Toothwalker wrote:
kosmoskatten wrote
About the 200/3.5 versus 200/4:
Yes, I was very surprised that the newer 200/4 would not be as sharp as the older 3.5. All three lenses, i.e. the TT 3.5/200, TT 4/200, and S 2.8/180, are 'old designs' saddled
with a fair degree of lateral color. This was already evident on film, and will only be more
conspicuous on today's sensors. In this regard the later zoom lenses VS 4/80-200 and
VS 4.5-5.6/100-300 perform better. However, if the OP is already disappointed with the
Canon EF200mm f/2.8 L USM mkII, he will not be convinced by any of these Zeiss
lenses.
Toothwalker - I don't think there is one number which can say how good lens is. For some of the qualities of Canon 200/2.8L are very ok (lack of CA, high contrast) however it lacks the look in images what I see in images taken with Zeiss lenses. Most of this look comes from 3 components, in which Canon doesn't perform well to my eye, below some comparison in my mind between CZ Makro-Planar 2/100 and Canon 200/2.8L:
1. transformation from focal plane to blur/bokeh (Zeiss = in middle of DOF there is small region which is really sharp, border between DOF and blur is pretty sharp - Canon = almost whole calculated DOF is having uniform sharpness, border between DOF and blur is very soft)
2. bokeh itself needs to have some edginess but not in unpleasant way (yeah, and really easy to explain this in my 2nd language - Zeiss = edgy bokeh but for example bokeh highlights are nice, Canon = soft bokeh, but bokeh highlights have annoying circles where the edge is slightly brighter than the center, overcorrection causes this sometimes)
3. microcontrast (Zeiss = delivered, Canon = not there, not with any aperture)
Of these typically Leica/Voigtlander delivers just #3 - they have then other characters which interest me - for example based on small number of available samples from Voigtlander APOs seem to have greater contrast on focal plane, and small contrast at bokeh. This character of the lens drawing definetly interests me. All the APO lenses do seem to represent colors better. Leicas' have their smooth bokeh and wide apparent sharpness inside DOF, which could be useful in some cases.
So I hope you understand that the reasons why I'm disappointed to 200/2.8L are not ones that one simple photodo measurement can tell. In contrary I'm sure 200/2.8L succeeds very well in all reviews which do not take lens character into account. For me it's the main reason to shoot with alternative lenses, Canon lenses are OK, build to succeed in there review giving just one number as the end result.
Lateral color (I assume you mean CA) can be corrected with software but like any aliasing (in princible it's just scaling of different color layers separately => scaling=aliasing) it will cause image to blur more, so it may mean that microcontrast is lost in process of correcting (depending amount of lateral color and correction). This will in practice mean that these Zeiss lenses may not deliver what I want, but I guess I have to find out that by myself since DSLR samples of these lenses are rare.