I've been thinking about the comparison and the discussion now for several days.
A possibility comes to mind that might skew the "test" results.
The possibility is that perhaps (worth a thorough check) the Canon has severe spherical aberration or curvature of field where the center and sides won't come into focus on the same plane (the strong CA is an indication), but will focus and "join" at smaller apertures for overall superb performance. Wide open shooting could make it hard to get accurate focus across the frame, and if the center were focused, the sides might not be, and vice versa. Then again, it may have been the Canon is fine, but was misfocused on the shots.
The Zeiss does look to be fairly well corrected for CA and seems to have better flatness of field, but those also need more testing and confirmation, and again it starts out slower at f/2. With such a limited sample base, it is really hard to tell anything conclusive.
I liked KR's review and it seems fair, and also sort of humorous -- better IQ than the 28/1.8 and a few zoom lenses! (I like the 16-35L II very much, but would expect the ZE 25/2 to trounce it handily.) He seemed to have his tongue in his cheek. Like many others, I would prefer to see the ZE tested against the fastest Nikon, Canon or other 24mm lenses. It does seem a shame that Zeiss hesitated to build a 1.4 lens.
A heads up for those interested in tech isues associated with the 25/2 is CZ's latest newsletter, which looks in depth at Distagons among other lens types.
Thanks for the link to Dr Nasse's tech paper.
Very interesting read. He of course just talks about the good things about ZE 25/2.
Doesn't't mention corner performance stopped down to f8.
edwardkaraa wrote:
The weak corners are only at infinity. No problem at closer distances.
Knowing Zeiss glass very well, I wouldn't be surprised if corners become excellent if the focus is slightly pulled.
I haven't followed the discussions closely the last few months, did we get a conclusion about the field curvature for the ZE 35 1.4? Is it possible to get sharpness across the frame at infinity at f/8 for that lens?
alundeb wrote:
Wasn't there an example in the paper showing perfect corner for the ZE 25/2 at f/8, but at close distance?
Yep, the Zeiss paper shows perfect F8 corners at 25cm! compared to the very soft ones at that distance for the 25 2.8 - sort of an odd, extremely close distance to do a comparison at but it does show the close to Macro abilities of the 25 2. The floating element design of the 25 2 is cited for allowing great performance "even at close range". What is not explained - and I don't understand - is why the performance at the corners does not carry though to greater distances, say at 100 feet or so to infinity. Isn't the main advantage of a floating element design great performance at both close and distant distances? In the case of the 25 2, it seems the lens is still optimized for one range of distances - close - whereas the older, non floating design of the 25 2.8 is optimized for longer distances.
Tariq Gibran wrote:
In the case of the 25 2, it seems the lens is still optimized for one range of distances - close - whereas the older, non floating design of the 25 2.8 is optimized for longer distances.
Tariq, I certainly agree with the above statement, but only if we are talking about the extreme corners. It seems to me that the f2 is superior to the f2.8 at infinity on almost the entire frame. At close distances there is no doubt about the outstanding performance of the f2. So the only weakness is at infinity and in the extreme corners. If the corners are really important, then I would say the f2.8 would be the better choice. But when you disregard these infamous cornersat infinity, the f2 has better resolution and sharpness at all distances, including the corners at anything but infinity. Not a bad performance at all in my opinion.
alundeb wrote:
I agree that only the extreme corners, the last ~2mm, are affected. If you crop to an aspect ratio of 5:4, it is completely gone.
That's a lot to crop - to a 5:4 ratio- if your doing so solely to correct this issue. I'm hoping someone here will thoroughly test the 25 2 Zeiss and show corners at longer distances as might be used for landscape so we can see if it's really an issue or if perhaps the sample Lloyd had was not representative.
Or just use the 21 for landscapes.
What I want is the PC apodistagon 3.5/25 lens which was designed along with the 21 but not productized. Hey Zeiss, make me a ZE version. It can be $3k, we will buy it!
wayne seltzer wrote:
Or just use the 21 for landscapes.
What I want is the PC apodistagon 3.5/25 lens which was designed along with the 21 but not productized. Hey Zeiss, make me a ZE version. It can be $3k, we will buy it!
Yeah I am actually amazed that they don't release it. I think it will be expensive but so does the 24 TSE from Canon and the 24 PC-E from Nikon. Also they will be even more comparable due to no AF. May be the quality of the current Canon and Nikon offering is already high and they thought that the Zeiss edge in quality may not justify its higher price tag
wayne seltzer wrote:
Or just use the 21 for landscapes.
What I want is the PC apodistagon 3.5/25 lens which was designed along with the 21 but not productized. Hey Zeiss, make me a ZE version. It can be $3k, we will buy it!
I suspect the price today would be more like $5000. It is essentially a 25mm medium format lens with a complex mechanical lens barrel. The current Zeiss ZV 50mm lens for Hasselbald costs about $4000. A much more elaborate PC Apo-Distagon 25/3.5 could easily cost $5000.
Lotusm50 wrote:
I suspect the price today would be more like $5000. It is essentially a 25mm medium format lens with a complex mechanical lens barrel. The current Zeiss ZV 50mm lens for Hasselbald costs about $4000. A much more elaborate PC Apo-Distagon 25/3.5 could easily cost $5000.
No, it was a 17 lens retrofocus lens with lots of special low-dispersion glass:
Maybe they are out to create some product differentiation, but it seems odd to make a wide angle that excels at very close range IF the trade-off was a loss of all-of-frame infinity excellence. Maybe my joke of using the 21mm and cropping to ~25mm is not such a bad idea, but I would hate to lose the extreme corners, they really help the 21's images, I feel.
For APS-C this 25mm is going to be great lens, however, a little longer than 35mm FL. But it's full frame Weird or what?
Weird or what?