DX 9 test

[Guest odog]

Hey, Interesting banter guys. I really would like to know how come dxdiag shows such an amazing increase in speed, but its not supposed to have any effect on my rig that is not 'configured' for DX9? Is this just coincedence? Being a simple minded guy, to me, it appears to be having some effect...what's up with that? Ive been playing with it all night and I see no problems as far as stability goes. Performance is still the same, but something changed to make textures a little crisper and I dont notice the bogging I usually see around Miegs...?

[Guest jase439]

While I am interested in seeing your results (please do post them), I don't think your observations will prove more or less conclusive than those made by others here unless you have a wide gamut of systems with which to test and some standard with which to measure against. This kind of benchmarking needs *many* data points - more than any one person here, I think, can provide.Someone tried to do this several months ago with FS performance with the intent of ending the "AGP vs No-AGP vs AGP aperture size" 'debacle'. There were no definitives to be drawn from that conclusion other than, "Try it for yourself, and see what happens" (insufficient/inconclusive data). Likely, the outcome here will be the same.FS performance is so varied from one system to the next - and the factors which affect it are numerous. So unless you're prepared to run numbers with anti-liasing, without anti-aliasing, with T&L, without T&L, with AF, without AF, different levels of AF, different CPU speeds, DX8 compliant card vs DX9 compliant card, etc...it will be difficult to conclude much of anything. Not sure it's a worthwhile investment your time other than to quell a sense of curiosity :).JPS. BTW, is this really a debacle for you? I thought it was just an interesting thread so I jumped in, tried it, and posted what I saw :^) *sigh* people get far too worked up over this stuff...I thought Paul here was going to get run through with hot pokers for disputing Dan's observations (which, like everything else said here, disputable). No disrespect here, but DX9 performance benchmarking is not a debacle...solving world hunger - THAT is a debacle.

[Guest jase439]

It's really hard to say. DX9 has a whole slew of new features for fast video streaming - all that is based in DirectDraw and DirectShow. Without really knowing WHAT was changed and what part of that code base is shared, it's hard to say if your observations can be attributed to this or not. But its possible (in my mind anyway - perhaps not others), you may be picking up some of the side-effects of those underlying improvements.JPS. I'm waiting for someone to jump in here and cry, "You're lying, your textures aren't crisper - that's impossible" :)