Upload & Sell: On
| p.6 #1 · Zeiss 135mm f/2 Apo Sonnar vs Canon EF 135mm f/2L |
This sentence from article gets at my point:
"Compared to the average (mean) MTF50 values for the Canon 135mm f/2L, the Zeiss is better wide open across the frame, as shown in the table below. That’s very impressive as the Canon is one of the sharpest lenses around."
The Zeiss shows somewhat higher MTF values wide open than the Canon, but the Canon is "one of the sharpest lenses around." This encapsulates the point that I - and perhaps a few others - have tried to articulate. Let me try it as a list:
1. The Zeiss lens is a very fine optical performer. At the largest aperture a difference between its "resolution" and that of the Canon can be measured on test equipment.
2. The Canon 135mm prime is "one of the sharpest lenses around" and an outstanding performer in a number of other ways.
3. The measured resolution improvement at f/2 on the Ziess comes at more than double the price of the Canon.
4. If a measurable difference plays out as a real difference in the photographs we produce, a follow-up question has to do with the value/cost of that difference. If the measurable is not visible outside of the test lab, there is no photographic benefit to this difference.
5. Since the 135 f/2 is already, according to the same source reporting the performance of the Zeiss, "one of the sharpest lenses around," it is reasonable to ask whether the value of sharper than the sharpest available at f/2 is going to make a difference in one's photographs.
6. From the evidence (apart from very subjective and ultimately unresolvable beliefs about magical character, etc) it won't be different at average to smaller apertures, so there would not be any image quality advantage for those who tend to shoot at smaller apertures than f/2. As an example, typical landscape photographers who may rarely if ever shoot wide open will not see a difference in the resolution of their prints.
7. The evidence of a measurable resolution difference between these two excellent lenses at f/2 makes it reasonable to ask in what circumstance (and whether) the measurable difference might play out in those photographs made at f/2. Some subset of those photographs are likely shot handheld, so they will not take advantage of the very tiny measured difference in resolution at f/2, since it is impossible to hold a camera perfectly still - which would be necessary to generate an image in which such levels of resolution would be produced. So we could imagine a smaller subset of f/2 images that are made from the tripod - these could include some kinds of portrait work (though there maximum resolution is not often the issue), perhaps some other kinds of tripod-based wide open work.
8. With that limited group of potential photographs - wide open aperture, working from the tripod, static subject - that might create the maximum resolution image, the next question is how large would we have to print in order for the difference in these photographs to be visible, much less noticed by a person not doing objective testing. I do not have a fully objective answer to that question. However, I know that I can produce very large and very sharp prints - equalling or exceeding that of classic landscape work, for example - using lenses like certain other L and non-L primes and the better L zooms. I know I'll never see agreement on this - some will claim the difference will be clearly visible at 8" x 10" - but I'm quite comfortable with the idea that the difference would not be relevant at, say, 24" x 36" and probably beyond... getting to the point where other factors impair resolution and are the limiting factors.
I know that by writing out this thought process in some detail I have handed some of you an oh-so-fun opportunity to nit-pick through it and look for any small logical error and I fully expect that to happen - comes with the territory.
In any case, I think what we are asking in the case of this very excellent Zeiss lens on a more practical level might be: "Is the measured increase in resolution at f/2 relevant in the sense that a) much of my photography is done at f/2 and of high resolution subjects and b) I regularly print at extremely large sizes and am confident that the results from these f/2 photographs will be visibly different to the observers who matter and c) the increment of improvement in the f/2 shots that are printed this large will be significant enough to be worth the added cost and some loss of functional functionality.
That's my perspective. I also know that some will want the Zeiss lens simply because it comes from a company called Zeiss, that it can claim to measurably exceed some parameter on some other fine lenses, and because lenses as objects given them pleasure apart from their function as photographic tools. I don't see it that way - but I hear you.