• Welcome to Overclockers Forums! Join us to reply in threads, receive reduced ads, and to customize your site experience!

What are everyone's dual core opteron SuperPi 32M/1M times?

Overclockers is supported by our readers. When you click a link to make a purchase, we may earn a commission. Learn More.

LeAd_Poisoning

Member
Joined
Feb 11, 2006
Location
top gear
i thought it odd that my 165 was outperformed by my old single core A64 3700+ @ 2.75 Ghz. Is something wrong with my setup? What are everyone else's times so i can see if my times are normal or not

Mine are:
32M: 28m:26s
1M: 31s

@ 2.9ghz
32M: 26m:36s
1M: 29s

imalso running dual channel now which might have helped
 
Last edited:
My 3700+ @2.97 is 29 sec. I have an opty 175 on the way so I can let you know in two or three days what that does.

LeAd_ , You are running a RAM divider on the opty overclock, were you 1:1 with the San Diego? That might make up the difference.


EDIT: 32MB 26:24:8
 
Last edited:
fabulouscoops said:
LeAd_ , You are running a RAM divider on the opty overclock, were you 1:1 with the San Diego? That might make up the difference.

i thought there was no performance loss on asynchronous setups with A64 chips?
 
www.hwbot.org allows you to search for results by CPU. Or you can search the apropiate threads in our system benchmarking section.

Superpi is a Single threaded app that is all about speed, great bandwidth, and tight timings. So the best AMD CPUs are going to be running at high speed with 1mb cache, for 939 they will be running 250+mhz 2,2,2,5 1t timings. For AM2 500+ 4,4,4,8 2t.

i thought there was no performance loss on asynchronous setups with A64 chips?

There is a performance drop its just not as bad as Athlons or P4s were. It especially shows up in superpi and 3dmark01.
 
Opteron 165 @ 2.75 GHz (306x9 7:10) / 214 MHz 3-3-2-5 1T (2x1gb) g.skill liteblue zx

Just under 31sec (~30.812sec) for 1MB.
28min 7sec for 32MB.

You sub 2.9 guys really need to do 32MBs.

Either way looks like clocks determine 99% of the score and yours is no exception =p
 
Max0r said:
Opteron 165 @ 2.75 GHz (306x9 7:10) / 214 MHz 3-3-2-5 1T (2x1gb) g.skill liteblue zx

Just under 31sec (~30.812sec) for 1MB.
28min 7sec for 32MB.

You sub 2.9 guys really need to do 32MBs.

Either way looks like clocks determine 99% of the score and yours is no exception =p

true, especially given the difference in our RAM settings and spec. We have the same score with roughly similar clocks despite significant RAM differences
 
Yar ^_^ That's what I like about s939... clocks count the most =D As opposed to my p4 system where RAM freq makes a big difference.
 
I was doin 32s with a 3500+ clocked at 2600MHZ (10 x 260 at 1.5-2-2-5) and rediculous voltage on the core and memory. its not all clock speed!

the memory was the OCZ ELVX3200 that was rated at 2-2-2-5 at 3.2V and I was giving it 3.7V to do that speed for benches, but it was REALLY hot to the touch.
 
Sorted by GHz, colored by result class.

2.60 GHz - 32.000 sec ~ [autolex84]
2.70 GHz - 31.000 sec ~ [gigabit]
2.75 GHz - 32.000 sec ~ [moz_21]
2.75 GHz - 30.816 sec [Max0r]
2.77 GHz - 31.000 sec ~ [Spacemonkey]
2.80 GHz - 31.000 sec ~ [LeAd_Poisoning]

2.90 GHz - 29.000 sec ~ [fabulouscoops]
2.90 GHz - 29.000 sec ~ [quityourbitchin]
2.91 GHz - 28.844 sec [The Wicker Man]
2.95 GHz - 29.000 sec ~ [68Injunhed]
3.00 GHz - 29.000 sec ~ [Excalibur1982]
3.00 GHz - 28.640 sec [smokie mcpott]

3.00 GHz - 27.906 sec [Bender]

Have too much time on my hands ^_^ Still waiting to see which chip Memphis was talkin about.

You are right of course, autolex. Of course so am I. Our statements are not contradictory :D

I'm really jealous of you high RAM freq guys. I've been having nothing but trouble with RAM in my board.
 
Last edited:
27.906 seconds for my Opteron 165 @ 3GHz. My ram is probably the reason for the slightly faster time than others.

superpi.jpg
 
Last edited:
Back