ATTENUATION_LogReverse implementation is reversed
The implementation is backwards, as you get farther from the source the source gets louder and as you get closer it goes to 0. This is the original implementation:
Assume Distance = 1000 and Falloff = 20000 then this will evaluate to 0.5 log(1 / (1  0.05)) = 0.5 log(1.05) = 0.01 Assume Distance = 10000 and Falloff = 20000 then this will evaluate to 0.5 log(1 / (1  0.5)) = 0.5 log(2.00) = 0.15 Assume Distance = 19000 and Falloff = 20000 then this will evaluate to 0.5 log(1 / (1  0.95)) = 0.5 log(20) = 0.65 As you can see here, this is not the same as it is here: Wolfram Alpha plot of the above function: Wolfram Alpha UE4 Documentation: Epic/UE4 Documentation I think this was the intended implementation: 0.5*log(1x) + 1
Assume Distance = 1000 and Falloff = 20000 then this will evaluate to 0.5 log(1  0.05) + 1 = 0.5 log(0.95) + 1 = 0.99 Assume Distance = 10000 and Falloff = 20000 then this will evaluate to 0.5 log(1  0.5) + 1 = 0.5 log(0.5) + 1 = 0.85 Assume Distance = 19000 and Falloff = 20000 then this will evaluate to 0.5 log(1  0.95) + 1 = 0.5 log(0.05) + 1 = 0.35 This is much better: Corrected version on Wolfram Alpha
(comments are locked)

Follow this question
Once you sign in you will be able to subscribe for any updates here
I had to retype that twice and submit it 3 times... That sucked that it never saved its state in Chrome. It is best from now on to write these outside of the browser and then copy them over.