10-09-2011, 03:42 PM
[quote name='thw' timestamp='1318115806' post='12232']
I believe most decentering issues, low resolution problems etc are caused by poor mechanical assembly rather than temperature differences.[/quote]
Define poor mechanical assembly. I think it is not only that, although my remark about temperatures was slightly tongue-in-cheek. Roger Cicala did mention temperature as a factor in his article.
With current sensors and the demand they put on tolerances, is such that certainly in the MF field it means that virtually any camera and any lens really ought to be manually calibrated for best performance. With smaller sensor cameras we are getting very close. Neither does it help that for AF, lenses need to be constructed in a relatively loose way, so that AF is possible without very large and heavy motors, and that AF is fast. In short, it is a balancing act between AF speed and possibilities, camera and lens construction, and optimal alignment of all parts.
Silly is relative; at least it is a tool available if and when all else fails. And AF on the imaging sensor still won't work 100% - we are dealing with stepper motors or a stepped approach for AF lens adjustment, plus there still isn't a guarantee that optical plane, lens mount plane, and sensor plane are 100.0000 % aligned correctly. That is the kind of precision you are looking for. At current manufacturing and engineering standards this can only be approached, not realized, short of increasing costs at least tenfold.
Furthermore, as Roger Cicala also indicates, AF is a statistical numbers game. This is partly due to the fact that we are dealing with stepper motors, or a stepped AF approach, partly because of minute variations caused by the lens wobble unavoidable due to the demands of AF, and a variety of other reasons.
On the bright side, AF performs better, faster, and under more difficult circumstances than the human eye was ever capable of.
Kind regards, Wim
I believe most decentering issues, low resolution problems etc are caused by poor mechanical assembly rather than temperature differences.[/quote]
Define poor mechanical assembly. I think it is not only that, although my remark about temperatures was slightly tongue-in-cheek. Roger Cicala did mention temperature as a factor in his article.
With current sensors and the demand they put on tolerances, is such that certainly in the MF field it means that virtually any camera and any lens really ought to be manually calibrated for best performance. With smaller sensor cameras we are getting very close. Neither does it help that for AF, lenses need to be constructed in a relatively loose way, so that AF is possible without very large and heavy motors, and that AF is fast. In short, it is a balancing act between AF speed and possibilities, camera and lens construction, and optimal alignment of all parts.
Quote:I do not know if current phase AF problems can be resolved by having the AF sensors on the imaging sensor itself. If it really works, then that's a far better solution than the silly AF micro-adjustment approach.
Silly is relative; at least it is a tool available if and when all else fails. And AF on the imaging sensor still won't work 100% - we are dealing with stepper motors or a stepped approach for AF lens adjustment, plus there still isn't a guarantee that optical plane, lens mount plane, and sensor plane are 100.0000 % aligned correctly. That is the kind of precision you are looking for. At current manufacturing and engineering standards this can only be approached, not realized, short of increasing costs at least tenfold.
Furthermore, as Roger Cicala also indicates, AF is a statistical numbers game. This is partly due to the fact that we are dealing with stepper motors, or a stepped AF approach, partly because of minute variations caused by the lens wobble unavoidable due to the demands of AF, and a variety of other reasons.
On the bright side, AF performs better, faster, and under more difficult circumstances than the human eye was ever capable of.
Kind regards, Wim
Gear: Canon EOS R with 3 primes and 2 zooms, 4 EF-R adapters, Canon EOS 5 (analog), 9 Canon EF primes, a lone Canon EF zoom, 2 extenders, 2 converters, tubes; Olympus OM-D 1 Mk II & Pen F with 12 primes, 6 zooms, and 3 Metabones EF-MFT adapters ....