i have the opportunity to "upgrade" from the zeiss 28mm f2.8 to the 28 f2. i'm getting a pretty good price on the 28 f2 (only $150 more than i can sell my 28 f2.8 for). i'm just wondering if anyone has had any experience with these lenses together? a site where they have been tested against eachother, etc. any help would be great. i have to make the decision today, before i meet the guy on friday.
i'm in the same boat with the 50mm (owning the f1.7, but getting the opportunity to get the f1.4...i know this is a big debate already going on in another thread, so i don't think it would be suitable to touch the topic here).
No personal experience with the 28f2. It has a floating design and vaguely I remember reading that it performed much better with near objects but infinity was not quite as good as the f2.8 (not 100% sure of this). I do have the 28f2.8 and find it nice but for some reason I prefer the 25 - esp as a street lens.
The 2.0 has the floating element, making for sharper corners at closer distances, and more even illumination. I have compared the 2.0 to the LeicaR 28/2.8 (believed to be the best 28mm lens for an SLR), and I thought my CZ 28/2.0 performed better.
The MTF charts may show the 2.8 to be able to resolve better in the center, but I like my 2.0. I did have the 2.8, and that is a fine lens as well. The 2.0 is a more deliocate mechanism, so you may want to invest in some Zeiss service to inspect and adjust if necessary. The 28/2.0 AE can be found anywhere from $600~$1000, and MM about double that as they are much rarer. I'm surprised you found one at such a reasonable price. Is it AE?
There are two versions of Contax CY mount lenses, the original AE (auto exposure), and MM (multi-mode) It refers to how the lens transmits information to the body for exposure information. MM lenses have an extra tab that protudes out from behind the aperture ring, and the position of that tab tells the body what the maximum aperture of the lens is so more enhanced program modes can be employed than on previous RTS bodies.
A few lenses underwent some changes when MM was inroduced, some had the rear shrouding re-designed, some had coating changes. Some lenses remained AE even after MM was introduced.
For EOS users, MM or AE is for the most part irrelevent, except that some 5D users report mirror clearance differences between AE and MM models.
This lens is about twice as long as the 28/2.8 model. Another way to visualize it is if you have seen the 21/2.8, it is like that without the wide element on the end.
jjlphoto wrote:
For EOS users, MM or AE is for the most part irrelevent, except that some 5D users report mirror clearance differences between AE and MM models.
Which one clears the mirror better?
You cannot use the 28/2.0 Dist on a 5D without mirror surgery.
It protrudes more than any other CY lens construction and completely blocks the mirror at infinity. (You need to press the shutter again to be able to close the mirror after it got stuck, not just turn back from infinity as with other CY lenses that collide with the mirror!)
We have extensively tested the CY 28's with a shaved 5D and a 1DsMkII just recently (sorry I have no image upload admission here).
There is actually three lenses to be discussed:
Distagon 28/2.8 AE first version
Distagon 28/2.8 AE & MM second version after 1984
Distagon 28/2.0
It is not widely known, that Zeiss improved their optical calculations for some CY lenses without any public notification. (This happened f.e. also with the Distagon 25/2.8).
The first 28/2.8 displays a steep and dramatic decay of resolution in the extrem corners. This is especially noticeable in combination with the sensor characteristics of the 5D.
Otherwise the center region of the first version 28/2.8 is unbeatable in resolving power.
The second version has a considerable better evenness of resolution to the corners and elsewise behaves similar to the older one. Sweet spot center f 4, corners f 8.
The "Hollywood" Distagon 28/2.0 possesses even higher corner resolution and better micro contrast that adds to an effect of sharpness and the famed Zeiss 3-D effect. The rendering from centre to borders is much more uniform. However the technical line recognition in the center is below the qualities of the 28/2.8 Distagon. Due to the floating element construction the close up performance is articulate compared to the 28/2.8 CY lenses. The 2.0's colours are warmer and yellowish next to the cool neutral 28/2.8 look.
Sweet spot at f 5.6.
A considerable degree of attention must be attended to the notorious and erratic adapter question:
The 28/2.8 Distagons are clearly better with a 1.48 registration. (At least our 3 copies.)
The 28/2.0 Distagon needs something thinner than 1.42.
(There is no other explanation to this problem than high tolerances in the manufacturing process on different Zeiss and Kyocera sites. Beside Zuiko, Leica and Canon 28's we tested all CY lenses with five different custom made adapters at infinity, 20m and 25cm just a week ago in a university study.)
I found the Zuiko 28/3.5 an interesting alternative to the CY 28./2.8. More overall consistency at all apertures, perfect corners and just a little bit weaker in the middle. Plus: No adapter hassle.
Edited by Arianne Dubois on May 12, 2007 at 07:42 PM GMT
Edited by Arianne Dubois on May 12, 2007 at 07:42 PM GMT
Arianne, my 28/2.8 goes to infinity only with the 1.42 Adapter. My custom made adapter turned out to have exactly the same thickness as the happygeak adapter which I bought recently.
It is also unlikly that the 2.8/28mm Distagon performs better with different adapters. This lens has no floating element and adapter thickness does not influence performance as long you can focus the lens correctly (an adapter which is too thick will not allow IF of course).
Ther 28/2.8 was manufactured from 1975 all the way up until the end. (Perhaps not right up until April 2005, but is stayed in production for the duration of the RTSxxx series.) It was always a Japan manufactured lens.
jonboring:
The well known Italian site with the CY MTF's is one source for the tacit optical calculation update of some Zeiss constructions. But we did not find a list of serial numbers that allows to draw a conclusion to certain production periods up until now. (We already asked Oberkochen and wait for an answer.) We are using one obviously late MM version bought in late 2004 (when Kyocera stopped the production line) that has a slightly higher weight (20 g) and a more consistent corner to corner performance than older AE copies.
Andi Dietrich:
I forgot one thing. I'm very sorry. You are right: Since the 28/2.8 lenses were the first ones we used for some experiments in 2005 before adapter thickness became a widely discussed topic, we have changed the internal infinity calibration to be able to work with the existing adapters. That is why they are perfect with 1.48 now. Again: I'm sorry, I simply forgot, when we tested it against the 28/2.0 and some other lenses last week. We never touched the 28/2.0 calibration. (On the other hand, there are clearly CY lenses that need less then the 1.42mm adapters. Distagon 21mm f2.8, 25mm f2.8, even the 135mm f2.8...)
This is our data comparing both lenses:
CY Zeiss Distagon 28mm f 2.0 vs. 28mm f 2.8 MM
Focal lenght: 28.8mm vs. 28.5mm
Lens/groups: 9/8 vs. 7/7
Min. focus: 0.24 m vs. 0.25 m
Angular field: 74° vs. 75°
Size: 62.5 x 76mm vs. 62.5 x 50mm
Weight: 530 g vs. 280 g
Distortion: -1.9% barrel vs. -2% barrel
Vignetting: 0.55 vs. 0.7
Year: 1981 vs. 1. version 1978, 2. version 1984
I will try to upload some paradigms for the corner difference of the CY 28's by tomorrow.
