WTF is PlanetX doing comparing heat of 2 cards using single slot coolers versus two using dual slot coolers. Would've made more sense to compare a GF8800GT to the single slot ATi cards or an HD4870 to the dual slot nV cards. 

Also who runs Crysis on 'Medium' ? Run it on Very High DX10 if you're going to bother using is as a benchmark.

As for which would be better for an HTPC, do you value the giant cooler or weake HD3850 (with only slightly lighter temps and power) over the HD video and audio features of the HD4850?

The review had little to tell people that a bazillion other reviews didn't already show more thoroughly.
"Overclocking was quite interesting. Daniel took a Phenom X4 9600 to just under 2.8GHz and an X2 6400+ to over 3.2GHz"

Wouldn't the definition of overclocking read something along the lines of pushing a device past factory performance specs? An X2 6400+ is 3.2 Ghz stock..

I'd have to agree that his "overclock" is phenominal.. O_o
WTF is PlanetX doing comparing heat of 2 cards using single slot coolers versus two using dual slot coolers. Would've made more sense to compare a GF8800GT to the single slot ATi cards or an HD4870 to the dual slot nV cards. 

Also who runs Crysis on 'Medium' ? Run it on Very High DX10 if you're going to bother using is as a benchmark.

As for which would be better for an HTPC, do you value the giant cooler or weake HD3850 (with only slightly lighter temps and power) over the HD video and audio features of the HD4850?

The review had little to tell people that a bazillion other reviews didn't already show more thoroughly.
"Overclocking was quite interesting. Daniel took a Phenom X4 9600 to just under 2.8GHz and an X2 6400+ to over 3.2GHz"

Wouldn't the definition of overclocking read something along the lines of pushing a device past factory performance specs? An X2 6400+ is 3.2 Ghz stock..

I'd have to agree that his "overclock" is phenominal.. O_o