I have to say that these videos are underwhelming. Having said that, I'm not pre-judging the 5D Mark III by them. I have no real information about the conditions of these videos or what they would have looked like with another camera.
I know what I don't see: moire. That's good. On one of the D800 videos, I seemed to see some moire, which would concern me if I had pre-ordered one.
I have a 5D Mark III on pre-order with B&H. I ordered it late morning after the intro, so I'm not sure if I'll be in the first wave of shipments. I selected overnight shipping, so I'm hopeful that next weekend will be spent putting it through my own tests. I am planning to shoot a resolution chart and conduct my own moire test (I use my wooden deck for my moire standard, which induces terrible moire in my 600D). I will do a "shootout" with the 600D, 5D Mark III, and my Sony PMW-EX1 (sorry - I don't have a Mark II with which to compare it).
Here's what I plan to base the 5D Mark III test on: https://vimeo.com/35445053
If anyone has suggestions on how to better conduct the resolution test, I'm open to suggestions. I'm not a professional - just an enthusiastic hobbyist who takes this hobby seriously.
This time, I'll upload the full 1920x1080 mp4 and let Vimeo downrez to 1280x720 (ughhh). This way, anyone on Vimeo can download and compare at Full HD.
Having said that, I don't think that this particular resolution test is the be-all, end-all of the performance of anything. My EX1 looks better on this test, but I find myself prefering the video from my 600D (except when there's moire!) in many situations, even though it clearly has less than full HD true resolution. Thus, the test is simply that - a test along one dimension that may or may not be informative for a particular project. That's the way I see most of these videos. They are examples in certain shooting situations that may or may not be relevant to me.
Would I like for these sample videos to blow me away with video quality? Yes. But, I have yet to see any YouTube video of anything do that. I'd really like to see some well-done examples on Vimeo. The "low light test" of fabrics was pretty good for what it was. Had it been steadier (not handheld), I would have found it more informative.