CV 35/1,4 Nokton - Sean Reid review, Part 1

[sreidvt]

The focus shift is clearly evident in ordinary pictures at F/2.8 and F/4.0, as I discussed in the review.

[lct]

Originally Posted by sreidvt

The focus shift is clearly evident in ordinary pictures at F/2.8 and F/4.0, as I discussed in the review.

OK but at which distance?

[scott kirkpatrick]

Brian, and some others,

Focus shift (see the Van Walree mini-encyclopedia for a very understandable discussion) in a fast lens usually means that the rays of light that pass through the outside half or 3/4 of the lens come to focus a tiny bit in front of the image plane that is focused by the center of the lens. Since so much of the light is affected the apparent sharpest focus when the lens is wide open is governed by the outside of the lens, with a slight blurring from the center rays. Stop the lens down three stops, and the center rays are the whole story, with no blurring added, so things get sharper, but for objects a bit farther away. For faces in dim light, close in, motion blur could well cover the extra softness. For street scenes with adequate light, focusing a little deeper would give great detail. Sean's tests will provide the information to get the most out of this lens, if you like its rendering overall.

This is a fairly symmetric lens, doesn't stick out very far, so i would expect some vignetting, since the image that is formed at the edge of the frame simply doesn't see the rays passing through the outside portion of the lens on that side. This will make those details a bit sharper, if they are focussed successfully.

TomA's examples and your review, which shows daylight medium distance scenes, seem to be showing this lens under its best conditions.

scott

[lct]

Sorry to insist but at which subject distance are your focus tests done gents?
If at f/4 a given lens has a focus shift of say 10 centimeters at 2 meters who cares really? It will be compensated by DoF no?

[sreidvt]

Originally Posted by lct

Sorry to insist but at which subject distance are your focus tests done gents?
If at f/4 a given lens has a focus shift of say 10 centimeters at 2 meters who cares really? It will be compensated by DoF no?

Hi LCT,

First, given your deep interest in these cameras and lenses (across many years) it would really be helpful to you to actually read the reviews themselves. I summarize some points when you ask me to (and have for years now) but the original source really is the best place to look. Sometimes, to really answer your questions, I'd almost need to restate an article and many of the questions you ask are covered there already.

The focus shift can be seen with subjects at five feet, ten feet, twenty feet, etc. When I say that it is a problem with normal pictures at F/2.8 and F/4.0 that's based on reviewing hundreds of pictures made in a variety of settings. DOF starts to cover for it at F/5.6 and then, more successfully, at F/8.0. I always base my reviews on a combination of field testing and controlled testing. The two together give one a fuller picture of "the thing itself".

I doubt that the focus shift in the two samples I tested is the result of a manufacturing defect. If it is not, no amount of sugar-coating or rationalization will change how the lens performs. That doesn't mean, however, that the lens will not be useful or desirable to many people. I love the Canon LTM 28/2.8 and it mostly crashes and burns when tested for optical performance.

I would imagine that many photographers will be able to say of this lens, "I know it has X,Y,Z weakness but I like it and its working well for my purposes." And who can argue with that?

At the same time, a review of a lens, that purports to really describe it, needs to look at all aspects as objectively as possible. Many lens reviews are really anecdotal reports. Those have their place, of course, but they can also be somewhat limiting. I think its best for a review to describe pros and cons, across various performance aspects, and then let the photographer decide how that mixture of strengths and weaknesses suits him or her.

Cheers,

Sean

[terrycioni]

Originally Posted by lct

Not everybody is interested in testings pictures of walls and rulers at one meter or so.
A 35mm lens begins to get some wide DoF at f/4, for instance 0.7 meter at 2 meters or 1.5 meter at 3 meters subject distance.
It is then possible that focus shift is hidden in such cases and that photographers who don't shoot at the shortest distance of their lens do not notice it in day to day photography.
Agree?

LCT

Agreed!

This pretty much described my situation. I had Tom's SC for about a week (exactly a week). You have hit the nail on the head - as I said in an earlier thread I just shot with it. I can not really tell you what distance I shot at but in simple as possible terms all was fine as I went through the images!

Sean does very thorough examinations of lens performance - and I didn't get a chance to check-out some of his excellent suggestions as I had to return the lens.

That said I doubt I would find any difference in what Sean has already pointed out. He has also stated that in general use most of this can be managed and the lens is fine.

I simply love the size and price.

[scott kirkpatrick]

Originally Posted by lct

Sorry to insist but at which subject distance are your focus tests done gents?
If at f/4 a given lens has a focus shift of say 10 centimeters at 2 meters who cares really? It will be compensated by DoF no?

Probably not. But it depends upon how carefully you look at the results. And what you are shooting.

scott

edit: didn't see that Sean had pretty much answered this already.

[lct]

OK folks, let me take an example if you don't mind.
Let's say that the subject distance is 20 feet or 6 meters for instance.
At f/4 with an M8, DoF is then about 7 meters wide, i.e. 2 meters in front of the subject and 5 meters behind it in round figures (details hereunder).
If one sees focus shift then, it would mean that the lens front focusses by 2 meters or back focusses by 5 meters more or less.
Is that right?

[SJP]

The numbers I get are very similar:
So-max 10.91 m So-max 11.05 m
So-min 4.145 m So-min 4.125 m

Circle of confusion depth of field Overall depth of field
CC-DOF 6.76 m DOF 6.92 m

In my DOF calulator I include the effect of diffraction.

So focus shift problems only make sense for longer focal lengths and low f-values.

See my text on DOF here

[lct]

Originally Posted by SJP

...So focus shift problems only make sense for longer focal lengths and low f-values...

Or wides with short distance subjects if i follow you.
But what about focus shift in your calculations if i may ask? Should we take for granted that a lens like the CV 35/1.4 has such a focus shift that it might be considered as front focussing by 2 meters or back focussing by 5 meters in my example above?

[SJP]

Originally Posted by lct

Or wides with short distance subjects if i follow you.
But what about focus shift in your calculations if i may ask? Should we take for granted that a lens like the CV 35/1.4 has such a focus shift that it might be considered as front focussing by 2 meters or back focussing by 5 meters in my example above?

Exactly! This amount of focus shift would seem unlikely to the extreme so I do not understand if there would be focus shift problems at f4 & 6 metres. Maybe at close range?

Take 35mm, COC 23, f4 at 1 metre then you get:
So-max 1.08 m
So-min 0.933 m
Overall depth of field
DOF 0.14 m

[lct]

Originally Posted by SJP

...This amount of focus shift would seem unlikely to the extreme so I do not understand if there would be focus shift problems at f4 & 6 metres. Maybe at close range?...

I chose 6 meters because the lens suffers from focus shift at 20 feet according to Sean's post above, but i might ask the same question for a 10 feet (3 meters) subject distance as well.
Should we take for granted that the CV 35/1.4 has such a focus shift that it might be considered front focussing by 50 centimeters or back focussing by 90 centimeters at f/4 & 3 meters?
Edit: Asking this because i have never seen that in 30 years so far except with one lens (M-Rokkor 40/2) which had apparently focussing problems with an Epson body.

[sreidvt]

There's not really not much more for me to explain about this than I've said already. I understand the interest some may have in a theoretical discussion and I hope that it is productive. Just a reminder, however, that depth of field is a theoretical construct with many, many interpretations and variables. Peak focus, however, occurs only at one distance. When in doubt about theory, trust empirical results.

Cheers,

Sean

[lct]

Originally Posted by sreidvt

...I understand the interest some may have in a theoretical discussion and I hope that it is productive...

So do i as i've just ordered this lens for my Epson's.
My bet is a good lens designer should compensate focus shift with DoF but my fear is CV lenses are designed for film where DoF is wider for a same focal length, which could explain the issues you've noticed with your M8 to some extent...

[bqw]

Originally Posted by sreidvt

BTW, I just read and enjoyed your article on the lens. Your two keys points about the lens performance, as I understood them, were: 1) The lens resolves well on center wide open and 2) the OOF rendering is only somewhat like that of the 40/1.4. I agree that the lens resolves fairly well on center wide open and that its OOF areas are somewhat, but not entirely like those of the 40/1.4.

Basically yes

Focus shift, field curvature/corner resolution, etc. are not aspects of the lens that your article addressed.

This is true as the I tested the lens by putting it through my "normal usage" pattern, which is how I generally decide weather or not to buy a particular lens.

My results and yours, so far, don't contradict each other even though we've used different copies of the lens. If one man says an elephant is large and another says it's grey, those are not contradictions. If you end up doing focus shift and field curvature/corner res. tests, please let know. I'd love to discuss them. I'd also love to look at samples from those tests.

Actually after reading through this and a few other threads out there I took a series of photos yesterday at f2-f8 to try and identify backfocus issue.

But even with a focus shift issue it doesn't seem to be something I encounter based on my normal usage. It's all really a trade-off...you get a extremely compact lens that's also fast (usually contradictory) and sharp in the center at/near wide-open for a very reasonable price. The trade-off is some edge distortion, corner softness, and backfocus at f2.8-f4. For me personally, it's an acceptable exchange.

[sreidvt]

Originally Posted by bqw

. It's all really a trade-off...you get a extremely compact lens that's also fast (usually contradictory) and sharp in the center at/near wide-open for a very reasonable price. The trade-off is some edge distortion, corner softness, and backfocus at f2.8-f4. For me personally, it's an acceptable exchange.

That kind of analysis, in my mind, is exactly what things come down to when one is choosing a lens or a camera, etc. and its a large reason why I always try to simply *describe* the thing I'm reviewing without giving number ratings, levels of recommendation and all that. Every lens involves some kinds of compromises and different photographers have different ideas about what mix of pros and cons works best for them. I can never assume what a given photographers priorities might be.

Cheers,

Sean

[bqw]

Okay, I've put up some test images from f1.4 to f8. I really can't identify any backfocusing so maybe you guys can help?

The images are at: the unofficial Cosina-Voigtlander User Community - CV 35/1.4 SC Nokton Classic

All are 1500px in size and un-edited beyond scanning into CS and saving. The focus point is the flower. Placing the pointer over different areas will allow you to "zoom" that area and you can use the +/- keys to "zoom in/out".

Same roll of film and the shots were taken in sequence. Tri-x in hc-110 (b) for 6:30 @20C.

Thanks!

[lct]

Must be due to your wider DoF i guess. Makes me fear awful results with my Epson's the DoF of which is even shallower than that of the M8 with the same lenses.

[Paul Hart]

Sean has now posted Part 2, for which he had the use of several examples of the lens. It's as fair and even-handed as ever, and I'm still not reaching for the credit card.

[bqw]

Originally Posted by lct

Must be due to your wider DoF i guess. Makes me fear awful results with my Epson's the DoF of which is even shallower than that of the M8 with the same lenses.

Originally Posted by Paul Hart

Sean has now posted Part 2, for which he had the use of several examples of the lens. It's as fair and even-handed as ever, and I'm still not reaching for the credit card.

I'm beginning to think that as a film user it's a non-issue