Re: [Flightgear-devel] Final Diff for Anisotropic Filtering

[Curtis O. <cur...@gm...>]

On 5/20/07, Martin Spott wrote:
>
> While I'd certainly vote for inclusion of such an option, I'd like to
> note that the scenery does not always look _that_ crisp in real life,
> especially at distance  :-)
> Leaving the choice to the user is certainly the best decision,


It doesn't hurt to leave the choice to the end user, especially if there are
platforms out there where this causes performance problems (?)  But I think
it's fine to let this option apply globally to all textures.  In almost all
situations, we will not have anything close to real life resolution.  I'm
talking about arcseconds of field of view per pixel here.  This is what
impacts things like sign readability.  Also, Martin: it's clear that Olaf
was *way* zoomed in on this screen shot to emphasize the difference.
Anisotropic texture filtering is a win-win in all cases.  It avoids/reduces
a technical problem in mip-mapping which overly blurs textures that are
viewed nearly edge on ... such as runway polygons on landing and approach
and often terrain polygons.

My original objection was over a patch that only turned this on for some
textures.  I want it on for all textures.  Makes a big difference in visual
quality.

Curt.
-- 
Curtis Olson - University of Minnesota - FlightGear Project
http://baron.flightgear.org/~curt/  http://www.humanfirst.umn.edu/
http://www.flightgear.org
Unique text: 2f585eeea02e2c79d7b1d8c4963bae2d