Technical Question: Is a 64-bit CPU better than a 32-bit CPU?

[LiQiCE]

Architecture efficiency aside, why is a 64-bit CPU better than a 32-bit CPU? I understand that a 64-bit CPU is more precise than a 32-bit CPU in terms of floating point calculations. But what other advantages does a 64-bit CPU have over a 32-bit CPU? I've been arguing with somebody who claims that 32-bit CPUs need to split numbers into two registers when they make calculations to get 64-bit precision but I was always under the impression that the CPU would just round off a calculation to 32-bits to get it to fit into a 32-bit register. Additionally he claims that 64-bit CPUs can send data faster because it can queue up 64-bits of data to be sent to memory.

From what I remember from my Computer Architecture class, a 64-bit CPU doesn't give you any critical advantages over a 32-bit CPU but perhaps I'm wrong since both AMD and Intel are moving to 64-bit architectures (and obviously they're doing it for some good reason... although Intel's big thing is getting away from the old x86 architecture).

To give you a little more insight, the arguement is actually over the X-Box's 32-bit Pentium III CPU versus the GameCube's 64-bit (or is it 128-bit?) Gecko Processor ... I don't see any advantage that the GameCube holds over the X-Box simply because the X-Box's CPU is 32-bits while the GameCube's CPU is 64-bits except for precision in floating point calculations. Please don't turn this into an arguement over X-Box versus GameCube, I just want to know some facts regarding why a 64-bit CPU would be better than a 32-bit CPU. Thanks for anyone who can help me out.

 

[Zwingle]

AMD Athlons are 40-bit CPU, Pentium 4 is a 32-bit.

 

[LiQiCE]

Yeah I know, but AMD's Hammer and Intel's Itanium are both 64-bit solutions right? I would guess that the new architecture will eventually replace the Athlon and the Pentium 4.

 

[thomsbrain]

For every bit you add, you get twice the data. A 64 bit CPU can handle like a trillion more numbers in one chunk of data, right? Of course, you need software designed to take advantage of it. Correct me if I'm wrong, guys.

 

[ILikeSprite]

OT, but the Xbox uses a variation of the Celeron processor.

 

[charrison]

64 bit cpus are only better than 32 bit cpus if you have a need for keeping track of more than 4billion things at a time.
32bit is good enough for almost everything.

Gameboxes are not 64 or 128bit cpus. They are 2x32(64bits) or 4x32(128bits). They are using the same type of SIMD that 3dnow/sse/sse2 use.

 

[Hamburgerpimp]

Dude, you just posted in Off Topic. We're the degenerates of Anandtech. Take it to "Highly Technical Forum"!

 

[ILikeSprite]

<< Dude, you just posted in Off Topic. We're the degenerates of Anandtech. Take it to "Highly Technical Forum"! >>



hahaha!

 

[psteng19]

more bits = more memory to address/access, i think.

 

[LiQiCE]

thomsbrain,

Well my understanding is that a 64-bit Processor will give you a more precise answer and can handle numbers much larger, but I don't know how that helps you out. In older days the jump from 8-bit to 16-bit CPUs gave you a much larger addressable memory, and allowed you to process numbers which were feasible. I can see a server's need for 64-bit processing but I don't see the need for it in a console or a desktop right now. Any help guys?

ILikeSprite,

The X-Box uses a modified Pentium III or Celeron processor, however you want to figure it ... It has only 128kb of cache but it maintains the Pentium 3's 8-way associative cache which gives it a 10% performance boost over the standard Celeron, and it runs at a 133Mhz FSB. So its technically a separate processor from both the Pentium 3 and Celeron, but realistically its a Pentium III with 128kb of cache.

 

[LiQiCE]

Hamburgerpimp,

HAHAHA! I posted in the highly technical forum too, but there are a lot of smart guys on ATOT too, so I figured I'd get some input from here as well

charrison,

So then by that definition, the Pentium III is just as much a 64-bit processor as the Gamecube's Gecko or the PS2's CPU because SIMD allows for 64-bit floating points right? I didn't realize they considered Gecko and PS2 64-bit or 128-bit or whatever just because of SIMD type calculations.

 

[jaydee]

I wouldn't take it to HT unless you have an EE/CE or similar backround, cause pm and WingnutPez will run circles around your brain if you haven't.

 

[LiQiCE]

jaydee,

hahah, actually somebody provided a really good answer on the HT forum that I was able to totally understand. It actually makes a lot of sense and sortof explains why Intel and AMD haven't simply jumped to a full 64-bit platform for desktop computers yet.

Here's a link to the thread I posted in HT: Link

 

[OS]

How many bits a CPU is does not make it inherently better or worse than another CPU. Think of the bits as like how many lanes on a freeway. Would you want to be in a 6 lane freeway which is bumper to bumper all the way, or a 2 lane freeway where everyone is going mph? Similar idea here. Higher bits does however mean more memory addressing, but for most applications, this is not a critical issue.



<< From what I remember from my Computer Architecture class, a 64-bit CPU doesn't give you any critical advantages over a 32-bit CPU but perhaps I'm wrong since both AMD and Intel are moving to 64-bit architectures (and obviously they're doing it for some good reason... although Intel's big thing is getting away from the old x86 architecture). >>



You have that more or less correct. I think 32 bit cpus have a 4 gig memory allocation limit which would be one major factor in the move over to 64 bits.

 

[Cessna172]

I don't know about the 64 vs 32-bit thing, but since your discussion with your friend dealt with the game consoles, it is worth noting that despite whatever advantages 64 bits has over 32, the Xbox has much more power than a Gamecube or PS2 for producing graphics. There is a really good article you might want to look at detailing the technical specs of the Xbox and Gamecube here:

Extremetech