6800 Ultra vs 9700 Pro - Heaps Of Benchmarks Inside!

Page 2 - Seeking answers? Join the AnandTech community: where nearly half-a-million members share solutions and discuss the latest tech.

BFG10K

Lifer
Aug 14, 2000
22,709
3,004
126
Looks like you took the (H) approach BFG.
Honestly I'd like to see apples-to-apples numbers as well (e.g., run everything at, say, 1280x1024x4aax8af, or some other static setting),
I thought about how to present the results for a while and (H)'s approach does have some merit provided it's done properly. If I just used the old settings on my 9700 Pro the bulk of those benchmarks would've just turned into a CPU test. Instead I chose to show the old settings I played the games at and the new settings I can play them at as a result of the upgrade.

would you say the filtering overall is at least better then the R9700Pro?
Roughly: with all optmizations on they're about the same, with optimizations off the 6800U has the edge.

Good work BFG. I'm not sure I needed to see Descent and Quake, but very comprehensive and well done.
Personally I think it's important to include old games if for no other reason than for interest. Even Quake II can slap around the 6800U at 1920x1440x8.

woohoo! crusher is back
I usually do GLQuake too but at the moment I'm looking for a replacement as the current version can't handle if the system is reporting too many resolutions. I might just change the GLQuake source code and recompile a new build that handles resolutions properly, but I'll see how I go. I definitely plan on putting the game back into benchmarking action.
 

Jeff7181

Lifer
Aug 21, 2002
18,368
11
81
Originally posted by: DonVito
I appreciate this, particularly since I have a 9700 Pro, and plan on getting a 6800GT and OCing it.

Honestly I'd like to see apples-to-apples numbers as well (e.g., run everything at, say, 1280x1024x4aax8af, or some other static setting), to get a clearer sense of the overall difference in performance. Still, great job!

I agree... not that I don't appreciate it, but what good does it do to know how a card performs at 1920x1440? I think it's safe to say less than 1% of people have a monitor capable of that resolution, let alone actually play a game at that resolution.
 

rbV5

Lifer
Dec 10, 2000
12,632
0
0
Excellent job! I have a good share of those games and considering stepping up from my 6800standard. Thanks for the higher resolution numbers especially.

Do you have a games list that kept you from running Cat 4.7 for all the tests?
 

apoppin

Lifer
Mar 9, 2000
34,890
1
0
alienbabeltech.com
good job!

the 2-year-old 9700p is still holding on as a practical "poor man's card" for a crt at 10x7 . . . if you are running a hi-res LCD, the 6800 is a "must" . . . ;)

. . . i see that i can 'hang on' till r500/nv50 :)

you DON'T wanna see my old 8500 vs my new 9800xt benchs :p
(the 8500 'runs' doom . . . even at 8x6 - but it ain't "pretty") ;)

:roll:
 

BFG10K

Lifer
Aug 14, 2000
22,709
3,004
126
Just out of pure curiosity, no need to start flame war or anything but why did you disable shadows???
In most cases disabling shadows only disables the dynamic shadows that appear under monsters' bodies but the rest of the world shadowing effects are untouched. Of course when a monster comes along where are you looking? At the head/torso because you're trying to blast it to pieces.

In my opinion dynamic shadows (most commonly manifested as character shadows) are a needless waste of resources.

A lot of games, one in particular, look horrible (not as good i mean) without shadows.
If you're referring to Doom III then again, you don't miss out on much. Apart from monster shadows certain flashlight effects are reduced a tad but the bulk of the shadowing still works as normal.

I agree... not that I don't appreciate it, but what good does it do to know how a card performs at 1920x1440?
I understand where you're coming from but again, if I kept the old settings the results would have basically started turning into a CPU test. I needed to use settings that strain the 6800U to show what it can really do yet at the same time make sure the settings are realistically fast enough for someone to play the games at.

Do you have a games list that kept you from running Cat 4.7 for all the tests?
  • Descent III doesn't work at all under OGL. It's also never worked with AA under any driver version.
  • COD hardlocks since 3.10 though I can stay up long enough to finish the demo.
  • While the Far Cry demo ran, the entire coastline had missing waves and instead had jagged saw blades.
  • The control panel fails to force AF in Unreal though I can work around this by editing the game's ini.
 

rbV5

Lifer
Dec 10, 2000
12,632
0
0
quote:

--------------------------------------------------------------------------------
Do you have a games list that kept you from running Cat 4.7 for all the tests?
--------------------------------------------------------------------------------



Descent III doesn't work at all under OGL. It's also never worked with AA under any driver version.

COD hardlocks since 3.10 though I can stay up long enough to finish the demo.

While the Far Cry demo ran, the entire coastline had missing waves and instead had jagged saw blades.

The control panel fails to force AF in Unreal though I can work around this by editing the game's ini.

I can't even finish the first level of the demo without hardlocking COD, seems to be getting worse with each driver. My 8500 has no such issue, 9700+COD is a definate problem. Thanks for the others.
 

Genx87

Lifer
Apr 8, 2002
41,091
513
126
I need to get more impartiality like bfg10K :)

Maybe when the NV50 and R500 go toe to toe.