OpenGL Fur Benchmark

[cdawall]

the awesome power of an 8400GS

not a bad increase over stock though

 

[Cold Storm]

Cdawall, you just want to have the LOWEST score in here! lol.. Sweetness!

 

[cdawall]

823/1044 on the cards not bad for stock ATi single slot coolers

 

[CrackerJack]

hey cdawall, have you tried lower your cpu volts. The lowest I can do at 3.2Ghz is 1.375 stable. It's little less heat.

 

[cdawall]

yea best i could get was 1.41v

 

[Spirou]

Slightly undervolted due to my PSU:

Spirou
3866
Sapphire Atlantis HD 3870 Silent OC
noiseless @ 43° C on air, no voltmod



Had to bench at 945/1350 because my 12V line drops to 11.3V when benching at higher clocks :-/

Surface Deformer: 16206, Soft Shadows nonbranching: 4192, Soft Shadows branching: 8197



Stop scratching your heads, all ye victims of mass psychology and marketing bollocks. As a matter of course my gaming performance is just better than that, surely 50% faster than 8800 GT, about 30% faster than 8800 GTX, and at least 20% faster than Ultra OC with no doubts. So stop daydreaming and get yourself better gaming cards

 

[lemonadesoda]

@cdawall,

somethings not right with your 2x3850's. I have a SINGLE AGP 3850 and I'm getting a better score. OpenGL FurMark is not CPU bound... so its 99% GPU. I can confirm this because I get results with 1% for the same card on a P4EE and on a Q6600.

One of your cards is bad, or the drivers arent working properly.

 

[CrackerJack]

@cdawall,

somethings not right with your 2x3850's. I have a SINGLE AGP 3850 and I'm getting a better score. OpenGL is not CPU bound... so its 99% GPU. I can confirm this because I get results with 1% for the same card on a P4EE and on a Q6600.

One of your cards is bad, or the drivers arent working properly.

i believe he's got the same problem as me. It's a driver issue.

 

[Wile E]

@cdawall,

somethings not right with your 2x3850's. I have a SINGLE AGP 3850 and I'm getting a better score. OpenGL FurMark is not CPU bound... so its 99% GPU. I can confirm this because I get results with 1% for the same card on a P4EE and on a Q6600.

One of your cards is bad, or the drivers arent working properly.

Crossfire isn't working on this bench on the ATI cards. Not sure if it's all of their lines, but the HD lines are for certain.

 

[largon]

Updated.

Stop scratching your heads, all ye victims of mass psychology and marketing bollocks. As a matter of course my gaming performance is just better than that, surely 50% faster than 8800 GT, about 30% faster than 8800 GTX, and at least 20% faster than Ultra OC with no doubts. So stop daydreaming and get yourself better gaming cards

Heh-heh. Yep, Furmark has little to do with realworld gaming performance.


Wait a minute.
Are you serious?


Nah, you can't be.


Anyways, you just have to vmod that card you know...

 

[lemonadesoda]

Crossfire isn't working on this bench on the ATI cards. Not sure if it's all of their lines, but the HD lines are for certain.

Yep, I just noticed

CyberDruid -------- 2× HD3870X2 ----------- 2963

scores much lower than my AGP 3850.

 

[JrRacinFan]

Here's what I don't understand at all one bit, here I will let the pictures do some talking . . .

Now I thought there isn't much of a difference between the 2 resolutions. But with 1440x900 being my native res, would that make all the difference in the world? Hmmmm . . .

 

[Spirou]

thx for updating

I can't do a voltmod cause my PSU is already on its knees with my P4D at 4000 MHz on its back.

Have to wait until july cause my budget ran dry with that IFX-14 and Scythe Kama PWM i just ordered to get that lazy CPU (still boxed) on par with my new Maximus Formula without beeing roasted instantly

 

[largon]

JrRacinFan,
1440x900 will shrink the donut ~10% in diameter, less stuff to render -> higher score.

 

[JrRacinFan]

Ahhh I see now. Thanks!

 

[Cold Storm]

So, Largon, when we give you the screen shot, does the res have to be at the 1280x1024? Or if we get a better screen shot using a different res. will you use that?

 

[DOM]

So, Largon, when we give you the screen shot, does the res have to be at the 1280x1024? Or if we get a better screen shot using a different res. will you use that?

I think its just stock settings like all other benchmarks

 

[Cold Storm]

Yeah, but the reason I ask is that if someone does it at the res that gives them the best, will there be the use of it or not? I would keep it at stock settings myself.

 

[DOM]

i dont think he will but I cant answer that ? im sure it must just be stock than ppl will run it at 640X480 lol

 

[largon]

I always check the screenshot and make sure the bench was run at 1280x1024.
If other res was used I don't add it in the chart and I note the poster about it.

 

[Cold Storm]

I always check the screenshot and make sure the bench was run at 1280x1024.
If other res was used I don't add it in the chart and I note the poster about it.

Thats the way to do it man. Thanks for the heads up on how you check it all.

 

[CrackerJack]

CrackerJack
X1950 Pro 512mb
1369

 

[JrRacinFan]

Got my new card . . .

 

[mlee49]

Daaaamn, those 3870's kick ass!!

 

[JrRacinFan]

Just think, I have not overclocked it either. It's still at Turbo clocks and sitting at 37C idle.

EDIT:Well I do take that back. The mem clock was up a little as you can see in the GPU-z screenie.