3dmark05 around the corner. what settings should be default?

[Pandora's Box]

i was wondering what settings everyone should use as a standard in 3dmark05 in the nvidia control panel?

im thinking:

0x AA
0x AF
Mipmap forced to bilinear
ALL optimisations ON
vsync off :bash:

[therapture]

That's why those synthetic scores are so...worthless, except to test changes. I play my games wth slider to QUALITY, TRILINEAR, and quite often AA and AF. That's the way a card should be tested... :ORDER: :)

[aAv7]

def. trilinear and the slider set to quality. I basically test it under the default settings my driver is at when it's installed.

[saturnotaku]

Raise your hand if you couldn't give two ***** of a rat's behind less about the next iteration of e-pen0sMark.

[Distraught]

Jesus, why must you people always crap on other people's threads. He ask a valid question. If you don't like 3dmark create another thread and b!tch about it there.

As for your question, I not sure about the setting but I'd bet the later drivers (65.xx and up) will perform better than the current 61.77s.

[Subtestube]

Edit: Whoops... stupid mistake delete this post!. Appy-polly-logies

[ChrisRay]

Come on saturn ;) lets play nice. Personally i dont care either, The benchmark may be useful. However I find it incredibly unlikely.

[sgio2]

The only reason I care is because it uses PS/VS3 , and I want to see it in all of its glory on my 6800 :D

[Jedi Rainman]

The only reason I care is because it uses PS/VS3 , and I want to see it in all of its glory on my 6800 :D


I couldn't agree more and it does give you bragging rights to say hey, can your video card do that? I didn't think so...go get a new one

[Blacklash]

My opinion is to bench default and how you play.

I really do not see a point in turning mip map to lowest detail, defeating AF and using 'performance' driver settings just to inflate a score.

I plan on using it with my driver on high quality, and AA/AF set to application controlled, and I may bench by forcing various levels of AA/AF as well.

Benching with MS paint quality settings on your system doesn't prove much. As long as people make it known they are setting things to maximize thier score it's fine with me. It's fairly obvious we have had some people posting up scores with nulled shaders. I did the same thing myself to mock them by posting a 17k 3dmark03 score. I think it's disingenuous to present a score as 'default' if you over ride what the program wishes to do when it runs, aka defeating AF etc.

[Shamrock]

The only reason I care is because it uses PS/VS3 , and I want to see it in all of its glory on my 6800 :D

as far as I know, the only way you will get the PS/VS3.0 support, is if you have the "pro" version...in other words...the PAID subscription...

[OWA]

Also, I don't think it'll look any different. In one of the other threads, the FM guy said things will look the same. The way they're using it, SM 3.0 should just make it a little faster.

[fivefeet8]

Also, I don't think it'll look any different. In one of the other threads, the FM guy said things will look the same. The way they're using it, SM 3.0 should just make it a little faster.

Yeah, which is a little dissappointing. They could have atleast used some HDR FP16 floating point blending. ;)

[dante_uk]

Raise your hand if you couldn't give two ***** of a rat's behind less about the next iteration of e-pen0sMark.

Consider my hand raised.

e-pen0sMark is a tool to tell me if something I've changed has improved or screwed my system. Even then it only checks DirectX so what's the f**king point? most ( all at the moment ) of the games I'm playing are OpenGL!!

Still we all have to see those pretty pretty graphics, just give me a video to download so I can admire all that eyecandy.

[oldsk00l]

Consider my hand raised.

e-pen0sMark is a tool to tell me if something I've changed has improved or screwed my system. Even then it only checks DirectX so what's the f**king point? most ( all at the moment ) of the games I'm playing are OpenGL!!

Still we all have to see those pretty pretty graphics, just give me a video to download so I can admire all that eyecandy.

Oh yaeh?? Well my pen0s is biggar tahn yuours!!!11
3dmark05 gets teh chicks

[dante_uk]

Oh yaeh?? Well my pen0s is biggar tahn yuours!!!11
3dmark05 gets teh chicks

Maybe, maybe not (duel)
I'm married, so "gets the chicks" isn't a good thing or I might lose my non-e-pen0s (pb)

[oldsk00l]

Yeah, I'm in the same boat.

I wouldn't dispute the accuracy of a JeffK 3dmark05 review though.

I bet it'd hit the nail dead on.

[Nandro]

I use 3d03 extensivly when overclocking and tweaking. I know that even though coolbits will let me clock my 6800 ram over 1.16 at that point or slightly over it starts having negative effects. This is true of alot of tweaks, so why not test them out. Even testing the optimizations is worth it to see what kind of effects they have. Also to see frame by frame differences. As far as settings, most people when comparing in bragging situations will use high performance all opt on at default res.

[noko]

Since it is sm2 and above based why not have AA at 2 or 4 and AF to 8x as default? The images do look very impressive, considering it is rendering in real time, maybe slide show rates but still impressive. My plastic is waiting for release and then I can download and have some more fun :D.