Google| Subject | Re: Any Minolta/Sony users using UFRaw and GIMP? |
| From | Eric Stevens |
| Date | 04/20/2014 02:48 (04/20/2014 12:48) |
| Message-ID | <0t06l95pq02sg4chac60969vm5f8amkp3b@4ax.com> |
| Client | |
| Newsgroups | rec.photo.digital |
| Follows | nospam |
| Followups | Alan Browne (13h & 6m) > Eric Stevens nospam (22h & 56m) |
On Sat, 19 Apr 2014 09:59:30 -0400, nospam <nospam@nospam.invalid> wrote:
nospamI think it was nospam who first caught mt attention by some time ago saying that all color spaces are converted to Lab mode in their passage through Photoshop. Elsewhere I have seen mentions that the Adobe color engine works in Lab mode. I have just now gone hunting through Google and found all sorts of woolly references including references to the use of CIE XYZ or it's derivatives such as Lab.
In article <j0l3l9hp5d82kr0djjqinnghltbn07uo98@4ax.com>, Eric Stevens <eric.stevens@sum.co.nz>wrote:Eric Stevensnospam
My questions are intended to learn why you think working in Lab mode requires making a conversion twice. Your quotation above doesn't address that question at all and certainly it doesn't explain why you think converting an image to Lab mode in PS needs twice as many conversions as leaving it in RGB. After all, PS processes the image in Lab mode. All the conversion to Lab mode actually does is give the user a set of controls which work more directly on the color engine.
again:
Some users are under the impression that Photoshop does all its conversions to and from LAB, converting on-the-fly. this is untrue as it would greatly slow down performance. Instead, Photoshop uses LAB as a reference when conducting many operations. Photoshop is not actually converting pixel data between color spaces unless you, the user, actually ask for this. None of these issues should be interpreted as implying that a conversion from working space to LAB is bad. Just be aware of the issues involved with this kind of conversion and whenever possible, try to use similar techniques that can be conducted in the RGB working space.
in other words, if you convert to lab and back, there are two conversions that otherwise would not have been done.
the conversions are also not lossless, something which is trivial to prove. make the conversion and subtract from the original. if they're identical, the result will be zero, which it definitely is not, and on an image i randomly picked, it's noticeable without subtracting.
I've looked through various books and found a definite statement in "Fine Art Printing for Photographers" by Uwe Steinmuller and Juergen Gulbins.
Begin quote: ------------------------------------------------ A color management system (CMS) is a set of program modules that mediate color translation among different devices. These modules are often part of a computers operating system, and also are usually provided by software companies (e.g., Adobe). If an application is used to display, edit or print a color image, it initiates the proper function(s), e.g., displaying a particular image, generates the correct ICC profile information, and then tells the CMS what function should be performed. The central part of the CMS is a color management module," which performs the calculations needed to translate (transform) a color from color space A to color space B. Here's how it works:
i. First, the CMM translates the device- dependent color values of the image to a device-independent color space, using the description of the source ICC profile. Now the color values of the image are in Lab color space, which is device-independent. This intermediate space is called transfer color space or profile connection space (PCS).
2. Next, these Lab values are translated to color values that will produce a color on the output device that is as close to the original color impression as possible. If the output device cannot produce the very same color, the CMM will try to find the closest match. Finding the best match is determined by the translation intent (explained in section 3.4 at page 62), also called rendering intent. ---------------------------------------------------- End quote.
They are quite definite in stating that the Adobe CMS uses Lab space for colour. https://dl.dropboxusercontent.com/u/31088803/CMM.jpg is the diagram included with that section of the text.
I don't know but I suspect that the various modes made available by Adobe to the user only affect the various controls, filters, tools etc made available to the user. In the end they all work on the profile connection space.
If my understanding is correct then it doesn't matter what mode you choose to work in: your image comes in and is converted to Lab. When the image goes out it has to be converted to RGB, CMYK or what ever it is you want. Your choice of mode only affects the various tools etc with which you manipulate the data in the CMM. If that is the case the idea is incorrect that working in Lab mode requires twice as many conversions as working in other colour modes. --
Regards,
Eric Stevens
04/05/2014-
