No time for posting tests, they often don't show things anyway, ones sharper than the other, just sharpen the other more, see now that ones sharper. (but what does that do to the overall look and feel?) See you need to look at it not as a pixel peeper looking at jpegs but as an artist looking at big images on screens or prints with the eye of an artist. I wish it could be more easily digestible for all to see but you really need to use one of these things yourself.
Well, I find that the over all look and feel of any of my big images on print or screen is largely dependent on the amount of processing I do. The finished image will almost always be very different in tonality, contrast, sharpness etc than the way it first appears in my RAW converter before any adjustments are made.
The idea that some subtle 'look' or 'quality' is going to survive the many adjustments I make and shine through, so to speak, to make it clear the original file was a P21 and not a 1Ds3, simply does not seem credible to me.
The few RAW files from Phase backs that I've looked at so far, initially have had a very different appearance to Canon files in ACR. For some reason the AWB (as shot) is very different. However, a simple adjustment of temperature and tint can remove 90% of that initial difference in appearance.
This argument you're presenting that it takes a fine artistic eye to discern these differences and that one either sees them or one doesn't, reminds me very much of the audio amplifier arguments that used to be common in the 1970's and 80's.
For a couple of decades or so I was fascinated with hi fi matters and used to read reviews in hi fi mags of the latest loudspeaker and amplifier designs. I have no trouble understanding lens MTF charts because I was already familiar with loudspeaker and amplifier frequency response charts.
One thing that used to puzzle me greatly were the claims by apparent experts in the audio field that differences in amplifier design were very audible. Even though two power amplifiers might have a similar RMS power rating, a similarly flat frequency response to beyond the limits of human hearing and similarly low harmonic distortion way below the threshold of human hearing, the expensive amplifier with an exotic design was often claimed to sound better.
Reviews of such amplifiers would often refer to a 'punchy' bass with lots of 'kick', or a 'sweet' treble, or a sense of 3-dimensionality that the cheaper amplifier just lacked. No matter how hard I tried, I just couldn't hear such differences. I could certainly hear major differences between 2 pairs of loudspeakers, but the subtle differences inherent in the different classes and desgns of power amplifiers escaped me. Provided the amplifiers had enough power to drive the speakers and were at least of reasonable quality, they all sounded alike to me.
Could it be I have defective hearing? Not likely, since I do (or did) play the piano.
Could it be that to hear such differences one needs above normal hearing, or golden ears?
Well, one day this was put to the test in a very thorough manner by one of the leading Hi Fi magazines of the day that seemed to specialize more in objective testing methods than some of their rivals. I suppose you could call them the Dpreview of the audio world.
They invited a number of leading audio experts to a listening session. All of them were chosen because they had avowed with great certainty that they could easily tell the difference between the sound of a valve amplifier, a hybrid valve/transistor amplifier, a Class A amplifier, Class C, MOS VFET and so on. These were guys who knew what they were doing; who had had years of experience in the audio industry. They were asked to bring along their favourite listening material and/or recordings they were particulalrly familiar with.
Now, I can't remember the precise details of the equipment used, but the loudspeakers and turntable would have been the best available. The amplifiers ranged from a $400 Pioneer model to a $10,000 Mark-levinson Valve/Hybrid amp the size of a small fridge. All the amplifiers were connected to the main speakers through a switching system so that at any given time whilst the music was playing, the amplifier driving the speakers could be changed. The listeners were never told which amplifier was in use at any particualr time.
The objective of the exercise was to identify the amplifier in use at any particular time. Do you think these experienced experts, so confident of their abilities, were able to do that? Not at all. The results were no better than tossing a coin. In fact they were slightly worse because the el cheapo Pioneer amp was confused with the expensive Mark-Levinson more than 50% of the time.
The reason I'm going to this trouble is to give a hint as to the sort of procedure that would be necessary to clear up this matter of the P21 versus the 1Ds3. Most of us can see subtle differences of hue, tone, temperature, saturation etc, otherwise we wouldn't be able to edit our images. Just a small touch of an adjustment curve makes a visible change. If you can't see the change, you're in trouble with regard to editing images.
What I would propose is that some competent photographer who owns both cameras, and a selection of good lenses for both cameras, does a thorough and careful job of shooting a still-life with both cameras, paying particular attention to precise matching of FoV, DoF and exact focussing on the same spot.
He then makes the RAW images available to anyone who's interested so they can process the images to the best of their ability. The processed images could then be made available in their original but equalised size, stripped of their EXIF information and metadata, and we could have a poll on which was which.
How does that sound? Too much trouble I'm sure, so we can continue to have these fruitless discussions about some mystical quality of DBs, only visible to the finely tuned artistic temperament.