Sun Jun 21, 2015 9:59 am
I haven't seen a definitive explanation of how it works, but clearly it must be making adjustments at the micro-contrast level. I believe it is also doing the selectively - it uses color contrast to determine which areas are haze affected and those that aren't. Further more apparently it can distinguish between surface detail and shapes. As far as I can make out no NR is being added as part of the process.
So yes you could probably replicate the results with a combination of levels/contrast and unsharp mask BUT you would also have to do some clever graduated masking to get the same results.
Some people are saying results are variable depending on the nature of the image - I've just tried it on some shots that were taken in actual fog, and while hidden detail was revealed, the results weren't at all usable, however shots that looked flat because of slight atmospheric haze responded well.
Its another tool in the tool box, useful for some things but with a risk of being misused or over applied.
Cheers,
Colin