Thanks for your help Juergen. I'll try tomorow to acclimatize my munki and to use the black drift compensation and the ambient light sensor. Until now i used to set only main features of dispcal : white temp, white level and gamma. About your link, what am I suppose to see on the jpg? Currently I see number from 1 to 8 with the 1 and 2 very sligthly visible. Best regards, Emmanuel 2010/11/21 Juergen Lilien <jp.lilien@xxxxxxxxxxxxxx> > Hello Emmanuel, > > Emmanuel Michonneau wrote: > > > Why response curves of colormunki does not start to zero? > > Maybe the black point of your monitor has a slight color cast, or there > is a problem with temperature related black point drift of the Colormunki. > > It is a good practice to give the Colormunki some time at the screen > (even an hour is not too long, if you are very demanding/critical) to > acclimatize to the screen temperature, before starting the measurements. > You can also try to enable the black level "drift compensation during > measurements". > > This brings me to the point, that you should tell us about the > calibration and profiling settings you use in dispcalGUI. > > Have you tried the dispcalGUI "Verify profile" function (tools menu) to > check the calibration/profile? > > > To illustrate my problem here are 3 screenshot of the same photo with > > respectively : > > - no profil : http://emmanuel.michonneau.free.fr/pub/withoutprofil.jpg > > - Spyder profil : > http://emmanuel.michonneau.free.fr/pub/spyder_argyll.jpg > > - Munki profil : > > http://emmanuel.michonneau.free.fr/pub/colormunki_argyll.jpg > > There are still deep blacks (0,0,0) in the Colormunki sample, so it > seems to be a different gamma correction applied, but the picture does > not look necessarily bad to me. > > Maybe it is better to compare a more artificial picture, have a look at: > > http://www.fotocommunity-prints.de/fileadmin/daten/einser.jpg > > > Does the problem comes from my display and its TN panel? > > The 8 Bit (per channel) color resolution is only archived with a FRC (in > the time domain) dithering stage, so it should be better to enable the > adaptive mode to use the longest possible integration time. > > By the way, have you disabled the ambient light sensor of the monitor? > > Best regards, Juergen > >