Provocative Statement

[PaulB]

I just read this Deal of the Day at CNet. Part way down, it states:

I think my only gripe with the 625 is that it comes with the 32-bit version of Windows 7 Professional. Granted, not many users reap the benefits of 64-bit Windows anyway, but I still prefer to have the faster of the two operating systems.

Can anyone here verify that W7 64 bit is indeed faster than its 32 bit counterpart and, if so, how much faster is it?

[BobH]

Hello Paul!

While I cannot offer evidence that 64-bit Win7 is faster than 32-bit, my guess is that we are approaching - and, indeed, may have passed - the point where the difference in speed between the two is perceptible. Without asking whether a tree falling in the forest makes a sound if there is no one to hear it fall , let me say that, all else being equal, a 64-bit OS must be faster than a 32-bit OS simply because the bandwidth is greater, both for physical input/output and for the increased ability to address and use more memory. The question remains whether all aspects of the hardware and the OS software, and even of application software, are sufficient to exploit the speed offered.

I don't even know how one would go about making such a measurement except by benchmarking one against the other under the same operating conditions and workload. Many of the variables in such a measure must be held constant in order to measure any one of the others.

It will be interesting to hear what others know of the matter.

[HansV]

There isn't a lot of software specifically written for the 64-bit platform yet, although the numbers are steadily increasing. Most software is still 32-bit. This means that there is a priori no great advantage for a 64-bit OS yet.

BUT 64-bits Windows can handle much more RAM than 32-bits. 32-bits Windows is limited to an address space of 4 GB, of which about 1 GB is taken up by video memory and some other stuff, so the maximum application RAM is about 3 GB. That is plenty for most applications, but if you either have a lot of applications open at the same time or use memory-intensive applications, the extra memory possible in a 64-bit environment would not only be very welcome, but make a significant difference in performance - RAM is much faster than the swap file on hard disk.

[PaulB]

There is much food for thought and many variables to consider here for sure. I was wondering, however, if there was anything specific that would lead the author to state so conclusively that 64-bit W7 would be faster on this particular computer than 32-bit W7? Would the same hold true for any given computer?

[BobH]

There isn't a lot of software specifically written for the 64-bit platform yet, although the numbers are steadily increasing. Most software is still 32-bit. This means that there is a priori no great advantage for a 64-bit OS yet.

BUT 64-bits Windows can handle much more RAM than 32-bits. 32-bits Windows is limited to an address space of 4 GB, of which about 1 GB is taken up by video memory and some other stuff, so the maximum application RAM is about 3 GB. That is plenty for most applications, but if you either have a lot of applications open at the same time or use memory-intensive applications, the extra memory possible in a 64-bit environment would not only be very welcome, but make a significant difference in performance - RAM is much faster than the swap file on hard disk.

That's what I was trying to say, but couldn't make as clear as did Hans!

[HansV]

There is much food for thought and many variables to consider here for sure. I was wondering, however, if there was anything specific that would lead the author to state so conclusively that 64-bit W7 would be faster on this particular computer than 32-bit W7? Would the same hold true for any given computer?

On a PC with 2 GB of RAM, I don't think the 64-bits version would be faster than the 32-bits version, if at all... But I don't have quantative evidence.

[StuartR]

There is some possibility that 64 bit Windows would be slower than 32 bit Windows on a system with less than 2.5 or 3GB of memory. This is because all memory addresses need 64 bits instead of 32, and programs have many pointers to memory addresses, so use up more memory storing them.

[Rus]

I would think that the comparison would be relative to the difference between 16 bit and 32 bit systems, a generation ago. Today, as then the speed difference depended on CPU, RAM, and Bus speed so comparing 32 to 64 in the raw is not relevant. If you have the hardware and the software to take advantage of a 64 bit system there is no good reason not to take advantage of what one has.
Don't forget that to go from 32 to 64 bit one must do a full/clean install, there is no upgrade path.

[hlewton]

Admittedly not knowing much about this subject but wondering about doing a full/clean install on a computer that could possibly handle Windows 7 after testing it for compatibility at a site like MS. Wouldn't you need the computer to have the hardware (Bus especially) that would be compatible with the 64 Bit version or could any computer be upgraded to Windows 7 64 Bit OS even one having an older Bus that tests compatible for installation of Win 7? Sorry if it is a stupid question but I am under the assumption that a Bus is designed to handle different things like speed and either 16, 32, or 64 Bit Operating systems.

[HansV]

See the section "How do I tell if my computer can run a 64-bit version of Windows?" in 32-bit and 64-bit Windows: frequently asked questions.

[StuartR]

Even if your computer is capable of running Windows 7 64-bit you may prefer to run the 32 bit version.

• Many programmes are not yet compatible with 64 bit.
• If you have less than 4GBytes of memory then 64 bit won't be any faster, and may even be a bit slower.

[HansV]

Stuart, while I agree with your second point, I'm not sure about the first one. I've had a PC with 64-bit Windows 7 for almost a year now, and I haven't found a single program yet that won't run on it unless it's incompatible with Windows 7 as such. Microsoft appears to have done a pretty good job of enabling 32-bit software in 64-bit Windows 7.

[StuartR]

Hans,

You may well be right, I don't have a 32-bit Windows 7 system on which to test the programs I have that won't run on my 64-bit Windows 7 PC.

[PaulB]

Stuart, I won't disagree with your second statement, but it does seem to be diametrically opposed to the CNet quote in my original post above. So far I haven't seen an answer here that addresses why CNet says that 64 bit Win7 is faster than 32 bit on that computer.

[HansV]

The laptop as offered has "only" 2 GB of RAM. While that is enough for Windows 7 32-bits, it is on the meager side for Windows 7 64-bits. If you frequently have several applications open at the same time (and who doesn't, nowadays), I think 64-bits would actually be slower than 32-bits.

The laptop is capable of having RAM exanded to 4 GB. With that addition, 64-bits would be an attractive option, but the 2GB add to the price, of course.

[PaulB]

I'm not looking at buying this particular computer. I'm just intrigued by the statement made by the author that seems to say: Win 7 64 bit is inherently faster than Win 7 32 bit regardless of the computer. I've never read anything that validates that statement. What does he know that I don't?

Personally, I tend to agree with you and Stuart. Win 7 32 bit is probably the better choice for lower end computers. Win 7 64 bit requires a bit more horsepower.

[HansV]

As a blanket statement, "Win 7 64 bit is inherently faster than Win 7 32 bit" is just plain wrong. So the author doesn't know anything you don't...

[PaulB]

Can't tell you how relieved I am!

[BobH]

Many people write poorly, as anyone who has read my posts can attest.

I think that if the terms "potentially" and "inherently" are examined for the concepts they either do, or might, convey in common, one can see that the author might have meant that the 64 bit has within its nature the ability to be faster than a 32-bit system. That statement is, to the best of my knowledge, a true and correct one. The fact that a 64-bit system contains potential to be faster might lead one who is not too careful with words to state that the capacity is inherent - contained within its potential for use.

At least, that is the only way I could justify the author's assertion . . . requiring me, perhaps, to swallow a camel but gag on a gnat.