• Some users have recently had their accounts hijacked. It seems that the now defunct EVGA forums might have compromised your password there and seems many are using the same PW here. We would suggest you UPDATE YOUR PASSWORD and TURN ON 2FA for your account here to further secure it. None of the compromised accounts had 2FA turned on.
    Once you have enabled 2FA, your account will be updated soon to show a badge, letting other members know that you use 2FA to protect your account. This should be beneficial for everyone that uses FSFT.

2400T'Bred vs. 2500Barton

S

Sepsis

n00b
Joined
Jun 3, 2004
Messages
19
I don't put a lot of stock in benchmarks, but I do run them just for kicks 'n giggles, as Austin Powers would say. The following is very odd.
All benches were in Sandra 2004.2.9.140 and the comp was exactly the same; except for, of course, the CPU.
============================================================
Athlon 2400 at 136x16.5=2194.5MHz/nVidia nForce2 SPP(Ultra400) Asus A7N8X Deluxe
============================================================
CPU Arith:
Dhrystone ALU 8425 MIPS
Whetstone FPU 3495 MFLOPS

CPU MM:
Integer aEMMX/aSSE 20345 it/s
Floating-Point aSSE 20945 it/s

Memory:
RAM int Buffered aEMMX/aSSE Bandwidth 1852 MB/s
RAM Float Buffered aEMMX/aSSE Bandwidth 1774 MB/s
============================================================
Athlon 2500 at 200x11.0=2200.0MHz/nVidia nForce2 SPP(Ultra400) Asus A7N8X Deluxe
============================================================
CPU Arith:
Dhrystone ALU 8281 MIPS
Whetstone FPU 3435 MFLOPS

CPU MM:
Integer aEMMX/aSSE 20619 it/s
Floating-Point aSSE 21947 it/s

Memory:
RAM int Buffered aEMMX/aSSE Bandwidth 2797 MB/s
RAM Float Buffered aEMMX/aSSE Bandwidth 2565 MB/s

The 2400 is running at 5.5MHz LESS than the 2500 is yet it scores a little bit higher in CPU Arith.
In CPU MultiMedia the 2500 pulls ahead by a very small margin.
In memory the 2500 has a massive lead due to the much higher BUS Speed.
What puzzles me is that the 2400 performs only slightly better in the Arith and slightly worse in the MM tests, yet way below in the memory bandwidth.
It seems to me that the 2400 should be performing below on ALL tests.
Logically it should perform massively slower on memory (as it does), a lot slower on MultiMedia, and quite a bit slower on Arithmetic benchmarks.
Like I said before, I don't put a lot of stock in benchmarks as they all seem to say something else and are always in conflict, but they do tell us something;
I'm just not sure what that is, or if any of us should even care. One thing I will say for sure is this: I'd rather perform a little bit slower on the Arith
and just a little bit better on the MultiMedia benches if it means the massive memory bandwidth improvement.
 
IIRC the barton has doubles the cache, which should allow it to pull ahead in most things (why its rated higher anyway, along with the bus)
 
You must log in or register to reply here.
Back
Top