On 17 Oct 2003 at 19:57, Ryan K. Brooks wrote: > Can you explain this more? I don't see how large bokeh features > interact with the bayer pattern.
If the bokeh of the lens/subject/lighting combination produces bright edges (similar to the effect of unsharp masking on a high contrast transition) then it's likely that the demosaicing matrix will produce quite bad colour distortions as can be seen in the rings behind the birds leg on imgp0037.jpg. > On these pictures, I see lots of big (in the pixel sense) ugly sticks > and stuff, and at the reduced scale I don't see how the bayer v. foveon > could be any different. Foveon will more faithfully reproduce pixel to pixel colour transitions as the colour information for each pixel represents a direct measurement not interpolated data. Rob Studdert HURSTVILLE AUSTRALIA Tel +61-2-9554-4110 UTC(GMT) +10 Hours [EMAIL PROTECTED] http://members.ozemail.com.au/~distudio/publications/ Pentax user since 1986, PDMLer since 1998
