Image Engineering has a device for characterizing cameras - slide projector with filter caroussel- and you can improvise one. My belief however is that it's useless for old gen colorimeters because quality control was so bad, and you need accuracy for RGB LED device measure. On Fri, Jun 24, 2011 at 2:59 AM, Graeme Gill <graeme@xxxxxxxxxxxxx> wrote: > Rishi Sanyal wrote: > >> this before for individual colorimeters. Actually, why can't they do >> it now for existing colorimeters? Say you send your colorimter in& >> they measure the transmission spectrum of each filter? I guess you'd >> need to then put that in the firmware, etc. which might get >> complicated? >> > > A thought that's crossed my mind (even before I heard about the > X-Rite colorimeters) is that it would be possible to generically > characterise various colorimeter models, and therefore be able to > create correction matrices for all colorimeters from a single > display measurement (ie. it turns the current ccmx from > an N x M combinatorial problem to an N + M measurement problem). > Characterising spectral sensitivity of input devices takes a > good deal of equipment though. > > Graeme Gill. > >