On Tue, May 31, 2005 at 12:45:33PM -0400, Collin Brendemuehl wrote: > > Because the two lens types step down with different mechanical > ratios between the arm and the aperture blades, the camera > (in its current iteration) does not know how to move or release > the arm. > > But this could be fixed in software, by having the camera be > smart enough to detect non-A lenses (all pins shorted, and > no "A") controlling the stepper position would be accomplished > in an alternate fashion.
You're making a very big assumption here - that the response to any given actuating arm movement is the same on any two old pre-A lenses. There is no requirement for this to be true (the actual response is controlled by the lens aperture ring). You'd probably not be off by more than a stop or so in most cases, but there again that's true even if you use the same movement as you would use for "A" lenses. The worst case is when you are trying to use small apertures - any small error in sensing the lever position leads to large exposure errors. Ask yourself this question: Why did Pentax change lens design? They already knew how to create non-linear mechanisms; there's one in (almost) every single "A"-or-later lens. But instead of keeping their existing lens mechanism, and producing a body with an actuator mechanism capable of controlling such a lens in the way you propose, they changed to a design that allowed much better control at small apertures, and made sure that all lenses built to that design had a calibrated response.
