POV-Ray : Newsgroups : povray.general : color_maps? : Re: color_maps? Server Time
4 May 2024 07:20:30 EDT (-0400)
  Re: color_maps?  
From: Bald Eagle
Date: 27 Sep 2017 18:40:00
Message: <web.59cc27f4471bf705cafe28e0@news.povray.org>
"Leroy" <whe### [at] gmailcom> wrote:

> Does anyone comments, Ideas?
>
> Would anyone use it?

Hi Leroy,

I approached the same topic, from another direction:
http://news.povray.org/povray.binaries.images/thread/%3Cweb.58cfc0cf857600e8c437ac910%40news.povray.org%3E/

I'd say that when we are dealing with color - and more importantly, the
_perception_ of color - we delve into a very complex topic that has almost as
much or more to do with what our brains completely make up, as what is there on
the screen.

http://www.grumpyoldastronomer.com/articles/illusions.htm
https://www.ted.com/talks/al_seckel_says_our_brains_are_mis_wired

That said, I think that any tool that can help a user explore and experiment
with procedural textures and color maps is a useful contribution that will be
used if it makes it easier to make a good looking scene.
My approach was based on the idea that if I had a photo / graphic file that I
already knew looked good, then if I could sample it and create a color map from
it, them my render with the same color map would look just as good.

I think that there is also a lot of "art" to making a truly good, and therefore
useful color mapping tool.  There are many ways of determining and representing
color - thus, the different color spaces such as rgb, srgb, hsv, etc.

My experiments were coded in SDL, and are kind of slow.
I'm sure there could also be many improvements in the execution, and the
algorithms used.

If you have a color mapping tool, and can make it work with 3.7-official or
3.8-alpha, then I'd say work it out and post a link for people to download it.

I'd also say that (from my own personal experience) that the thing that might
inadvertently prevent an otherwise excellent tool from being used is the lack of
good documentation and ease of use / learning curve.

Although many of the things we do here are digital, I think that a guiding
principal is to let the user be able to guide the usage of the tool.

[I hate when an "update" is rammed down my throat by some nameless faceless
programmer without my input or consent.   I ought to be able (to the extent
allowable) to use the same program with the same commands / menus / keyboard
shortcuts in both the version I had and the new version.  Make changes an
option, not a arbitrary dictate.]

Allow the user to modify and get feedback from the tool to maximize their
comfort and have it work for THEM.
The late Eric Sloane wrote some excellent books on physical hand tools, and how
craftsmen would modify the handles of things like chisels and gouges to fit
their hands, and they could _feel_ how it would cut through the wood.

So, after all of this blathering, I would say that whatever your code does, the
three most important things that are incidental to that, but central to it being
used, are
1.  Good, explanatory documentation that provides clear illustrative examples of
usage
2. Feedback from the program about what it expects, what it encountered, how it
processed the information, and what it spit out as a result of all that
3. The ability to modify and tweak the tool so that it's easy and intuitive for
the end user to use as they see fit

Given that the subject of color is an ongoing interest for many people here, and
it's something that's so central to producing a good, or excellent render, I'd
say that even if what you post is a prototype or intermediate iteration of what
you have in mind, you are likely to generate interest, and
perhaps spark collaboration in further development.

Thanks for your offer, and I hope to see what you've got if you decide it's
worth pursuing!


Post a reply to this message

Copyright 2003-2023 Persistence of Vision Raytracer Pty. Ltd.