Originally Posted by Fast Glass
I agree as well - we're already on the same page here. My point was not that you shouldn't compare them like that (as you say, there's no way around it), but that you cannot compare them like that in a meaningful way by using different lenses. The best compromise would be to use a zoom lens on both cameras and change the actual F and f so they give the same effective F and f on the different bodies.
In other words, when comparing the systems it's unfair to compensate for the effect of sensor size by changing lenses to match effective F and f - you need to compensate for the effect of sensor size by changing F and f on the same lens, for the purpose of an IQ comparison. (For other purposes, sometimes you just cannot compare, i.e. you cannot get 85mm f/1.2 FF DoF on a crop body - there isn't a 50mm f/0.75 lens!)
Here's what I mean, based on the what's available from the body+lens combos in Bryan's comparisons (limited number of non-EF-S lenses tested on crop bodies):
EF 50mm f/1.2 on FF/1DsIII @ f/1.2 vs. EF 85mm f/1.8 @ f/1.8 on FF/1DsIII- similar effective F and f, but different lenses - invalid comparison
EF 28-135mmon FF/1DsIII@ 85mm f/8 vs. EF 28-135mm on 1.6x/50D @ 50mm f/5.6 -similareffective F and f, same lens - mostly valid comparison (but still not the same as comparing 5DII with 7D, for example).
In a valid comparison the IQ edge will still go to FF, of course. For sharpness, I think it's an edge only, not a blow-you-away 'huge' difference - that's apparent from the 28-135 comparison. ISO noise is another matter entirely - there the difference is major.
But the overall performance had damn well better go to FF - the 7D's sensor is essentially the same one used in the consumer-level T2i, meaning I could get that level of sensor performance for 1/3 the cost of a 5DII!