ARM, Microsoft Collaborating On 64-bit Windows Version

← Back to Stories (view on slashdot.org)

ARM, Microsoft Collaborating On 64-bit Windows Version

Posted by Soulskill on Friday November 2, 2012 @07:26AM from the champing-at-the-bit dept.

angry tapir writes "ARM is working with Microsoft to tune the Windows OS to work on processors based on ARM's 64-bit architecture. Ian Forsyth, program manager at ARM, could not comment on a specific release date for the 64-bit version of Windows for ARM processors, but said ARM is continuously working with software partners to add 64-bit support."

19 of 93 comments (clear)

Min score:

Reason:

Sort:

So, the next MIPS? by TWX · 2012-11-02 07:34 · Score: 2, Insightful

So, is this the next MIPS, or other non-Intel architecture flavor of the day, to fade into obscurity in a few years?

--
Do not look into laser with remaining eye.
1. Re:So, the next MIPS? by Anonymous Coward · 2012-11-02 07:40 · Score: 3, Insightful
  
  Yup. Once Intel gets its power consumption in line with ARM, that'll be it. Good bye (again) ARM, say hi to Itanium, MIPS, ALPHA, and for all intensive purposes PPC and SPARC.
2. Re:So, the next MIPS? by Anonymous Coward · 2012-11-02 07:44 · Score: 5, Funny
  
  intensive purposes
  OMG, kill me now.
3. Re:So, the next MIPS? by Anonymous Coward · 2012-11-02 07:47 · Score: 5, Insightful
  
  I hope this is a joke. ARM has a vastly stronger market foothold than anything related to MIPS ever had. In terms of supported software, there is just as much, if not more stuff supporting ARM than x86/Intel, so what reason would hardware makers have to go back to Intel?
  I'm pretty sure ARM is here to stay for a long time.
4. Re:So, the next MIPS? by Chibi+Merrow · 2012-11-02 07:47 · Score: 3, Insightful
  
  No, because as the desktop becomes less and less relevant and people do more on mobile devices, backwards compatibility with software written for a 80386 just stops mattering. Even if Intel chips had exactly the same power usage as ARM (highly unlikely), what's the benefit of having an Intel chip in your mobile device?
  
  --
  Maxim: People cannot follow directions.
  Increases in truth directly with the length of time spent explaining them
5. Re:So, the next MIPS? by MightyMartian · 2012-11-02 07:57 · Score: 2
  
  WTF? ARM has massive penetration in mobile devices. Beyond that, ARM's huge advantage is custom SOC, which, along with low power consumption, is another reason ARM has become such a monster platform. As much as anything else, ARM is a philosophy.
  
  --
  The world's burning. Moped Jesus spotted on I50. Details at 11.
6. Re:So, the next MIPS? by Lemming+Mark · 2012-11-02 08:01 · Score: 2
  
  Do you mean ARM 64 bit, or ARM in general? The latter has at least got itself deployed more widely than I am aware of MIPS ever getting. The former ... well, I hope it takes off.
7. Re:So, the next MIPS? by tlhIngan · 2012-11-02 08:01 · Score: 4, Informative
  
  So, is this the next MIPS, or other non-Intel architecture flavor of the day, to fade into obscurity in a few years?
  Well, given that ARM is probably the #1 shipped architecture out there, probably not. I'm fairly certain for every x86 CPU Intel ships, several ARM SoCs are shipped, probably a few in said PC (WiFi, Bluetooth, drive controllers (optical, SSD, spinning rust), network cards, etc).
  Intel rules only on one aspect - high power computing. ARM has pretty much taken over the low end embedded side - the processors and controllers needed for everything else to get that big beefy Intel fed.
8. Re:So, the next MIPS? by bemymonkey · 2012-11-02 08:13 · Score: 3, Informative
  
  Have you been keeping an eye on Cloverfield? The CPU performance pretty much kills the fastest ARM tablet chips, and power usage is similarly impressive - The first early adopters are reporting 10h+ battery life on the Samsung ATIV Smart PC, which seems to be right up there with all the ARM based tablets.
9. Re:So, the next MIPS? by MightyMartian · 2012-11-02 08:15 · Score: 2
  
  Most x86 software currently in existence will not run well on a tablet. Two different kinds of interfaces. The same applies to Linux-based tablets. While one can certainly compile older GUI software to run on Androids, without the nice interface extensions, it's just plain ugly.
  For all intents and purposes PCs and tablets/smartphones/smart devices are totally different ecosystems. Certainly Microsoft is betting on at least a partial merger, but I'll wager that the PC version of MS-Office will always be a different critter to the mobile/tablet version.
  
  --
  The world's burning. Moped Jesus spotted on I50. Details at 11.
10. Re:So, the next MIPS? by AcidPenguin9873 · 2012-11-02 08:17 · Score: 4, Insightful
  
  Even if Intel chips had exactly the same power usage as ARM (highly unlikely), what's the benefit of having an Intel chip in your mobile device?
  Why do you think this is unlikely? On the contrary, Intel has a massive fab/manufacturing advantage over any ARM chipmaker - they are at least 1 process shrink (node) ahead of any other foundry, not to mention their process at any given node is better than all competitors (TSMC, GloFo, Samsung). This fab advantage directly translates to lower power usage, and by all accounts, Intel's advantage there is only getting larger - ask AMD how it feels to be on the receiving end of this advantage. Intel needs to put out a microarchitecture which targets about the same performance range that ARM Cortex-A9 (or maybe A15 [1]) does, and in all likelihood, Intel's chip will be lower power because it will be manufactured on either one shrink ahead of any ARM equivalent, or the same node but using Intel's superior process at that node. In fact, Intel is doing just that - the next-gen Atom chip (Silvermont/Valleyview) is targeting right around where A15 is in terms of performance, area, and power.
  ARM isn't magic; there is nothing in the ARM ISA that makes it inherently lower power than x86. Yes, I'm counting all the decode hardware and microcode that x86 chips need to support legacy ISA. There just isn't much power burned there compared to modern cache sizes, execution resources, and queue/buffer depths which all high-performance cores need regardless of ISA. If you have an x86 processor that targets A9 performance levels, it will burn A9 power (or less if Intel makes it, given Intel's manufacturing advantage). If you have a ARM processor that targets Sandy Bridge performance levels, it will burn Sandy Bridge (or more) power.
  [1] I say maybe A15, because from Anandtech's latest review here, Samsung's Exynos 5250 using A15 cores does not have a prayer of getting into smartphones using 5W at load. Your smartphone will be dead in an hour of web browsing with that kind of power draw. Yeah yeah, Exynos 5250 is on a 32nm process and the 28nm A15's are right around the corner, which should be lower power. But still.
11. Re:So, the next MIPS? by Anonymous Coward · 2012-11-02 10:45 · Score: 5, Funny
  
  I would of said the term of speech "can't see the forest through the trees" is a mute point when they are by enlarge one in the same, but case and point unless it happened on accident irregardless of whether you were chomping at the bit or could care less, I wouldn't call you escape goat per say, but getting to the crutch of the matter being straight as a narrow is better than being on tenderhooks since time in memorial for all intensive purposes to hold down the fort, so I'll take another tact and won't cut off your nose despite your face because after all is set and done the proof is in the pudding and since you have a long road to hoe with a myriad of something or rather to do of upmost importance I have the courage in my conviction to be internally grateful it all goes well
12. Re:So, the next MIPS? by OdinOdin_ · 2012-11-02 13:42 · Score: 2
  
  Did you read what was written ? How many transistors does it take to make a 1Mb ARM CPU Cache ? Then a 1Mb Intel CPU Cache (with has a stronger memory model) ? Now how many transistors for a core ?
  I know I have a suggestion, you know those picture of silicon that Intel marketing put out with nice coloured lines painted on them to point out where each functional unit is. Compare the area of the cache to the area used for a core.
  So while it is possible for a RISC CPU to use less transistors to implement a whole useful ISA. It is not possible to also make it go fast (and keep the "less transistor" power advantage). This is the struggle ARM now faces. Intel are already ahead and fighting back with power reduction capabilities (i.e. switching whole functional units off when not in use) and in process size reduction around 10 to 20nm.
  How much power does a PowerPC chip use, these are the closest competitor in the RISC family to be targeting at the kinds of systems Intel has market share in. Take a look at the industrial cooling solutions for some of those systems then compare heat output against performance and Intel clearly wins. Intel already have chips that will work reliability through a lifetime warranty in a consumer workstation situation. PowerPC doesn't have such a product, they do have slower chips such at those used in gaming consoles and other such devices.
13. Re:So, the next MIPS? by Anonymous Coward · 2012-11-02 14:07 · Score: 2, Interesting
  
  Technically what occurs with switching on demand for cache transistors causes surging. Any time you power up and power down a transistor there is a micro spike in current demand. So there is a price you pay for powering down and then powering up on demand as well. This is an inherent problem with micro circuits do you leave them on or do you cycle them to save power.
  There might be an advantage to being able to being able to shut down unused cache resources on demand but on servers that require constant high demand this does not necessarily save anything in power consumption and therefore heat.
  This is why IBM is really spending money big time on learning how to put the DREADED nanotubes on a substrate. Create a transistor cache out of this and the power consumption is down to ridiculously low levels. But that is for the future. http://www.hardwarezone.com.my/tech-news-ibm-announces-carbon-nanotube-chip-breakthrough and no doubt a few posts and article on /. with a pile of ridiculous so what comments.
Will it still run IE6? by cryfreedomlove · 2012-11-02 07:37 · Score: 2, Funny

I know lots of corporate desktop police still subjugate their users by forcing them to still use IE6.
1. Re:Will it still run IE6? by corychristison · 2012-11-02 07:56 · Score: 3, Insightful
  
  Tell them they are putting their company at risk by forcing IE6 and that you do not want to be the one to blame when shit hits the fan (assuming you are either the person performing the work or the guy in charge of the person doing the work).
  As a web developer I can tell you very few developers still cater to IE6. Many have even dropped IE7. Most now use conditional comments to display a "Upgrade your browser" or "use a standards compliant browser" error message.
2. Re:Will it still run IE6? by swb · 2012-11-02 08:05 · Score: 2
  
  It really all boils down to some fat fuck in management or ownership and only gives a shit about how much cash he can wring out of the operation.
  They keep the browser at some old version because their application doesn't work in newer browsers and licensing a new version or doing the development on a new version would cost money.
  I see this all the time in consulting.
Re:Not enough coffee, this morning by Desler · 2012-11-02 08:00 · Score: 5, Informative

Except Microsoft has had software running on ARM since their 1997 release of Windows CE 2.0. They have also been working together with ARM since for those last 15 years as well. Don't let the facts get in your way, though.
Re:Must muscle in by Desler · 2012-11-02 08:03 · Score: 2

No, they just missed modernizing their mobile platform. They've been doing mobile on ARM for 15 years.