performance and much more efficient than Atom will be in the same technology... but this will probably turn out to be false.
Put another way... are MIPS or PowerPC cores dramatically more efficient than x86 at similar performance levels? No. They have most of the same architecture benefits that ARM does... more, in many ways, because they have about double the number of useful registers. But they're on basically the same efficiency/performance curve as everyone else.
You could probably do an x86 implementation that was similar to ARM11/A5... no floating point, no SSE, just the basic 386 instruction set. Give it a short pipeline and turn down the frequency, and it will probably compete relatively well on energy efficiency with those low-end ARMs.
The thing I DON'T understand... why does ARM marketing get an article on slashdot every week or so?
Re:Fast is not always best (Score:3, Insightful)
When waiting for windows to open its start menu it is fun to estimate the number of clock cycles which have elapsed. Two billion clock cycles? What was it doing?
Re:ARM/Linux in the Tesla Roadster (Score:1, Insightful)
Half Life, Company of Heroes, World of Warcraft, etc. No games is a deal breaker for a lot of people.
100% familiarity with the computer one uses all day at work is another deal breaker for a lot of people. If you have more money than time a few hours wages to mean your home computer is not ever going to deviate from what you know is money very well spent.
The family tech support guy only knows Windows so buying anything else means figuring stuff out on ones own.
There are plenty of reasons that may feel insignificant to you, but too many any one can be non-negotiable.
Re:So, where are ARM netbooks? (Score:4, Insightful)
Re:ARM/Linux in the Tesla Roadster (Score:3, Insightful)
Seems like Linux will fill the bill with a browser, maybe a PostScript app and a media player. Text editing isn't such an elaborate thing these days. And only a few people even know what to do with Excel.
And most peoples "MS Office needs" can be met with Open Office. That people "require" MS Office has mostly to do with laziness (too much work learning a new program) and myths (you must have MS Office, otherwise you're not compatible).
The funny thing is, going from MS Office 2003 to MS Office 2007 has a steeper learning curve than going MS Office 2003 to Open Office and has more or less the same compatibility issues.