2900xt owners\benchmarks thread

Tom|Nbk · 29 Jun 2007 at 19:16

Tom|Nbk said:
Another benchy to add to the mix,

Doom 3 time demo

1680x1050

8xAA

16xHQAF

Ultra details everything maxed out.

65.9 FPS.

8.38 RC3 Drivers.

Just thought I'd bump this with an updated Doom 3 bench with the latest Catalyst 7.6's

Same settings as before :

Anyone else want to update there benchies so we can see what improvements have been made?

TaKeN · 29 Jun 2007 at 19:24

So are these cards beating 8800 GTX's yet ?

Tom|Nbk · 29 Jun 2007 at 19:26

TaKeN said:
So are these cards beating 8800 GTX's yet ?

Hard to say really some cases they come really close though.

Tom|Nbk · 9 Jul 2007 at 21:13

Battlefield 2142 2900 Perf

1680x1050
8xAA
16x HQAF

Min Max Avg
27 109 55.955

Benchmark recorded performance for 10 minutes while I played in a 64 man server on Camp Gibralter, all in game options maxed fully. Any GTS owners want to compare?, pretty sure the GTS wins in this game though as it is a nvidia optimised title.

Oblivion 2900 Perf

1680x1050
8xAA
16x HQ AF

10 Minute fraps benchmark, and I went for a walk around the path near Skingrad including walking through forrested areas, bear in mind I also have the 200% vegetation mod on, It's as smooth as butter faster than my GTS ever was. Here are the results from a 10 minute bench from Fraps.

Walking round Chorrol (SP?) during the day (Bring sunlight)

Min Max Avg
36 78 52.461

Walking round the forest surrounding Skingrad (Sunset)

Min Max Avg
31 90 46.914

As you can imagine with these averages its just incredible to see it so smooth at such a high detail.

Tom|Nbk · 9 Jul 2007 at 21:38

GRAW 2 Single Player Demo Perf

1680x1050

Edge - Smoothing

Everything Ultra/High

16x HQAF

Min Max Avg
27 324 42.979

Ignore the unusually high max I paused the game for a split second and it recorded the menus FPS

Shocky-FM · 9 Jul 2007 at 21:42

Tom|Nbk said:
GRAW 2 Single Player Demo Perf

1680x1050

Edge - Smoothing

Everything Ultra/High

16x HQAF

Min Max Avg
27 324 42.979

Ignore the unusually high max I paused the game for a split second and it recorded the menus FPS

Then we should also ignore the Min and Avg as they would take inacount the unusually high max.

Tom|Nbk · 9 Jul 2007 at 21:45

Shocky-FM said:
Then we should also ignore the Min and Avg as they would take inacount the unusually high max.

Doubtful It would change the avg at all it was only on for a split second but I will test again tommorow just for you

Shocky-FM · 9 Jul 2007 at 21:49

Tom|Nbk said:
Doubtful It would change the avg at all it was only on for a split second but I will test again tommorow just for you

Thats true but max of 300+ is going to have a nice impact on your average regardless of how long it was there.

Tom|Nbk · 9 Jul 2007 at 21:54

Shocky-FM said:
Thats true but max of 300+ is going to have a nice impact on your average regardless of how long it was there.

Possibly a frame or so, I'll do it again tommorow for you.

Tom|Nbk · 10 Jul 2007 at 16:15

BF2 64 Man Server Gulf Of Oman Perf

1680x1050
8xAA
Everything Ultra/High
16x HQAF

Min Max Avg
43 100 80.336

Re-doing my GRAW2 Test in a bit.

Tom|Nbk · 10 Jul 2007 at 16:39

Shocky-FM said:
Thats true but max of 300+ is going to have a nice impact on your average regardless of how long it was there.

Re done the bench for you as I promised

GRAW 2 Single Player Demo Perf

1680x1050
Edge - Smoothing
Everything Ultra/High
16x HQAF

Min Max Avg
27 75 42.033

FreeStream · 10 Jul 2007 at 16:46

Tom|Nbk said:
BF2 64 Man Server Gulf Of Oman Perf

1680x1050
8xAA
Everything Ultra/High
16x HQAF

Min Max Avg
43 100 80.336

Re-doing my GRAW2 Test in a bit.

Hmm, to compare my X1900XTX @ 709/845 @ max @ 1600x1200 @ 6xAA, 16xAF did 80-100...

Tom|Nbk · 10 Jul 2007 at 16:47

Concorde Rules said:
Hmm, to compare my X1900XTX @ 709/845 @ max @ 1600x1200 @ 6xAA, 16xAF did 80-100...

2142 / BF2 definetly goes to nvidia and most likely always will, both games are mega shader intensive and because the 2900 resolves the AA in the shaders obviously theres a hit more noticeable specially in the BF series.

Tom|Nbk · 10 Jul 2007 at 17:50

R6V second level perf

This one is for Mav

1680x1050
0xAA

It isn't supported at least not for ATi card's, Looks amazing without.
Everything Ultra/High
16x HQAF

10 Minute gameplay.

Min Max Avg
32 107 56.76

Anyone got any feedback or wants to compare their results. A little would be nice

Vogon · 13 Jul 2007 at 01:51

Does 3DM06 recognise the card now, or will I need to add a command line to get it working, and if so what is it ?

Sone · 14 Jul 2007 at 22:13

i had to delete the file, adding command switch not to scan run tests but wouldent give me a score :confused:

Sumanji · 15 Jul 2007 at 01:40

Hey,

Sorry I haven't read the entire thread, but have subsequent driver revisions done anything to boost performance in games? And is there still hope for any additional increases in future drivers?

Cheers,

Suman

Vogon · 15 Jul 2007 at 10:48

From what I recall on earlier tests, STALKER has been improved at least.
I don't fully understand the AA settings meaning, but can tell you wide tent looks very good AA, 8x with 16x wide tent is very nice.
Any amount of AA still has a large performance hit (in 1920 x 1200 at least), but the actual performance hit between 2x (box) (the lowest) and 8xwith 16xwide tent (the highest) isn't as big as you may imagine.

Some figures from my testing HD2900XT 1024mb @ Stock, E6600 @ 3.5 ghz:

IL2 Sturmovik 1946:
Res 1920 x 1200
AA setting 4xaa 8xwide tent (control panel)
AF setting App controlled (control panel)
In game setting maxed, perfect settings with water mode 2.
Avg: 53.629 - Min: 19 - Max: 118

FEAR
Res 1920 x 1200
AA setting 4xaa (in game)
AF setting 16x (in game)
In game setting maxed (except soft shadows)
Avg: 53 - Min: 21 - Max: 119

STALKER
Res 1920 x 1200
AA setting mid way (in game slider)
AF setting max (in game slder)
In game setting maxed
Avg: 61 - Min: 26 - Max: 247

3DMark06
Default settings for app and control panel, IE no LOD reductions to score more

11315 (I think 12k+ stock shoulkd be possible, but there is a major slowdown for a second or so on the flying boat test, when the dragon comes out of the water, I presume a driver update will cure that soon).

3DMark05
Default settings for app and control panel, as above.
19894

X3 Reunion Demo/Bench
Res 1920 x 1200 4xAA 16xAF in demo
Settings maxed
Average 58 fps

I did need to add "-nosysteminfo" to the command line to get 05 and 06 to run.

Other opinions so far, the drivers still need some work IMO, but past experience shows normally ATi do well sorting drivers.
I am dissapointed with IL2 Sturmovik performance, I would hope that can be improved in newer drivers, it is an older title, but is pretty demanding still with all the eye candy, but those figures are not much above the X1900Pro.
Other games obviously are much faster than the X1900.
Overall, in comparison to my 8800GTX OC2 it works out around 20-25% slower, as it cost 25% less thats fair enough.

Sumanji · 15 Jul 2007 at 11:55

Those numbers still look a bit poo... for example in STALKER, what's the use of getting 247fps in one bit, if it's then going to drop to 26?

TheRealDeal · 15 Jul 2007 at 12:01

That 26 could have been 1 frame.