nteresting blog below by Bill Dally, Professor at Stanford University and Chief Scientist at nVIDIA... With WINDOWS now supporting ARM, existing ARM cores can power tablets and netbooks, but there is opportunity to develop a high-performance ARM core/chip for desktops, higher-end notebooks and servers. nVIDIA is doing it today, and I expect others (eg TI, Marvell, maybe even Apple or IBM, in partnership with ARM) to do the same. Processors for Windows PCs is a huge, profitable market!
Vaporware - more hot air from a company that excells at hot air, smoke, and mirrors. Microsoft's bloated and buggy code base will hardly run on a 3 GHz dual core desktop PC. They will never strip out enough of the bloat to make it operate on an ARM processor in a reliable manner. Microsoft's software is a nightmare of unreliability and security holes that should be kept from spreading like the virus that it is. As far as presuming that Microsoft's next living dead software will be called Windows 8. Ha! Maybe it will be called Vista 2 or Windows You. Their marketing people are not only deceptive, ala the unveiling of the latest XBox which was modified for the demonstration, they also have no concept of what a positive image consists of.
Regardless of what we feel about Microsoft and it's products, the statement: "Increasingly, people want a Windows experience on all of their devices" is largely true for the vast majority of computer users. I wouldn't extend that statement to phone users, but for future tablets and netbooks, I certainly would.
Given that ARM devices are approaching traditional PC territory, Microsoft doesn't really have a choice but to port Windows to the ARM platform. It will likely not be the most efficient, most secure or most reliable, but I would bet that it will be the biggest selling.
Re: "The next version of Windows—presumably to be called Windows 8"
As far as what it will be called, I wouldn't put money on "Windows 8" right now. Let's see. We had: Windows 1.X, Windows 2.X (renamed Windows 286), Windows 386 (renamed Windows 3.0), Windows 3.11, Windows 95, CE (renamed Windows Mobile at somewhere around V6), 98, NT, 2000, Me, XP, Windows Tablet PC, Windows Media Center, Vista and 7.
Given that sequence of names, I think it's pretty obvious that the next name in line is "Windows Bunny Rabbit."
I have never actually met anyone who wanted a Microsoft Windows experience at all. People greet the names "Microsoft" and "Windows" with curse words. No we don't want anything with Microsoft in it. What we want is for any and all other companies to continue to thrive in the tablet, netbook, or smart phone market. We can learn to use whatever interface is provided but what we can't stand is for the company that has destroyed worldwide productivity for over a decade to continue their shameless act.
'Bloatedness' is caused by user expectations...eventually a platform must adopt new technologies and then it becomes bloated. Just consider how trim and secure Firefox was at a time and now I refuse to use it. Viva Android..
This is big for ARM and may be for MS too. With Android throwing them off the ring they don't want to be an also ran and have put massive efforts and come up with an OS from ground up for ARM. They bungled on the internet and they almost did it in the mobile space.
This could open up a huge potential for companies writing windows based third party applications. They get their leg in the mobile market.
"Microsoft had for years resisted calls for the company to add support of ARM-based devices ...."
"However, the emergence of Google Inc.'s Android operating system—and its subsequent momentum—changed the equation, and likely persuaded Microsoft to act to stem Android's . "
Another reason could be that Intel,who was faithful to WinTel partnership for long time started flirting with Apple and Android OS. Also they teamed with Nokia to bring their own Meego OS in x86 Atom platform to market & directly challenge MS mobile OS.
I would hope that MS understands and works with the next gen windows "codename whatever" to boot fast and be slim. ARM processors are low cost with reasonable performance,adding in a lot of OS overhead might not be the best move market wise. I look forward to the roll out with some interest, more to the point the next generation ARM devices that are spawned..
A couple things about this move that I don't understand:
1.) According to Engadget, existing Windows applications won't run on the ARM version of Windows. What good is Windows without the ability to run existing Windows applications?
2.) The ARM architecture is relatively lightweight, especially compared to the current Intel Core i3/5/7/9, and is designed to provide exceptional battery life at the expense of raw performance.
As is, Windows barely runs on the Atom series CPU's, how crappy is the performance going to be on ARM?
When Apple wrote the iOS for the iPhone/iPad, they started with the base MacOS X, but gutted everything not required for mobile use, then optimized the hell out of the code to get decent performance out of the ARM CPU.
Yet, M$ is porting a full blown desktop OS to ARM and expects good performance?
Plus, Microsoft has a poor history of optimizing their code for the dominant, and powerful x86 platform.
Also, they've got a crappy track record of supporting platforms other than x86. Anyone remember NT on PowerPC? Back when PowerPC was new and had a lot of buzz, M$ ported NT 3.51 to to it, only to discontinue support a version or so later.
I use Windows on almost every computer I have, so I'm not a MS hater - but what's the point of doing this? Only some software will be recompiled for the ARM architecture so you won't have the huge range of software we already have for the x86 platform. We already have numerous Linux and Android builds that run well on ARM and x86, with a huge amount of software available. This is just wasting time and resources.
So anyone who complains about anything is a "hater"??? How odd. Or is this reserved for those of us who want to see Microsoft fail? Either way, Microsoft has made thousands of claims and statements about what markets they are going to enter and dominate, and most of what they have said has been shown over time to be false. Microsoft is destroying the US and world economies with their do nothing operating system software that costs $320 a license. Microsoft is destroying the US by their tight fisted monopoly that is taking software development in the wrong direction. Microsoft's illegal business practices have cost the US and world economies trillions and I'm not happy about it. The media is also to blame. For years they claimed Vista was a great success, but now we find out Vista never had more than 14% of the desktop PC market. And for all the current lies out of Redmond, XP still holds supreme on over 60% of desktops. So if I'm a hater, then let's just apply the label "liar" to Microsoft.
@Duane Benson: I have to disagree on your observation that most people would like to see window's environment on tablets. Windows including CE-version is an OS that was never optimized for haptics-based instructions to the processor. As some one else remarked in this forum, I hope Microsoft guts most of the bloated OS blocks and writes new, optimized (& more importantly crash-proof!) code for the handhelds.
The last thing I want to see is a blue screen with text on my tablet! My Windows7 on a Toshiba laptop crashes so often that I use my backup desktop with XP!
Dr. MP Divakar
Dr. MP Divakar, I could be wrong about the desire for a Windows environment. But what I am very confident about is that users will not want to up and down convert files going from one platform to the next. Regardless of what the environment is (HW or SW), users will want complete interoperability.
I used a number of CE devices in years past and I really found them useful, except for the fact that the formats had to be converted and the CE applications were significantly feature stunted. Not just reduced feature set, but stunted. The version of "Word" for CE was only slightly more usable than is notepad. That's not going to cut it.
A Book For All Reasons Bernard Cole3 comments Robert Oshana's recent book "Software Engineering for Embedded Systems (Newnes/Elsevier)," written and edited with Mark Kraeling, is a 'book for all reasons.' At almost 1,200 pages, it ...