Half of Windows 7 Machines Running 64-Bit Version

nk497 writes "Microsoft has said that nearly half of machines running Windows 7 are using the 64-bit version, up from just 11% of PCs running Vista. The 32-bit version is limited to 4GB RAM, while the 64-bit version allows 192GB, as well as added security and virtualization capabilities. While Microsoft is pushing 64-bit as a way to gain performance in the OS, it earlier this year advised users to install the 32-bit version of Office 2010, 'because currently many common add-ins for Office will not function in the 64-bit edition.'"

Comments

Statistics, statistics

[mstefan]

The reality is though that 10% of Windows systems are 64-bit (there's actually still more systems running Vista than Windows 7 out there, although the gap is shrinking). The vast majority of Windows desktops are still running the 32-bit version of Windows XP, and that's not going to change until businesses decide they have a compelling reason to upgrade.

Re:Statistics, statistics

[Monkeedude1212]

The vast majority of Windows desktops are still running the 32-bit version of Windows XP, and that's not going to change until businesses decide they have a compelling reason to upgrade.

And my guess is that'll happen when they stop supporting XP P3 - which if my memory serves correct is 2014? Can someone back me up on that?

64 bit isn't too far off. As a developer you'd be better off getting a copy soon and work on merging your projects over to work on 64 bit now, rather than wait for crunch time.

Re:Statistics, statistics

[alta]

And considering so many companies are moving towards web based interfaces for their internal applications, this is going to take a long time. Sure, MS can stop making IE for XP, but get SP3 on the machines and it's pretty damn secure. Add to that Chrome or FF to run the business applications and you have a machine that's going to last for many years to come. Want to go faster? Get faster/more servers! XP can essentially become a dumb terminal as for as those enterprises are concerned. I think that about the only thing they could do is to make new versions of office not run on XP. That'll make a few companies switch because they can't be without outlook+exchange/word/excel.

Re:Statistics, statistics

[Zixaphir]

Unless, of course, they are the majority and it doesn't look like that is going to change anytime soon. I mean, you're talking about being behind the curve, but I also bet you're running an x86 derivative. A 64-bit instruction set doesn't fix the gaping problems of x86. Yet, x86 remains the defacto standard.

Re:Statistics, statistics

[drsmithy]

64 bit isn't too far off. As a developer you'd be better off getting a copy soon and work on merging your projects over to work on 64 bit now, rather than wait for crunch time.

Pro-active developers ? You've got to be kidding. It took the "annoyance" of Vista's UAC before developers finally started changing their Windows applications not to needlessly require admin privileges. They're not going to be implementing 64-bit support one second before "crunch time" arrived.

Re:Statistics, statistics

[VGPowerlord]

A 64-bit instruction set doesn't fix the gaping problems of x86.

...but it does fix some of them, such as increasing the number of registers and support for SSE2 (I think) extensions on all x86-64 chips.

Re:Statistics, statistics

[mstefan]

Businesses don't tend to have that short of an upgrade cycle when it comes to operating systems. They typically prefer to stay on the trailing-edge of technology as long as possible -- "if ain't broke, don't fix it" is the mantra of most IT departments, particularly in larger companies. If you look at a lot of the "droneware" business desktops out there today, they're sold with 2-4GB RAM and downgrade rights to XP 32-bit. So while the system may be "sold" with a Win 7 64-bit license, that doesn't mean it's ultimately how it's being used.

Re:Statistics, statistics

[Sycraft-fu]

It truly amazes me how lazy developers are when it comes to supporting new things. They whine and bitch and drag their feet and blame MS, rather than just admitting they have to learn something new and doing it.

For example back in the Windows 2000 days I had a professional sound card. It had Windows NT drivers, but had some limits since NT wasn't good at sound. 98 was the preferred OS. Well when 2000 came out, the refused to release 2000 drivers. They claimed that the kernel mixer introduces 30ms of delay that you couldn't get around and that was unacceptable for pro work, etc, etc. Just use the NT drivers even though that caused some problems.

I (and probably many others) found the MSDN page on kernel streaming and sent it to them, showing them they were full of shit. Finally, many months later, they released a WDM driver... That supported only 2 of the interfaces 10 inputs and outputs. They claimed that WDM could only support one stereo set, that's it. A built in limitation by MS, nothing you can do about it. In frustration I e-mailed MS and I think they were sufficiently surprised by the stupidity of the statement that a developer actually responded and showed me where to find the docs. Turns out that WDM support lots of audio devices, either enumerated as multiple stereo pairs (as old style drivers did) or as a single multi-channel output. in fact you could do both at once.

So that went off to them and they ignored it for a long time and finally got out a real, full, WDM driver that was buggy as shit. The proceeded to work on the bugs and eventually had a nice driver. They decided they really liked WDM at that point, and quickly stopped supporting the older formats. It went from "We can't do it," to "This is the only way to do it.

But it took like a year and a half.

Many, perhaps most, developers are extremely, EXTREMELY lazy at updating to new technologies and fixing up their code. They want to keep doing shit the same way they always have, no matter how outdated that is.

Re:Statistics, statistics

[mstefan]

The latest stats that I read (which was for June) was that in terms of marketshare, Vista is at around 14.5% (down from almost 20% when Win7 was released) and right now Win7 is at around 13.5%. So if you factor in the percentage of 64-bit systems all combined, it puts 64-bit Windows platforms at roughly 8.25% give or take. These are all rough numbers, and there's varying estimates out there (some higher, some lower in terms of overall marketshare) but I think it's safe to say that 64-bit Windows systems are around 10% overall.

Windows XP usage has dropped as well since Win7 was released (by about 7%, it's now at about 62% of systems overall). So for the most part, Win7 has cannibalized both Vista and XP installations fairly equally, but the effect has been much more profound for Vista because there were far fewer systems out there running it. I would exepct that by the end of the year, Win7 will clearly surpass Vista. Long term, I'm looking for when XP drops below 50% marketshare; when that happens, I think that'll be the indicator that the transition is accelerating.

Re:Statistics, statistics

[Bigjeff5]

It truly amazes me how little the average person understands about economics.

A business exists to make money. A company doesn't sell a sound card out the kindness of their heart, and they don't write drivers for said sound card out of the kindness of their heart either. They can't survive if they do. If being pro-active does not generate more income than it costs, then it is economic suicide to be pro-active.

It's called return on investment, and any halfway decent developer is going to do at least a rudimentary ROI analysis before beginning even a small upgrade, let alone a complete overhaul of their code.

Basically what you are asking developers to do is exactly the same as your boss telling you you need to work evenings and weekends for no pay. You'll do it if it means losing your job, but otherwise you'll tell your boss to either pay you for it or fuck off. Getting an updated driver is no different. If it's free, expect it to take a while - they'll have to spread the cost over a long period of time to make the investment feasible. If it means losing customers because they didn't write the driver, well then you'll get it pretty quickly.

Why, oh why?

[palegray.net]

Is there a good technical reason for 32-bit Windows 7 not supporting more than 4 GB of RAM, period? PAE has been in use for a long time now, and while you can't have a single process that exceeds 3 GB in Linux (tunable, I'm given to understand, can also be a 2 GB per process limit in some installations), you can definitely go past 4 GB of total system memory. Windows Server 2008 Enterprise supports 64 GB per 32-bit system...

Re:Why, oh why?

[0racle]

Why enable a workaround when there is a native way to support it? PAE does also technically have a performance impact, your average desktop user isn't exactly going to understand that.

Re:Why, oh why?

[shutdown -p now]

Is there a good technical reason for 32-bit Windows 7 not supporting more than 4 GB of RAM, period? PAE has been in use for a long time now, and while you can't have a single process that exceeds 3 GB in Linux (tunable, I'm given to understand, can also be a 2 GB per process limit in some installations), you can definitely go past 4 GB of total system memory.

PAE can break badly written drivers, which are more common on desktop versions of the OS than they are on server versions.

Re:Why, oh why?

[mpfife]

Well, the first off, PAE only gets you to 64gb of memory. While that should be plenty for most people for the foreseeable future, we all know that setting arbitrary and somewhat lower limits turns you into the most quoted man in history (640k should be enough for everyone....) Also, as I recall, the 64-bit memory manager in Vista was quite a bit different (and faster) than the one in the 32-bit version. Legacy support(?) However, there is more than just addressable memory to consider with a 64-bit operating system. If you use PAE, your APPS are still running 32 bit. Apps need to be recompiled or even reworked/rewritten to utilize the new 64-bit operating system features. That's probably what they're hoping for more - to get folks to thinking and writing in 64-bit. While I don't think it was a totally cool move - if I'm not mistaken Microsoft has some features available only in the 64-bit api's. In Vista, certain secure driver signing modes didn't exist except in the 64-bit version (not that this is a good thing - they were terrible - but it does show a difference).

Re:Why, oh why?

[Skuld-Chan]

For the 64 gig support on a 32 bit machine you often need special servers with chipsets that bank the memory appropriately and special system drivers (Serverworks is/was famous for this) on top of that - its really only something you need to do if you were running Metaframe (I think its called XenApp server?) because most Windows apps won't go past 2 gigs of allocation anyhow.

My understand is the reason for this is just special hardware/driver support - many consumer motherboards for instance map real world pci resources in the 4 gig address range. Its probably easier on quality assurance to only support what they do on server OS's.

64 bit system doesn't have any of these limitations and you can address all the memory in one chunk without any work-arounds - hence the wider support for more ram there.

Re:Why, oh why?

[TheRaven64]

PAE adds another layer to the page tables (as does 64-bit addressing), which makes TLB misses more expensive, so you don't want to enable it on systems that don't have more than 4GB of RAM. Given that very few machines ship with more than 4GB of RAM, but a 32-bit processor, it's likely that this would be a configuration that would get very little or no testing (especially from driver developers) so would be potentially very buggy.

A lot of PCI devices are 32-bit, so drivers need to use bounce buffers to do DMA transfers to physical memory over the 4GB line. This is something that device drivers designed for 32-bit systems won't do, because they can just pass 32-bit physical addresses straight to the device on the systems they were written for.

Re:Why, oh why?

[bernywork]

You try accessing more than 2GB of RAM (or 3GB of RAM with the /3GB switch in boot.ini) in a single process. What you end up having to do is (firstly) your own memory management (Which sucks) and having to manage multiple 2GB "windows" so if you want to read data you have to swap in an out of these "windows" to be able to read them as the kernel itself is only 32bit and can only directly address 4GB of RAM.

So you end up coding in what is known in Windows as "Addressable Window Extensions" and they are a pain in the arse. Doing this on SQL server and Oracle was basically a necessity, and when PAE was first thought of, this is exactly what was being thought about, database systems. They have been able to use PAE in VMware etc and other places as they give the upper and lower limits for memory address directly to the operating system (Windows, Linux whatever else is actually running in the VM) and then they address via the hypervisor that memory address space, meaning that the hypervisor doesn't have to do a lot of memory management (Certainly nothing like protected memory)

So in effect, the biggest reason for 32bit Windows 7 not supporting more than 4GB of RAM is because the kernel itself is a 32bit app and doesn't have the 64bit address space to directly address more than 4GB of RAM itself.

In the long term it's just too hard, and it's easier to code for 64bit than to deal with what are effectively kludges to make this work.

If you need to know more about this, I would suggest Mark Russinovich's Windows Internals book.

Re:Why, oh why?

[thegarbz]

Desktop user? Why would they care? Also the performance impact is tiny compared to the impact of seeking on the harddisk. The article itself should come with a knowledge disclaimer:

"Half of the windows 7 Machines Running 64-bit Versions, 90% of users don't even realise"

The average desktop user doesn't know what defragment is, has 50 apps that load on startup, and think computers naturally get slower over time and simply require replacing.

limits

[StripedCow]

192GB ought to be enough for anyone...

32 at work, 64 at home

[Monkeedude1212]

We use the 32 bit at work for the compatability with old the old MS Access databases (don't ask... I just work here...)

I use the 64 bit at home - even though it causes some odd glitches with various games here and there, for the most part it runs everything much smoother. I decided that I'd need more than 4 Gigs of RAM to run Visual Studio to Debug my modified Source game.

Re:32 at work, 64 at home

[NJRoadfan]

64-bit should be fine for most. For those 32-bit apps that glitch out (or those random Win16 apps, like the old Windows Entertainment Pack games), just run them with XP Mode.

In another news

[miknix]

I've been running 64bit Gentoo Linux since I bought one the first models of Athlon 64, which was almost 7 years ago!

Re:In another news

[Anonymous Coward]

Compile time doesn't count.

Same boat here

[RingDev]

Just got the word that the desktop team is pushing out Window 7. Unfortunately, there are "a couple" of printers that they couldn't get working 64b drivers for. So they are pushing the 32b version out to everyone...

Blows my mind... It would cost at most a $5000 to replace those printers, compared to the cost of 600+ copies of Windows 7. Crazy.

-Rick

Why 64-bit is ready now

[Robotron23]

One of the most obvious things about the development of 64-bit architecture is how woefully behind its 32-bit counterpart just a few years ago. I had a spare rig which I put XP 64 on in around spring of 2007. The low level of development together with the fact that powerful machines did not often have more than 4GB back then meant little incentive for devs to put time in 64-bit. I couldn't even find hardware drivers, and this led my disgruntled self to format that particular rig, whilst telling myself to research this stuff in future.

I believe Microsoft deserves some cred, along with certain hardware firms like AMD/Intel, with bringing 64 bit to the fore. Not to mention the PS3 and some Macs being of that architecture too. Ironically Microsoft's most stunted OS since the much maligned Windows ME was the first one that could run 64 bit convincingly despite a laundry list of flaws that haunted the entire life cycle of that particular product. Windows 7 is in some ways like XP was to ME a whole 9 years ago.

Whilst it's great for RAM purposes, and thus demanding things like gaming which will soon require 6GB or more for popular titles there are drawbacks. A file in 64 bit takes up more memory, mainly due to alignment padding. Thus one needs a fairly good set of chips to cache efficiently in future years as the levels of memory inevitably increase. However with the amount of progress going on I daresay all but the most budget hardware solutions will tackle drawbacks very well.

That the figure is now 50% compared to about a fifth of that not long ago is indicative that 64-bit has finally become established in the mainstream.

Re:What about flash?

[shutdown -p now]

We've only been waiting FOREVER for a Win64 version of Flash from adobe...

They probably don't feel like they need to rush it, because, as it is, Win7 x64 still ships with 32-bit IE as a default browser (due to need to preserve plugin compatibility), and all other mainstream browsers only release officially supported Windows versions in 32-bit.

Artificial limits R US (tm)

[Twinbee]

Is there a reason they can't go above the artificial limit of 192 GB?

64 bit CPUs should be able to access up to 18,446,000,000 GB of memory space, so I see no reason for the arbitrary limit.

Re:Artificial limits R US (tm)

[Zerth]

Is there a reason they can't go above the artificial limit of 192 GB?

Because then Windows Server wouldn't look very impressive.

Re:Artificial limits R US (tm)

[MobileTatsu-NJG]

Is there a reason they can't go above the artificial limit of 192 GB?

Is there a realistic way of testing it past that amount?

Re:Artificial limits R US (tm)

[Anonymous Coward]

18446000000 / 16GB = 1,152,875,000. Note at 400MHz(the minimum DDR3 speed) light travels about 1 meter per clock cycle. Such a memory array would be much larger than 1 cubic meter, making DDR latency numbers impossible to meet even with lightspeed interconnects.

Yes, IAAAD(I am an ASIC designer)

Re:Artificial limits R US (tm)

[KarmaMB84]

x86-64 can only address 48-bits and Windows only addresses 44-bits (16 TB). The Windows limitation is interesting because no windows release to date can even touch that address limitation.

My best guess would be that OS releases are artificially limited to the amount of memory they actually test internally against. Home Premium probably doesn't get serious testing beyond 16GB while Ultimate might get tested against 192GB workstation hardware. High end server releases probably get tested with up to 2TB (probably the maximum amount of hardware available at time of testing). 32-bit desktops probably don't even get tested with PAE enabled at all since chances are desktop hardware drivers will crash and burn so they get a 4GB limit.

Only half?

[JustNiz]

Everything I've ever thrown at 64 bit windows runs just fine, and usually somewhat smoother than 32 bit. Even some really old stuff. The only software I ever found that don't run on 64-bit are some really old dos games and utilities, but then they didn't even run under 32-bit XP either.

It boggles my mind why so many people with 64-bit hardware would still install a 32-bit version of windows. I wonder how much of this is actually ignorance and/or just force of habit rather than actual knowledge that they have something that actually doesn't work under 64 bit.

why I'd pick 32 bit

[ILuvRamen]

I had to work on someone's Vista 64 bit machine and I hated it. Not only were half the programs running in 32 bit mode but almost none of my virus removal tools worked so I couldn't completely disinfect it. Three different antivirus programs wouldn't install properly on it either. Almost no software I had ran on it and for some reason, Java 32 bit was installed and 64 bit wouldn't install. If I wanted a computer that no software ran on, I'd buy a mac.

Re:What about flash?

[Joe Snipe]

Did you just say that we don't have 64-bit flash because Windows uses 32-bit IE because we don't have 64-bit flash?

Blame MS's Pricing Plans

[EXTomar]

There are multiple version with multiple flavors at different price points that confuses "people". Add to this the finicky way upgrades behave and "upgrade upgrade" software and it is no wonder a lot of people don't care or realize a 64-bit version exists.

A Mac seems to run everything just fine

[SuperKendall]

If I wanted a computer that no software ran on, I'd buy a mac.

Ha Ha. But what do you mean? The 64 bit transition there has been much smoother. And these days what doesn't run on a Mac? You can after all even run Steam now...

Not sure why they do that

[Sycraft-fu]

But yes, it is an artificial limit. Their lower end products are limited on purpose. For example Home Premium is limited to 16GB which is as much as you are likely to see in a home system, but pros might want more. Hence 7 pro (and ultimate) has a higher limit.

As to why they choose to limit it to 192GB? No idea.

The actual limit for The Windows NT 6.1 setup is 2TB. I don't remember the particular technical reasons for that, but they are there. There is no reason to process addressing for the full 64-bits of memory when no system exists that can take it. You'll also find that CPUs have memory limits lower than the 64-bit cap. They don't have all 64 address lines because it is not needed. I don't know what it is currently, but it is still below 64-bit. Again, no system could possibly have that much (never mind space, a memory controller couldn't handle the electrical load) so no sense in adding hardware you don't need.

Thus are ARE real limits below the actual 64-bit space but you are correct, 192GB is not one of them. That is an artificial limit and I don't know why they chose it. Doesn't really matter, I do not see people using more than that in a desktop system (144GB is the most I've seen workstation hardware support) and they can always up the limit.

However you are correct that it seems odd.

I'll never get it either

[Sycraft-fu]

That is the most common reason I hear for not using 64-bit: "My printer and/or scanner don't work with it." My response is "Then throw that piece of shit out and get a new one." However for some reason, that never seems to be an option considered. They want to keep their 10 year old scanner with their brand new $1000 desktop, even though $50 would get a better, newer scanner.

Here at work we do both. I work for a university so we have to be flexible. I push 64-bit as the default but I'll give them 32-bit if they bitch or if there is a compatibility issue.

Yes

[Sycraft-fu]

It gets circular like that because developers are lazy fucks. So MS includes 32-bit IE, 32-bit Windows Media Player and so on as compatibility measures. Your old plugins will keep working in your new OS. They also have 64-bit versions so that when they upgrade you can use that, but they maintain old versions for compatibility. So, the lazy fucks at Adobe say "Well 32-bit is there, so we'll just keep that since that's what people use." People then say "Well Flash isn't out for 64-bit so I'll keep using 32-bit." You do get a circular situation. You can't blame the users, they use what is available, you need to blame the lazy shit devs.