Align I hate to be the one to toss a rock in your gears but I see an inherent problem with your theory. On most zoom lenses the f stop changes as you change the zoom setting. Thus one reason why your testing proved better on a 50 mm prime. You also list that your test was done successfully on another zoom however. If the zoom is set on the shortest focal length then, and only then with the exception of some lenses that claim f stop stays the same, the cameras shown f-stop will equal what it says on the aperture ring. I guess I would question the accuracy of doing this type of test in this manner on any zoom lenses.
I've done some testing by setting the rig up on a tripod aimed at a steady unchanging scene with a constant midtone color for a large area. I would use aperture priority to select an aperture with the ring set on "A", making sure it was not in between stops but at a full stop setting corresponding to one available on the dial. If it's a zoom use again only the shortest focal length, primes again work better. Note the corresponding shutter speed selected by the camera. Change the aperture ring setting to the same one shown in the camera, and again note the shutter speed settings for any change. I would think theoretically they would be the same if all is well. > > Hi everybody, > > Startled by a recent mention of bodies unable to precisely set the > diaphragm aperture, I conducted a small test to see for myself. Not > that I care much as I set the aperture from the lens most of the > time, but I wanted to know how reliable exposure is for those rare > situations when I recourse to program or shutter speed priority. > > Here's what I found: the MZ-5N consistently sets the aperture on > FA 28-70/4 half of stop higher across the entire range, from 4 to 19 > (that is 4.5 instead of 4, ...22 instead of 19). > On the contrary, it is very accurate with the FA 50/1.4 and FA > 80-320/4.5-5.6, where the size of the aperture set by the body > matches exactly the one set on the lens. So it seems it depends on > the lens, and not surprisingly in this case the poorer built lens > performed the worst. > > If you're willing to test your lenses and report back on the list to > see if we can find a pattern, here is how to do it. Note that this > works only with bodies capable of electronic depth of field preview. > On the bright side, the error is easy to detect. > > - set the aperture ring on the lens to A; > - set the aperture value on the body (with bodies like 5N lacking > this feature just set the shutter speed to some value, measure the > light and record the aperture selected by the body); > - press the DOF button to close the diaphragm; > - while keeping the DOF pressed, disengage the aperture ring from A > and rotate it towards the target aperture value while looking > through the lens at the size of the physical aperture; the blades > should maintain their position all the way down from A to and > including the target aperture; the true aperture set by the body > is the one before the aperture shape has changed. > > Servus, Alin > >
