I found it to be quite an interesting read, especially this part:
Intuitively you would expect far pixels to be white (depth of 1) and closer pixels to be darker. This is not the case here: GTA V seems to be using a logarithmic Z-buffer, reversing Z. Why do this? Due to the way they are encoded, floating point numbers have much more precision when closer to 0. Here, reversing Z leads to more precision when storing the depth of very distant objects, hence greatly reduces Z-fighting. Given the long draw distance of the game such trick was necessary. It’s nothing new though, Just Cause 2 for example was using a similar technique.
The whole article goes on for quite a while, explaining virtually everything in the following two parts.