Great!

I run my light bias at .02 most of the time. I've been doing that so long I forget what it looks like if I don't.

The lighting is much more clear now. I forgot to say earlier why the white light matters.

When GC is enabled it changes how color values are handled not just on the output, but also on the input. All incoming material is converted from sRGB color space to linear via anti-gamma correction. This includes the little color chips for parameter on nodes as well as any texture maps you're using. Even the color chips on light nodes are converted.

You said you had a 40% light with light gray in it, but didn't say exactly what value of gray. But just as an example, suppose the gray was RGB 200,200,200. That's about 78% luminance. Combined with the 40% intensity, you multiply them to find the effective luminance before GC was 31%. But when GC is enabled, the 78% gray was anti-gamma corrected to about 59%. Coupled with the 40% intensity settings, the effective luminance after GC was 23%. So by turning on GC, you dropped your main light effective luminance from 31% to 23%. That's a decrease of 25% in brightness. That's why I saw less improvement by turning on GC the first time than I expected. Even thought the final gamma correction raised it back a bit, the net result wasn't very different.

This is why it is very important to get in the habit of only using white (RGB 255) lights. The only time you should not be using white is if you're actually trying to make a colored light. By that I mean something that it isn't color neutral, like blue or red. Only change light intensity using the intensity parameter. Don't do it via a darker shade of white, i.e. a gray.

---

Renderosity forum reply notifications are wonky. If I read a follow-up in a thread, but I don't myself reply, then notifications no longer happen AT ALL on that thread. So if I seem to be ignoring a question, that's why. (Updated September 23, 2019)