Mackinaw said:
Somebody on Photo.net discovered that proper back focus (or whatever the term is that defines proper lens-to-film plane adjustment) is key with this lens. Quite a few of these lenses seem to be off a few tenths of millimeters which can drastically affect sharpness. Mine apparently is right on (when used on my MP) and delivers very useable images. I'll be curious to what other folks think once they take a few pics with their 50mm F1.2's.
Hmmm, interesting. I own a 50/1.2 (although it's out on loan right now) and when I tested it on my Epson R-D 1 I noticed it seemed to "front-focus" by several mm... in other words, at an intended focusing distance of 1 meter (100 centimeters) the actual plane of best focus was about 99-point-something centimeters.
I thought this might be a simple mismatch between this particular lens and camera body (and this
still may be a possibility) but if enought people are encountering it to be noted on photo.net, then maybe there's more to it than that.
I can see how critical this adjustment is on this particular lens (not only because of its wide aperture, but because out-of-focus areas go blurry so dramatically) and easily can imagine the possibility that "back in the day," Canon simply aimed for a statistical middle ground in setting the focus position, and assumed that a critical user would have his/her individual lens select-fitted to his/her camera body!
Incidentally, the lens-to-film distance on this particular lens is set by a single ring-shaped brass shim that fits between the rear lens barrel and the mating surface on the focusing unit. I'm guessing that Canon's assemblers had a range of shims of different thicknesses from which to choose, and selected the one that provided the correct distance for that specific combination of lens barrel and focusing unit, so this sort of fine-tuning wouldn't have been difficult for anyone with access to the factory parts stash.
It would be tougher today, but I suppose a talented technician could adjust this by either removing a very small amount of thickness from the shim (taking great care to keep it parallel) if the distance needs to be decreased, or possibly adding a very thin additional shim (thin shim stock is easy to cut) if it needed to be increased.
Also -- One thing I noticed when gazing at this setup was that the shim sits in a shallow recess in the focusing unit. The outer diameter of this recess isn't much bigger than that of the shim -- so it would be all too easy for a DIY tinkerer to reassemble the lens with the shim NOT seated fully in the recess, but sitting up on the edge. The effect would be to move the optical unit outward by a significant fraction of a mm. It would be interesting to know what sorts of back-focus errors are being reported on photo.net -- if almost all of them are on the "long" side, this might well provide a partial explanation!