Hi Hal, hi list, > In addition for > monitors this package does only calibration, has no interface for a > spectrometer or any way to input spectrometer data and does not > generate a monitor profile. And then in Windows I don't even know > what it does with the data that was generated during calibration.
It will create a profile in your profiles directory, named e.g M20040907.icm. This one does not contain monitor calibration data, therefore no need for AdobeGamma and friends. > Does it use AdobeGamma or does it install it's own utility to update > the LUT? How would I use this in Linux to update the LUT? see above. > I think using Windows based profilers are fine if you are profiling > scanners, cameras and printers for Linux as long as you do all of the > image IO in Linux using the same drivers and setting that you will be > using normally. I have done this for my printer and my scanner using > Profile Prism with good results. You can profile everything with Win32 for any OS. But only if the drivers do not come with extra stuff for "better color reproduction by automatism". So if your Win-driver does send raw-colors to your printer just like Linux, they are valid. What you could do is (see below for more comments on this): Scan/Print the test targets using Linux and use ProfileMaker (comments below!) to analyze the results (tiff or printout) on Win32/Mac. > I just generated a monitor profile using course profile option and all > default values with lprof and when I used xcalib to try to set the LUT > using that profile my display went all blue. So there is clearly a > problem with the vcgt-tag being generated by lprof. Anyone else tried > this? Is it just me or are others seeing this? lprof does not create a vcgt-tag since Marti decided not to do so (he weighted the drawbacks more than I do, what is okay, because there are real problems using the Video-LUT with todays video cards that do only support 3*256 entries (8bit resolution) - why did they stop selling good 2D-cards ;( ). >>> > But my display still goes totally blue when I invoke xcalib with this > profile. I had expected that xcalib would not do anything if it did > not find a valid vcgt. So I guess the question is should xcalib issue > a warning and keep going or should it issue an error message and stop > execution or something else? this is a bug in xcalib which will be fixed in the next version. There should be an error, not a warning. > The bigger issue is how do I get a valid profile for my monitor in > Linux. Of course this is not your issue to deal with but a larger > issue for the Linux community. I now have the source code for lprof > so I may dig into that code to see if I can have it generate profiles > with a valid vcgt tag and data. I don't think this will be trivial. It is'nt trivial at all. But maybe one could analyze the TRC-tags and use the error from a known gamma ramp. The error could be used to create a vcg-table or vcg-formula (=gamma). Valid profiles can be made with argyll (for the brave!) or the old lprof under Linux. And all Win32-profilers should work as well. I don't know about Mac profilers, but they should work if you use a gamma of 2.2. > How did you get your test profiles that are part of xcalib? Is there > a way to get good monitor profiles in Linux that I don't know about? The gamma profiles are from Bruce Lindblooms website ( http://www.brucelindbloom.com ). For table testing I used a profile created with Gretag's/Logo's ProfileMaker 4.1.5 together with an Eye-One Pro. The results were very satisfactory and I can recommend it over every other profiling solution (if you have the bucks). The ones that are located in Germany should have a look in a magazine called "c't special: Digitale Fotogrfie". I got it yesterday and it explains nearly everything about vcgt-tags and low-cost hardware monitor profilers (starting from 100$). There were only two which created a table and not a formula (which is the best way to calibrate your monitor). b.t.w.: does someone know about spectrophotometers or colorimeters that have a complete interfacing description (maybe even source code)? Stefan -- Stefan Döhla Regionales Rechenzentrum Erlangen Martensstr. 1 91058 Erlangen --------------------------------------------------------------------- mailto:[EMAIL PROTECTED] ------------------------------------------------------- This SF.Net email is sponsored by BEA Weblogic Workshop FREE Java Enterprise J2EE developer tools! Get your free copy of BEA WebLogic Workshop 8.1 today. http://ads.osdn.com/?ad_idP47&alloc_id808&op=click _______________________________________________ Lcms-user mailing list [EMAIL PROTECTED] https://lists.sourceforge.net/lists/listinfo/lcms-user
