Leadtek 6800nu 3DMark03 benches.

[Keysplayr]

P42.8C o/c'd to 3GHz (143FSB)
ASUS P4P800
1GB PC3200 (dual channel)
Leadtek GeForce 6800nu
200GB Western Digital IDE drive
Audigy2
Windows XP SP1
DirectX 9.0c
Forceware 66.81 for all tests
RivaTuner 15.2

3DMark03/ 6800 12x1,5vp 325/700: Score 8683

3DMark03/6800 16x1,5vp @ 325/700: Score = 9362

3DMark03/6800 16x1,6vp @ 325/700: Score = 9488

3DMark03/6800 16x1,6vp @ 350/800: Score = 10208


I have not tried to o/c any higher at this point but the card ran flawlessly in all my games and in 3DMark03. I don't usually put any stock into 3DMark benches when one type of video card is compared to another, but benching the same card against itself is a great way to see if your improving.

Enjoy. If anyone has a similar setup to mine here, post your results.

 

[MercenaryForHire]

Please enjoy a cold :beer: for all your hard testing.

And that's just sad ... you gained more from the OC+softmod to 16/6 than my 9500Pro gets.

- M4H

 

[Marsumane]

Originally posted by: keysplayr2003
P42.8C o/c'd to 3GHz (143FSB)
ASUS P4P800
1GB PC3200 (dual channel)
Leadtek GeForce 6800nu
200GB Western Digital IDE drive
Audigy2
Windows XP SP1
DirectX 9.0c
Forceware 66.81 for all tests
RivaTuner 15.2


3DMark03/6800 12x1,5vp @ 325/700: Score = 8683

3DMark03/6800 16x1,5vp @ 325/700: Score = 9362

3DMark03/6800 16x1,6vp @ 325/700: Score = 9488

3DMark03/6800 16x1,6vp @ 350/800: Score = 10208


I have not tried to o/c any higher at this point but the card ran flawlessly in all my games and in 3DMark03. I don't usually put any stock into 3DMark benches when one type of video card is compared to another, but benching the same card against itself is a great way to see if your improving.

Enjoy. If anyone has a similar setup to mine here, post your results.

Impressive. Not too bad for a 6800nu huh? I would push it further if itll go. The luck seems to be with you. Spend all of it

 

[nRollo]

Originally posted by: Marsumane

Originally posted by: keysplayr2003
P42.8C o/c'd to 3GHz (143FSB)
ASUS P4P800
1GB PC3200 (dual channel)
Leadtek GeForce 6800nu
200GB Western Digital IDE drive
Audigy2
Windows XP SP1
DirectX 9.0c
Forceware 66.81 for all tests
RivaTuner 15.2


3DMark03/6800 12x1,5vp @ 325/700: Score = 8683

3DMark03/6800 16x1,5vp @ 325/700: Score = 9362

3DMark03/6800 16x1,6vp @ 325/700: Score = 9488

3DMark03/6800 16x1,6vp @ 350/800: Score = 10208


I have not tried to o/c any higher at this point but the card ran flawlessly in all my games and in 3DMark03. I don't usually put any stock into 3DMark benches when one type of video card is compared to another, but benching the same card against itself is a great way to see if your improving.

Enjoy. If anyone has a similar setup to mine here, post your results.

Impressive. Not too bad for a 6800nu huh? I would push it further if itll go. The luck seems to be with you. Spend all of it

Easy for Marsumane to say when it's your money down the drain if it burns. I wouldn't go any higher than that.

 

[Keysplayr]

Yup, I was just willing to see what the card was able to do. Was willing to take a little risk, I seldom do with my computer hardware. The softmod was irresistable for me. I think I'll leave my CPU at 3GHz though. only 200MHz over, very mild. I got the card from a vendor my company uses for computer components and they have an employee purchase plan. So the discount was nice.

Very VERY nice copper front and back heat sinks on this Leadtek. Nice and quiet.

On a side note, I can't tell a difference in 2D between my PNY 6800GT and this Leadtek 6800nu.
I heard that Leadtek had the best 2D, and it is very nice. So I guess the PNY is just as good. All reference cards I suppose besides the Heatsink on the Leadtek.

 

[IceMole]

Makes me realize my cpu is limiting my video. XP2200 running eVGA 6800NU 16x1,6 @ 380/939 barely getting over 9k Gonna return it tomorrow and get in on the dell deal.

 

[Keysplayr]

Originally posted by: IceMole
Makes me realize my cpu is limiting my video. XP2200 running eVGA 6800NU 16x1,6 @ 380/939 barely getting over 9k Gonna return it tomorrow and get in on the dell deal.

You should try backing off a little on that o/c and see if your score goes up a little. Sometimes too much o/c can hurt performance. Just a suggestion.

 

[Pete]

Congrats on your good luck, keys. That's a nice card (an unlocked 6800 in general, but that quiet Leadtek in particular), and it sounds like you got a good deal on it. Probably best bang/buck of this gen, especially if you'll stay below 12x10 with AA ("only" 128MB).

 

[Keysplayr]

Thanks Pete. I can deal with 12x10 and AA. CoD is what I find myself playing most these days and 16x12 2xQ 8xAF is a piece of cake. Keeps the frames pegged at 60 and almost never drops from that. This was before the softmod and o/c.

 

[Jhatfie]

Very nice keysplayr2003. I bet you could get some more still.

I did a bunch of testing the other day with my Chaintech AA6800 (6800nu) and noticed some pretty huge performance gains with my overclock and added pipes. After I changed the throttle voltage in the bios, I am able to get a 24/7 overclock of 430/920, which is what I used to run my tests. I did some additional tests and verified that CPU power makes a huge difference as well. There is a fps gain of nearly 10% with my A64 going from 2.2ghz to 2.55ghz.

FarCry Bench was the training level timedemo @ 1280x1024 with Ultra detail and 2xAA/4xAF, average of three runs. Doom3 was just the demo1 timedemo at 1280x1024 with high detail (vsync off) with 2xAA/8xAF, 2nd run. I also dropped my card down to standard 6800nu speeds (325/700) to see how big a difference it all makes with overclocking compared to leaving it alone.

card config,,,,,,,,,,12p/5,,,,,,,,,,,,,,,,,,,,,16p/6,,,,,,,,,,,,,,,,,,,,,,,12/5:6800nu clocks
3dMark03:,,,,,,,,,10676,,,,,,,,,,,,,,,,,,,,,12549,,,,,,,,,,,,,,,,,,,,,,, 9164
FarCry:,,,,,,,,,,,,, 38.27fps,,,,,,,,,,,,,,,,,46.75fps,,,,,,,,,,,,,,,,,,,31.79fps
Doom3:,,,,,,,,,,,,, 70.1fps,,,,,,,,,,,,,,,,, 79.6fps,,,,,,,,,,,,,,,,,,,, 58.7fps

 

[Marsumane]

Originally posted by: Rollo

Originally posted by: Marsumane

Originally posted by: keysplayr2003
P42.8C o/c'd to 3GHz (143FSB)
ASUS P4P800
1GB PC3200 (dual channel)
Leadtek GeForce 6800nu
200GB Western Digital IDE drive
Audigy2
Windows XP SP1
DirectX 9.0c
Forceware 66.81 for all tests
RivaTuner 15.2


3DMark03/6800 12x1,5vp @ 325/700: Score = 8683

3DMark03/6800 16x1,5vp @ 325/700: Score = 9362

3DMark03/6800 16x1,6vp @ 325/700: Score = 9488

3DMark03/6800 16x1,6vp @ 350/800: Score = 10208


I have not tried to o/c any higher at this point but the card ran flawlessly in all my games and in 3DMark03. I don't usually put any stock into 3DMark benches when one type of video card is compared to another, but benching the same card against itself is a great way to see if your improving.

Enjoy. If anyone has a similar setup to mine here, post your results.

Impressive. Not too bad for a 6800nu huh? I would push it further if itll go. The luck seems to be with you. Spend all of it

Easy for Marsumane to say when it's your money down the drain if it burns. I wouldn't go any higher than that.

Rollo, i know uve pushed things to their limits. Uve been around doing this for quite some time so i'm quite sure of it. I'm sure u also know that if he does it right, he wont mess up the card. Aka: VERY small increments and heavy testing each notch. I'd actually be more abpt to do it just because i got such a good ocer and would like to see how high it would go. Dangerous territory gives some of us a rush when we get above 95% of all other similar cards and are still stable, and still going. I'm sure u can atleast understand that, if not relate.

Oh, and im not saying u should oc it if u dont want to, just that I personally would because i know i wouldnt burn it.

 

[effee]

What about 3dmark05?

 

[Keysplayr]

Originally posted by: effee
What about 3dmark05?

3DMark05 Is probably more geared toward the next generation of cards. When a 6800GT and a X800 pro falls below 5fps in a bench, the difference in 1 fps can result in a 20% advantage or disadvantage. This is just my opinion. I will run 05 if you request it. But Id rather not mess with my card too much more.

 

[Pete]

The relative gains in overall score will probably be the same, tho the extra VS may prove more useful, as 05 is supposed to be significantly more vertex-limited (altho its pixel shading is also supposed to favor ATi's texture-friendly architecture, whereas 03 favored straight computation--this may just be for the discrete PS test, tho).