HL2 40% faster on X800 compared to NV40?

[Evildeus]

No, for posting only this . Sorry for OT guys

 

[Neeyik]

Perhaps you'll also from here

for posting pics comparing iq? i dunno 'bout you, but that affects MY purchases

Please read your PM.

 

[Xmas]

8 > 6
So what exactly makes them "comparable"?

Highest mode vs. highest mode would be obvious, if we want to get really technical then lets use ATIs 2xFSAA vs Nvidia's 4X FSAA on a FX. As ATIs 2XAA approaches Nvidias 4XAA.

"Consumers don't have a freakin clue what Multisampling vs. Supersampling is, all they care about is higher number of AA samples must mean better AA."

That was your argument. 8 > 6, so the 8xAA of NV40 is "better" than 6xAA of R420, and may damn well be slower...

And no, highest mode vs. highest mode is not a sensible approach IMO. What if NV still supported 16xOGSS, like they did with Det3 drivers? Yes, readers should be informed of all aspects of a card, but not all aspects are comparable.

 

[Joe DeFuria]

And no, highest mode vs. highest mode is not a sensible approach IMO. What if NV still supported 16xOGSS, like they did with Det3 drivers? Yes, readers should be informed of all aspects of a card, but not all aspects are comparable.

I take the middle ground.

"Highest mode against the highest quality mode" for the sake of argument is not particularly useful. There needs to be some common basis for the comparison.

So, doing something like [H] has done with their past few reviews (Highest mode that in their estimation, gives playable performance), is perfectly legitimate. They are using the common basis for "playable frame rates" as the anchor. The simple fact that nVidia's 8X FSAA never emerges as one of the settings used for "best quality with playable framerates", whereas ATi's 6X does...is what is telling.