[EMAIL PROTECTED] wrote: > On 14/10/06, Gonz <[EMAIL PROTECTED]> wrote: > >>I bet you'll find that due to mass production, the sample variation is >>not significant. Even though the "A" lenses might be calibrated that >>way, its still the same basic concept. I've found that while using the >>aperture ring is more accurate even for "A" lenses, its not off by that >>much, maybe 1/3 stop on worst case. Its probably similar to the K/M >>lenses. Besides you could autocalibrate your lens and write the table >>back into the firmware. Problem solved. > > > There was a post on the dpreview forum a couple of weeks back, someone > had mapped the accuracy of multiple exposures made using the lens > aperture setting vs the on body aperture setting for a number of > lenses. The results were plotted and the deviation of accuracy when > using the on camera aperture selection was in the order of double, but > it wasn't consistently negative or positive so it couldn't be mapped > out in reality. >
I have no doubt that the hard stop of the aperture setting is the most accurate way, this is what I've noticed also. But I think that what some people are saying about not being able to map a position of the actuator with a stop consistently for a K/M is IMO equivalent in saying that you cannot set the stop from the body consistently on "A" and newer lenses. I suspect that the deviation of accuracy would probably be about the same for K/M as "A" lenses (using body). > Johns suggestion of positional feedback on the actual in lens aperture > mechanism is likely the most practical way that it could be remedied > in the future. > Would that change the order of the way the mechanism works now? Because if I extrapolate correctly what you are saying, the body would take a meter reading right after it stopped down and adjusted the position to match the exposure reading wide open. This would imply that it would have to keep the mirror up during this period I think, because once the mirror is up it cannot meter. > Currently the more complex the aperture activation lever > couplings/cams the more error there is when aperture is set by the > body. When I shoot panos now I prefer to use lens based aperture > selection due to this fact. > Anything that is (relatively) static I do this. -- PDML Pentax-Discuss Mail List [email protected] http://pdml.net/mailman/listinfo/pdml_pdml.net
