I'd venture to say all photographers run up against situations like this. I do all the time. My point is that there's always a disconnect between lab tests and field results, and right now, I'm having hard time actually seeing the 3 stop DR advantage of the D800 in the sample images that are rolling out. Your results may vary
Come on, three stops of DR is HUGE. The difference should be very obvious outside of the lab, should it not?
It is a big difference and there are demos comparing the exmor vs the the other sensors, real world demos, and it does show up easily (if your scene has more than 10 or stops of dynamic range, certainly plenty don't and then it makes no difference at all, of course plenty do), that is my point, my point is also that the DR tests I, others and DxO do, DO very much match up to real world use. I don't know why you don't see the difference. I've seen some pretty stunning demonstrations of the difference.
Getting back to F1, Ferrari has gone from mopping everyone one up in the early '00s to struggling to keep up with McLaren and team Red Bull since the FIA banned off-season and in-season testing. They have their own freakin' track in Fiorana, Italy that they can't even use anymore. I'm sure if they could get back to doing some on-track, real world testing, their F1 program would be much more competitive. Furthermore, one of the new scrub F1 teams designs their cars entirely on computer. They don't do much testing at all, because they don't have the funds to do so. And guess what? They suck, and are always getting lapped within the first half of a race.
1. on track testing, is still testing, not 'real world' races and sure it is important, once you get down to a car on the road and drivers you are into a super complex scenario, far beyond measuring single things which is more like what dynamic range is
2. you can bet all the teams would be wayyyy behind where they are if not for lab tests, so many more parts and specific little bits would take so much longer and waste sooo much more money to develop
Many people are now raving about the awesome 5D3 AF, and quite a few of the ones who are now going on about how awesome it is are the very same people who called those of us who called out the prior non-1 series AF stuff like pathetic Nikon trolls or silly little people who just need to learn how to shoot properly or told us that is was absurd to ask for top AF in something that wasn't a 1 series. So us know-nothing Nikon trolls perhaps helped get them the very thing they are raving about now and said was impossible to even think of hoping for.
This is the body rumor forum, not a serious how to take better photos forum, so I don't know that it is a bad thing to call Canon out here
I have no problem calling Canon out when it needs to be called out. Before the karma system went away, I had the smites the prove it
I just find it odd that people are talking about the 5DIII like it's something to wipe your @ss with when you run out of toilet paper based solely on the lab tests conducted by a single group of tech heads. That's not to say lab tests are useless, but rather they're not nearly as useful as determining the strengths and shortcomings of your equipment based on how you actually use it in the field.
I'm sorry, but when some posts images that they took of their girlfriend's hairy arm pit, and marvels at how much detail they can see in every last pungent follicle at 100%, I can't take them seriously. The D800 appears to be a terrific camera, but I doubt that's the intended use Nikon engineers had in mind when they designed it. Maybe that's your idea of creating art if you have a strange armpit hair fetish, but I other strange fetishes to attend to that don't involve photography
Not quite sure how this came onto hairy armpits, just recall which of the two of us is the only one to have mentioned "hair armpits" and "fetishes"

.
I hardly think the 5D3 is some trash to chuck in the garbage!
I did spend half of my original post hear going on about all of the early positive reports about the AF.
But the low ISO dynamic range is wayyy behind Exmor sensors though and even quite a bit behind the best non-Exmor stuff Nikon is doing (D4), I really thought they'd have at least gotten it to D4-level.... I do find that quite disappointing. For the landscape-type shooter, the 5D3 really adds virtually nothing over the 5D2 (for stuff relying on AF or better body response it should be much better than the 5D2 though). Some won't care, but some will and there is nothing any less valid about their caring about that than someone not caring about it. (and once again, I DO see a very large difference real world that matches what the lab results suggest)
And the video is in some ways awesome now, free from moire and aliasing messes, but it still ain't the promised 1920x1080p, every video blog I read is getting on it like crazy for this which is not what Canon needed to continue the low-end film-making domination.
It has less reach than stuff like 7D/1D4/D800 for wildlife/sports.
But yeah the 6fps are not bad for the class and the AF is sounding better and better, it's no doubt a better cam the 5D2, no doubt, it certainly has a lot of great stuff going for it.
But their sensor tech does appear to be falling behind, in ways that may matter for some and appear to be doing a bit of video protectionism for the C300 now, which is a shame, why not a C300 2x2 sampled crop video for 5D3 at perfect 1920x1080 if they can't do that over the entire frame? Probably to not step on the C300.... although it's hard to know for sure.
But once again the improved AF does sound pretty awesome going by early reports and coupled with 6fps that's not a bad thing at all! If the old sensor did what you need, then it should be awesome. 22MP, FF, 6fps, top of the line AF, small form factor, video without moire/aliasing, great user interface, certainly not a bad mix. I think it is one step shy of great though, as is, plus more MP, or plus noticeably more dynamic range, or at least with true 1920x1080p video then it's great IMO for $3500, without at least one of those added in $3500 seems steep. It'll be interesting to see if it drops to nearer $3000 in a couple months.