Guys 11/8/03 6:40 pm william.curwen <william.curwen-at-virgin.net> wrote
>I think it is also important to bear in mind the use of working in different >colour spaces. > >As has been said, monitors are capable of reproducing CMYK colour very well >which is good if we want to be able to get as close as we can to our printed >work on screen. how so? many CMYK spaces are well beyond the capability of a screen which excels in Red Green and Blue for obvious reasons, not in CMandY. > >However the default space for images coming in from cameras and scanners is >RGB as the sensors in these devices measure the primary colours of light >(Red, Green, Blue), although it is possible to scan straight into a CMYK >format. but not really possible, there are no CMYK scanners > >•Possible to convert and save in CMYK, not possible to scan in CMYK as >nobody makes CMYK scanners. right > >When I am manipulating images, I want to work in the largest colour space I >can until I am happy with the result. I am likely to keep changing an image >and it's colours many times and would not want to be limited by the CMYK >colours until as late in the process as possible. a good policy, ideally you want a space which just contains all image colours (or those desired - and to be introduced in edits). For this assessment and many other useful features I recommend the excellent Colorthink by friend and colleague Steve Upton. go to : http://www.neilbarstow.co.uk/profilgear.html >>>As we all know there are >>>many RGB colours that cannot be reproduced in CMYK but they (some, not all) >can be seen on a monitor. This means that we can work with as much >information as possible and then window the best CMYK colours when we >are ready to print. > >•Best to work in a capture colourspace suitable for the subject >matter being recorded. absolutely. >For example sRGB colourspace is the default colourspace for the >web....and records white/pink/yellow/red/black/brown skintones incredibly >well for - most purposes. Another extreme is Joseph Holmes 'Ektaspace' which >he created for scanning his awesome 6 bath dip and dunk 5x4in E6 trannies - >which is not very good for skintones but brilliant for extreme landscapes. see answer from Joseph below. > >For specific needs, the fact that CMYK gives us the use of black as well, it >is true we can create tones in CMYK that we cannot in RGB. Indeed there are >also some colours that can only be achieved in CMYK and not in RGB (Browns & >turquoises etc) but these represent a far smaller proportion than the extra >information that can exist in an RGB colour space. > >William Curwen There was a comment above about Joseph Holmes EktaSpace and Ektachrome Space being unsuitable for skintones. I thought I'd ask the man himself, he did say he's not willing to enter into any more discussion about this tho. I have a download link on my site http://www.colourmanagement.net/downloads.html for those who would like to try Joe's free ektaSpace, also to read up about it's advantages and find out more about Ektachrome Space too. ====== Forwarded Message ====== Date: 11/8/03 5:31 pm Received: 12/8/03 1:31 am +0100 From: Joseph Holmes 11/8/03 5:31 pm Joseph Holmes wrote >Neil (feel free to post this response if you like, but be assured that >I will not entertain any continued conversation on this endless >discussion of the merits of color spaces, or at least not this aspect >of them -- anyone who would like to receive Ekta Space PS 5 can get it >from you, Neil, can they not? > > >The contention that Ekta Space or other large RGB working spaces are >not suitable for skintones, per se, is nonsense. However, it is true >that with 24 bits worth of data, that the narrower an RGB working >space is, the finer the steps of color balance adjutment (and other >similar adjustments) are, albeit over a more limited range of color. >Thus in the sense that humans are hyper-sensitive to skin tones (as >we are hyper sensitive to the three-dimensional contours of faces) >narrow working spaces may afford a more pleasingly precise adjustment >environment for tweaking the color balance. > >To see if you might prefer this advantage with any given picture (I >work very precisely with subject matter of subtle coloration quite >regularly with Ektachrome Space with no such trouble whatsoever), >simply map a 16-bit scan of a picture with skin tones into each of >the two spaces, convert them to 24-bit (or not, since the tools will >work as 24-bit in any case) and proceed to perform color balance >adjustments on them in Photoshop using either the Color Balance tool >or the Curves tool, to see whether shifts of a single point are >sufficiently fine in the larger, coarser-unit space and so on. >Adjustments done in sRGB to color balance will be a little over half >the magnitude, per unit of adjustment, as adjustments made in Ekta >Space. And they will be perhaps a little less than a third the >magnitude as adjustments made with files that are in Lab. > >But have fun in any case, and do good science! > >Joe Holmes > >Kensington, California Regards, NeilB. Apple Solutions Expert colourmanagement.net :: Consulting in Imaging & Colour Management custom scanner and printer profiles, training on Imacon Scanning supply Gretag + eyeOne, ColorSoloutions basICColor : Display etc. XRite www.colourmanagement.net/ :: www.apple.com/uk/creative/neilbarstow/ =============================================================== GO TO http://www.prodig.org for ~ GUIDELINES ~ un/SUBSCRIBING ~ ITEMS for SALE
