Q6600 Benchies Please

SiriusB

SiriusB

SiriusB

Soldato
Joined
16 Dec 2005
Posts
14,443
Location
Manchester
Hey guys

Has anyone got any benchies for a stock Q6600. I have two clients running on mine and both are now crunching 3059s. I have a total PPD of just below 2K...

I was getting better PPD on my E6600 - what gives? :p
 
My C2Q 6600 at 2.7Ghz runs two 2653's to give a ppd of around 3600.
My C2D 6600 at 3.0Ghz gave me 2100 ppd on the same unit.

-- SMP2 --

Min. Time / Frame : 13mn 47s - 1838.74 ppd
Avg. Time / Frame : 13mn 49s - 1834.31 ppd
Cur. Time / Frame : 14mn 31s - 1745.86 ppd
R3F. Time / Frame : 14mn 35s - 1737.87 ppd
Click to expand...

A 2652 will only get 2 X 1300 ppd on the quad core, 1000 ppd less :(

I haven't had a 3059 to compare.
 
Last edited:
Hmm, I was getting great PPD before on a 1760 pointer, so it's probably just a case of bagging two really awful WUs!

Still, I have high hopes for this mr G0, so I might be able to make my PPD more respectable with these WUs.
 
I have two quads on native Ubuntu installs, each running two clients @ 3.2Ghz. both rigs are averaging 3,300 ppd. Im using DS3 mobos and 3.2Ghz is an easy overclock.
Im not 100% sure on this but using VMware in XP to run Linux seems to allow the quads to run the 2653 WUs which would give a higher ppd.
My quads only get the 1440 and 1167 pointers on a native install but Im happy with an extra 1000 points compared to my E6600 runing at 3.2Ghz also.
 
I stopped one of my clients and let the other do a few frames. Looks like these are just horrible WUs. And I go and get two of em :(

OCing should help this considerably :)
 
I don't drop below 3100ppd on my 3.5GHz Q6600's. My 2.4GHz Q6600 gets about 2200ppd. My 1.6GHz T5500 in the laptop does 800ppd.
 
Hmm, it would appear that Stanford are curtailing the quad's performance [or PPD] by giving them really awful WUs.

I was halfway through a 1760 pointer when I booted up with my quad, and it was getting very similar PPD to my [email protected] - would have loved to have seen the PPD if I had a chance to OC my quad.

Hopefully I can get my E6600 server up and running some time soon so it can pick up the slack.
 
SiriusB said:
Hmm, it would appear that Stanford are curtailing the quad's performance [or PPD] by giving them really awful WUs.
Click to expand...

The P3059 is a quad-only WU, though not sure why it's effectively valued so much lower when the benchmark machine is also a quad - have you tried running a single SMP client to see how it scores by itself?
 
Havent had one of those thar nasty ones. :)

This is what I'm getting currently

Code: 
 Project : 2653
 Core    : SMP Gromacs
 Frames  : 100
 Credit  : 1760


 -- Gemini - [email protected] --

 Min. Time / Frame : 3mn 15s  - 7798.15 ppd
 Avg. Time / Frame : 9mn 22s  - 2705.77 ppd
 Cur. Time / Frame : 10mn 28s  - 2421.40 ppd
 R3F. Time / Frame : 9mn 49s  - 2581.73 ppd
 Eff. Time / Frame : 9mn 45s  - 2599.38 ppd


 -- Pluto A64 [email protected] --

 Min. Time / Frame : 5mn 26s  - 4664.54 ppd
 Avg. Time / Frame : 24mn 30s  - 1034.45 ppd
 Cur. Time / Frame : 25mn 45s  - 984.23 ppd
 R3F. Time / Frame : 25mn 29s  - 994.53 ppd
 Eff. Time / Frame : 25mn 59s  - 975.39 ppd


 -- Bigjobbie- [email protected] --

 Min. Time / Frame : 10mn 51s  - 2335.85 ppd
 Avg. Time / Frame : 12mn 43s  - 1992.98 ppd
 Cur. Time / Frame : 12mn 59s  - 1952.04 ppd
 R3F. Time / Frame : 12mn 57s  - 1957.07 ppd
 Eff. Time / Frame : 23mn 21s  - 1085.40 ppd
 
SiriusB said:
I stopped one of my clients and let the other do a few frames. Looks like these are just horrible WUs. And I go and get two of em :(

OCing should help this considerably :)
Click to expand...

Even with just one client running the PPD wasn't great, so just a bad WU basically. Like I said, an OC will sort it out.

One of the WUs is about to finish, so lets see if I get a nice one this time :D
 
There's good ones -

Code: 
Project : 2653
 Core    : SMP Gromacs
 Frames  : 100
 Credit  : 1760


 -- CPU1 --

 Min. Time / Frame : 11mn 37s  - 2181.69 ppd
 Avg. Time / Frame : 20mn 58s  - 1208.78 ppd
 Cur. Time / Frame : 21mn 12s  - 1195.47 ppd
 R3F. Time / Frame : 21mn 08s  - 1199.24 ppd
 Eff. Time / Frame : 22mn 00s  - 1152.00 ppd


 -- CPU2 --

 Min. Time / Frame : 11mn 36s  - 2184.83 ppd
 Avg. Time / Frame : 21mn 01s  - 1205.90 ppd
 Cur. Time / Frame : 21mn 19s  - 1188.93 ppd
 R3F. Time / Frame : 21mn 15s  - 1192.66 ppd
 Eff. Time / Frame : 34mn 28s  - 735.32 ppd


 -- CPU3 --

 Min. Time / Frame : 11mn 24s  - 2223.16 ppd
 Avg. Time / Frame : 20mn 50s  - 1216.51 ppd
 Cur. Time / Frame : 21mn 26s  - 1182.46 ppd
 R3F. Time / Frame : 21mn 17s  - 1190.79 ppd
 Eff. Time / Frame : 40mn 46s  - 621.68 ppd


 -- CPU4 --

 Min. Time / Frame : 11mn 16s  - 2249.47 ppd
 Avg. Time / Frame : 21mn 01s  - 1205.90 ppd
 Cur. Time / Frame : 21mn 15s  - 1192.66 ppd
 R3F. Time / Frame : 21mn 15s  - 1192.66 ppd
 Eff. Time / Frame : 37mn 04s  - 683.74 ppd

and not so good ones

Code: 
Project : 2652
 Core    : SMP Gromacs
 Frames  : 100
 Credit  : 1148


 -- CPU1 --

 Min. Time / Frame : 13mn 18s  - 1242.95 ppd
 Avg. Time / Frame : 18mn 45s  - 881.66 ppd
 No Cur. Time / Frame
 No R3F. Time / Frame
 No Eff. Time / Frame


 -- CPU2 --

 Min. Time / Frame : 13mn 44s  - 1203.73 ppd
 Avg. Time / Frame : 18mn 49s  - 878.54 ppd
 No Cur. Time / Frame
 No R3F. Time / Frame
 No Eff. Time / Frame


 -- CPU3 --

 Min. Time / Frame : 12mn 21s  - 1338.56 ppd
 Avg. Time / Frame : 18mn 47s  - 880.10 ppd
 No Cur. Time / Frame
 No R3F. Time / Frame
 No Eff. Time / Frame


 -- CPU4 --

 Min. Time / Frame : 11mn 47s  - 1402.93 ppd
 Avg. Time / Frame : 18mn 49s  - 878.54 ppd
 No Cur. Time / Frame
 No R3F. Time / Frame
 No Eff. Time / Frame

Thankfully far more good ones than bad ones.
 
You must log in or register to reply here.
Back
Top Bottom