• We’re currently investigating an issue related to the forum theme and styling that is impacting page layout and visual formatting. The problem has been identified, and we are actively working on a resolution. There is no impact to user data or functionality, this is strictly a front-end display issue. We’ll post an update once the fix has been deployed. Thanks for your patience while we get this sorted.

9600 pro: doom worthy?

depends on your settings, and the fact that its very un optimized now, it will run much much faster on all cards when its done, In fact, when its done, ill be running at a medium settings at 800X600 on my 9000 pro at over 30 fps. Thats an estimate, but your 9600 pro is about the same in performance as a 9500 in some things, but the 9600 pro will excell in this game over the 9500 pro quite well, for this game useds gpu power more then any other part of the card, and your gpu is much faster then a 9500 pro, no need to overclock, I will guess, that with a decent system, you could run doom 3 at 1024X768 highest settings, at maybe 40 fps, but dont us anti aliasing, or AF.
 
I think there is still quite some way to go before the figures that Doom3 produces are worth looking at especially when you consider this statement:

"The Nvidia card will be fastest with "r_renderer nv30", while the ATI will be a tiny bit faster in the "r_renderer R200" mode instead of the "r_renderer ARB2" mode that it defaults to (which gives some minor quality improvements)"

Remembering that the R200 is the Radeon 8500/9100 chip, it wouldn't surprise me if that meant that the demo was not taking full advantage of the DX9 features available to the R9500 or higher cards, whilst it clearly shows that the demo is already more optimised for the GeforceFX and higher (nv30) cards.

The slightly worrying thing is that id still has a lot of speed to find in the game if it is to live up to the inspirational statement made long ago about being playable on a Geforce2 card with a <1GHz processor.
 
Anand's latest NV35 review has a comparison between various graphics cards and their Doom3 performance. The 9600 Pro performed pretty well IIRC. Anyway, the game is far from complete and is obviously not optimized yet. The codepath seems to favor NV designs since they have specific codepaths for the NV2x and NV3x cards but none for the R3x0s, just the default ARB2 and R200 code paths.

Despite this, the 9600 Pro should still be "worthy", considering the targetted minimum specs are only a GF3 class card.
 
Originally posted by: vss1980
I think there is still quite some way to go before the figures that Doom3 produces are worth looking at especially when you consider this statement:

"The Nvidia card will be fastest with "r_renderer nv30", while the ATI will be a tiny bit faster in the "r_renderer R200" mode instead of the "r_renderer ARB2" mode that it defaults to (which gives some minor quality improvements)"

Remembering that the R200 is the Radeon 8500/9100 chip, it wouldn't surprise me if that meant that the demo was not taking full advantage of the DX9 features available to the R9500 or higher cards, whilst it clearly shows that the demo is already more optimised for the GeforceFX and higher (nv30) cards.

The slightly worrying thing is that id still has a lot of speed to find in the game if it is to live up to the inspirational statement made long ago about being playable on a Geforce2 card with a <1GHz processor.
Click to expand...

There's no r300 path because ATI has remained close enough to ARB2 not to warrant it. JC did nV30 path because nVidia has many specific extensions to be used beyond ARB2.

So I think Radeon 9500,9600,9700,9800 will all use R200.

As far as playable on Geforce2 with 1GHz... "playable" is a vauge term 🙂 It probably means 20fps at 640x480 or something 😉
 
Theoretically speaking, it is possible that the ARB2 path should be faster than the R200 path for the newer DX9 Radeon cards as it should be able to exploit the higher level of programmability that the R8500 and 9000 chips cannot support. But its not really important at the moment as the game is still months away.

"As far as playable on Geforce2 with 1GHz... "playable" is a vauge term 🙂 It probably means 20fps at 640x480 or something 😉"

Yeah, I am fully expecting the same thing for that kind of level of speed...... except that I doubt I'm the only one who would not class that as 'playable'.
 
Remembering that the R200 is the Radeon 8500/9100 chip, it wouldn't surprise me if that meant that the demo was not taking full advantage of the DX9 features available to the R9500 or higher cards, whilst it clearly shows that the demo is already more optimised for the GeforceFX and higher (nv30) cards.


What DX features will Doom3 use considering it is an openGL based game?!?!?!?!?!?!?!?!?!?!?!?

 
I dont explicitly mean DX features, however what new avenues are opened up by new DX features spills over into OpenGL...... for example every time a new Geforce or Radeon comes out you hear about its new DX features and what they do graphicallly, however these same things are usually available under OpenGL also.
Example, Doom3 will make use of vertex/pixel shading which is a DX8+ feature but is also available in OpenGL by using the various new extensions that support it.

Note the fact that I said 'DX9 features', just cos its part of DirectX doesn't mean it cant be done in OpenGL. Unfortunately it seems to have become a standard to quote things in terms of DX spec... such as original T&L is DX7 although OpenGL had support for it before.
 
I don't think we can judge ATi performance in Doom 3 based on the latest nVidia-sponsored benchmarks. It seems to me ATi wasn't aware and/or ready for the tests.

I also don't think ATi's DX9 cards will gain/lose much speed moving from the ARB to R200 rendering path, as their cards seem to perform FP as fast as INT--whereas nV's FX cards require their own path, as they're a "twitchier" architecture, in Carmack's words. nV cards seem to perform much faster in INT than in FP--at least, this was the case with the NV30--so Carmack was forced to code a separate path to extract full performance out of it.
 
You must log in or register to reply here.

TRENDING THREADS

Back
Top