On 25 Jan 2004 at 22:58, Butch Black wrote: > This begs another question. What then would be the ideal pixel size and > density to get the least noise, widest exposure latitude, with enough > resolution to keep up with the best (or at least the very good) lenses? > Would this number change as the sensor got larger as larger format lenses > resolve slightly less. Would this be offset by the need to "enlarge" less > for any given print size?
I'd put my money on 6-7�m being around the optimum pixel size. Any smaller and the losses due to focus errors and lens anomalies would swamp any advantage excluding the underlying noise limits of the current technologies. Any larger and the best lenses are rendered average in performance. The *ist D sensor (Sony ICX413AQ) features 7.80�m square pixel cell size, if this cell size was scaled to 24x36mm the new sensor would provide just over 14MP (like the Kodak 14N). The major advantages in making the sensors less dense would be to provide wider exposure range, lower noise and higher reliability, this the part of the rational behind the designs of the Mars Rovers digital cameras. http://gsfctechnology.gsfc.nasa.gov/laisdocs/spie_paper.pdf Cheers, 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
