Upload & Sell: Off
I don't know about the quality of this guys measurements, but it looks like dynamic range is a touch greater on ISO 200 than ISO 100.
Where is Skibum5?
One thing to keep in mind, is that above ISO100 his chart makes the 5D2 look better than it is. The 5D2 applies less gain than the 5D3 and most likely less than the 1DX for the same named ISO. You'd need to shift 5D2 graph a solid 1/3 stop to the left for sure.
His banding test, IMO, makes the 5D3 and 1DX look a bit closer than they really looked to me, looking at full screen chunk of sensor, at least with the two copies I looked at. My copy of the 5D3 is unfortunately on the very worse end of read noise and vertical banding for 5D3 (OTOH it seems to have less high iso banding than StanJ's).
Anyway, as I said, it's not breakthrough for maximum DR, basically same old. But his high iso measurements do show that the high ISO improvement is not just in SNR but also DR so it appears that one you get into over ISO1600 it should be a full match for D4 and D3s's record breaking high ISO DR. Those three are in a total class of their own up there. The 5D3 comes is next, but noticeably behind and then come all the rest.
It is a little shitty, that for the $3500 price, Canon couldn't at least have used their best new high ISO technology in the 5D3 too.... That sure would've helped distinguish itself more from the D800. At least it could have clearly beaten the D800 for highiso then instead of more like a tie until you get to super, super high ISO.