I hope that somebody who has acquired this lens would shoot a few shots to really detail what this lens can do at the closeup end of its range. The closeup test shot on husband's website was shot at 2 feet. I want to know if anything funky happens when this lens is shot at the range of 6-18 inches. There are a few zooms that go down to around foot, and of course a few of the sigma wideangles. But the sigma wideangle macros are not very good because at the close end of the focusing range they have some real funky distortion in addition to poor sharpness and CA. I know as I have tested several copies of the sigma 24 and 28mm lenses with macro mode. I am interested in a 24mm lens that focuses below 12 inches to use for some serious image stacking, near-far photos that you can't do with the canon 24mm TS or any other 24mm lens. Putting a slim extension tube or closeup lens on a 24mm lens loses the infinity focus making it unuseable for near-far image stacking.
I was really surprised by the results of this test and could not quite thinking that this particular sample of the Zeiss was a lemon or somehow defective with the corners never really getting tack sharp even by F8. I really had the impression all through the review that the Canon was the winner for full frame use from the posted crops. I don't think this particular Zeiss would trump the Olympus and really have my doubts that it would be better than a Nikkor 24mm 2.8 AIS(which I do not believe has actually been tested at all). I have the Nikkor and it does not come into its own until F8 as far as resolution is concerned and the extreme corners never do get tack sharp, not too unlike the Zeiss. The Nikkor does have CRC - Close Range Correction/Floating elements which makes me wonder if it would traunce all over the Zeiss in that area. Just my 2 cents.
Given the cost, and problems, associated with the adapters, which is in another current thread, I can't imagine why someone with a Canon digital body would buy the Zeiss 25mm.
httivals wrote:
Given the cost, and problems, associated with the adapters, which is in another current thread, I can't imagine why someone with a Canon digital body would buy the Zeiss 25mm.
I agree, especially when you can get the 28mm Distagon for $200-$300 used! I love that little lens!
Hrm. Looking over the pics from the ZF25, I see that my CZ28 performs exactly the same: from the "B" area outward, it's just poor, until I hit f/8 or so. This is the case at infinity. At close-range, on something like a newspaper test, it gets demolished. Oh, I'm using a CQ adapter.
I'm trying to minimise my subjective involvement in the editorial, but actions speak louder than words: I'm keeping the 24L and returning the ZF25.
If I had unlimited funds and an appetite for a vast menagerie of lenses, I would keep both, because in the f8-f16 range, which I believe to more important than f2.8-f5.6, the Zeiss will give better shots if you've got time to use it.
If you subtract out Zone C, as all Nikon digital users will, you're left with a very impressive performer. What makes sense on a Nikon body, however, doesn't necessarily work as well with a full frame sensor.
Though I really would rather use a flare-free, low CA, Zeiss lens at 25mm / f11, the 24L is no slouch in the f8-f16 range, either. The convenience argument weighs heavily, then – as does the big two stop advantage. My recommendation is unreserved for Nikon users: it's so much better than the 24/2.8 AIS I had, but Canon users already have a great (if not perfect) lens at this focal length, so I agree it's hard to see the appeal.
It's interesting to see how much Zeiss has optimised this lens for Nikon users: with a little DX sensor you won't see the mess made of the 'real' corners at close range, so why include a floating element?
I've update the diagram on the second page to show exactly how much of the image circle is visible to a Nikon user: basically, none of Zone C and only half of Zone B apply, so the lens should be fine on a D2X.
Given the huge improvement over the original Contax 25mm suggested by the new lens's f5.6 MTF (especially at 40lp/mm), I am amazed at how crap some of these crops look, particularly the area B ones.
I haven't really done exhaustive tests like these, but I'm pretty sure my Contax 25mm performs better than this. Either the MTF is exaggerated (and Zeiss are known for their accurate and honest lens data) or something is wrong with that lens/adapter.
Saves me considering a new purchase though, and makes me even more happy with my good old Contax 25mm. Maybe the ill-informed internet myth about its performance will be reconsidered now.
deshojo wrote:
Given the huge improvement over the original Contax 25mm suggested by the new lens's f5.6 MTF (especially at 40lp/mm), I am amazed at how crap some of these crops look, particularly the area B ones.
I haven't really done exhaustive tests like these, but I'm pretty sure my Contax 25mm performs better than this. Either the MTF is exaggerated (and Zeiss are known for their accurate and honest lens data) or something is wrong with that lens/adapter.
Here is the published MT's for the Zeiss Contax 25mm Distagon. I follow it with the MT's for the Zeiss ZF 25mm Distagon http://gallery.photo.net/photo/5522380-lg.jpg http://www.16-9.net/lens_tests/zf25_canon24/mtf_zf25.gif
The ZF lens, at least from the MTF's does suggest a significant improvement over the old lens. Further, the ZF 25mm 's MTFs suggest that it would be pretty sharp in the corners at f5.6 (and really not all that bad at f2.8). The test results showing the lens only reaching some measure of corner sharpness by f11 is wholly inconsistent with the MTFs -- and yes, Zeiss IS known for their accurate MTF's (based on actual lens testing and not theoretical calculations). Quite frankly in trying to resolve this inconsistency, I would question the informal test rather than jump to the conclusion that the MTF's are wrong or at best "optimistic". (There is the remote possibility that had one really bad ZF 25mm sample, but it is highly unlikely that Zeiss' QC would allow a lens that bad out of the factory door).
Two adaptors were tried: the normally faultless Fotodiox Pro, and a cheaper, thinner Chinese version. Performance was identical with both – which either means that the ZF25, like the Nikon 28, is relatively unfussy about adaptor matching and that either version would be OK; or, they're both wrong.
I have asked Zeiss, who supplied the lens directly, to comment on the review. I'll post an update if anything helpful transpires.
Hi Mark,
I'm reading with very much interest your tests and, as usual, I really appreciate your efforts.
What I see, here, is that we've a variable in this system that is named "adapter": let me explain cause I wouldn't like to be misunderstood. This is NOT a critique to your hard work, absolutely!
As far as we know, distagons (especially the ultra-wides like the 15mm) always had problems with adapters so my guess is that if you receive an answer from Zeiss, it could be that you're not using the lens on a Nikon but on a Canon's body. This imply that they cannot guarantee on adapter's qualities (quite obvious).
So, my very humble suggestion would be to try the same lens on a Nikon body that, although with a cropped sensor, should provide the demonstration that the lens has the issue in the "B" area: infact it should be soft in the corners with a Nikon DSLR.
If the test is positive and the lens has soft corners on a Nikon in the same manner in which it has the middle of the frame on a FF, the only other thing to consider is copy-to-copy variation.
At that point, after the test on a Nikon body, I would try another lens and see how it performs...
MY opinion and NOT a critique to your invaluable review!
For a lens with linear focus group like the ZF25, all the talk about adapter issue is not very convincing. The problem I think is the ability for Cosina to bring out the potential from the design. I do not know if it is a good decision for Zeiss to choose the least capable Japanese lens producer as its new partner. I have experience with a Cosina made APO Lanthar 125/2.5 before, the quality control issue is all over the map.
marcwilson wrote:
just a thought..if schenider can make their 28mm Pc in differnent mounts.canon nikon, etc why could zeiss not have taken this approach?
Patents. Nikon F is out of patent now, hence ZF. With the onset of electronic product enabling codes and the Digital Millenium Copyright Act, in the future even patent expiry won't help, as manufacturers will be able practically and legally to ensure the incompatibility of any competitor's products. It's here now (printer cartridges, iPod) but it's coming harder, so sign up to your brand religion or bend over.
I think there would be a way to test the ZF25 on a full frame body without adapters ... Film! If you shoot the ZF25 on a Nikon film body and the EF24 f/1.4L on a Canon film body, using the same film, scanning it with the same scanner, you should have a good test. If the Zeiss turns out to behave a lot better in the corners than in Hubsand's test, we know that it's an adapter problem ... if not, then we know that it's a problem with the Zeiss lens.
