Anisotropic Options with 9700

BoddoZerg said:
hmm. Why is 4x so much less curved than 2x or no aniso? Even looking at the furthest-away mip levels (even on 2x aniso!), they are a lot more curvy than 4x... ?!?

Funny things tend to happen when calculating LOD. I wouldn't be too concerned about that. Although it isn't clear the last anisotropic shot is curved at all.

No, the Radeon 8500's boundaries, to be perfectly accurate, always consisted of two straight lines intersecting at the center, basically a rough approximation of the semicircle that should actually be calculated.

TNTx cards did it like that, but I think Radeon 7500/8500 cards are more depth based (as opposed to the range-like 9700 or any Geforce). Most of the time the boundaries on a Radeon 7500/8500 look straight horizontally. TNTx cards had the two colliding boundaries as you describe.
FYI, I know at least the first gen Geforce cards weren't using perfectly curved surfaces, they just used the junction of something like 5 lines. I haven't tested my GF3, yet.
 
Back
Top