Nanh wrote:
On the WA side I think the symmetrical ones are free from lateral CA. Not sure about longitudinal, but since the DOF is big it tend not to show up either.
I can't recall an image, and I've done thousands, where CA was apparent with the ZM21 f/2.8 on the M9. It's possible this characteristic is also shared by the f/4.5 version. Same impression with the ZM35 f/2.8. Maybe someone else can confirm this?
Bifurcator wrote:
Perfect RangeFinder Lenses (We're ready for ya!):
(Contax G) Zeiss Planar 45/2
Leica Summicron-M 28mm f/2 (but with FF vignetting WO) - About $4.3k new
Leica APO-Summicron-M 90/2 ASPH - About $4k new
I would remove the 28 Cron. It has very slight CA, along the lines of the 50 Lux ASPH. In fact, in LR3 I had initially set up CA correction identically for it and the 50 Lux, but then noticed it needed a slight tweak of additional yellow/blue correction as well.
While I can't claim comprehensive experience with the full range of Zeiss and Leica M rangefinder lenses, one impression has been that Leica may sacrifice CA perfection for other characteristics. All of my four Leica M lenses (21 Lux, 28 Cron, 50 Lux ASPH, 90 Summarit) exhibit some degree of CA, while I can't recall it being an issue with any of my three Zeiss (21 f/2.8, 35 f/2.8, 50 f/2 - though IIRC the 50 does have 'bokeh CA').
Zeiss (C/Y) Planar 85/1.2 - perfect really? No LoCA?
There is some CA visible in the so popular Cosina Voigtländer Lanthar 125/2.5 APO Macro around the focus plane, and sometimes around high contrast parts also further away from the focal plane. The Leica Macro Elmarit-R APO 100/2.8 could also show some LoCA. (Both the examples from own experience.)
That make me wonder a little about the list. Are all the lenses on the "perfect" list really perfect, or just nearly perfect and fully acceptable with regards to CA?
Yeah all the Nikon 200/2, the contax g45 f2 always has loca even at f4, an I'm sure down that list a couple more but I don't have time to look through it. In a brief scan it seems that perfect list is just a list of nice but not perfect lenses.
To the original question, "What lenses lack CA," there's a handful of APO LF lenses with maximum apertures of f/5.6 that lack any evident CA regardless of aperture setting. It's probably true of the Contax 645 120 macro and the R 280/4, but if lacking CA even wide open in all shooting scenarios is the criteria, none of these lenses belong on this list:
Voigtlander APO Lanthar 125mm f/2.5 Macro (1:1) - About $2.3k
Nikkor 200/2 AF-S VR I and II.
Canon EF 70-200mm f/2.8 L IS II USM
(C/Y or ZE/ZF) Zeiss Distagon 21/2.8
Zeiss (Contax N) 85/1.4
Zeiss (C/Y) Planar 85/1.2
Angenieux 180/2.3 APO
Leica Elmarit-R 28mm f/2.8 V2
Leica APO-Summicron-R 90/2 ASPH
Leica APO-Macro-Elmarit-R 100mm f/2.8
Leica APO-Elmarit-R 180mm f/2.8
Leica APO-Summicron-R 180mm f/2
Sigma 150/2.8 macro
Leica APO-Telyt-R 180mm f/3.4
Pentax 200mm A* f/4 ED Macro
(C/Y) Zeiss Planar 85mm f/1.2 (Anniversary edition) - About $4k
Canon 200/1.8, 200/2
(Contax 645) Zeiss Distagon 35mm f/3.5
Even the Distagon 21/2.8 has some even though some people here like to say that it's "almost apochromatic".
The CY is somewhat better, but I see the barest CA trace from time to time - tree branches backlit is the torture test of choice, more so than water reflections IMO.
The 35-70/3.4 does indeed have a small amount at the wide end, at wide (for it) apertures. I was sad to see the mighty 100-300 CY show just a smidgen also, but like most great zooms it is FL dependent. The Mamiya 120/4 is a beauty but I cannot be conclusive, having not subjected it to the tree torture test.
I agree entirely with Steve - CA is critical to fine lenses for edge definition, colour presentation and image depth. I not thrilled that s/w removal is likely to become commonplace, permitting a fudge for poor lenses similar to the distortion removal algos.
All the better ones over 100mm should get close enough - there is a threshold of acceptability at play, and it is quite simple...if I see much of it in more than a few percent of images, that lens leaves my possession.
For a long time in the film era, quite a few people thought that all later Zeiss lenses (this would therefore be CY and G) did not produce CA - modern monitors, high Mp cams and software changed that notion, but as a range they are still very good, even the 28/2.8 CY is excellent. The new 2013 Zeiss series, like the 55/1.4, will not have CA, I predict, and the RX1 will be very good also, as a fixed lens should be.
Good Thread ... reminds me of one reason why I have chosen many of my lenses. Not that they are "perfect", but "well corrected" (distortion.ca/etc. yuck). It's easy to get lured into looking mostly at resolution/mtf, etc as lenses come out touting their greatness ... but resolution times come at a price, either in the form of $$$ or optical compromise of one form or another. Getting all the pieces of the puzzle to work well together is a tricky proposition for lens designers ... especially for those who are trying to do it on the cheap.
I've got a couple that might make the "A" list, but my beer budget will only go so far toward "perfect" ... but I still like "pretty darn good" from the "B" list.
I am sure that some APO lenses might lose CA control ability while you add something like TC, filters on them, and the digital sensors may cause the CA also.Until today, the best ever optically made lenses are for the film era, sorry to see Cosina to produce lenses at that level and at very low pace.
There is not much deserved consideration for camera makers to do research on sensor CA?.
Interesting. The best work around is to not focus your 1:1.5 macro closer than 1:3 and then use extension tubes when shooting at small apertures to avoid the reflections.
I don't think that answer made many D800e macro shooters very happy.
contas wrote:
I am sure that some APO lenses might lose CA control ability while you add something like TC, filters on them, and the digital sensors may cause the CA also.Until today, the best ever optically made lenses are for the film era, sorry to see Cosina to produce lenses at that level and at very low pace.
There is not much deserved consideration for camera makers to do research on sensor CA?.
Digital sensors don't cause CA, they're just worse at hiding it than film.
Back to add the Minolta 200mm f2.8 APO to the list of very good telephotos. I have not seen any CA in that one but have not looked at ultra high contrast scenes with it.
philip_pj wrote:
Back to add the Minolta 200mm f2.8 APO to the list of very good telephotos. I have not seen any CA in that one but have not looked at ultra high contrast scenes with it.
It's definitely a good lens. I have not used one myself. But from what I have seen, it's definitely one of Minolta masterpieces. I think I have seen some CAs in Kurt's review. But again, every APO would exhibit CA if you really want to see it. I wonder how the color of this Minolta is compared to the Contax's you have.
philip_pj wrote:
Back to add the Minolta 200mm f2.8 APO to the list of very good telephotos. I have not seen any CA in that one but have not looked at ultra high contrast scenes with it.
I've seen a smidgen of defocus CA from that lens, though small compared to many less-well-corrected designs. For lateral (in-focus-plane) CA, the lens is exceptionally well corrected. For comparison, the Voigtlander 180/4 SL has no defocus CA that I have seen, but more lateral CA than the Mamiya 200/2.8. The Zeiss 250/5.6 Superachromat is the only lens I've personally seen yet that doesn't show CA of any type under the most adverse conditions (though the level of CA in lenses like the Mamiya 200/2.8 or CV 180/4 has never come anywhere close to being relevant/annoying for actual photographic purposes).
