I think I’ll throw in my $0.02…
I believe that Joseph James ended up being banned from DPReview forums due to his near evangelical promotion of the concept of equivalency (link to his work was provide earlier) and his forceful defense of it from detractors. It certainly is an idea that seems to pull the curtain back to reveal some sort of deep truth about sensor size that the manufactures are hiding from us. Personally, I question its value.
As has been discussed already, two images from two different cameras can be considered equivalent if they have the same perspective, framing, DOF, and shutter speed. When the images are examined, they’re found to have the same noise, hence, sensor size made no difference.
This deduction, however, is missing a crucial point. It’s like the 2012 doomsday alarmist that say there will be a planetary alignment that year. They’re right…it happens every year. That tidbit casts new light on the effects that the alignment is supposed to cause.
Such is the case with this concept of equivalence. What people don’t seem to pick up on is that both cameras capture exactly the same amount of light. This is critical because it means that both images have exactly the same shot noise. And if the shot noise is the same, the images will be the same. So it is essentially a rigged comparison…the camera with the larger sensor simply has no method for flexing its low-noise muscle. It is exactly the same as if you held a race between a Ferrari and a Yugo, but set a 20mph speed limit (yes, I know…the Ferrari might still win :P)
I always felt that the equivalence test was unfair to the large sensor, just as the speed limit is unfair to the Ferrari. And in the same way that you can’t really justify a speed limit in such a race, I don’t think there’s any justification for the impositions of equivalence. So what if my DOF is more shallow, or I changed my shutter speed a bit. I got a big sensor…so I can! (rhetorically speaking, of course…I don’t really have a big sensor
)