8800GTX @ 60841 3d03 on DI , 3d01 @ 92490 , 3d05 @ 28132 , AM3 @ 280K

[Gorod]

Spent about 2 or maybe 3 hours for that score and just noticed the "publish" button is greyed out .... Needless to say how it feels that all that work and time was for nothing
Had somebody experienced that problem , maybe theres a way to fix it ? Any tips will be highly apreciated

Thank you for looking


QX6850 @ 5211MHz , 434x12.0 , 1.84V
Mouse Pot Rev.3 @ LN2 ~ -130C
DFI LanParty UT P35-T2R
2x1024Mb Corsair C3DF , 543MHz @ 4-4-4-9 , 2.49V
Win XP SP2 , untweaked , FW 162.18 WHQL

8800GTX @ 891/2052/1188MHz , 1.65V GPU / 2.20V Mem
Cooled by Mouse Pot Rev.4 @ Dry Ice





All other submisions i did today with that card today have publish button grey'ed out also

[Viss]

Nice QX Gorod Is that 03 Build 3.6.0 ?

[Gorod]

Nice QX Gorod Is that 03 Build 3.6.0 ?

Yeah its 3.6.0

Now after looking and trying to figure out for a long time i see that processor model was not detected in System field (next to Tests > Settings > System) it just show the clock speed but no model

Will try to figure out how to fix that tommorow , hope card will clock good again

[Gorod]

Small update , 3d03 still cant publish but all other 3dmarks are ok .
Did a quick test with a single 8800GTX , but it started to have condensation problems very quick so had to stop for now and let it dry

QX6850 @ 5.3GHz (but low fsb hurst score a lot ) , 8800GTX @ 891/2052/1188 under Mouse Pot Rev.4 @ DI , 1.65V for GPU and 2.20V for Mem

Seems that card can do 900/2100/1188 for 3d01 - will try next time . With 900/2100/1188 Nature score was just a few points below 1000

http://service.futuremark.com/compare?2k1=9258907

[Stapler]

Lookin' good Gorod, let's see if you can't get a get a few more K out of that '01.

[Gorod]

Lookin' good Gorod, let's see if you can't get a get a few more K out of that '01.

Thanks
Yeah 01 is first priority for now Hope i can figure out how to stop condensation on PCI-E contact area , it's forms realy quick , only can bench for ~ 30 min and then it gets unstable . I got PCI-E slot all covered with Luberex , same thing with cards PCI-E contact area , the thing is that it gets covered with ice and then looses stability I am not very experienced with running VGA with subzero cooling , can someone help and suggest something to prevent ice forming on contact area ? Or its just the way it works with VGA - no long multi hour benches like with CPU ?


A few photos , you can a lot of ice forming around PCI-E contact are :

[harleybro]

Gorod: Nice stuff. Hopefully I will have a go again at breaking 92k. As for the slot I have a tendancy to fill the slot well but I use dielectric grease. Also make sure the holes on the top of the slot are covered. This might be the problem area. Lastly I pack paper towels all around the PCI-e slot to absorb any condensation that might drip off while benching. I have gone for a good two hours without issue using this method.

[Gorod]

I have gone for a good two hours without issue using this method.

Ah ... thx a lot for valuable tip Harley I will try to insulate PCI-E and cover the small holes on the top and on the sides like you said next time i bench tommorow , hopefully it will fix the problem
And good luck with your 01 runs m8 , hope you get a good score

[ZoLKoRn]

Nice benching Gorod, great result congrats.

hey..Gorod for your problem with pci-e contact i'm ever too, i fix by use
Petroleum jelly in the slot and contact on the card

[G H Z]

Damn nice 01 & 03 Gorod

Maybe something like JC did would work for you, I gonna try a Kapton heater

[Gorod]

Thank you ZoLKoRn
But where to find that Petroleum jelly ? Got any pics of how it looks ?

Damn nice 01 & 03 Gorod

Maybe something like JC did would work for you, I gonna try a Kapton heater

Hey G H Z
Oh , very good idea , thx for finding that , i think that gasket around PCI-E slot is going to help , will try it ASAP And + some other tips Harleybro suggested and i think it will fix the problem

[GraduS]

congrats! good results)
Блин...чё ты сделал со своей видюхой если скоро появица 8800....попытаюсь побить...

[harleybro]

Gorod I didn't get a chance to take photos last night but as for Petroleum jelly (vaseline ) I think dielectric grease cleans up easier but costs more. One small tube is more then enough and think I picked it up at pep boys (any auto store should work).

In these links of older pics you will see the towels:
http://i63.photobucket.com/albums/h1...o/IMG_0074.jpg
http://i63.photobucket.com/albums/h1...o/DCP_1417.jpg

[Gorod]

GraduS

Gorod I didn't get a chance to take photos last night but as for Petroleum jelly (vaseline ) I think dielectric grease cleans up easier but costs more. One small tube is more then enough and think I picked it up at pep boys (any auto store should work).

In these links of older pics you will see the towels:
http://i63.photobucket.com/albums/h1...o/IMG_0074.jpg
http://i63.photobucket.com/albums/h1...o/DCP_1417.jpg

Just finished reinsulating area around PCI-E + made some neoprene gaskets so hopefully it will stop condensation problems for now . Going to gas station to get some more DI and will start benching that card soon Wish me better luck this time
Will post some photos later today if my insulation's any good of course and all goes well

[Gorod]

It didnt went too well with 3dmark01 , fsb too low and biggest botleneck it seems

Oh well next time with high fsb C2D 6850 might be better will see



http://service.futuremark.com/compare?2k1=9259441



http://service.futuremark.com/compare?3dm05=3468784





5460MHz for AM3 run , 455x12 and RAM - 569MHz @ 4-4-4-9 with Perf. Level = 5 . PCI-E only 125MHz (forgot to bump it a bit )

[Gorod]

Btw , it seems that insulation wasnt a problem last time , it realy looks like some sort of OVP protection kicking in , i might be wrong but looks like it .
What happens is that some benches can run with 1.65V and some can't .
Lets say AM3 , if 1.65V wont run any speeds over 800 and if 1.60 it can run ~ 865MHz . 3d03 can handle 1.65V and @ 890MHz
3d05 likes 1.60V @ ~ 865 , 3d01 can handle 1.65V and run some tests 890MHz (nature) but other subtests will fail (Lobby and CarLow) , Car Low was very picky in particular

V GPU for 01 run : 1.65V

Nature @ 890/1188
LL , LH , DL , DH , CH @ 865MHz / 1188
CL @ 837/1134 (Failed with 865/1188 one time so i lowered just to be safe because i was runing out of LN2 )

That 8800GTX is very strange when it comes to VGPU And even with 1.65V it will drop to 1.50V while loaded sometimes and mostly stays ~ 1.53-1.55 thats a huge Vdrop

[harleybro]

Nice stuff Gorod impressive runs. Real strange about the issue with CL.

[Gorod]

harleybro thx I guess i am done with that GTX for now as she acts very weird with vgpu over 1.55~1.60V Maybe will have better luck with my Ultra i hope

[ZoLKoRn]

wow very nice ! congrats

[G H Z]

Very nice Gorod

I have the same problems with my GTX, OVP is never in the same place depending on the bench your on.

[nachthymnen]

Congratz for the great scores!!!

[hipro5]

VERY NICE........

[Gautam]

Those scores are no joke. Awesome work Gorod!

[Gorod]

Thank you very much everybody

I have the same problems with my GTX, OVP is never in the same place depending on the bench your on.

Yeah very strange cards those GTX are , one bench can take almost 1.70V and another will fail 1.60V But good to know thats normal at least , almost started to worry about my card at one point

[T_M]

What happens is that some benches can run with 1.65V and some can't .
Lets say AM3 , if 1.65V wont run any speeds over 800 and if 1.60 it can run ~ 865MHz . 3d03 can handle 1.65V and @ 890MHz
3d05 likes 1.60V @ ~ 865 , 3d01 can handle 1.65V and run some tests 890MHz (nature) but other subtests will fail (Lobby and CarLow) , Car Low was very picky in particular

AM3 is very picky like that, you really need to dial in your voltage down to 3 decimal places to peak sometimes.
Mine always fails at the smoke/particle tests if too many volts.