Stephanie Brim said:
I'm strange this way, but I kind of like the vignetting that people have been getting. I'd actually *love* to get one of the ones with the more prominent vignetting problem.
I should clarify that the vignetting seen in the picture I posted was
artifically induced. When I make prints from film, I almost always burn in the edges of the print a bit, to concentrate attention on the center. When using the R-D 1, I often do the same thing by using the "Vignetting" slider in Adobe Camera Raw. The intended purpose of this slider is to
reduce vignetting, but it works just as well to
add vignetting for aesthetic purposes.
But all that pertains to digital photography... Stephanie, you don't want to go there, and we don't
want you to go there!
I really figure that if I want to pick up a faster lens I'll end up with a Canon 1.4 or 1.2, but the Nokton looks better and better. If CV ever gets a faster 50 out I'd probably finally sell my soul. 😛
Extending your Canon range would be a tough call, because there's more to it than just numbers. Others disagree (as we've been doing on another thread) but I don't feel there's much of any performance difference between the 50/1.8 and 50/1.4; the speed difference is only about 1/2 stop, which is only going to make a difference if you spend a lot of time shooting hand-held, or with moving subjects, in low light. (If motion isn't an issue, you can just choose a slower shutter speed.)
The 50/1.2 would give you a full stop more maximum aperture than the 50/1.8, and that does make a more significant difference -- but, IMO, this lens is distinctly lower in sharpness, especially at the edges, than either the 50/1.8 or 50/1.4. It produces a look that a lot of people find very appealing, but if you want to make bitingly sharp images at full aperture, it's not going to be as good a choice as the others.
The Nokton, in my limited experience so far, seems to have just a slight edge over the Canon 50/1.4 in terms of fine-detail sharpness. But that wasn't the reason I bought it; for sharpness, I'm sure I could have gone on happily with the Canon (which has now gone back to living on my Canon 7s, by the way.)
What I'm really hoping to get from the Nokton is better control of rear-element reflections. Again, this is a digital-camera-specific problem: the camera's sensor is flat and shiny, so it reflects some of the image-forming light back toward the lens, like a mirror. If the lens' rear element is also fairly flat and fairly reflective, it can re-reflect some of the light back to the sensor again, at a slight offset from the original image. The result is that a very bright object against a dark background can appear to have a "double edge." (All you film users may now smirk smugly about the advantages of your camera's darker, matte-finished, less-reflective "sensor.")
This effect has spoiled a few of my R-D 1 pictures made with the Canon 50/1.4; I haven't noticed the problem with my other C-V lenses, so I'm hoping that the 50/1.5 Nokton will do the same.
About the idea of a faster C-V 50 sometime in the future: I wouldn't hold my breath, unless C-V also brings out a camera with a substantially longer RF base (which I feel they're unlikely to do, since it would compete with the camera they manufacture under contract for Zeiss-Ikon.) So far, it seems that C-V has tried to keep all its lenses within the limits of what their own cameras can focus accurately.
I was taught as a crude rule of thumb that, all other things being equal (which I admit they often aren't) the rangefinder-accuracy requirements of different lenses depend on their
numerical aperture (actual size of the "hole" in the lens diaphragm; obtained by dividing the focal length by the maximum f/number.) The larger the numerical aperture, the greater level of RF accuracy required.
To put it another way, lenses of different focal lengths will require the same level of rangefinder accuracy if their numerical apertures are the same.
By this measure, in the current C-V lens line, the 50/1.5 is already the most "demanding" lens in terms of RF-accuracy:
Lens---------------------------------------Numerical aperture
50/1.5--------------------------------------33.33mm
75/2.5--------------------------------------30.00mm
35/1.2--------------------------------------29.17mm
40/1.4--------------------------------------28.57mm
90/3.5--------------------------------------25.71mm
35/1.7--------------------------------------20.59mm
28/1.9--------------------------------------14.74mm
[The rest are less; you can do your own math.]
This table also explains why I don't expect ever to see C-V offer, for example, a 90mm f/2 (45mm numerical aperture) or 75/1.4 (53.57mm numerical aperture); these lenses would be far beyond the current cameras' RF-accuracy capabilities.
It also shows why I can't really count on close-up, full-aperture focusing accuracy when using my Bessa with my Canon 100/2 (50mm numerical aperture) or 85/1.5 (56.67mm numerical aperture.)
If I want to use these lenses on this camera, I have to be a bit prudent -- either by stopping down the lens when shooting up close, or avoiding the closest limits of focusing when shooting at full aperture. Both the 85/1.5 and 100/2 focus down to 1 meter; at this distance and full aperture (and assuming an 8x enlargement viewed from 12 inches) both lenses have a total depth-of-field of somewhat less than 2/3 inch! A 50/1.5 Nokton under the same conditions has a DOF of nearly 1.9 inches, which should give you some idea of what C-V considers safe in terms of the RF accuracy of its camera bodies.
