Forum: Poser - OFFICIAL


Subject: What's the big deal with gamma correction?

inklaire opened this issue on May 23, 2010 · 242 posts


Sentinelle posted Sun, 30 May 2010 at 3:55 PM

Quote - > Quote - > Quote - Pardon the stupid question, but is it a hard and fast rule that GC has to be set at 2.2 if you're using a PC?

Couldn't it be set at 1.2 or 1.3 if less of the effect were desired?

Good question actually.  There's no hard and fast rule that GC has to be set at 2.2 if you're using Windows.  We use the gamma values of 2.2 for Windows and 1.8 for MacIntosh because that's what most monitors have.  I don't see why you can't experiment with other gamma values.
 

Here's why.

Thanks for the valuable wiki page on gamma correction.  After reading this page and a few more on the web, I've come up with the following simplistic understanding of gamma correction.  Please don't hesitate to correct me if I'm wrong.

Computer monitors do not display materials linearly. If the input luminance is 0.5 and a monitor's gamma is 2.2, then the output luminance is not 0.5 but 0.5 to the power of 2.2, which is 0.2176. Therefore, for an image whose luminance is 0.5 to appear with correct brightness on a monitor that has a gamma of 2.2, the brightness of this image must be raised to the power of 1/2.2 so that the output brightness produced by the monitor is correctly set at 0.5. When a JPEG is saved on a monitor that has a gamma of 2.2, the JPEG algorithm encodes the file with the appropriate gamma correction by raising the image's brightness to the power of 1/2.2 (most Windows monitors).  On a MacIntosh, however, the file will be encoded with a gamma of ~1/1.8.

A JPEG saved on a MacIntosh looks darker on Windows because it is decoded with a gamma of 2.2 instead of 1.8.  The inverse is true when a JPEG saved on Windows is dipslayed on a MacIntosh.  The image will end up looking brighter than it should.

If we do not know exactly where a JPEG comes from, we can only guess that the gamma value applied to the image must have been 2.2.  It could have been anywhere between 1.8 and 2.3.

In such cases we need to experiment with different gamma values.  Am I somewhat correct?  Please do not hesitate to correct me if I'm wrong.