michaelwatkins Offline Upload & Sell: Off
|
p.4 #4 · p.4 #4 · DotTune: New AF tune technique, no photos required | |
Jay968 wrote:
When you 'tuned by eye' was this by looking through the viewfinder (or LCD), or did you view on a larger monitor, or better yet make some prints?
I tried two different approaches; I was using an inclined target similar to a ruler or the commercially available lens tuning aides. The problem with this approach - my fault probably - was I was shooting too close to the target, in addition to judging results by eye.
The second effort involved shooting a planar subject square to the camera. My guess is I was far too close, again, for this to work well, but still I was able to determine a range of acceptable values. Had I made notes of the upper and lower limits of that range, and then picked the middle point, maybe I'd have got the same results as FoCal or DotTuning. Instead, I sifted through 40 different tuning values times 5 or 10 images per tuning value, eliminated groups which had no or few in-focus images, eventually whittling them down to just a few from which I picked the group that returned the best focus quality and the most in-focus images.
I did not keep notes of that process but I thought it was sound. It is very likely that by coincidence a group near the outer range happened to look best to me at the time, probably by virtue of the test conditions. Again, had I marked the lower and upper bounds of what seemed to be generating acceptable (but not necessarily "the best") images, and then divided that range in half and used that value, maybe I'd have achieved the same results as FoCal or DotTuning. Instead I was too focussed on picking groups which appeared to achieve maximum focus quality. I think this may be a common trap by those new to autofocus tuning, like me.
|