I'm pretty sure that I am not, and will not be, the only one who gets
confused about the new (I suppose better) approach at gamma correction in
POV-Ray 3.7, which differs from the way 3.6 did it, and which results in
images looking different, and different approaches having to be used.
This will create a lot of questions related to this subject, repeated
over and over, especially when the final 3.7 is published.

  Thus I think it would be a great idea if we created a concise, clear and
easy-to-understand section to be included in the official documentation so
that we can simply direct people to it rather than having to explain things
over and over again. I was hoping we could come up with something in this
thread.

  Preferably this tutorial would be somewhere in the documentation that makes
it prominent and visible, rather then buried under hundreds of similar
subsections.

  My ideas about what this section should have:

  1) A simple and clear explanation behind the reasoning and motivation behind
the new gamma correction approach, why it makes sense and why it's better
than what POV-Ray 3.6 did.

  Also: Why POV-Ray 3.7 will render scenes differently than 3.6, when using
default settings and "#version 3.6" is not used.

  2) Why "Display_Gamma" and "File_Gamma" should always have the correct
values for the user's system (ie. usually 2.2, or in some Macs 1.8) and
why they shouldn't be abused for visual effects (because that's not their
purpose).

  3) The difference between PNG and other output image file formats in this
regard.

  4) How to render an old 3.6 scene using POV-Ray 3.7 so that it will look
the same (basically, by adding "#version 3.6" to the scene file; I also hope
that the small problem of the gamma metadata being written to the PNG file
regardless of that #version directive being used gets fixed, for full
backwards "emulation").

  5) How to pre-gamma-correct input images (if and when this feature is
implemented in POV-Ray 3.7) and why this has to be done (and why it doesn't
need to be done for input PNG images which already have a proper gamma
metadata in them). Likewise, how to pre-gamma-correct color values from
other programs (such as color pickers).

  Might not be the proper place for this, but perhaps also:

  6) If the user needs to do it, how to create an image which has no gamma
correction at all, only non-corrected raw pixels. The example of rendering
an image to be used for creating a heightfield could be mentioned.

-- 
                                                          - Warp

Post a reply to this message