SWM writes:
DxO aims to provide similar benefits for the rest of the pack,
specific
Point Spread Functions for each lens and aperture.
No email access at work last night except my phone--fortunately that is
largely corrected.
It very much appears Canyon takes this approach for DLO including
inverse functions for the AA filter and to mitigate effects of
diffraction. Some aberrations are focus distance dependent which may be
one reason why Canyon re-added focus distance to exif after the 5Dc.
Curiously it was in the 30D then gone for awhile.
Early on it was very likely DXO used at least in part R-L deconvolution
applied in a gradient fashion (assuming a Gaussian PSF that does indeed
work quite well) where required by the pattern of lens aberration. I
suspect they do the same now given how fast they crank out the lens
modules, but they keep thing closer to there chest now and will not
reveal this type of info anymore. Perhaps you saw something I missed?
They have stable full of optics Ph.D's so anything is possible, but the
former approach is much easier and faster. Canyon almost undoubtedly
have the lenses modeled prior to construction anyway. Canyon
reformulated the DLO package for the 24-105 but I was not real able to
improve on a couple DXO corrected images ported to PS followed my
minimal FM in selected spots and I still don't like the version of DPP
I have. The newest one for the OM 5DIII may have a better interface.
Convoluted Mike
--
_________________________________________________________________
Options: http://lists.thomasclausen.net/mailman/listinfo/olympus
Archives: http://lists.thomasclausen.net/mailman/private/olympus/
Themed Olympus Photo Exhibition: http://www.tope.nl/
|