Aniso quality changes between NV driver relases

[Sharkfood]

You didn't post any evidence about what I asked for.

I posted links showing clear and present effect of the anisotropic filtering on the GF3 between 4x and 8x, along with clear, precise, detailed instructions on how to reproduce and prove this. lol.

Looks like your "obvious" and "just knowing where to look" are all strawman after all. Can't tell the difference, huh?

Back to square one as has been clearly illustrated through examples, by screenshots and with data.. versus a dissent with none of the above.

 

[Sharkfood]

That's the ticket. Spend a few days with these in Photoshop at 800% magnification... maybe you'll be able to find SOME pixel pattern that may designate 8x. hehe.

Looks like "obvious" isn't so obvious at all as such a trivial task for the "obvious" would have been answered same-day.

Oh well, I wonder if NVIDIA will fix this issue any time soon in the next revision or so of drivers. I'd *really* like to see the quality illustrated by the 12.XX->23.XX on the GF3 duplicated, even if this does mean a larger hit for AF at 8x.

 

[tb]

If NVIDIA did some tricks (filter only one texture layer or something like this) with the new drivers, than they could easiely deactivate this feature, if someone makes a screenshot. You would not see any difference on a screenshot. The other way around is also possible - imagine the driver bumps up the image quality on a screenshot.

How to detect such a trick?

Regards,
Thomas

 

[Bambers]

I'd say that A is the 8x picture in sets 1 and 10. (that was without zooming in but the differences are small). The rest look the same in either, esp the NFS5U shots.

 

[noko]

Reverend,

Looks like it maybe a GF4 thing more so then a GF3 change. Well in any case the performance and quality between 28.32 and 29.42 is virtually identical in all respects. Here is a graph doing a performance comparisons using anisotropic filtering between the 28.32 and 29.42 drivers on my Gf3 Ti200. Using my own configuration for all test which stayed the same, this is what I got. SSSE, The Elephant Atrium, modified Max Addon configuration (same in each test except for anisotropic filtering settings), 245/530 core/mem.

Here are some comparison images for anyone who is interested, good luck in finding anything significant.

28.32 8x anisotropic
29.42 8x anisotropic

28.32 8x anisotropic, colored mipmaps
29.42 8x anisotropic, colored mipmaps

 

[K.I.L.E.R]

That graph is not strait (or is it my eyes?) so the performance hit is not linear. Isn't that good to know?

 

[noko]

The minor variance could be my hard drive being access or what not. It does look kinda linear come to think about it, hmmmmm. I was hoping the newer drivers would have a bigger impact vice virtually zero for my rig. Well a full update for my setup should start in a few months .