The diffraction limits of optical lenses are independent of focal length and absolute aperture. Diffraction limits are inversely proportional to f-ratio (f-stop) and the approx formula is 1500/f-ratio in line pairs /mm. This means at f6.3 lens would have to first resolve ~240 lp/mm to be considered diffraction limited and that would be quite an impressive lens indeed. Most lenses dont reach true diffraction limited resolution performance until f11 or smaller. e.g. they are WORSE than the diffraction limit as f11 or wider. I dont think you are going to find any 14mm lenses for 35mm FF or even APS that are truly reaching diffraction limited performance across entire format ( as good as phyics will allow ) at an f-stop as wide as F6.3. Aint going to happen on such a wide angle lens, even on APS where its not as wide as 35mm. jco
-----Original Message----- From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] On Behalf Of Godfrey DiGiorgi Sent: Thursday, December 14, 2006 11:14 PM To: Pentax-Discuss Mail List Subject: Re: K10D diffraction limiting Well, the diffraction limit for a 14mm lens, without considering sensor pixel pitch at all, is around f/6.3 or so. I always consider lens diffraction limits to be when the iris gets down to around 2mm in size, whatever f/number that works out to be for a given focal length. Some lenses are more sensitive than others due to placement of the iris in the light path. I haven't seen the formulae that factor in pixel pitch and aliasing. Godfrey On Dec 14, 2006, at 6:14 PM, Digital Image Studio wrote: > Does anyone want to have a stab at estimating the practical > diffraction limit for the K10D? My *ist D resolution seemed drop off > from about f16, so I assume for a Bayer filtered sensor the practical > limit would be something like twice the airy disc diameter plus a bit > to account for the AA filter? > > The *ist D sensor has 7.8µm square photosites so the aperture to > produce an airy disc of twice that is f11 plus one stop fudge. Given > that the K10D sensor has 6.05µm square photosites applying the similar > factors would yield a drop in resolution after about f13. > > I haven't done any serious comparisons as yet but I'll be keen to see > if the theory fits the practical. > > If you think it's all a crock then you are quite entitled to your > opinion :-) -- PDML Pentax-Discuss Mail List [email protected] http://pdml.net/mailman/listinfo/pdml_pdml.net -- PDML Pentax-Discuss Mail List [email protected] http://pdml.net/mailman/listinfo/pdml_pdml.net
