Hi Thorsten,
Stéphane is probably not interested in visually 'pleasing' results but also
to improve the performance of ppg and amaze. As mentioned particular amaze
expects it to be absent and uses the color gradiënts to get superresolution
(sub Nyquist) details. CA screws up these gradient estimates (like high iso
noise) and you can get funky demosaic artifacts. Both were discussed in in
his previous mails. Subpixel accuracy matters for that.
Lensmakers try to make it easily corrected. As some lenses have mustache
shape distortion no doubt some require more than linear compensation
particular wide angle and fish eyes.
>From his post I understand he also investigated different orders and their
respective fitting errors. Thpigh he did not state how he tested that. I
assume an adaption of fitting half the points and calculating errors in the
other half of points not in the fit. A good approach for grids as errors
halfway fit points are generally worse than errors near fit points.
I also think my argument to allow non-radial factors for special lenses is
still valid.
Best,
Gert
------------------------------------------------------------------------------
Learn Graph Databases - Download FREE O'Reilly Book
"Graph Databases" is the definitive new guide to graph databases and their
applications. Written by three acclaimed leaders in the field,
this first edition is now available. Download your free book today!
http://p.sf.net/sfu/NeoTech
_______________________________________________
darktable-devel mailing list
darktable-devel@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/darktable-devel
