However, my reading of what you have written above (which I have edited in my reply) tells me that:
1. You and your friends felt that the images were not as sharp as might have been expected.
2. You attributed this, incorrectly you felt later, to the AA filter.
3. You revised your opinion in your report.
Notwithstanding your change of mind on the reasons, the fact is that the images were either not as sharp as they might have been, or not. You do not appear to have revised your opinion on the sharpness, only the reason for it. There is an implication in your text above that it may have been simply comparing them with MFDBs being used on the same day, but you don't make this clear.
So, were the images disappointingly lacking in sharpness? Is there no marked improvement on previous SLRs? Or was it all down to the presence of the MFDBs?
Dinarius, While Michael is entirely capable of responding to the questions you put to him in his own way, I have a couple of things to say about all this, because I was participating in the Bill Atkinson/Charles Kramer workshop over a Saturday and Sunday at Michael's studio when he was analyzing images and drafting his preliminary report on the 1DsMkIII.
The camera was right there, and during a couple of coffee breaks I had an opportunity to take the camera out of doors and make some trial shots. On Saturday, the camera had a 16~35 lens on it. We put the images up on the monitor and I have to say they were under-whelming. I was more than a bit concerned because I have one of these cameras on order. So we discussed it, and the thought occured to me that perhaps it's the lens, not the body.
So on Sunday I brought MY 24~105 f/4 L lens, which I know is tack-sharp from what it does with my 1Ds (11MP). First coffee break I went out of doors, took a number of shots of gritty stuff with lots of texture, and a shot of Bill Atkinson talking to some students. We transferred them to my USB key, I brought them home, and subjected them to nothing more than the bare-minimum I would do for any image: normalize the histogram in CR4.1, a bit of contrast enhancement with the Parametric Curve, and PK Capture and output sharpen in Photoshop.
I printed the photographs on Epson Enhanced Matte A3 size. I found the results to be very sharp, very high resolution, e.g. for hair, skin texture, brick walls, hose pipes, paint peeling off wood, etc.
But to make sure I wasn't fooling myself, I brought the prints to New York City when I attended Photo Plus Expo several weeks ago and showed them to three of the foremost digital imaging experts in the country (who will go un-named to protect the guilty). They confirmed that the resolution and overall image quality were fine. So the conclusion I came away with from this limited personal test is that the sensor is very unforgiving of second-rate lenses. (Sorry Michael, your 16-35 just doesn't cut it!)
Then, as I mentioned in my previous post, I had the opportunity of seeing some of Michael's prints of the images in the final field report, and commented on those here: [a href=\"http://luminous-landscape.com/forum/index.php?showtopic=20877]Previous Post 1DsMKIII Prints[/url], post #19.
Now all that said and done, suppose you can find someone like MaxMax
to remove the filter for you. Then you will need some kind of post-capture processing software to deal with the issues the filter is designed to mitigate. Maybe it's easier to manage for MF cameras than it is for DSLRs, and presumably Canon has researched this alternative, because if it can occur to a photo-technics nonentity like me, it most likely occured to them - ions of time ago. Why would they want to market cameras with sub-optimal resolution? I think it is time to put the AA filter issue to rest.