I have just read a posting elsewhere on the Net which I am not sure I understand. It relates to the new REI/SOS ISO determination which I mentioned in an earlier post
It seems possible that "the exposure index values used by Canon are perhaps 65% of the value that would be determined under the ISO procedures for (saturation-based) ISO speed
For your upcoming review of the 40D is there any chance (or point in?) taking a shot of a suitable test card at a fixed ISO and aperture and recording the shutter speed set by the camera's metering system. Then take a shot of the same card (same lighting) and the same ISO and aperture (same lens) and record the shutter speed from a Canon camera released before REI/SOS was thought of (30D or 5D)? Obviously, I'm hoping you have a suitable earlier model available and one would hope that the final images from the 40D and the control would have the same brightness in order that comparing shutter speeds in that way is meaningful.
This isn't just idle curiosity as there have been some conflicting opinions about the high ISO noise performance of the 40D and it occurs to me that if Canon have substantially tweaked their ISO baseline then that might be part of the answer.
I know you have many demands on your time so I will, of course, understand if this proposed test can't be done.