Windows 8 and 9 may support 128-bit architecture?

With Windows 7 still not available for everyone, news has surfaced that Microsoft may be researching and working on some of the architecture for Windows 8/9, which includes "128-bit architecture compatibility with the Windows 8 kernel." This is surprising, seeing as we're at least 3-4 years away from the next release, and still on the brink of the release of the latest version of Windows.

Microsoft Kitchen reports that they found a listing on a Microsoft employee's Linked-In profile (which has since been taken down - but hastily screenshot thanks to Microsoft Kitchen) with a job description that explained the employee, Robert Morgan is:

"Working in [a] high security department for research and development involving strategic planning for medium and longterm projects. Research & Development projects including 128bit architecture compatibility with the Windows 8 kernel and Windows 9 project plan. Forming relationships with major partners: Intel, AMD, HP and IBM."

Also, on the same LinkedIn profile page, Robert's status goes on to point out that: "Robert Morgan got called into the office, alarms in the morning are great. Someone didn't set the test for the new AMD 128bit properly."

Based on Microsoft's repetitive claims that Windows 7 is a "Major" release and that their release cycles alternate from being "Major" to "Minor", Windows 8 should be a minor release, even though it doesn't sound like it will be if it incorporates this feature. It also sounds like Microsoft is already getting involved with some of the big names in the hardware industry, namely Intel, AMD, HP and IBM already, which is impressive this early on. Interestingly enough, AMD is actually working on a 128-bit processor currently, codenamed "Bulldozer".

Another status update, according to Windows 8 News the following was posted by Robert earlier this year:

"Robert Morgan is working to get IA-128 working backwards with full binary compatibility on the existing IA-64 instructions in the hardware simulation to work for Windows 8 and definitely Windows 9."

The sightings of more than one of these updates seem to indicate that 128-bit might actually be in the next version of Windows, although it's still far too early to tell whether or not this will be the case, as Microsoft is quite well known for removing features from Windows in the recent years. We'd take this "news" as a maybe if that.

Update: Robert Morgan does not work at Microsoft and 128bit Windows 8 does not exist.

What does Neowin think? Do you think that 128-bit processing will be needed by the time Windows 8 is released?

Report a problem with article
Previous Story

Microsoft changes browser ballot screen, testing begins

Next Story

Microsoft announces new ways to try and buy Office 2010

169 Comments

Commenting is disabled on this article.

Windows 8 128-bit Edition will probably be like Windows XP 64-bit Edition (the one for Itanium processors, not the x64 version) - an OS for special, specific and super expensive workstations, not for mainstream use.

Can't you see? With intel working on GPU like CPUs and NVIDIA working on CPU like GPUs, we already have processors on the market today that sport 240 cores and 512-bit memory data paths (Quadro FX 5800). Now all we need is an OS that can run on them and make proper use of the architecture!

A GPU is of course an entirely different beast than anything that runs x86 code but with new virtualization technologies and some clever merging of existing and future technologies, we could see ourselves running Windows 8 or 9 on a virtualized platform on top of a "do-it-all" CPU/GPU/PPU hybrid with umpteen gigs of ram and modular upgradability within the next 10 years.

Seriously? SERIOUSLY?! I hope you guys redact this article -- officially, and soon. IA-128 DOES NOT EXIST. This is fake news.

Well.. its a good idea to stay ahead of the game.. but thats the only advantage to this. more than half the world still run on 32bit and 80% of applications out there are still 32bit.. so it will be a loooooooooooooooong time before we see a 128 bit application..

in any moment we will move from a commodore 64 to a commodore 128. When amiga comes that would make a big breakthru

Those were named for the amount of installed RAM though, 64Kb and 128Kb respectively. They both used 8-bit processors. I wish Amiga was still around though, the 500 was so far ahead of anything else that was available. Why did we have get stuck with boring IBM PC clones?

TRC said,
Why did we have get stuck with boring IBM PC clones?

Because something thats proprietary in the computer worlds equals bad.

The IBM PC was proprietary too once, until Compaq reverse engineered it and won the ensuing lawsuit. Then the market was flooded with clones. There were Apple clones once also, but they were authorized and Apple later decided to stop allowing them to be made.

Great - maybe in the next millennium, we'd have 1024 yotta-bits (I think it's one xona-bit) processors!!!

z0mFG wTF L0LCAT I must have a super-uber-1337 computer!

Ha not.

Keep in mind, he said he was working on 128bit COMPATIBILITY, that doesn't necessarily mean that Windows 8 itself will be released in 128bit. I mean, the XP kernel had 64bit compatibility, yet never got a proper 64bit release (it did, but it wasn't a "main" version you could buy in stores or anything), however the SERVER versions did.

The first 64 bit microprocessor was released in 1991 (or 1992) depending on your definition of what 64 bit means. It was not until 1999 that AMD released specs on their upcoming x64 chip, which was released in 2003. Individuals were writing code well before there was a 64 bit version of Windows. It was at the least theory and at the most experimentation (simulated run-times).

I think you might have a point about 128 bit compatibility. It does not have to be a CPU. FPU, arithmetic or integer library. It would lay the groundwork. It is not much of a breaking story, however.

My question, where is the 128 bit CPU? Not a prototype, not 128 bit operations but a true CPU. We could see one in the next decade, but I am not currently sure where the benefit is for a fully 128 bit CPU.

Seeing how 32bit CPUs are fading away, I wonder what would happen to the license deal between AMD and Intel. 32bit being no more a key player on the field, especially with 128bit architecture around the corner, it seems Intel has no more grip on AMD and AMD can soon be either taken over by some major player to give Intel a good run for their money or just do what the heck they want to irritate Intel. Surely emulating 32bit on 64bit CPUs would not fall under the license, I would assume. Maybe I am missing something, but could AMD finally have some breathing room in forthcoming future... o_O

Ah, bits. Remember when that was the important thing to look for? All the game consoles touted how many bits they were, we got up to 128 I think before that marketing tactic faded away. Now we have the Xbox 360 for example and it's 32-bit, yet somehow it doesn't look at all like a Super Nintendo. It's like the old "clock speed" marketing for CPUs; it's not really that important in the grand scheme of things. For consumers 128-bit computers would be ridiculous.

Ci7 said,
Nintendo was 64bit

The Nintendo 64 was, but there were others before it. The one I mentioned, the Super Nintendo, was 16-bit. The original NES and Famicom was 8-bit.

I thinks 128 right now is (snipped). 64 bits it's still a mess, there's snow leopard, windows 7 and linux distros x64, but tell me one that fully works, all drivers that fully works! kernel and all that works at x64, I think that the day to be working on 128 is when 32bits OSs become history!.. In about 5 to 10 years!

Migra said,
I thinks 128 right now is (snipped). 64 bits it's still a mess, there's snow leopard, windows 7 and linux distros x64, but tell me one that fully works, all drivers that fully works! kernel and all that works at x64, I think that the day to be working on 128 is when 32bits OSs become history!.. In about 5 to 10 years!

MS work with hardware vendors on new hardware and their specification long before they hit market - you think ATI defined the DX11 spec for their HD5870 alone? It's common sense for mutually supporting vendors to work on products behind closed doors to get them ready for market - Intel, AMD and Microsoft all have a long history of working together simply because their interdependant in their market ecosystem.

Windows 7 is like "Vista R2", or "Vista: Second Edition". I don't see how its a "Major" release. Major in name only, as they are wanting to distance themselves from Vista.

Xenomorph said,
Windows 7 is like "Vista R2", or "Vista: Second Edition". I don't see how its a "Major" release. Major in name only, as they are wanting to distance themselves from Vista.

This is where you are wrong. I actually have used both and 7 is nothing like Vista. I'm using 7 on an Acer Aspire One Netbook with only 1 gb of RAM and it runs smoothly.

There is a chicken called a Vista, but in the case of the OS it is more like a turkey. The services packs were just too late to save.

Foub said,
This is where you are wrong. I actually have used both and 7 is nothing like Vista. I'm using 7 on an Acer Aspire One Netbook with only 1 gb of RAM and it runs smoothly.

There is a chicken called a Vista, but in the case of the OS it is more like a turkey. The services packs were just too late to save.


+1

I'm sure there is development for 128bit - but I think its going to be at least 8 years before it even becomes considered within retail sector. How long did it take for people to move from 16-bit to 32-bit completely? Almost 8+ years if memory serves correct. We are just seeing the transition from 32-bit to 64-bit and even that it has been very slow e.g. limited 64-bit plugins, take Flash for example. Until a complete transition to 64-bit is made i.e. all programs are developed for 64-bit specifically, 128-bit will be ultra niche.

MistaT40 said,
I'm sure there is development for 128bit - but I think its going to be at least 8 years before it even becomes considered within retail sector. How long did it take for people to move from 16-bit to 32-bit completely? Almost 8+ years if memory serves correct. We are just seeing the transition from 32-bit to 64-bit and even that it has been very slow e.g. limited 64-bit plugins, take Flash for example. Until a complete transition to 64-bit is made i.e. all programs are developed for 64-bit specifically, 128-bit will be ultra niche.

It'll be much more longer then 8 years. Although 64bit has been around since 70's in RISC the reason it came into play much sooner was because of memory restriction. Even with progressing technology atm maximum on core i7 you can get is like 24GB ram which is nothing compared to 16EB limit 64bit has or 128GB limit that vista/7 has on OS which i highly doubt will get reached anytime soon within 8 years.

I could be wrong but that i7 ram limit is more to do with the memory controller Intel decided to use than core CPU itself. The server versions of the i7, the xeons, iirc, can access way more than 24GB. I'd have to double check though.

Owenw said,
Lol wut?

They were talking about 128 filesystems, like OpenSolaris' ZFS. - NTFS will be enhanced with 128 bits to support bigger filesystems.

blehh 128bit? advertising scheme?
Why do consumers need 128 bit in the near future (10-15 years).
If its for memory, consumers mostly use 2^33 at most 2^34 because of hardware limitations.
As far as number crunching, I can see how 128 bit registers will be helpful especially for space applications. (go go floating point accuracy!)

What's the percentage of 64 bit systems over 32 bits currently being used? How long has 64 bit been around? If it ain't broke..... Maybe this is also why netbooks are gaining in popularity as well?

as customer system

as early as Athlon 64 in 2003

--------
Enterprise systema

that another story (in '90s)

Foub said,
What's the percentage of 64 bit systems over 32 bits currently being used? How long has 64 bit been around? If it ain't broke..... Maybe this is also why netbooks are gaining in popularity as well?

So since current gen "aint broke.." we should just stick with it forever. Sounds like a plan genius.

Ci7 said,
as customer system

as early as Athlon 64 in 2003

--------
Enterprise systema

that another story (in '90s)

Yeah, and how many of them are in use now compared to 32 bit?

Harbinger said,
So since current gen "aint broke.." we should just stick with it forever. Sounds like a plan genius.


New is not always better. You have a consumer mentality. Not everyone is addicted to bright and shiny like you. When its time comes it will come.

Foub said,
New is not always better. You have a consumer mentality. Not everyone is addicted to bright and shiny like you. When its time comes it will come.

You miss the big picture obviously. We're talking about 3-4 years down the road the very least. In the end, no one will force you to upgrade.

"Based on Microsoft's repetitive claims that Windows 7 is a "Major" release and that their release cycles alternate from being "Major" to "Minor", Windows 8 should be a minor release,"

So that would mean that Windows Vista was a minor release after all! :P


Anyway, seeing how slowly 64-bit got adopted and how both Apple and Microsoft still offer full support for 32-bit with their latest OS releases I don't see 128-bit becoming mainstream before the next big Windows version hits. That is unless it faces major delays again just like Longhorn/Vista did.

This is future-proofing more than anything else. The Windows 8 kernel might be capable, but I'd be willing to bet they're not going to ship a 128-bit flavor of Windows 8...maybe even in the Windows 9 timeframe.

Keeping in mind that the 32-bit capable 80386 came out in 1985-6, it took what, nearly 20 years before we started running into the 32-bit address space limitations? We've just started being able to justify the need for a 64-bit capable consumer OS...

_dandy_ said,
This is future-proofing more than anything else. The Windows 8 kernel might be capable, but I'd be willing to bet they're not going to ship a 128-bit flavor of Windows 8...maybe even in the Windows 9 timeframe.

Keeping in mind that the 32-bit capable 80386 came out in 1985-6, it took what, nearly 20 years before we started running into the 32-bit address space limitations? We've just started being able to justify the need for a 64-bit capable consumer OS...

if we keep doubling the size of ram
as of 2009 there are 16GB desktop systems

by around 2029 would have system which well hit the limitation of 64bit

I wonder how many 128-bit CPUs Intel & AMD have under development in 'Top Secret Labs'? Must be an 'Area-51' development for sure.

Is it just me or does this Robert Morgan guy seem like an absolute moron? He has to have some balls to post commercial-in-confidence research plans in status updates.

Why? This has to be some sort of misunderstanding. There'll be no 128-bit hardware architecture out by the time either Windows 8 or 9 is out, mark my words. Sure, maybe the internals will be validated to be able to run on 128-bit, but potentially for a 128-bit release? You have to be kidding me. There'll be nothing in at least the consumer market for it to even run on.

Based on Microsoft's repetitive claims that Windows 7 is a "Major" release and that their release cycles alternate from being "Major" to "Minor", Windows 8 should be a minor release

No, Windows 7 is a minor release, that was just said in defense of the 6.1 moniker in the kernel version string. Windows 7 is basically Vista with speed enhancements, a new taskbar, and UI polish as main features. From all I've heard on Windows 8, that should be a major release. UXExperience has already posted some resume indicating a new UI platform.

Agree!! Windows 7 is a minor release not a major release!!
Windows 8 should be a major release; let's hope vista story won't repeat again!!

Jugalator said,
Why? This has to be some sort of misunderstanding. There'll be no 128-bit hardware architecture out by the time either Windows 8 or 9 is out, mark my words. Sure, maybe the internals will be validated to be able to run on 128-bit, but potentially for a 128-bit release? You have to be kidding me. There'll be nothing in at least the consumer market for it to even run on.


No, Windows 7 is a minor release, that was just said in defense of the 6.1 moniker in the kernel version string. Windows 7 is basically Vista with speed enhancements, a new taskbar, and UI polish as main features. From all I've heard on Windows 8, that should be a major release. UXExperience has already posted some resume indicating a new UI platform.


That's it.

I don't from where they got that information, 128-bit, are you kidding me?, AMD Bulldozer, are you kidding me again?.

I love AMD but they've to much troubleright now to even think in an architectural change.

We are living in the 32 to 64 transition and that transition will take a lot of more time than 16 to 64.

Well said Jugulator, BTW I love the Jugulator album by Judas Priest

Windows 8 won't come out with a "x128 edition". I would assume what is meant is that some of the code is optimized so that when an x128 processor reaches market, Windows 8 can still run on it.

Again, NT6.1 is simply for compatibility reasons, NT7.0 is the actual kernel. NT6.1 is used to allow legacy applications be fooled into thinking that they are installing onto Vista. Drivers designed for Vista can install on Windows 7 for exactly this reason.

No. Windows 7 is a minor release.

It's designation as NT 6.1 or NT 7.0 doesn't matter. Those numbers have no concrete effect on anything.

You proved my point yourself by saying that Vista drivers work fine on 7. If most kernel-level drivers are still generally cooperative with the Windows 7 kernel, that must mean not much has changed for it to retain that fined grained compatibility.

Even Microsoft thinks its a minor version. Thats why they called the server edition of Windows 7: Windows Server 2008 Release 2. Thats right. A minor naming update for a minor OS update.

Stop the misinformation. If you don't have non-marketting derived evidence of your cause, stop.

Erm... Tell me exactly WHERE I said that it wasn't an incremental release? Nothing I said in my post was incorrect. Windows 7 allows both WDDM 1.0 and WDDM 1.1 drivers to install, but most graphics card driver installers check the kernel version to make sure the user is installing the correct driver onto the correct OS as opposed to whether the WDDM version is supported. This also applies to some applications too.

Fagutish said,
You proved my point yourself by saying that Vista drivers work fine on 7. If most kernel-level drivers are still generally cooperative with the Windows 7 kernel, that must mean not much has changed for it to retain that fined grained compatibility.

Stop the misinformation. If you don't have non-marketting derived evidence of your cause, stop.


So your definition of "major" is that drivers must stop working? How insightful.

Major or minor, it really depends how you look at it. If you go with the UI changes win7 is kinda major imo. Looking under the hood it's minor since it's mostly performance improvements and so on.

The fact vista drivers work isn't a shock, they share the same driver model. MS doesn't change driver models every other windows version.

Kirkburn said,

So your definition of "major" is that drivers must stop working? How insightful.

Precisely. So if Windows 8 has the same basic driver model then it'll be 'minor' too. Guess what kids, we don't get to decide if it's minor or major - MS do - they don't need to meet some random checklist of what you think constitutes a major release of their operating system. Each and every MS OS is derivative in some form - even when they do a major rewrite the underlying APIs remain much the same for compatibility, though are often extended (e.g. ITaskbar3 in 7).

While I agree that device driver framework changes immediately warrant a major OS change, (or the lack thereof) from past microsoft behavior, we know that every major OS update (4.0->2000->Vista) has broken kernel-level drivers in certain ways and required rewrites. User-mode drivers for basic devices such as mice etc. are (mostly) unaffected.

Not that this matters, that was a subset of my argument. Also, you know you're fighting a pointless cause when the opposition will take down logical arguments made by an actual MS ex-employee. (shoutout to raindrop)

Pointless.

Fagutish said,
While I agree that device driver framework changes immediately warrant a major OS change, (or the lack thereof) from past microsoft behavior, we know that every major OS update (4.0->2000->Vista) has broken kernel-level drivers in certain ways and required rewrites. User-mode drivers for basic devices such as mice etc. are (mostly) unaffected.

Except kernel drivers for 7 have also required changes, as have user mode drivers - and subset or not this still fails to demonstrate this triggering the need to fling 'major' around. As I said, if MS rewrote the internals of the model and yet kept the same API to the outside world then what does that count as? As for raindrop, he's said nothing significant to warrant him being ex-MS (or even if he was nothing to show he did more than make cups of tea). The 'pointless' arguement is that you or I get to designate the criteria for 'major' or 'minor' - and that's my er.. point

Of course MS are going to claim it is a major milestone, the best piece of coding since sliced silicon, but not much has changed since Vista has it? Essentially some spit and polish on a fairly new foundation. This is much more of an XP to Windows 2000 than a Vista from XP.

Jugalator said,
No, Windows 7 is a minor release, that was just said in defense of the 6.1 moniker in the kernel version string. Windows 7 is basically Vista with speed enhancements, a new taskbar, and UI polish as main features. From all I've heard on Windows 8, that should be a major release. UXExperience has already posted some resume indicating a new UI platform.


Uhh, Windows 7 is a "major" release and already has a new UI platform. Actually, more than one (Direct2D, Scenic, etc).

If 128-bit architecture was in windows wouldn't that mean that 32 bit software would not be compatible? I mean on a 64 bit system you can't run 16 bit programs, is it the same way with 128 bit?

I don't think so, unless they totally drop and remove the 32bit legacy support. You could just run it in a hypervisor and get 32bit app support that way. At the speed these CPUs are going through data now, you can emulate/virtualize older apps and never know they're not native.

Look at what they're doing now with XP mode for Win7 and how you can run apps in win7 so they look and feel like a native win7 app. It's signs of things to come.

You'll have a 128bit OS with a 128bit CPU etc, and running 32bit apps without even knowing the difference. That's the goal in the end.

miguel_montes said,
What's so great about 128-bit architecture? What are its big advantages over 64-bit?

Moving larger memory blocks per CPU instruction? I can't think of much else...

Well aside from the more addressable memory, you can add more CPU instruction, which can speed things up. Aside from that though, until someone implements an x86-128 CPU, we won't really know.

It's all about number crunching. Having 128bit x86 CPUs would really be nice for supercomputers and servers in general. Anything to get more GFLOPS than what you can get now.

If you have two scientific apps sifting through data, if the apps are written right the 128bit will give you better results and faster than the 64bit one.

You can also get more processing power with fewer physical CPUs/Cores as well when you work with specific apps that can take advantage of 128bits.

miguel_montes said,
What's so great about 128-bit architecture? What are its big advantages over 64-bit?

Adding CPU instructions, mainly useful for stuff like SSE/CUDA so that they can calculate stuff very very fast! That's about it really, it can also address more ram but 64bit takes care of that nicely.

64 bit computing isn't really mainstream yet, and they're talking abouyt 128 bit computing.

My opinion: let just the -main- software programs (like even MS Office, and I'm not talking on the to-be release of Office 2010) run on 64 bit systems (and thus 64 programming) before jumping to 128 bit.

You do realize that Office 2010 will be 64-bit? Also, 32-bit programs run just fine on a 64-bit system as long as they don't require special legacy drivers. Actually, when moving to 64-bit for office, everyone will probably have to rewrite their extensions (even adobe!) since you can't load 32-bit extensions in a 64-bit app (not without the help of a proxy, which they could do). So that means that you could see less third-party functionality for Office 2010 for the first bit of it's release.

kiddingguy said,
64 bit computing isn't really mainstream yet, and they're talking abouyt 128 bit computing.

My opinion: let just the -main- software programs (like even MS Office, and I'm not talking on the to-be release of Office 2010) run on 64 bit systems (and thus 64 programming) before jumping to 128 bit.

Actually it is probably more mainstream than you might think. My last laptop (which I owned for 3 years) had a 64-bit processor and my current one does as well. I currently run Windows 7 64-bit and I can see myself doing so for quite some time. Just about everyone I know runs 64-bit OS's right now.

Gibwar said,
You do realize that Office 2010 will be 64-bit? Also, 32-bit programs run just fine on a 64-bit system as long as they don't require special legacy drivers. Actually, when moving to 64-bit for office, everyone will probably have to rewrite their extensions (even adobe!) since you can't load 32-bit extensions in a 64-bit app (not without the help of a proxy, which they could do). So that means that you could see less third-party functionality for Office 2010 for the first bit of it's release.

Actually 16-bit code doesn't run on a 64-bit platform. You can only go one generation back on a processor. Therefore when 128-bit OS's come out then 32-bit code won't run on them when the processor is running in native 128-bit mode. If they are running emulated 64-bit mode then they will do 32-bit but otherwise.

Adapting OS on architecture would be great, i.e. the OS should be able to recognise archtecture and install itself in the right type. No need of separate installation media. If you still have 32bit it will install it. If you have 64Bit, will do it also. Why does anyone can't design something like that. I would say kind of "Adaptive Code". One OS for all architectures. Or I may be dreaming...

Apple has done this type of install for ages. Several releases were compatible with both 68K and PPC architectures, and a couple of versions of OS X were compatible with PPC and X86 architectures.

Because the apple installs dumped all the binaries for both versions in at the same time. As far as OSX goes, with it's Rosetta? Or w/e, that had fat binaries to support both x86 and PPC till they decided to dump PPC fully.

It's not hard to do, but then if MS does something like that people would yell BLOAT, and so on.

The closest thing we have is JIT compiled code, like .NET and Java. But JIT'ing has a cost and building an OS that way doesn't make any sense. Your OS is tied to your hardware, people don't go around taking out their Intel CPUs and putting in PowerPC ones. So the penalties of just-in-time compilation for an entire OS simply aren't justified.

As insider, let me put a few things straight:
Windows 8 will have both server and client OS releases.
The tech. and features has not yet been determined. And I cannot discuss the suggested features list.
Windows 7 was a minor release, a move from NT 6.0 to NT 6.1,
Therefore windows 7 will have less compatibility issues, as most of the software and drivers compatible with Windows Vista, will also be compatible to Windows 7.

Windows 7 was a major upgrade to NT 7.0, but it identifies itself as NT 6.1 to legacy applications so that they can be fooled into thinking that they're running on Vista.

raindrop said,
As insider, let me put a few things straight:
Windows 8 will have both server and client OS releases.
The tech. and features has not yet been determined. And I cannot discuss the suggested features list.
Windows 7 was a minor release, a move from NT 6.0 to NT 6.1,
Therefore windows 7 will have less compatibility issues, as most of the software and drivers compatible with Windows Vista, will also be compatible to Windows 7.

Insider my ass - your lack of understanding of the "NT6.1" designator shows that very clearly lol :P

raindrop said,
As insider, let me put a few things straight:
Windows 8 will have both server and client OS releases.
The tech. and features has not yet been determined. And I cannot discuss the suggested features list.
Windows 7 was a minor release, a move from NT 6.0 to NT 6.1,
Therefore windows 7 will have less compatibility issues, as most of the software and drivers compatible with Windows Vista, will also be compatible to Windows 7.

Total fail. I wouldn't come back around these parts if I was you--we tend to remember you fake 'insiders'.

dangel said,

Insider my ass - your lack of understanding of the "NT6.1" designator shows that very clearly lol :P


Except that Windows 7 IS a minor update - MS has publicly stated so, they are not making any major changes, just cleaning it up, speeding it up... if that's not minor, I don't know what is!

greenwizard88 said,

Except that Windows 7 IS a minor update - MS has publicly stated so, they are not making any major changes, just cleaning it up, speeding it up... if that's not minor, I don't know what is!

Ok - show us all the quote where MS state it's "minor". That's right, you just got on the FAILbus.

dangel said,


Insider my ass - your lack of understanding of the "NT6.1" designator shows that very clearly lol :P


Please watch your language
As someone who worked for the client devision at microsoft for the last 3 years, I consider myself as insider.

After Windows Vista Criticism, Microsoft didn't want to have major changes, the specs for windows 7 were mostly, to have almost full compatibility to Windows Vista and improve performances.

Windows 8 will most probably be a major change.

Majesticmerc said,
Windows 7 was a major upgrade to NT 7.0, but it identifies itself as NT 6.1 to legacy applications so that they can be fooled into thinking that they're running on Vista.

Where is this information coming from? I doubt that they would do something as stupid as that. What "legacy" apps are coded for NT 6 but are not made using the recommended coding practices? Could I get a link to where Microsoft said this? Windows 7 in my opinion is a very minor release. It is extremely similar to Vista but with a new UI and some speed improvements.

raindrop said,

Please watch your language

Ass worries you? Grow up. And whilst you're at it post something that'd make us believe your 'ex-MS' claim to fame - nothing you've said so far backs that up in any way shape or form.

mrp04 said,
Where is this information coming from? I doubt that they would do something as stupid as that. What "legacy" apps are coded for NT 6 but are not made using the recommended coding practices? Could I get a link to where Microsoft said this? Windows 7 in my opinion is a very minor release. It is extremely similar to Vista but with a new UI and some speed improvements.

Not an official article, but it quotes someone here one Neowin who works for Microsoft.

http://windows7center.com/news/windows-7-k...anged-still-70/

dangel said,


Ok - show us all the quote where MS state it's "minor". That's right, you just got on the FAILbus.


You obviously dont work in IT. If you did, you would know the article is wrong and that windows 7 is a minor release. I was just at a talk yesterday with Microsoft and they said just that. The underlying kernal of win7 is the same as vista. So who's riding the failbus now? Idiot.

Server 8 will be 64bit only (I hope our NDA didnt cover that tidbit...I cant see how it would). Thats a fact. There was mention of supporting 128 way processors, but nothing about 128 bit. I wouldnt be surprised if windows 8 was 64bit only too since there will be no 32bit server.

BrewNinja said,

You obviously dont work in IT. If you did, you would know the article is wrong and that windows 7 is a minor release. I was just at a talk yesterday with Microsoft and they said just that. The underlying kernal of win7 is the same as vista. So who's riding the failbus now? Idiot.

Aww bless, upset are we? Listen, sweetheart, I didn't quote any article (how's that bus ride going?) and yeah, I most certainly do work in IT (not that I need justify myself one way or t'other - it's you that need to back up your claims). Oh, and BTW, I was just in a talk yesterday with Steve Balmer, and Bill Gates popped in for dinner - you people make me laugh my ass off.

mrp04 said,
Where is this information coming from? I doubt that they would do something as stupid as that. What "legacy" apps are coded for NT 6 but are not made using the recommended coding practices? Could I get a link to where Microsoft said this? Windows 7 in my opinion is a very minor release. It is extremely similar to Vista but with a new UI and some speed improvements.


Actually the biggest reason for 6.1 is due to applications which hardcoded checks for XP such as (if MajorVersion =>5 && MinorVersion >= 1) - this works on XP and Windows 7, but fails on Vista (MinorVersion=0).

Brandon Live said,
Actually the biggest reason for 6.1 is due to applications which hardcoded checks for XP such as (if MajorVersion =>5 && MinorVersion >= 1) - this works on XP and Windows 7, but fails on Vista (MinorVersion=0).

That was part of my understanding but here is how I understood it. They basically had stripped down Vista to it's kernel (check "minwin"). Once they did that they optimized it and had it work as well as they could. Afterwords they added back all the needed components and this time went with a modular route which allows for some of the changes that we see like being able to remove IE from the system. This ended up producing a faster and leaner OS.

In essence you could say that Windows 7 is both a major and minor release. They did enough to display itself as a major release but since they still used a kernel based on Vista it still only got a minor version nudge on the NT release. So rather than move it from NT6 to NT7 they just made it NT6.1. The same thing happened with XP. I would say that XP is a major upgrade from Windows 2000 (it's base) but since it still had the Win2k kernel base they just nudged the NT version to 5.1. Vista got a bump from NT5.1 to NT6 since it uses an entirely new kernel that is not based on any prior kernel.

P.S. - Minwin isn't the kernel but what they used as a guideline.

shinji257 said,
That was part of my understanding but here is how I understood it. They basically had stripped down Vista to it's kernel (check "minwin"). Once they did that they optimized it and had it work as well as they could. Afterwords they added back all the needed components and this time went with a modular route which allows for some of the changes that we see like being able to remove IE from the system. This ended up producing a faster and leaner OS.

In essence you could say that Windows 7 is both a major and minor release. They did enough to display itself as a major release but since they still used a kernel based on Vista it still only got a minor version nudge on the NT release. So rather than move it from NT6 to NT7 they just made it NT6.1. The same thing happened with XP. I would say that XP is a major upgrade from Windows 2000 (it's base) but since it still had the Win2k kernel base they just nudged the NT version to 5.1. Vista got a bump from NT5.1 to NT6 since it uses an entirely new kernel that is not based on any prior kernel.

P.S. - Minwin isn't the kernel but what they used as a guideline.


The Vista kernel is the Vista kernel. It is based on the prior NT kernels (namely, it grew from Server 2003 / NT 5.2), but has significant changes.

The Win7 kernel is the Win7 kernel. It is based on the prior NT kernels (namely, Vista) but has significant changes. That's the main reason Win7 / 2008 R2 can scale to 256+ cores where Vista / 2008 couldn't do that very well.

Why is this is a surprise? MS has always had at least a handful of people working at least one release ahead. It's good that this is in the planning phase, the sooner the better.

@g8crash3r, Dell isn't on the list because they don't make CPUs. HP does.

The guy can kiss his career in MS goodbye if he was indeed divulging future plans of the company which are under the NDA that goes along with his employment contract (common practise for many employers).

Nave said,
The guy can kiss his career in MS goodbye if he was indeed divulging future plans of the company which are under the NDA that goes along with his employment contract (common practise for many employers).

Funny that you seriously believe he is editing his own page. I don't see why that's such a great secret anyway.

64 bit systems support up to 16 GB of RAM ? with a 128 bit system it would theoretically support up to 256 GB of RAM
will we ever need that amount of RAM in the near future ?!
besides 64-bit software don't show great advantage over 32-bit ones afaik .. i wonder it would be the same case for 128-bit compared to 64-bit

The 64-bit bus width supports 16 Exabytes of RAM. A 128-bit operating system could (theoretically) support up to 281474976710656 yotabytes of RAM, a number so high that my calculator couldn't work it out.

Thats not to say that it wouldn't be useful. Even as recently as 10 years ago, the idea that 4GB of RAM would be commonplace at the end of this decade was proposterous.

15 years ago, we were pushing the limits of the 16-bit address width, 5 years ago we were pushing the limits of the 32-bit address width, I don't see why we couldn't be pushing the limits of the 64-bit address width by 2020.

Majesticmerc said,
The 64-bit bus width supports 16 Exabytes of RAM. A 128-bit operating system could (theoretically) support up to 281474976710656 yotabytes of RAM, a number so high that my calculator couldn't work it out.

Thats not to say that it wouldn't be useful. Even as recently as 10 years ago, the idea that 4GB of RAM would be commonplace at the end of this decade was proposterous.

15 years ago, we were pushing the limits of the 16-bit address width, 5 years ago we were pushing the limits of the 32-bit address width, I don't see why we couldn't be pushing the limits of the 64-bit address width by 2020.

even assuming memory size double every year
there would be still quite headroom
let say
this year memory size 16GB
by 2020 that would be 32TB which would still wont scratch 64bit headroom of 16exabyte

Guys, it's not just a case of being able to access more physical RAM. A 128bit CPU with a 128bit compiled app will bang through data way faster than any 64bit app even if they use the same amount of RAM in the end.

This support is more for early servers than anything for home use. The home always falls a few steps behind big business etc.

Why have a datacenter/mainframe with, say 16 64bit CPUs to run some DB/Simulation w/e, when you can do the same work and more with something like 4 128bit CPUs? (My numbers could be off but you get the idea).

Going to 128bit is more about pure raw GFLOPS (which is why there is this big move with GPGPU atm, CUDA, OpenCL etc) than just about being able to access more RAM.

Forgive my ignorance but is that how it works? I wasn't aware that you could push, say, two 64-bit integers down the 128-bit bus in a single instruction. I'm probably wrong, but my knowledge of this sort of thing isn't as good as it should be.

Majesticmerc said,
Thats not to say that it wouldn't be useful. Even as recently as 10 years ago, the idea that 4GB of RAM would be commonplace at the end of this decade was proposterous.


That's just wrong. Nobody is surprised that 4GB of RAM is commonplace. RAM sizes track nearly as well as Moore's law for CPU speeds.

15 years ago, we were pushing the limits of the 16-bit address width, 5 years ago we were pushing the limits of the 32-bit address width, I don't see why we couldn't be pushing the limits of the 64-bit address width by 2020.


Because it's exponentially larger. Memory capacities don't grow at an exponential rate.

GP007 said,
Guys, it's not just a case of being able to access more physical RAM. A 128bit CPU with a 128bit compiled app will bang through data way faster than any 64bit app even if they use the same amount of RAM in the end.

This support is more for early servers than anything for home use. The home always falls a few steps behind big business etc.

Why have a datacenter/mainframe with, say 16 64bit CPUs to run some DB/Simulation w/e, when you can do the same work and more with something like 4 128bit CPUs? (My numbers could be off but you get the idea).

Going to 128bit is more about pure raw GFLOPS (which is why there is this big move with GPGPU atm, CUDA, OpenCL etc) than just about being able to access more RAM.


None of that makes any sense at all.

It's pretty rare that you even need to do math with 64-bit precision, let alone 128-bit precision. Besides, any current CPU already supports 128-bit operations via MMX / SSE / etc, and they'll even be supporting 256-bit soon enough.

Brandon Live said,
That's just wrong. Nobody is surprised that 4GB of RAM is commonplace. RAM sizes track nearly as well as Moore's law for CPU speeds.

I was talking to a work colleague only yesterday that was talking about going into a shop to buy a 5.4" floppy many moons ago, and being told he had to buy a pack of 10. His response was: "Why would I ever need TEN of them?". 5 years ago, people laughed off terabyte hard drives on the desktop as totally unnecessary, and now we're seeing 2 terabyte hard drives come to the desktop market. I remember my second ever computer having a 4GB hard disks, which at the time was perfectly adequate for my needs, and now 4GB of RAM is commonplace.


Because it's exponentially larger. Memory capacities don't grow at an exponential rate.

Except that they have in the past. 0xFFFF (15 years ago) is exponentially larger than 0xFF, which is in turn exponentially smaller than 0xFFFFFFFF (5 years ago), and so on.

Forgot to add. In fairness, yes its going to take a long time to reach the limit of 64-bit memory addressing, but I'd say that its foolish to think its not possible in an ideal world. There are no technological issues (aside from possibly heat) with making MUCH larger chips, its just that theres no profit to be made in it. Welcome to capitalism I guess.

Majesticmerc said,
I was talking to a work colleague only yesterday that was talking about going into a shop to buy a 5.4" floppy many moons ago, and being told he had to buy a pack of 10. His response was: "Why would I ever need TEN of them?". 5 years ago, people laughed off terabyte hard drives on the desktop as totally unnecessary, and now we're seeing 2 terabyte hard drives come to the desktop market. I remember my second ever computer having a 4GB hard disks, which at the time was perfectly adequate for my needs, and now 4GB of RAM is commonplace.


Sorry, I should have said nobody in the industry or nobody who bothered to make an educated extrapolation is surprised. There are always people who don't think ahead. Plus, buying 10 floppy disks is silly no matter what capacity they are these days. Who buys 10 USB sticks? Most people have one or two at most.

Except that they have in the past. 0xFFFF (15 years ago) is exponentially larger than 0xFF, which is in turn exponentially smaller than 0xFFFFFFFF (5 years ago), and so on.


Even if memory capacities do grow an an exponential rate, it's a much slower growth than address space. You can easily extrapolate this out. If you follow Moore's law, memory capacity roughly doubles every 2 years or so. I find it hard to believe that can continue indefinitely or that it even needs to for personal computing devices, but that's another discussion. Let's say RAM capacity doubles every two years and use 4GB as the baseline. This isn't quite what happened recently since 2GB became standard more than 2 years ago, and 1GB was common more than 2 years before that, but I'm willing to go with an agressive timeline for the sake of argument.

Using that exponential growth rate, 20 years from now you'd have roughly 88,000 TB of memory. That's still pretty far from hitting that 16 Exabyte limit.

I'm not saying we'll never use 128-bit addressing, but it's going to take enough time to get there that it's impossible to say whether we ever will. Not because it's hard or unnecessary, but because the entire computing paradigm could change in that time and make our current memory addressing schemes irrelevant. Plus, you have techniques like PAE which could end up being far more effective in the 64-bit world (decades from now).

Anyway, 128-bit addressing would be a massive waste right now. The page table entries alone would be a ridiculous waste of your existing memory.

Article topic is misleading a bit Compatible and support is not the same thing. For example 32bit apps are compatible with 64bit OS, but does not support 64bit features. Same might the case here - windows 8 will/might run on 128bit CPU, but won't be able to use all of it's features.

we're at least 3-4 years away from the next release

I don't think so; it should be more like 2-3 at the most. The gap between XP and Vista, and even the gap between Vista and 7 were longer than usual for Windows. My goodness, between Windows 2000 and XP, there was only 20 months! (Feb 2000 - Oct 2001).
I can see how in the next 2-3 years, not just 64-bit but even 128 bit CPUs would start to be more used. I don't think the lag time between 64 and 128 will necessarily be the same as from 32 to 64.
By the end of 2010, likely all new computers will have 64-bit CPUs. Since AMD has had 64-bit processors for already some years now, I think 32-bit processors are going to quickly go the way of the dodo.

Windows 2000 wasn't a consumer OS, XP was. Think of 2000 as being like Windows Server editions, technically the previous OS to XP was Windows 98/ME, it just so happens that XP was when Microsoft decided to merge the consumer and business lines (and code bases) together.

Kushan said,
Windows 2000 wasn't a consumer OS, XP was. Think of 2000 as being like Windows Server editions, technically the previous OS to XP was Windows 98/ME, it just so happens that XP was when Microsoft decided to merge the consumer and business lines (and code bases) together.


Windows 2000 Professional was considered a consumer OS to an extent, although it was mainly marketed to businesses and IT pros. There were seperate server versions of 2000.

roadwarrior said,
Windows 2000 Professional was considered a consumer OS to an extent, although it was mainly marketed to businesses and IT pros. There were seperate server versions of 2000.

No, it wasn't.
Maybe by some geeks running pirated versions it was considered and used as such, but Windows 2000 was _not_ aimed at the home consumer.
All versions of Windows 2000 were aimed at corporate customers. It wasn't marketed mainly at businesses, it wasn't marketed at consumers at all. There were different versions because different (sized) companies have different needs.

Actually, Microsoft did intend to release a "Windows 2000 Home" edition. It was only when it became clear how bad the compatibility nightmare would truly be that they began a crash program to develop an enhanced version of Windows 98.

Thus was Windows Me born...

I don't find it surprising at all. The 32bit to 64bit transition was terrible. Microsoft of all people should be preparing for future architectures as soon as possible so they don't have to support old architectures. I personally believe that releasing Windows 7 on 32bit is a wrong move especially as it's classified as a major release, they'll have to support it for years to come rather than focusing efforts on other aspects. Whilst I don't see 128bit architecture being mainstream any time soon, it'll be extremely nice for MS to support this and fix any problems when there is a low userbase so that when 128bit does become mainstream it'll be a smooth transition.

I tend to agree with you; it should have been 64-bit exclusively. Even businesses that skipped Vista are going to upgrade to 7, and if they have the hardware to run Windows 7, they have 64-bit CPUs. Where can you find a new computer with a 32-bit processor anymore? Even on laptops they don't exist hardly.

devHead said,
Where can you find a new computer with a 32-bit processor anymore? Even on laptops they don't exist hardly.


Nearly every netbook in existance uses a 32-bit processor.

devHead said,
I tend to agree with you; it should have been 64-bit exclusively. Even businesses that skipped Vista are going to upgrade to 7, and if they have the hardware to run Windows 7, they have 64-bit CPUs. Where can you find a new computer with a 32-bit processor anymore? Even on laptops they don't exist hardly.

Well, I agree except one thing: Windows 7 Starter & Basic should have been 32-bit, and Home Premium, Professional, and Ultimate/Enterprise should have been 64-bit only. Or a compromise that Home Premium would be 32- and 64-bit, while the two enthusiast editions (Pro and Ultimate/Enterprise) be 64-bit only. ^_^ That might have helped move folks to 64-bit while still supporting those blasted netbooks.

I wouldn't be surprised if they do something like that for Windows 8 actually. Have the lower end versions be 32bit while Premium/Pro/Ultimate are 64bit only.

They're going to go 64bit only in time.

roadwarrior said,
Nearly every netbook in existance uses a 32-bit processor.

And that will change very quickly after Q1 of 2010 when Intel releases their dual core atom.

PCBEEF said,
And that will change very quickly after Q1 of 2010 when Intel releases their dual core atom.

Oooh, you mean all existing netbooks won't be 32-bit anymore?

Come on, not every person with a netbook is going to buy a new one.

@cm3k3: Most laptops made in the past 3 years will likely support a 64-bit OS. I'm pretty sure ALL of AMD's mobile Athlons are 64-bit-capable, as are all Core 2 Duos.

Are there that many laptops in use that are more than 3 years old?

32bit cpu's are virtually non-existent. In professional environments the lease date on a system are generally around two years, so every two years the system gets refreshed, so any existing 32bit cpu computers would be released with 64bit. Even if 32bit are still the majority (would like a source on that), the chances of it being the majority will quickly drop.

PCBEEF said,
32bit cpu's are virtually non-existent. In professional environments the lease date on a system are generally around two years, so every two years the system gets refreshed, so any existing 32bit cpu computers would be released with 64bit. Even if 32bit are still the majority (would like a source on that), the chances of it being the majority will quickly drop.

the VAST majority of CPUs are 32-bit. You'll be surprised how many devices have CPUs in them

PCBEEF said,
32bit cpu's are virtually non-existent. In professional environments the lease date on a system are generally around two years, so every two years the system gets refreshed, so any existing 32bit cpu computers would be released with 64bit. Even if 32bit are still the majority (would like a source on that), the chances of it being the majority will quickly drop.

You do realise that most if not all netbooks have 32 bit CPU's? And before you say it Windows 7 is being tested with Netbooks in corporate environments.

M2Ys4U said,
the VAST majority of CPUs are 32-bit. You'll be surprised how many devices have CPUs in them ;)

If you put it that way, I'm willing to bet that the vast majority of CPU's are 24-bit, 16-bit, 15-bit, or 8-bit.

PCBEEF said,
32bit cpu's are virtually non-existent. In professional environments the lease date on a system are generally around two years, so every two years the system gets refreshed, so any existing 32bit cpu computers would be released with 64bit. Even if 32bit are still the majority (would like a source on that), the chances of it being the majority will quickly drop.


You don't have read i said 32 bits OSs, no 32 bits CPUs

.fahim said,
You do realise that most if not all netbooks have 32 bit CPU's? And before you say it Windows 7 is being tested with Netbooks in corporate environments.

How many of those cpu's are relevant though? I'm sure there are more 16bit embedded devices on this planet than there are personal computers. Netbooks will all be transititioning to 64bit once intel releases their dual core atom chip in Q1 of 2010.

Aokromes said,
You don't have read i said 32 bits OSs, no 32 bits CPUs

Hmm, my bad in reading it wrong. I guess this is much due to the reasoning I made in my post after yours (#6). MS wants to make the transition as smooth as possible so that the world is not stuck in 32bit when the majority of desktop cpu's manufactured are 64bit.

Aokromes said,
128 bits? when most of world remains on 32 bits OSs even on professional environments?

It's called looking ahead. When Windows 8, let alone 9, becomes reality we're a couple of years further. The technology sector moves fast. If you don't plan now, you'll be behind when it actually becomes relevant.

Well even if they're working on it chances are it won't be in consumer OS, probably only Intel Itanium, Opteron and Cell Windows HPC servers support since they're only ones who maybe reaching heights of 64bit.

Imo, even under developing modern home/business computing and general use standards 64 bit still has at least a good decade or two in it if not more until consumers need to start considering otherwise specially with the move to merge GPU and CPU into a single processor even taking into account the fact that XDR or GDDR memory could come into play in that architecture on the main board.

It's very much a beyond the horizon of computing idea and technology in general though.

Digix said,
Well even if they're working on it chances are it won't be in consumer OS, probably only Intel Itanium, Opteron and Cell Windows HPC servers support since they're only ones who maybe reaching heights of 64bit.

Imo, even under developing modern home/business computing and general use standards 64 bit still has at least a good decade or two in it if not more until consumers need to start considering otherwise specially with the move to merge GPU and CPU into a single processor even taking into account the fact that XDR or GDDR memory could come into play in that architecture on the main board.

It's very much a beyond the horizon of computing idea and technology in general though.


I agree, i cant see home computing growing THAT fast int he next 6-7 years where well need 128bit

Maybe not home desktops, but on the server side? I see this move as just another step in the big MS server/services push. It's pretty simple, it also ties into them adding compute shaders to DX11 so you can tap into the GPUs for scientific work etc.

Adding 128bit support is another big step, If you need pure processing power and FLOPS a 128bit CPU will do it better than a 64bit one ofc, as long as the app uses it well. MS is just getting in on the game early.

Makes sense to plan for something like this, to futureproof your product. I'd say that Windows has had a reasonable transition to 64bit but it certainly hasn't been seamless and it's certainly not straightforward for the average consumer to understand the differences and make informed choices about it.

If they plan waaaay ahead for 128bit they can make the transition process much smoother.

Chicane-UK said,
Makes sense to plan for something like this, to futureproof your product. I'd say that Windows has had a reasonable transition to 64bit but it certainly hasn't been seamless and it's certainly not straightforward for the average consumer to understand the differences and make informed choices about it.

If they plan waaaay ahead for 128bit they can make the transition process much smoother.

It's not just MS fault for the 32 to 64 bit change this architecture that should have been in place by now, people have been reluctant to move on because of the lack of support from 3rd party software apps etc, 128 bit + terrabytes possibilities of RAM

If they plan waaaay ahead for 128bit they can make the transition process much smoother.


That and it really comes down to the whole "chicken and the egg" philosophy. AMD apparently has a 128bit processor in testing... now they need the OS to even make it worth producing.

Brandon Live said,
This is all pretty silly. Nobody understands what the heck they're talking about.

lol tell me bout it

Krome said,
lol tell me bout it

And you are both talking about whom? I know what I'm talking about!

32-bit processor use 32 bits to refer to the location of each byte of memory. 2^32 = 4.2 billion, which means a memory address that's 32 bits long can only refer to 4.2 billion unique locations that's 4GB! 2GB for the kernel 2GB for the app.

In a 32 bit world each app has it's own virtual 4GB mem space. Unless you talk about /PAE booted sys then each sys can access up to 64GB but still limited to 2GB per app.

You want to see the maths?

2^32 = 4,294,967,296 bytes
4,294,967,296 / (1,024 x 1,024) = 4,096 MB = 4GB

however for 64bit

2^64 = 18,446,744,073,709,551,616
18,446,744,073,709,551,616 / (1,024 x 1,024) = 16EB (exabytes)

with 128 bit

128-bit processors max memory addressing 340,282,366,920,938,463,463,374,607,431,768,211,456 bytes (~340.3 undecillion bytes or 281,474,976,710,656 yobibytes

As this exceeds the amount of available physical memory on the planet I referred to terabytes!

Maybe you should think before you slag other people off, or else they may come back and bitch slap the pair of you right upside your head and make you look like a pair of idiots! Not to mention provide the maths in figures you can't count! Some of us have PhD's, Master's Degrees, Degrees and an IQ of 174 so suck on that!!!!

Aw, Dell is not a major partner, hehehe, maybe I should transfer to HP, hmmm.
128-processing will be needed as long as the hardware is available and there is a demand for it, why not?
I think 3-4 years will be enough.

Dell doesn't make processors. IBM, Intel, AMD and yes, even HP do make them. Although HP isn't currently selling the PA-RISC processors, they are still supported for several more years, and they could be working on a new processor.

Neo003 said,
That would be great, but I wouldn't say 3-4 years rather 5-6 years away.



More like 100.

MS cannot even write 64 - Bit drivers for their devices 5 years after they announced 64 - Bit OPS'S.

So 128 - Bit maybe on planet next Millennium.

leesmithg said,
More like 100.

MS cannot even write 64 - Bit drivers for their devices 5 years after they announced 64 - Bit OPS'S.

So 128 - Bit maybe on planet next Millennium.


Huh? Do you honestly think that Microsoft writes all the device drivers for their 64 bit platform? Well guess what, they don't. It's the manufactures decision and responsibility to write drivers to support their hardware/devices/peripherals or what not.

leesmithg said,
More like 100.

MS cannot even write 64 - Bit drivers for their devices 5 years after they announced 64 - Bit OPS'S.


Don't confuse "can" with "want".

leesmithg said,
More like 100.

MS cannot even write 64 - Bit drivers for their devices 5 years after they announced 64 - Bit OPS'S.

So 128 - Bit maybe on planet next Millennium.


Strange, all my hardware works under X64. If you are referring to drivers for old, obsolete hardware then yep, you have a point Microsoft doesn't write drivers for it

Billus said,


Huh? Do you honestly think that Microsoft writes all the device drivers for their 64 bit platform? Well guess what, they don't. It's the manufactures decision and responsibility to write drivers to support their hardware/devices/peripherals or what not.

A bit of a harsh way of putting it butyou are correct... Although developers have to pay Microsoft for certification which a lot of manufacturers decided only to support 32-Bit and not 64-Bit drivers for Windows Vista 64-Bit and Windows 7 64-Bit.

Big manufacturers who will get their revenue back with sales will develop drivers for Windows 7 64-Bit. Smaller companies will not develop 64-Bit drivers because they cannot afford to do so.

EVANK said,

A bit of a harsh way of putting it butyou are correct... Although developers have to pay Microsoft for certification which a lot of manufacturers decided only to support 32-Bit and not 64-Bit drivers for Windows Vista 64-Bit and Windows 7 64-Bit.

Big manufacturers who will get their revenue back with sales will develop drivers for Windows 7 64-Bit. Smaller companies will not develop 64-Bit drivers because they cannot afford to do so.


And those who cannot afford to do so will fail once Windows switches entirely to 64-bit.

EVANK said,
Big manufacturers who will get their revenue back with sales will develop drivers for Windows 7 64-Bit. Smaller companies will not develop 64-Bit drivers because they cannot afford to do so.

Are there really that many small hardware companies? I don't expect so.

It sounds like a silly excuse to me. I think the hardware companies are looking at it like, why update stuff we already made with a 64bit driver when we can just as well not and have people upgrade to our new hardware that WILL have 64bit drivers?

It's just another way to get people to upgrade.

To be fair, leesmithg said for THEIR devices. As in, Microsoft keyboards, mice, etc. I'm assuming.

Now it's just a question of how true it is.

leesmithg said,
More like 100.

MS cannot even write 64 - Bit drivers for their devices 5 years after they announced 64 - Bit OPS'S.

So 128 - Bit maybe on planet next Millennium.

you really need to get a clue

EVANK said,
Big manufacturers who will get their revenue back with sales will develop drivers for Windows 7 64-Bit. Smaller companies will not develop 64-Bit drivers because they cannot afford to do so.

How about open-source drivers and modded drivers? Everyone's going to have to pay the MS Tax.