Archived posting to the Leica Users Group, 1998/06/29
[Author Prev] [Author Next] [Thread Prev] [Thread Next] [Author Index] [Topic Index] [Home] [Search]>Tom Kumagai wrote: >> >> The difference between between 1/250 and 1/236(which is 0.331038 ms off >>4ms, >> thus 8.275934%). See the table below,(HTML). I don't think it's a big >>deal either. >> > >that's what i think is so interesting, focus seems to have a fundamental >importance whereas exposure is so elastic. > >or is it ? > >the nikon manual for the f5 descibes the "lithium niobate >oscillator-controlled >speeds from 1/8000 to 30 sec (in 1/3 steps)....".... > >if they can accurately control the shutter between 1/8000 and 1/6666 then >we can >safely assume they are getting at least half 1/1333 (or 1/2666) degree of >accuracy >in their shutter speeds ? > >m No. They can probably get accuracy of 1/6 f-stop on all speeds, and that can be considered as _very_ accurate. Exposure is generally not as accurate as has been assumed recently on this list. Professional film is accurate to 1/3 f-stop, and as LF users know, Kodak has in the past often placed a little note in a film box stating that 100ISO film is actually 64ISO for a given batch, or, more rarely, 125ISO. Amateur film is also usually accurate to 1/3 f-stop. Today things are more accurate than they were in days past, but I don't count on anything being better than 1/3 stop accurate. Lens apertures are accurate to about 1/6 f-stop. Light transmission of lenses is often (especially in zooms) 1/3 stop less than stated, but at least that is consistent, so it doesn't matter as much. My light meters are good to about 1/6 stop, because I had them adjusted. Recent meters from the factory were all within 1/3 stop. Film processing at my local professional lab is within 1/6 stop. My metering from one subject to the next is probably no better than 1/2 stop, depending on what I'm metering with, what the contrast and light level situation is, and what I've shot recently. So in a worst case scenario I could easily be out 2 stops. Statistically that is unlikely, but it could happen. 1/3 stop variation on chrome film is considered generally to be barely perceptible, if the images are not viewed side by side, so the smallest bracketing increments are usually 1/3 stop. In many cases, if you bracket a series of 7 shots in 1/3 increments (total range, 1 stop over to 1 stop under) in an uncontrolled situation, you will usually find that you have 2 or even as many as 5 useable shots, in the sense that if you had only 1 out of that series in front of you, you would think that you got the exposure right. To deal with these vriables, including but not limited to subject contrast ranges, bracketing with slide film is generally adviseable for critical amateur or professional use. Bracketing with B&W film is not necessary in most cases. Bracketing with colour negative film is generally pointless; bracketing with lith film is absolutely essential. The variations in shutter speed that have been talked about are generally insignificant, and the ones that Dominique posted are _truly_ insignificant. A lot of my leaf shutters are more than 1/3 off, and some are a full stop off at some speeds. I get them calibrated regularly, and then I get repeatable results. That's all that matters. The sum of variations in subject matter, film batches, film gamma curves , film processing and 'LUGnut torquing error' will be more significant in exposure variation than shutter speed error in pretty much every case. ....IMHO * Henning J. Wulff /|\ Wulff Photography & Design /###\ mailto:henningw@archiphoto.com |[ ]| http://www.archiphoto.com