• We’re currently investigating an issue related to the forum theme and styling that is impacting page layout and visual formatting. The problem has been identified, and we are actively working on a resolution. There is no impact to user data or functionality, this is strictly a front-end display issue. We’ll post an update once the fix has been deployed. Thanks for your patience while we get this sorted.

how to compare cpu's speed

HiME

Senior member
is it the same way to rate 32 bit and 64 bit processors?

for example:

an athlon 64 4000+ 2.4Ghz, and a P4 3.2 Ghz


is there anyway to calculate if they're not the same??
 
It was first rumored to believe that 3000+ = 3.0 GHz for intel... 3200+ = 3.2 GHz, and so on. But after AMD 64 was released, it was pretty much soon figured that the different numbers didn't correlate with the clockspeeds of intel at all.
 
^^ that is completely false. amd's are better at certain programs, intels are better at others. comparing single cores.. intel cpu's with the same exact components as the comparable amd system would rip dvd's faster than amd's and run photoshop better. amd cpu's would run those apps slower but run cooler and run a few(not all) games with higher fps. but gaming doesn't really matter with cpu, more or less video card
 
Originally posted by: LOUISSSSS
^^ that is completely false. amd's are better at certain programs, intels are better at others. comparing single cores.. intel cpu's with the same exact components as the comparable amd system would rip dvd's faster than amd's and run photoshop better. amd cpu's would run those apps slower but run cooler and run a few(not all) games with higher fps. but gaming doesn't really matter with cpu, more or less video card

I don't see anything false with what he said. Seems like you are addressing a totally different side to it.
 
so are you telling me an AMD 3700($200) can even compare to an intel p4/xeon@3.6($400-700) or p4/xeon@($625-875) single core? google some benchmarks where the 3700 is even compared to those procs? highly doubt it..

i'm not a fanboy, just sick of all those who think amd processors are the best at everything, amd is still 1/2 the company intel is in value
 
Originally posted by: LOUISSSSS
so are you telling me an AMD 3700($200) can even compare to an intel p4/xeon@3.6($400-700) or p4/xeon@($625-875) single core? google some benchmarks where the 3700 is even compared to those procs? highly doubt it..

i'm not a fanboy, just sick of all those who think amd processors are the best at everything, amd is still 1/2 the company intel is in value

READING IS GOOD....

He said you used to be able to....which is basically true...

Then he said that it no longer applies with the advent of the AMD 64
 
Originally posted by: PurdueRy
Originally posted by: LOUISSSSS
so are you telling me an AMD 3700($200) can even compare to an intel p4/xeon@3.6($400-700) or p4/xeon@($625-875) single core? google some benchmarks where the 3700 is even compared to those procs? highly doubt it..

i'm not a fanboy, just sick of all those who think amd processors are the best at everything, amd is still 1/2 the company intel is in value

READING IS GOOD....

He said you used to be able to....which is basically true...

Then he said that it no longer applies with the advent of the AMD 64

AMD XP's naming scheme was supposedly supposed to equal the same performance as P4 Procs... AMD XP 3200+ = Intel P4 3.2 Ghz... Although with differences in different tasks, it would match sorta?...

As PurdueRy stated, it no longer applies to AMD 64's naming scheme at all...

Unless you'd like to find an ACTUAL article proving it, then go ahead LOUISSSSS. You just implied I was an AMD fanboy, when if you read my first post in this topic from a different perspective, I never said anything of the sort.


"i'm not a fanboy, just sick of all those who think amd processors are the best at everything, amd is still 1/2 the company intel is in value "
 
thanks for the replies

as far as i understand is 64bit cpu only differs with the memory addressing, as it doesn't affect anything about clock speed. a 32bit and 64bit cpu's processing speed should be rated at the same way, is it?

my friend was like telling me that his A64 cpu was actually more than 2.2 Ghz while compared with 32bit cpu, i don't think he's right..
 
Originally posted by: HiME
thanks for the replies

as far as i understand is 64bit cpu only differs with the memory addressing, as it doesn't affect anything about clock speed. a 32bit and 64bit cpu's processing speed should be rated at the same way, is it?

my friend was like telling me that his A64 cpu was actually more than 2.2 Ghz while compared with 32bit cpu, i don't think he's right..

A Athlon 64 is a hybrid 32/64 bit CPU. So you can't really compare it just off that. 64 bit "mode" will only kick in if you have a 64 bit OS. The net gains tend to be small at this point.

If you could clarify better what you mean by a 64 bit CPU it would be helpful...do you mean a 64 bit CPU or a AMD 64. They are different as the AMD64 can run in 32 bit "mode"
 
That's irrelevant, I'd rather see both processors running at the same speed and see the results. You're adding too many variables to support a sufficient experiment, aren't you?
 
Originally posted by: HiME
thanks for the replies

as far as i understand is 64bit cpu only differs with the memory addressing, as it doesn't affect anything about clock speed. a 32bit and 64bit cpu's processing speed should be rated at the same way, is it?

my friend was like telling me that his A64 cpu was actually more than 2.2 Ghz while compared with 32bit cpu, i don't think he's right..

Go under CPU/Chipsets on Anandtech's websites, and read the articles about AMD being compared to Intel in various things... Doom 3, Quake 4, etc.
 
Originally posted by: PurdueRy
Originally posted by: HiME
thanks for the replies

as far as i understand is 64bit cpu only differs with the memory addressing, as it doesn't affect anything about clock speed. a 32bit and 64bit cpu's processing speed should be rated at the same way, is it?

my friend was like telling me that his A64 cpu was actually more than 2.2 Ghz while compared with 32bit cpu, i don't think he's right..

A Athlon 64 is a hybrid 32/64 bit CPU. So you can't really compare it just off that. 64 bit "mode" will only kick in if you have a 64 bit OS. The net gains tend to be small at this point.

If you could clarify better what you mean by a 64 bit CPU it would be helpful...do you mean a 64 bit CPU or a AMD 64. They are different as the AMD64 can run in 32 bit "mode"

Yeap, AMD 64 is a hybrid, which IIRC, it has just an extension to support 64 bit, but technically it could be considered a 32-bit chip. Course, there are ACTUAL 64-bit processors that would only run 64-bit only.
 
If you want to compare an AMD and an Intel, you can get an approximate result by comparing the number of instructions per second (measured by MIPS).


P4's - 6 instructions/cycle
AMD64's = 9 instructions/cycle
Therefore:
3000MHz p4 = 3000 * 6 = 18,000 MIPS
2000MHz A64 = 2000 * 9 = 18,000 MIPS


So, a 3GHz P4 is about the same speed as a 2GHz A64 (So a 3GHz p4 = 2GHz A64 3000+). Otherfactors such as cache and memory controllers also affect performance, but in terms of raw power, the MIPS measurements are perfectly accurate.

RoD
 
Originally posted by: rod
If you want to compare an AMD and an Intel, you can get an approximate result by comparing the number of instructions per second (measured by MIPS).


P4's - 6 instructions/cycle
AMD64's = 9 instructions/cycle
Therefore:
3000MHz p4 = 3000 * 6 = 18,000 MIPS
2000MHz A64 = 2000 * 9 = 18,000 MIPS


So, a 3GHz P4 is about the same speed as a 2GHz A64 (So a 3GHz p4 = 2GHz A64 3000+). Otherfactors such as cache and memory controllers also affect performance, but in terms of raw power, the MIPS measurements are perfectly accurate.

RoD

Uhh... a 3200+ is 2 GHz... 3000+ is 1.8 GHz
 
Originally posted by: Ricemarine
Originally posted by: rod
If you want to compare an AMD and an Intel, you can get an approximate result by comparing the number of instructions per second (measured by MIPS).


P4's - 6 instructions/cycle
AMD64's = 9 instructions/cycle
Therefore:
3000MHz p4 = 3000 * 6 = 18,000 MIPS
2000MHz A64 = 2000 * 9 = 18,000 MIPS


So, a 3GHz P4 is about the same speed as a 2GHz A64 (So a 3GHz p4 = 2GHz A64 3000+). Otherfactors such as cache and memory controllers also affect performance, but in terms of raw power, the MIPS measurements are perfectly accurate.

RoD

Uhh... a 3200+ is 2 GHz... 3000+ is 1.8 GHz


I believe newcastle 3000+ were 2.0 GHz. I have a clawhammer 3200+ that is 2.0 GHz
 
Originally posted by: PurdueRy
Originally posted by: Ricemarine
Originally posted by: rod
If you want to compare an AMD and an Intel, you can get an approximate result by comparing the number of instructions per second (measured by MIPS).


P4's - 6 instructions/cycle
AMD64's = 9 instructions/cycle
Therefore:
3000MHz p4 = 3000 * 6 = 18,000 MIPS
2000MHz A64 = 2000 * 9 = 18,000 MIPS


So, a 3GHz P4 is about the same speed as a 2GHz A64 (So a 3GHz p4 = 2GHz A64 3000+). Otherfactors such as cache and memory controllers also affect performance, but in terms of raw power, the MIPS measurements are perfectly accurate.

RoD

Uhh... a 3200+ is 2 GHz... 3000+ is 1.8 GHz


I believe newcastle 3000+ were 2.0 GHz. I have a clawhammer 3200+ that is 2.0 GHz

The 754 3000+ is 2.0 GHz... Uhh... I might be wrong? lol.
 
Originally posted by: Ricemarine
Originally posted by: PurdueRy
Originally posted by: Ricemarine
Originally posted by: rod
If you want to compare an AMD and an Intel, you can get an approximate result by comparing the number of instructions per second (measured by MIPS).


P4's - 6 instructions/cycle
AMD64's = 9 instructions/cycle
Therefore:
3000MHz p4 = 3000 * 6 = 18,000 MIPS
2000MHz A64 = 2000 * 9 = 18,000 MIPS


So, a 3GHz P4 is about the same speed as a 2GHz A64 (So a 3GHz p4 = 2GHz A64 3000+). Otherfactors such as cache and memory controllers also affect performance, but in terms of raw power, the MIPS measurements are perfectly accurate.

RoD

Uhh... a 3200+ is 2 GHz... 3000+ is 1.8 GHz


I believe newcastle 3000+ were 2.0 GHz. I have a clawhammer 3200+ that is 2.0 GHz

The 754 3000+ is 2.0 GHz... Uhh... I might be wrong? lol.

http://geek.pricegrabber.com/search_getprod.php/masterid=2827542/

Like I said...newcastles are rated at the same performance rating but have a 200 Mhz higher clock to makeup for the loss of cache that the clawhammers have
 
Originally posted by: PurdueRy
Originally posted by: Ricemarine
Originally posted by: PurdueRy
Originally posted by: Ricemarine
Originally posted by: rod
If you want to compare an AMD and an Intel, you can get an approximate result by comparing the number of instructions per second (measured by MIPS).


P4's - 6 instructions/cycle
AMD64's = 9 instructions/cycle
Therefore:
3000MHz p4 = 3000 * 6 = 18,000 MIPS
2000MHz A64 = 2000 * 9 = 18,000 MIPS


So, a 3GHz P4 is about the same speed as a 2GHz A64 (So a 3GHz p4 = 2GHz A64 3000+). Otherfactors such as cache and memory controllers also affect performance, but in terms of raw power, the MIPS measurements are perfectly accurate.

RoD

Uhh... a 3200+ is 2 GHz... 3000+ is 1.8 GHz


I believe newcastle 3000+ were 2.0 GHz. I have a clawhammer 3200+ that is 2.0 GHz

The 754 3000+ is 2.0 GHz... Uhh... I might be wrong? lol.

http://geek.pricegrabber.com/search_getprod.php/masterid=2827542/

Socket 754..... 😛 😀

Socket 939 is a different story 🙂
 
Originally posted by: Ricemarine
Originally posted by: PurdueRy
Originally posted by: Ricemarine
Originally posted by: PurdueRy
Originally posted by: Ricemarine
Originally posted by: rod
If you want to compare an AMD and an Intel, you can get an approximate result by comparing the number of instructions per second (measured by MIPS).


P4's - 6 instructions/cycle
AMD64's = 9 instructions/cycle
Therefore:
3000MHz p4 = 3000 * 6 = 18,000 MIPS
2000MHz A64 = 2000 * 9 = 18,000 MIPS


So, a 3GHz P4 is about the same speed as a 2GHz A64 (So a 3GHz p4 = 2GHz A64 3000+). Otherfactors such as cache and memory controllers also affect performance, but in terms of raw power, the MIPS measurements are perfectly accurate.

RoD

Uhh... a 3200+ is 2 GHz... 3000+ is 1.8 GHz


I believe newcastle 3000+ were 2.0 GHz. I have a clawhammer 3200+ that is 2.0 GHz

The 754 3000+ is 2.0 GHz... Uhh... I might be wrong? lol.

http://geek.pricegrabber.com/search_getprod.php/masterid=2827542/

Socket 754..... 😛 😀

Socket 939 is a different story 🙂

He mentiioned nothing about socket 😉
 
IPC makes running them at the same speed irrelevent, P4's are not designed to get alot done per cycle on purpose..
thats what the extra frequency is for..

I say compair them by cost...
or compair them by cost and highest overclock possible.
 
Originally posted by: PurdueRy
Originally posted by: Ricemarine
Originally posted by: PurdueRy
Originally posted by: Ricemarine
Originally posted by: PurdueRy
Originally posted by: Ricemarine
Originally posted by: rod
If you want to compare an AMD and an Intel, you can get an approximate result by comparing the number of instructions per second (measured by MIPS).


P4's - 6 instructions/cycle
AMD64's = 9 instructions/cycle
Therefore:
3000MHz p4 = 3000 * 6 = 18,000 MIPS
2000MHz A64 = 2000 * 9 = 18,000 MIPS


So, a 3GHz P4 is about the same speed as a 2GHz A64 (So a 3GHz p4 = 2GHz A64 3000+). Otherfactors such as cache and memory controllers also affect performance, but in terms of raw power, the MIPS measurements are perfectly accurate.

RoD

Uhh... a 3200+ is 2 GHz... 3000+ is 1.8 GHz


I believe newcastle 3000+ were 2.0 GHz. I have a clawhammer 3200+ that is 2.0 GHz

The 754 3000+ is 2.0 GHz... Uhh... I might be wrong? lol.

http://geek.pricegrabber.com/search_getprod.php/masterid=2827542/

Socket 754..... 😛 😀

Socket 939 is a different story 🙂

He mentiioned nothing about socket 😉

In my post I was actually referring to the socket754 3000+. The socket939's get extra performance from the same clock speed, due to the integrated memory controller (i did mention about that in my post...).

OP: To answer you question - a 3000+ is APPROXIMATELY equal to a P4 3GHz. If you want to know exactly how well each perform under certain circumstances, the only way is through benchmarks.

RoD
 
Back
Top