Post your 3dmark05 results

[Rage187]

Apopin, when you place your cursor over the image, eventually a little box appears in the right hand corener. Click that box and t will resize the image.

 

[lordtyranus]

Originally posted by: Marsumane

Originally posted by: lordtyranus

The only thing i could think of, is that the clockspeed really matters on the xtpe and thus the 6800ue would be more close to it due to its slightly higher clock compared to the ultra. But what about the 6800GT?

Well, the x800 Pro and 6800 GT have nearly equal theoratical fillrates. The results seem to match Farcry performance rather closely.

I didnt think the FX series did THIS bad in farcry. I mean below the 9600xt w/ the 5950? U also must be refering to earlier benches in farcry because if i remember right nv caught up quite a bit on that game. Also the 6600 series killed the x700 series if i remember right on farcry. And I yea i wasnt sure on the pro vs gt fillrate because i dont know the pro's clockspeed off of the top of my head. I just assumed the GT's fillrate was a bit higher due to having 4 quads vs the pro's 3 quad setup multiplying the clockspeed.

This is the last roundup I could find including the 9600 XT and the 5950. They seem to be pretty much equal. Since 3dmark05 uses a lot more SM2 stuff, it's believeable that the 5950 could drop even more. Here is firingsquad's latest test of the x800/6800 cards; as you can see the x800's lead becomes quite large at 16x12.

The GT's 16*350 fillrate is actually 100 less than the x800 Pro's 12*475 fillrate. It looks to me like the 3dmark05 tests are mirroring the 6800/x800 clockspeeds. I don't know how SM3/2.0b affects the score, will test that soon when I finish my DL.

Weird that the URL died so fast, I captured it though so here is the project the guy posted text

7155

 

[DAPUNISHER]

Can't say if it's real or not brudda. He only ran the 3 tests it looks like since the rest were N/A and the scores were as follows

Game Tests

GT1 - Return To Proxycon 32.6 fps

GT2 - Firefly Forest 20.0 fps

GT3 - Canyon Flight 34.9 fps

 

[Rage187]

He also did not get any CPU marks.

 

[Marsumane]

Originally posted by: lordtyranus

Originally posted by: Marsumane

Originally posted by: lordtyranus

The only thing i could think of, is that the clockspeed really matters on the xtpe and thus the 6800ue would be more close to it due to its slightly higher clock compared to the ultra. But what about the 6800GT?

Well, the x800 Pro and 6800 GT have nearly equal theoratical fillrates. The results seem to match Farcry performance rather closely.

I didnt think the FX series did THIS bad in farcry. I mean below the 9600xt w/ the 5950? U also must be refering to earlier benches in farcry because if i remember right nv caught up quite a bit on that game. Also the 6600 series killed the x700 series if i remember right on farcry. And I yea i wasnt sure on the pro vs gt fillrate because i dont know the pro's clockspeed off of the top of my head. I just assumed the GT's fillrate was a bit higher due to having 4 quads vs the pro's 3 quad setup multiplying the clockspeed.

This is the last roundup I could find including the 9600 XT and the 5950. They seem to be pretty much equal. Here is firingsquad's latest test of the x800/6800 cards; as you can see the x800's lead becomes quite large at 16x12.

The GT's 16*350 fillrate is actually 100 less than the x800 Pro's 12*475 fillrate.

Weird that the URL died so fast, I captured it though so here is the project the guy posted text

7155

Alright, u are right on the pro's fillrate and the 9600xt seems to be around the 5950's marks on farcry. I stand corrected with your evidence proving to be more reliable then my memory

But on the other hand, you cant use aa+af numbers for the 6800/x800 series because in 3dmark its w/o aa+af. The 6800 is ahead in farcry by a notable, yet small amount in the benches you have provided w/ aa+af off. It does seem to be similar to farcry in the older archetectures tho.

 

[apoppin]

Originally posted by: DAPUNISHER
Can't say if it's real or not brudda. He only ran the 3 tests it looks like since the rest were N/A and the scores were as follows

Game Tests

GT1 - Return To Proxycon 32.6 fps

GT2 - Firefly Forest 20.0 fps

GT3 - Canyon Flight 34.9 fps

using Firefox (just recently) . . . and in a BIG hurry for work!
(no excuse, i know)

Thanks for the summary!
(and i found the magnifier)

:shocked:

i'm LATE!
(again . . .. i'm the boss, so what)

 

[kylebisme]

5604. Mighty pretty graphics as usual.

Originally posted by: Marsumane
the 9600xt has a better pixel shader then the 5950 but less fillrate, and bandwidth.

Actually the 9600xt has slightly more single texture fill rate than a 5950.

 

[DAPUNISHER]

Originally posted by: TheSnowman
5604. Mighty pretty graphics as usual.

Originally posted by: Marsumane
the 9600xt has a better pixel shader then the 5950 but less fillrate, and bandwidth.

Actually the 9600xt has slightly more single texture fill rate than a 5950.

Looking at your scores that guy's score with the PE would seem at a glance to fall in line for 7K.

 

[kylebisme]

I'd imagine the score is kosher. If you check the results database you will find a few people geting up there with the xt-pe's. Defenatly not a strong showing for SLI at this point.

 

[Megatomic]

2608

 

[obeseotron]

I get 5543 total, 24.5fps game1, 15.8fps game2, 28.2fps game3, 5309 CPU marks, 2.6fps cpu1, and 4.9fps cpu2.

My system is an athlon64 3500+ @ 10x250 (stock is 11x200), 1gb dual channel ddr500, and a gainward 6800gt @ 430/1.16 with the new 66.70 drivers.

3dmark05 compare url

If you look in the result browser its all overclocked x800 pe's at the top, a little over 7000. Seems odd that an sli 6800gt or ultra wouldn't do a lot more than that if I'm getting 5500....

 

[Marsumane]

Originally posted by: DAPUNISHER

Originally posted by: TheSnowman
5604. Mighty pretty graphics as usual.

Originally posted by: Marsumane
the 9600xt has a better pixel shader then the 5950 but less fillrate, and bandwidth.

Actually the 9600xt has slightly more single texture fill rate than a 5950.

Looking at your scores that guy's score with the PE would seem at a glance to fall in line for 7K.

Actually thats also true cuz of the 4x2 setup of the 5950. I keep thinking its an 8x1 until i remember the later findings of that blasted piece of hardware

 

[Marsumane]

Originally posted by: obeseotron
I get 5543 total, 24.5fps game1, 15.8fps game2, 28.2fps game3, 5309 CPU marks, 2.6fps cpu1, and 4.9fps cpu2.

My system is an athlon64 3500+ @ 10x250 (stock is 11x200), 1gb dual channel ddr500, and a gainward 6800gt @ 430/1.16 with the new 66.70 drivers.

3dmark05 compare url

If you look in the result browser its all overclocked x800 pe's at the top, a little over 7000. Seems odd that an sli 6800gt or ultra wouldn't do a lot more than that if I'm getting 5500....

I agree

 

[capodeloscapos]

I dont have it right now.
I have a question: Which is the difference between the Pro and the Trial Version?

 

[kylebisme]

Watch the demo part, wow.

 

[Marsumane]

Originally posted by: capodeloscapos
I dont have it right now.
I have a question: Which is the difference between the Pro and the Trial Version?

Trial you cant select as many options such as selective testing, looping, iq tests, demos, etc. Theres more things but i usually dont care enough about the other things to research them out. I think they are like website options like comparisons and what not. The pro version also costs money (obviously). For either one, u can generate the same score. It wont be any different if you just want to run the standard benchmark.

 

[obeseotron]

All of the synthetic (well, really synthetic) tests have been disabled, like polygon count and the individual shader tests. You can't do batch runs or basically any kind of configuration beyond pressing the run button.

Edit: yeah, what he said

 

[JBT]

Hmm I'll have to DL when I get home. THX.

 

[lordtyranus]

4594 stock.

http://service.futuremark.com/...=12&projectId=3271

 

[Marsumane]

2600 (9800p oced, 2255ghz clocked 2500+, gig of ram @ 410fsb, cat 4.8's)

 

[Tates]

2727

 

[acx]

Originally posted by: DAPUNISHER
Weird that the URL died so fast, I captured it though so here is the project the guy posted text

The picture says 3DMark03 Revision 0 Version 0 ?!

 

[Keysplayr]

Originally posted by: DAPUNISHER
What browser you using brudda? It will enlarge for clarity

Why does it still say "3DMark03 Revision 0 Build 0" ?....

 

[MercenaryForHire]

God damn it. My ISP is still blocking BT, even when I've set it up on the 20000+ ports. Feel that screaming 1.1KB/s ... asshats.

Anyone have a mirror that isn't 100% Fubar?

- M4H

 

[Mr. Lennon]

Man hardware cant keep up with software these days. Why did they release this benchmark so early? Maybe to make us feel bad that our 400 dollar cards run like a 9600 xt in 3dmark03? All I know is that my next set up will have to atleast score 14,000 in 3dmark05, which I am guessing wont happen for awhile. I don't purchase things based on a 3dmark score, but most of the time 3d mark is first to show us what will be seen in future titles.