I've seen lots of threads on t'interweb about the focus shift issues with this lens and some well respected members of this forum who say it has a problem.
BUT, I've also seen recently that Canon may have quietly fixed it in the past couple of years?
Here's my question;
Have they fixed it? Or is it a fault of the design (non floating elements) which means even Canon can't overcome the laws of optical physics with a bit of tweaking?
If Canon changed the optical design of the EF 50/1.2L, they would have announced it as a revision and branded it a Mark II. This would also be necessary to facilitate the correct servicing of these lenses.
My personal understanding of the focus shift issue is that it is really a tiny effect, one that is only noticeable near MFD and stopped down from f/1.4 to about f/2.8. What is much more likely is that inaccuracies in AF precision with this lens is the dominant factor explaining missed focus, not the focus shift itself. I've read posts online with missed focus images taken with this lens at f/1.2 at a subject distance of around 5-10 meters, and asking if this is due to focus shift. Such is the state of confusion over this phenomenon.
wickerprints wrote:
If Canon changed the optical design of the EF 50/1.2L, they would have announced it as a revision and branded it a Mark II. This would also be necessary to facilitate the correct servicing of these lenses.
Yes, I'm aware if the design was changed then it would have been announced and it'd be MkII. But fixing/tweaking a problem isn't a redesign.
I must take pains to point out that focus shift cannot be rectified by changing the way the lens autofocuses--indeed, for a manual-focus lens with focus shift such as the Zeiss ZE 85/1.4, the user can find critical focus with Live View and still miss when the exposure is taken, because focus shift is caused by the change in plane of sharpest focus due to the focusing aperture being different than the exposing aperture.
Furthermore, AF cannot compensate for focus shift because the degree of shift depends not only on the chosen f-number, but also the focusing distance, and this kind of complexity is not presently supported in the EOS system.
The only way to eliminate focus shift, if it is present, is to redesign the optical formula; hence my previous post.
Perhaps I should also add that focus shift is not a phenomenon that is ever unintentionally introduced into a lens of this caliber; that is to say, Zeiss, Canon, and any other major lens manufacturer are not so incompetent as to make thousand-dollar lenses that accidentally have focus shift. It was deliberately allowed into the design, and there are a variety of plausible reasons for doing so. One reason may be that the designers allowed it so that the lens renders the blur circles of distant background objects in a more pleasing fashion.
wickerprints wrote:
I must take pains to point out that focus shift cannot be rectified by changing the way the lens autofocuses--indeed, for a manual-focus lens with focus shift such as the Zeiss ZE 85/1.4, the user can find critical focus with Live View and still miss when the exposure is taken, because focus shift is caused by the change in plane of sharpest focus due to the focusing aperture being different than the exposing aperture.
Furthermore, AF cannot compensate for focus shift because the degree of shift depends not only on the chosen f-number, but also the focusing distance, and this kind of complexity is not presently supported in the EOS system.
The only way to eliminate focus shift, if it is present, is to redesign the optical formula; hence my previous post.
Perhaps I should also add that focus shift is not a phenomenon that is ever unintentionally introduced into a lens of this caliber; that is to say, Zeiss, Canon, and any other major lens manufacturer are not so incompetent as to make thousand-dollar lenses that accidentally have focus shift. It was deliberately allowed into the design, and there are a variety of plausible reasons for doing so. One reason may be that the designers allowed it so that the lens renders the blur circles of distant background objects in a more pleasing fashion....Show more →
That's the kind of answer I was looking for, cheers!
Sir_Loin wrote:
Yes, I'm aware if the design was changed then it would have been announced and it'd be MkII. But fixing/tweaking a problem isn't a redesign.
I suppose we'll never really know!
Glenn, I kinda doubt Canon might have fixed the problem on the QT.
They look at the revenues generated by the sale of that lens, and as long as the lens is making a buck or two of profit, they'll keep churning it out.
50 f/1.2 is a lens where Canon of today probably doesn't see much merit in showcasing their lens craftsmanhip.
My lens is OK in a very practical sense, yet its continuing existence in my collection is being threatened, nay, squeezed, by a couple of lenses from the opposite ends of the price scale: 50 f/1.0 and 50 f/1.8. Those latter lenses deliver without fault on their principal functions, and according to their price. Alas, 50 f/1.2 doesn't quite do that because I do not have enough confidence in its focusing accuracy.
This has been debated and discussed for some time. There is a focus shift under certain specific conditions (that doesn't effect many) with the 50L. This focus shift is a part of the optical design of the 50L (and some other 50mm super speeds).
Even though the 50mm is a very common FL, there are some optical design problems making a 50mm super speed (in contrast a 100mm FL is easy/successful). THERE ARE NO GOOD COPIES of the 50L! There are, however, some "Bad Copies" that exhibit typical Canon QC variations common to so many Canon lenses. Why not just skirt the issue and get an 85L?
Focus shift can be prevented with lenses that have a floating element in their design. The 35L, 85L (both MKI and MKII and the old nFD 85L), 50 f/1.0L, nFD 50 f/1.2L, and the FD 55mm f/1.2 Aspherical all have floating elements and no problems with focus shift. In my view it is too bad that Canon didn't include a floating element in the EF 50 f/1.2L, but to keep it in perspective I believe that there are no other fast (i.e., f/1.2 or faster) 50ish lenses made for a dslr that have one except the uber-expensive and uber-rare Zeiss 55mm f/1.2 anniversary lens. So, yeah the 50 f/1.2L has focus shift, but so do all the other modern 50 f/1.2 lenses made by other manufacturers. By the way, you can compensate for it using liveview if you also hold down the depth of field preview button.
I shoot at at F2 to 4, at 2 or 3 feet, and haven't noticed problems with focus shift, although I was looking for it. Perhaps the amount varies from lens to lens and in some cases is so small it's a moot point.
Canon could fix it electronically with a lookup-table inside the lens that refocused at time of exposure based on reference data. But it would require more time and effort than they probably feel would be warranted.
Ed Sawyer wrote:
Canon could fix it electronically with a lookup-table inside the lens that refocused at time of exposure based on reference data. But it would require more time and effort than they probably feel would be warranted.
Exactly. The time it takes for the diaphragm to close should be sufficient to make a small correction in focus without a noticeable delay, even in Tv mode where the aperture is not necessarily known at the time of focus lock. The correction could also be made focus distance dependent, as the focus distance is known by the lens, it is by the definiton of autofocus.
That makes perfect sense, but wouldn't it then be done as a camera firmware upgrade, rather than something changed in the lens? I was under the impression the firmware in the lens was hard wired.
Indeed, it would probably require changes to the camera, since the lens itself is rather "dumb" compared to the camera. Too bad as it would be easier and cheaper if it could be handled at the lens level, and would be able to solve the issue for more people (e.g. those with older bodies, and no firmware upgrade path).
