When installing a new version of Windows (7 or 10), what are the pros and cons of using the 32 bit or the 64 bit version, other than software compatibility? If I have 4 GB or less memory, does the 64 bit version give any advantage?
In 2017, there is very little reason to install any 32 bit OS. Unless you have some big reason to run legacy software, but even then I’d install 64 bit and put 32 on a different partition or older computer.
Among other things, your GPU is also affected by the limit.
I’ve used 64 bit since 2006, although that was Linux of course ( they got there first at the turn of the century ).
I have never understood why anyone still uses 32 bit on a modern PC for the last 10 years. It’s like dropping a T-Model Ford engine in a Ford SUV.
RAM addressability is the main issue, but 4G is not as much memory as it used to be and you may find that very limiting. Right now I’m using a midrange laptop with 8G. I know of no disadvantages to going to 64, except if you want to upgrade from 32-bit to 64-bit I believe you are required to do a clean install. Are you doing a clean install or an upgrade? What’s your machine spec and what is the current OS?
Although it feels like the 32-bit to 64-bit transition was much, much slower than the 16-bit to 32-bit transition. And I don’t really hear anyone talking about 128-bit processors in the mainstream. It does seem like 64-bit actually is enough for everyone.
Well, a vast number of Windows users deeply loved XP, and resented the succeeding Windows, and still do. I know some of them. However, although there was a version of XP 64 bit, it was not on most people’s machines, so I’d suggest the prolongation of XP accounts for the slowish uptake of 64 bit.
The Microsoft monoculture’s own success sometimes militates against it’s own best interests.
I’ve looked forward to 128 bit for a long time, not because it’s likely to do me any good, but because computing has to evolve. Otherwise people might as well have stayed with adequate old XP.
I think that was their advertising slogan: ‘XP: Entirely Adequate’.
Some software requires a 64 bit operating system and won’t run on a 32 bit OS.
On the other hand, some old software won’t run on a 64 bit OS. I personally run a virtual machine runing XP on one of my systems so that it can handle the older software, so there are workarounds.
Well, 64 bit addressing will handle over 17 BILLION gigabytes of memory. I don’t think there’s any need for a processor with 128 bit addressing anytime in the near future.
128 bit arithmetic registers are somewhat more defensible. But except maybe for some graphics applications, which are mostly handled by separate GPUs nowadays anyway, it’s hard to imagine applications where 128 bit arithmetic will really help much.
I’m not 100% sure, but 64 bit might give you access to the whole 4 GB instead of 3.2-3.5 GB on 32 bit.
the downside of 64 bit is that AMD removed the ability to access the “Virtual 8086” operating mode when the CPU is running in 64-bit. So old (old) 16-bit Windows and DOS software will not run under 64-bit Windows.
It’s not even that slow, considering how some industries like banking are using ancient hardware that skews things. 50% were 64 bit in 2010, and in 2017 among Steam users (granted, not the most Luddite group), almost 90% of Windows users were 64 bit (expand OS version link; some on the survey don’t specify whether 64 or 32 so I didn’t count them).
“640k ought to be enough for anybody.”
There are some limits to such things, a 512-bit machine could address every proton in the universe.
[ul]
[li]If you want to run Windows 7/10, 4 GB is just fine.[/li][li]If you want to run Windows 7/10 and run any application in Windows, the RAM sweet-spot starts at 8 GB. For most users, 16 GB RAM is the real minimum.[/li][/ul]
There are a number of security features that are only available in 64-bit editions of Windows (or are significantly better). If you care about security and and don’t absolutely need to run 16-bit apps natively (vs. in a VM if you even need them), you should always run 64-bit Windows on 64-bit hardware.
See https://support.microsoft.com/en-us/help/946765/a-description-of-the-differences-between-32-bit-versions-of-windows-vista-and-64-bit-versions-of-windows-vista (written for Vista but applies to all subsequent versions).
The OP may already know this, but the earlier versions of 64-bit Office, not Windows did not confer any advantage and were a real can of worms if you were using 3rd party add-ins.
So in the early days of 64-bit Windows (both XP & 7+) the advice was to stick with 32-bit Office unless you had an explicit need for some of the magic stuff that was only available in 64-bit Office.
I’ve been out of the industry long enough that I don’t know how this advice has aged. If you’re still trying to Run office 2003 or 2007 or even 2010 it’s applicable if you’re using add-ins as so many corporate users do. Office 2013 & subsequent I have no clue.
“Evolve” towards what? Evolution has to be measured in tangible utility. Sometimes it’s clearly and even desperately needed: Windows 3.1 was an important advance over MS-DOS and Windows 2; Windows 95/98 over Windows 3.x; and Windows XP over all of them. At which point Microsoft had achieved the perfect marriage of the stable NT kernel with important consumer OS features in a stellar package, and at which point they ran out of steam and started flailing about for ideas. Years later, they presented the world with Vista, whose motto should have been “trust us, you need this, though we can’t really tell you why”. Turned out, most of the world couldn’t figure out why, either.
The primary reason that Windows XP is no longer adequate is artificial and is due to lack of support: root certificates and associated protocols are out of date, applications require new versions of .NET framework that are not available for XP, things like that. If it was still adequately supported there would be no need for anything newer, which would be a problem for a software company trying to flog new operating systems.
Just simply not true. I run Windows 7 (32-bit) on an old laptop with 2 GB with some major large applications and have no problem whatsoever, and ditto with Windows 7 64-bit with 4 GB on a 3 GHz Core 2 Duo where it runs like a bat out of hell no matter what I throw at it. (Also, XP on a system with 512 MB, which is a bit starved for memory only because nowadays Firefox is a bloated pig.) Disclaimer: I am not a gamer.
Anyway, to the OP, the typical choice with any modern OS from Windows 7 on should be 64-bit absent some good reason to do otherwise. With XP, the good reason was that the 64-bit version had a number of problems and limitations, but that has not been the case since Windows 7.
Registers are already 256 bits (for the AVX instruction set). Some models are up to 512 bits. There’s no longer any strong relation between register size and memory addressing range (which isn’t really 64 bits anyway–no processor actually supports that much physical memory).
There are lots of advantages to a 64-bit OS, even with only 4 GB of memory. 64 bits is the size of the total address space, and can include things which aren’t system memory (like GPU memory, memory-mapped files, and other things).
Quad core /w 32 Gigs of RAM. SSD hard drives.
I screw up at insane rates.
Low performance PC’s & Laptop’s (2GB RAM, single/dual Core CPUs) perform better with 32bit, since the workload on the computer is lower.
The 64bit operating system on those systems is terrible slow, however a 64bit OS allows you to address more Cores and more that 4GB (3.25) RAM.
I downloaded and installed Office 2016 from Microsoft’s Office 365 portal just last month. They’re still advising clients to install 32-bit Office unless they have a compelling reason to do otherwise.