• Advertisement
Sign in to follow this  

The RGB Color Model

This topic is 427 days old which is more than the 365 day threshold we allow for new replies. Please post a new topic.

If you intended to correct an error in the post then please contact us.

Recommended Posts

Hi, community.

 

I wrote a new article about The RGB Color Model

 

The post is the following

 

https://nbertoa.wordpress.com/2017/02/14/the-rgb-color-model/

 

If you have any suggestions (things to add, modify, things that are wrong, corrections, etc) please let me know, because the intention of this post (in addition to helping people with the same doubts than me) is to learn if my knowledge is correct or not.

 

Hope you find it useful!

Share this post


Link to post
Share on other sites
Advertisement

You went a little fast on gamma correction. The fact that a grey (128,128,128) is in linear space 0.21, not 0.5. This is because humans are more sensitive in low brightness and we use a gamma curve for storage to put more bits in the dark tones. The fact that phosphore from old TVs match approximately the response curve of the human eye in term of brightness is a convenient byproduct that lead to the gamma ramp of 2.2. In fact a better approximation of luminance to brightness is b=pow(l,1/3). modernes LCD/LED/oLED monitors does not have the constraint, but emulate the gamma curve to fit in the antediluvian standard

 

In that context, your example of multiplying a color to get light attenuation is dangerous in gamma space. To get different colors equally half bright we need to divide not the brightness value, but the luminance by 2, because our perception of brightness is not linear to the luminance. If i take a pure white at 255, the half bright grey value in the gamma space that your monitor need to receive is 188, not 128.

 

 

This is also true for summing colors, let say you add a fixed amount of light, if you do it in the gamma space, the result ends to not be uniform and depends on the original value your are summing too, you have to convert everything to luminance, sum, and go back to gamma space for presentation or storage ( rec 709, srgb, ... ).

Edited by galop1n

Share this post


Link to post
Share on other sites

Thanks galop1n for reading the article!

 

You are right, I went a little fast on gamma correction, but that was because I added a hyperlink to an article I made about it (https://nbertoa.wordpress.com/2016/06/20/gamma-correction/). In that article, there is an explanation about why we need to work in linear space instead of gamma space when doing lighting computation, etc.

 

I agree with you that is not clear in RGB color model article; maybe I need to add a paragraph about that at the beginning of "What operations can be done with RGB color representation?" section.

 

Thanks again for your suggestions to improve the article.

Share this post


Link to post
Share on other sites
Sign in to follow this  

  • Advertisement