Heh, took me a moment to realize why did you post that :- D thought you mistaken thread for moment..
I dunno what they tried to show with their super-ugly example, all I see is noise at 15, even more noise at 10, and even higher noise at 5 minutes. It's faster !!
Why cannot they for once pay some attention to the examples. They only had 30 minutes to test it ?
2Deadclown: I don't know. I am pretty sure I and many others understand DMC quite well, but even then it's imho hoocus-focus magic with too many potato values, and behaves differently often. There are also too many workflows, with each professional adopting his own way around this and it just doesn't look right to me in such case. And when biggest pros like Bertrand Benoit struggle with it to reach satisfying solution ( the top solution, un-clamped, sub-pixel off, linear, with BF engine and complex scenes with every material being glossy) I doubt it's so powerful. But I am might be biased in opinion as I do actually hate it :- )