From: Alastair M. R. <bla...@fa...> - 2008-12-25 23:26:10
|
Hi :) Lorenzo M. Catucci wrote: > Now, what I'd like being able to do is printing a nice calibration sheet (or > a set of sheets) at some defined resolution with patches covered by well > known percentages of each color: in my case, I'd rather like being able to > have a calibration using both matte and photo black, since the latter can > saturate more the true blacks; As you've no doubt seen from the preview of GPLin, I currently treat the two blacks as distinct channels and will linearize each separately. In the fullness of time GPLin will be able to use a separate density for each colorant. > as I've read somewhere that some third-party > gloss-optimizer does really have a colour cast, I'd think there should be > some way to have the gloss-optimizer sprinkled along the "real" colours; at > best the g-o should get printed on each of the less saturated patches, > just like it does when printing; at least, there should be g-o patches too. Linearization can't correct for a variation in hue - so I think the best approach here may be to linearize with Gloss Optimiser turned off, then *profile* with it turned on. Having said that, GPLin only currently operates in Raw channel mode, but once I add support for composite (CMYK and RGB mode) then the Gloss-Optimiser can be used or not used as you please, just as with any other print job. > I don't know if Robert or someone else just tried Alastair's GPLin wizard; > the first couple of prints help to calibrate ink saturation and black gamma > (I think it could be of some use doing just the same for all colours' gamma); Well, if the gamma's going to be *used*, then yes - but currently its purpose is really just to ensure that the target patches are well distributed and the actual value isn't critical - the gamma will be kind of "baked in" to the final linearization curves, so will be individualised anyway. > 1. ad-hoc calibration: the user saves the calibration curves etc., and is > able to have a calibrated > print using ad-hoc utilities (like photoprint) or maybe send the ad-hoc > user curves to gutenprint > along to a standard sRGB printable file Yup - the plan is to achieve this much as a kind of "proof of concept". > 2. system global calibration: if someone does just put the new curves where > gp can find them, and there are good established conventions to tell the > ink-set name and the paper type, whenever > I choose to print on epson's Ultra-Glossy instead of Premium-Glossy, I > would just have the driver choose the right calibration data in an almost magic way! Again, that's the plan. The last bit isn't going to be easy to achieve though. > I really think 2. could be of general use; I could help with calibrating > Epson's papers with third party inks; someone else should do the standard inks stuff. > I think Alastair wouldn't object to someone "buying" the calibration data for a target > he sends and asking to put the final calibration data online I'll be more than happy to help with building a library of linearizations once we get the infrastructure done. > While I am at it, I think there should be some way within gp's proper to > print fully saturated colour > bars in all the available colours (gloss optimizer too), to help in keeping > the heads clean. Agreed - I find that's sometimes more effective and less wasteful of ink then repeated head-cleanings when I have problems. In my case it's usually slight cross-contamination between black and yellow after a cartridge refill / priming cycle. Being able to address individual nozzles or blocks of nozzles would be nice, too! Oh, and: Seasons Greetings everyone :) All the best, -- Alastair M. Robinson |