You may be interested to know that there is a difference between how SceneFixer handles the setting of gamma and the way the changeGamma script does it.

Specifically, they differ in their handling of cases where the same image map is used for both colour (e.g. Diffuse) and data (e.g. Displacement).

This is a problematic situation because each image map can only ever have one gamma setting. This means that if a map is plugged into both Diffuse and Displacement (for example) you have to decide which of those channels will have the correct gamma, and accept that the other channel will have the wrong gamma.

The changeGamma script will assign a gamma value of 1.0 in this case, resulting in the colour channel having the wrong gamma and looking awful.

However, SceneFixer will favour the colour channel and set the gamma to the current render settings value (usually 2.2). This means the data channel will have the wrong gamma, but I would argue that using a colour map for data is going to produce nonsense anyway so the fact that its gamma setting is wrong isn't so important - at least one channel (the colour channel) is correct!

With changeGamma, though, both channels will be wrong: the colour channel because it has the wrong gamma setting and the data channel because it is using a colour map.

Of course, the correct thing to do is to have separate maps for colour and data, so they can each have their own gamma settings. Unfortunately it is too often the case that even commercial products come with maps being pressed into dual service like this.

Note that having seperate imagemap nodes makes no difference at all -  it has to be physically seperate image maps for it to work correctly.

Free stuff @ https://poser.cobrablade.net/