A couple of points occur to me.
We are a bunch of nose-pressers so that needs to be taken into account.
"Intrinsic viewing distance"? There are two sorts of viewing distance. Firstly, there is the distance at which a print must be viewed to re-create the original perspective.
This can be calculated from the focal length and the magnification. It's important in a technical sense.
We don't usually view prints this way. We stand at a convenient distance from the wall and all prints have the same viewing distance, regardless of size or magnification. We might well choose to move closer (for nose-pressing) or further away from a mural of course, but in general we stick to the same distance. If we're handling loose prints, they'll all be held wherever our arm reaches.
So we have two criteria for judging perceived sharpness.
In Ben's excellent video, what impressed me was not the slight but perceptible increase in sharpness in the drum scan, but the improvement in saturation and the separation of small colour differences. This might be inherent in the two processes or it may reflect a difference in the software used. (Does drum-scanning include adjustments?) At normal print sizes, I doubt if we would notice the edge effects on the flat-bed scan. It didn't look like $200-worth of improvement to me, but Ben has his own reasons.
All this is judged from viewing on this particular screen, of course.
I suggest that if we saw either image alone, we would find it satisfactory. We seldom have such close side-by-side comparison. This is not just nose-pressing: this is eyelash-pressing.
If I may return to the original B+W image from Ian, he is to be commended for his heroic efforts, but I don't think it demonstrates the superiority of the projected image over the scan. (What we see is a scan anyway.) On this screen there is a jagged black area in the middle, free of detail, which I find uncomfortable viewing.
The eye does not see this way. It adjusts itself so that shadows in the real world have more detail than simple densitometry suggests. I can see no objection to black areas a such; they can have a strong and attractive graphic effect, but that does not seem to be the case here. Once again, I have to add the caveat that this is judged from one particular screen and the print itself may be much more satisfactory.
May we eventually have a report of the interesting conversations in Brittany. please?