Color Mapping and irradience map quality

[racoonart]

Hi,
I just made a few tests concerning LWF and Irradiance map quality. I expected similar results when rendering gamma either:
a, inverse gamma 1,0 ; gamma 2,2 and
b, inverse gamma 0,4545 ; gamma 1,0.
As far as I understand lwf I thought this would give similar results but inverse gamma seems to deliver better results. (GI settings are of course the same for both images). As you can see the inverse gamma setting does produce a more homogeneous distribution.

Thanks for any help =)

[racoonart]

Ok, again. Here is another example with much more difference: left one: gamma 2,2 ; right one inverse gamma 0,4545 (same as obove)
As far as i can say that from these two pictures, the inverse gamma sampling has lacks of detail in the dark spots, so i suppose it is NOT the right way to go. But I would really be happy if someone could explain what inverse gamma is meant to be for...