How Many Cores?!?!?!

Printable View

Show 100 post(s) from this thread on one page

09-09-2009, 09:57 AM
i found nemo

first venice and now this s7? good job!
09-09-2009, 12:16 PM
mursaat

Quote:

Originally Posted by St Devious

lol, somebody just came back from the future with a CPU and is teasing us all.

Exactly what I was thinking, welcome back s7e9h3n!

Have you been in the future? :rofl:

BTW I love the Task Manager screenshot with 24threads!

As someone said, where in Europe are you?
09-09-2009, 10:48 PM
Particle

Quote:

Originally Posted by justapost

Oh that was quick, thank you. Guess there is still some room, I assume the NB/IMC is running at <=1600MHz with 1650GHz core speed. The results look like IMC/L3 bottlenecked, can you run with DDR3 at 1066 which should roughly equal DDR2 800. Max possible for DDR3 1333 eight channel is ~85GB/s so this is ~48% efficiency atm, in comparison to ~80% on the quad socket istanbul system with DDR2 800 (theor. max ~51GB/s).
Particle here has an dual istanbul setup, maybe he can run a few benchmarks for comparison. Maybe stream at the same cpu and nb clock speed.

Quote:

Originally Posted by Mechromancer

Yes, it would be good to know if K10stat works for this CPU. I have a feelings it will work exactly like it does on Particle's Istanbul setup :(. I think all these server CPUs are hard wired for their clocks and voltages just like Istanbul is. Probably won't be able to overclock unfortunately.

Whoops--almost forgot about this. Here's a run at default for me. I think I must be doing something wrong. Not sure how strongly effected by ECC this is, either. I run with ECC set to maximum protection. Come to think of it, this may be with underclocked memory. Let me check...

http://www.pcrpg.org/pics/computer/opty2427/stream.png

Default: 12x2200, DDR2-800, ECC Off...not a whole lot better

http://www.pcrpg.org/pics/computer/opty2427/stream2.png
09-10-2009, 02:05 AM
justapost

Quote:

Originally Posted by Particle

Whoops--almost forgot about this. Here's a run at default for me. I think I must be doing something wrong. Not sure how strongly effected by ECC this is, either. I run with ECC set to maximum protection. Come to think of it, this may be with underclocked memory. Let me check...

Default: 12x2200, DDR2-800, ECC Off...not a whole lot better

Ohh thanks, those results look like single socket vs. dual channel. :)
Do you really have all four channels populated and ran with 12 threads? Beside that with six cores sharing two channels this benchmark is heavi imc speed limited. My 955be and also the AII 240 and 250 get best results if the nb runs one notch below the cpu. I hit a limit once the nb speed reaches 2xmem speed, so the optimum with ddr3 1333 is 2.8GHz/2.6GHz cpu/nb. If it's the same with DDR2 you should not be limited by the stock nb speed however since it's already above 1.6GHz.

Beside that I think only this ES has unlocked multis and voltages and the retail chips will be locked so using k10stats or similar to modify these settings on the chips msr register will not work and oc capability will be limited by the max possible ref HT the boards will offer.

@s7: I'm curious, what msr register did you modify? Was it the controll register or the p-state-0 register?
09-10-2009, 05:09 AM
SocketMan

FASN8ing :up: (times 3).
I really hope some of these new G34/C32 boards will be made with "XS
people" in mind.
Definitely sounds like there is a 2P market here.

Many thanks for sharing S7. :clap:
09-10-2009, 07:05 AM
Particle

Quote:

Originally Posted by justapost

Ohh thanks, those results look like single socket vs. dual channel. :)
Do you really have all four channels populated and ran with 12 threads?

- Dual Opteron 2427s (Istanbul) @ 2.2GHz CPU/2.2GHz NB
- Each CPU has two 2GB DDR2-800 DIMMs installed in dual channel configuration

The numbers are just lower than I'd expect.
09-10-2009, 10:44 PM
X47

can you run 3dmark 06? :)
09-11-2009, 07:02 AM
EniGmA1987

Particle, Maybe it has something to do with the chipset difference?
09-11-2009, 07:55 AM
justapost

Quote:

Originally Posted by Particle

- Dual Opteron 2427s (Istanbul) @ 2.2GHz CPU/2.2GHz NB
- Each CPU has two 2GB DDR2-800 DIMMs installed in dual channel configuration

The numbers are just lower than I'd expect.

Check your numa settings in the bios you should get around 20MB/s in that benchmark, just like the other dual socket shanghai/istanbul setups.
09-17-2009, 05:08 AM
Smartidiot89

So someone finally noticed...
http://www.fudzilla.com/content/view/15552/1/
09-17-2009, 05:45 AM
Particle

Quote:

Originally Posted by Smartidiot89

So someone finally noticed...
http://www.fudzilla.com/content/view/15552/1/

Hehe This thread started so long ago I'd completely forgotten to watch Fudo. I thought they'd pick it up quicker.
09-17-2009, 05:51 AM
Jakalwarrior

Didn't they get it wrong though? this is 2x CPUs right? And all of the benches were run with 2 CPUs right?
09-17-2009, 11:36 AM
s7e9h3n

Looks like the CPUs were definitely throttling:

http://img9.imageshack.us/img9/7927/wprime28.jpg

This run is over 1 sec faster than @ 3ghz.

Don't know how much I'll be playing with the setup today as I'm dead tired from my trip.
09-17-2009, 12:14 PM
ubuntu83

:clap: Great. Waiting for more.
09-17-2009, 12:17 PM
Don_Dan

Quote:

Originally Posted by s7e9h3n

Looks like the CPUs were definitely throttling:

Do you have better cooling on them now?

Could you please test with WPrime v1.55, then it's easier to compare to hwbot. :)
09-17-2009, 12:31 PM
FlanK3r

Quote:

Originally Posted by s7e9h3n

Looks like the CPUs were definitely throttling:

http://img9.imageshack.us/img9/7927/wprime28.jpg

This run is over 1 sec faster than @ 3ghz.

Don't know how much I'll be playing with the setup today as I'm dead tired from my trip.

can u try Cinebench run? And winrar benchmark?
wprime is better with the same HT: NB (2000:2000, 2400:2400 etc). Example with 2000MHz HT and 2600MHz at NB is run whorse. But i think, u know it:)
09-17-2009, 01:22 PM
s7e9h3n

Quote:

Originally Posted by Don_Dan

Do you have better cooling on them now?

Could you please test with WPrime v1.55, then it's easier to compare to hwbot. :)

WPrime 1.55 seems to have issues detecting the proper number of cores.....

http://img170.imageshack.us/img170/8762/wprime155.jpg

Quote:

Originally Posted by FlanK3r

can u try Cinebench run? And winrar benchmark?
wprime is better with the same HT: NB (2000:2000, 2400:2400 etc). Example with 2000MHz HT and 2600MHz at NB is run whorse. But i think, u know it:)

.....so does the Winrar bench:

http://img294.imageshack.us/img294/5459/winrart.jpg

Cinebench has the same problem as well...don't have a screenshot at them moment, but I'll post one later....
09-17-2009, 01:24 PM
s7e9h3n

Quote:

Originally Posted by Don_Dan

Do you have better cooling on them now?

Could you please test with WPrime v1.55, then it's easier to compare to hwbot. :)

BTW, same stock coolers...just less Vcore ;)
09-17-2009, 01:33 PM
LowRun

Quote:

Originally Posted by s7e9h3n

WPrime 1.55 seems to have issues detecting the proper number of cores.....

.....so does the Winrar bench:

Cinebench has the same problem as well...don't have a screenshot at them moment, but I'll post one later....

You're way too ahead of the pack with your gear, software can't cope with you :D
09-17-2009, 01:35 PM
btarunr

Quote:

Originally Posted by Nedjo

http://i.cmpnet.com/eetimes/news/09/08/AMD_chart.gif
probe filter takes 1 meg!
;)
we need new rev of CPUz ;)

That 1 MB is inaccessible so no point seeing it as 2x 6 MB L3.
09-17-2009, 01:42 PM
demonkevy666

Quote:

Originally Posted by s7e9h3n

WPrime 1.55 seems to have issues detecting the proper number of cores.....

http://img170.imageshack.us/img170/8762/wprime155.jpg

.....so does the Winrar bench:

http://img294.imageshack.us/img294/5459/winrart.jpg

Cinebench has the same problem as well...don't have a screenshot at them moment, but I'll post one later....

have your tried setting it Manually on both ?
09-17-2009, 01:45 PM
s7e9h3n

Quote:

Originally Posted by demonkevy666

have your tried setting it Manually on both ?

Lol...no...I guess I'll try that now...thought these programs would be a bit "smarter than that :p:
09-17-2009, 02:22 PM
chew*

Quote:

Originally Posted by FlanK3r

can u try Cinebench run? And winrar benchmark?
wprime is better with the same HT: NB (2000:2000, 2400:2400 etc). Example with 2000MHz HT and 2600MHz at NB is run whorse. But i think, u know it:)

I think he can't adjust it.

This is not an overclockers board......if he could adjust he wouldn't be writing scripts to OC all cores he would just flip multis in bios ;)

S7 advanced tab on wprime 1.55, set to 24 cores :)
09-17-2009, 02:42 PM
s7e9h3n

Quote:

Originally Posted by chew*

I think he can't adjust it.

This is not an overclockers board......if he could adjust he wouldn't be writing scripts to OC all cores he would just flip multis in bios ;)

S7 advanced tab on wprime 1.55, set to 24 cores :)

Yep...frequencies aren't adjustable. Only multi:

1.55 seems a lot slower:

http://img223.imageshack.us/img223/8140/wprime155a.jpg
09-17-2009, 02:47 PM
s7e9h3n

Any programs out there that adjust memory timings?

Show 100 post(s) from this thread on one page