Charles Poynton-Color technology
Charles Poynton-(updated 1997-02-27 - yes, this is stale!)
I have written answers to "Frequently Asked Questions"about Gamma and Color in video, computer graphics, and print.The documents are available in severalformats.
Permanent, easy solutions to many of the problems in tone andcolor reproduction in computing require assistance - even leadership- from the developers and manufacturers of hardware and software.Solvingthat problem is the primary goal of the Gamma FAQand Color FAQ documents.
I dedicated many weeks to developing the Gamma and Color FAQs,and I make efforts to keep them up to date and free of errors.The popularity of the documents is gratifying - they average severalhundred accesses a day. However, I receive one or two unsolicitedcolor or video queries each day. That may not seem like a lot,but that's 350 or 700 in a year. Owing to the requirement forme to work to pay my rent, I regret that I can't promise a responseto unsolicited queries, particularly those related to specificcombinations of devices. (I don't even own a Wintel machine!)For color science issues, you can try the newsgroupsci.engr.color;for computer graphics issues, trycomp.graphics.algorithms;for video, trysci.engr.television.broadcast(for broadcast video) orrec.video(for consumer video).
I would appreciate comments, criticisms, corrections, and contributions.In particular, I am looking for a qualified volunteer to contributea section on color management systems, and a qualified volunteerto write about UCR, GCR, and RGB-to-CMYK separation.
You may be interested in the companionFrequently-questionedanswers about gamma, the GammaFQA.
JamesM. Palmer was upset, as I am, concerning misuse of the termintensity. See his note,GettingIntense on Intensity; also, you might be interested in hisFAQ on Radiometry and Photometry.The most important thing for me to pointout to the computer community is thatintensity is, bydefiniton, proportional - that is, linearly related -to a physical measure of light power. A quantity isn'tnecessarily an intensity value just because you get more intensityas the quantity increases!
On 1998-02-25,TimoAutiokari published, in anopenletter in a Usenet posting, some inflammatory remarks concerningmy FAQs. These remarks are sufficiently grave that I feel obligedto respond. You can read about theTimoincident.
You can obtain data for theGretagMacbeth ColorChecker.
This note introduces the two main user adjustments of a videomonitor,Brightness andContrast (sometimes calledBlack Level andPicture). I explain the effect thatthese controls have on picture reproduction, and I explain howto set them. You canread the note online, or you can obtain a typographic-qualityPDF version (69126 bytes).
This note is available online, or you can obtain a typographic-qualityPDF version (101879 bytes).
I presented this paper at the SPIE/IS&T Conferencein San Jose, Calif., Jan. 26 - 30, 1998. The paperis published in the Proceedings of that conference, B. E. Rogowitzand T. N. Pappas (eds.),Proceedings of SPIE 3299,232-249 (Bellingham, Wash.: SPIE, 1998).
This article describes the theory of color reproduction invideo, and some of the engineering compromises necessary to makepractical cameras and practical coding systems. I presented thispaper at the SMPTE Advanced Television and Electronic ImagingConference, San Francisco, Feb. 1995. This is an edited versionof the paper published in the proceedings of that conference,New Foundations for Video Technology (pages 167-180).
Proceedings ofInternational Broadcasting Convention,1994 (Amsterdam, 16-20 September 1994), IEE Conference PublicationNo. 397, pages 218-222.
The following two articles are dated now, but indicate thestate of color management in 1992:
Color Management Technology for Workstations,Sun Expo '92, Manchester, U.K., Sepember 10, 1992.
RISC/UNIX Workstations in Desktop Color Prepress,Youngblood/IBEC'sDeskTop PrePress Today - Colour Conference 92, Toronto, March 26, 1992.
The International Color Consortium has agreed on a standardfor color device characterization. The standard, designated Specification ICC.1:2001-04,File Format for Color Profilesis available in AcrobatPDF format (1.7 MBytes).
PaulHaeberli has written an interesting set of notes,Graphica Obscura. In hisViewing Notes, Paul points out some problemswith gamma on the web. Paul has written a noteMatrix Operations for Image Processing thatexplains how certain image processing operations that people thinkhave to use HSB, HSL and so on can actually be accomplished moreefficiently in RGB space.
Charles© 2005-04-03(a)