Sonnar followup

mojojones said:
Can someone sum up for me? Is the focus shift only wide open at short distances or more pervasive? The well take images I've seen from this lens are just extraordinary. I'm thinking of selling my summicron to fund a sonnar purchase.
Click to expand...

Mojo, I suggest that you contact Rich Schleuning at Zeiss USA (RSchleuning@zeiss.com). He might considering providing you with a loaner. A lot of things can factor in with DOF this shallow including the accuracy of the rangefinder that is being used to test it. You want to know how such a lens will perform on your camera.
 
40oz said:
............ I think it is safe to say it isn't an issue as much with the lens as it is user error. With a depth of field in the range of a 1/4 inch or less, it would be very surprising if people didn't have problems........
Click to expand...
The focus inaccuracy is measureable. It's at least 100mm at a focus distance of 2metres. Given that the distance between a person's two eyes is about 70mm, that's a lot. You don't need any special equipment to see that if you focus on a subject's nearest eye, both eyes will be out of focus.

Has anyone had Zeiss adjust their Sonnar to focus at full aperture?
 
So is the argument that, as you open the aperture up, the plane of focus moves? Surely impossible? Or is it that the plane of focus is consistantly out and wider aperture reveals the problem? I can understand focal-plane shift wrt focal distance but not aperture...

Do any other lenses do this do any great degree?

Jamie
 
Last edited:
Yes, this does seem unfortunate. I agree, I would buy an f1.5 lens to use it wide open. I guess this is one of the faults of nor being able to look through the lens.
 
jamiewakeham said:
So is the argument that, as you open the aperture up, the plane of focus moves? Surely impossible?
Click to expand...

Jamie, this is exactly what happens. As the aperture changes & some light rays are eliminated wehn the lens is stpped down to a narrower aperture, the point of best focus changes (or "shifts"). Some lenses are corrected for this with "floating elements" but this makes the lens more expensive. This technology didn't exist when the original Sonnar was designed & apparently Zeiss wanted to reissue this as a "classic" design.

The focus shift is due to spherical aberration inherent in this design. So, another factor with accurate focusing of such a design is that depth of field is markedly asymmetrical in the presence of spherical aberration. It is possible that with this lens front DOF is much greater than rear DOF, creating an impression of front.

Zeiss has described the focus characteristics of this lens as "wandering" - so there may be a certain amount of unpredictability resulting from a number of contributing factors that leads to the conflicting reports.

Or is it that the plane of focus is consistantly out and wider aperture reveals the problem? I can understand focal-plane shift wrt focal distance but not aperture...

Do any other lenses do this do any great degree?

Jamie
Click to expand...

You can read more about this issue under the topic of spherica aberration here: www.vanwalree.com.

Best regards,
Huck
 
I continue to contend that the problem is most likely related to excessive sample variation. I have now shot 3 rolls of film in the last week with my Sonnar with no focus difficulty at all. I have not tried to compensate for any anticipated focus error and have simply gone by what the rangefinder on the ZI has shown me. My pictures have been quite sharp and pleasing to my eye.
I don't think I am any better a rangefinder photographer that those who have had difficulty, so the only conclusion I can draw is that some Sonnars have a problem while others do not. The issue of film vs. digital camera may also play a role, perhaps that deserves further investigation.
LJS
 
jamiewakeham said:
So is the argument that, as you open the aperture up, the plane of focus moves? Surely impossible? Or is it that the plane of focus is consistantly out and wider aperture reveals the problem? I can understand focal-plane shift wrt focal distance but not aperture...

Do any other lenses do this do any great degree?

Jamie
Click to expand...
There is focus shift with quite a few wide aperture lenses. Generally, if focus at full aperture is correct, at narrower apertures the increased depth of field takes care of most/all of it. You may be familiar with the geometrical diagrams showing how DoF actually works (the smaller aperture making the circle of confusion smaller). Well, it's not as simple as that. Have a look at Erwin's page about Dof at

http://www.imx.nl/photosite/technical/DoF/DoF.html

Although the second diagram is not easy to understand (at least for me) the first one is, and it does indicate that a claim that a lens has "more depth of field" than other lenses can actually be true. I'm planning to send my Sonnar to Zeiss to get the focus spot-on (or at least close) at full aperture, and I hope that the "extended depth of field" will give reasonable results at other apertures. I will report back.

Incidentally, I'm impressed with the communications that I've had from Zeiss. Prompt, courteous and helpful.
 
Last edited:
Wah. I'm supposed to be good at physics... I think I'm reading this as follows. At smaller apertures, only the centre of the lens is involved in bending light rays to a focus on the film (because the outer parts of the lens are covered over by big metal leaves). They bring the light to a focus at a certain distance, and this is what the RF mechanism will indicate as 'correct'.

But at widest aperture, the outer parts of the lens are also at work. And they converge the light rays at a slightly different point (not sure why this might be but I suspect some sort of abberation that just can't be corrected out of these very fast lenses - I could well believe that it would be a Sonnar-formulation-specific problem). At widest aperture, most of the light going through the lens is indeed travelling through the outer parts of the glass simply because it's got a greater area than the centre, so overall the lens appears to be focussed in a different place.

Now, that might be correct or might be utterly spurious. What I can't figure is why didn't Zeiss set the RF to get wide-open and close-up right, given that further away and stopped down will pretty much be covered by DoF?

Jamie
 
As far as I understand it, they can adjust it for close or far focus. Since they advertise it as a reporters lens they probably thought it would be used around two meters or more most of the time.

A loosely framed groupportrait seems to work just fine with the lens as it is, if you want it for a tightly framed head/shoulder portrait they adjust it for this purpose but probably with a slight tradeoff for the former use.
 
I know this is not a C Sonnar T* 50/1.5 ZM photo, but it seems that a Carl Zeiss Planar 50/1.4
lens for Contax RTS (C/Y -mount) doesn't suffer focus shift... :)



Contax IIa color dial + Zeiss-Opton Sonnar T 50mm/1.5 from 1950's. Shot with
Contax RTS + Carl Zeiss Planar T* 50mm/1.4 at f/1.4 wide open from 0,45m (1.5 feet).
 
Nice Contax :)

Conventional Planars don't suffer from focus shift and have exemplary field flatness too (something that Sonnars lack). Most of copy and better enlarger lenses are Planar derivatives for this reason.
 
jamiewakeham said:
Wah. I'm supposed to be good at physics... I think I'm reading this as follows. At smaller apertures, only the centre of the lens is involved in bending light rays to a focus on the film (because the outer parts of the lens are covered over by big metal leaves). They bring the light to a focus at a certain distance, and this is what the RF mechanism will indicate as 'correct'.

But at widest aperture, the outer parts of the lens are also at work. And they converge the light rays at a slightly different point (not sure why this might be but I suspect some sort of abberation that just can't be corrected out of these very fast lenses - I could well believe that it would be a Sonnar-formulation-specific problem). At widest aperture, most of the light going through the lens is indeed travelling through the outer parts of the glass simply because it's got a greater area than the centre, so overall the lens appears to be focussed in a different place.

Now, that might be correct or might be utterly spurious. What I can't figure is why didn't Zeiss set the RF to get wide-open and close-up right, given that further away and stopped down will pretty much be covered by DoF?

Jamie
Click to expand...

Jamie, you have this right . . .pretty much. What you suggest is what lens designers do for most lenses. Optimize for the shallowest DOF & increasing DOF will cover any focus errors at smaller apertures & longer distances.

There is no problem at f/4 & smaller apertures. I'm not sure why Zeiss chose to optimize for f/2.8 other than the fact that f/1.5 is a crap shoot for any lens with such challow DOF - especially at close range. The issue is that if the lens is optimized for f/1.5, you'll have problems at f/2.8. It can't be optimized for both.

You're are also correct that there are issues with the Sonnar design itself. Because it is an asymmetrical design, there are spherical aberrations which are both its charm & its source of problems. The design also results in asymmetrical DOF - most of it in front of the lsubject. I suspect that there is some truth as well to the speculation of sample variation. With DOF shifted to the front & with limitations inherent in the design, a small difference in adjustment an make for a big difference in results.
 
Last edited:
Huck, don't you mean the Sonnar's DOF is shifted in front of the 'subject', not the 'lens' due to the asymmetrical design?
 
awilder said:
Huck, don't you mean the Sonnar's DOF is shifted in front of the 'subject', not the 'lens' due to the asymmetrical design?
Click to expand...

Yes. Sorry, stupid of me. I was typing in a rush & had brain lock.:eek:

Thanks for the correction.:)

I've edited my post to make it easier to read correctly.
 
Huck your point about the asymmetrical optical design/ DOF shift and production tolerance reminds me of why I sent my ZM 50/2 Planar back a few years ago. The lens was potentially as sharp as any modern 6/4 Gaussian design but the lens had to be slightly refocused (towards infinity) in order to obtain maximum sharpness even though rf focus seemed spot on at infinity with my camera. I later learned from Erwin Puts review of the ZM lenses (seen here: http://www.imx.nl/photosite/comments/c016.html under MECHANICAL QUALITIES) that ZM lens design intentionally focus' more forward from the film plane than do Leica lenses due to Zeiss' philosophy on outward bulging of film. So, if that's the case, I would assume this might also help account for complaints of front focus with the Sonnar.
 
Last edited:
Good point. I assume it's Put's' theory to explain minor back focus difference when measured with his instruments. Overall he seems impressed ZM optics as much or more than Leica's in many cases.
 
awilder said:
Huck your point about the asymmetrical optical design/ DOF shift and production tolerance reminds me of why I sent my ZM 50/2 Planar back a few years ago. The lens was potentially as sharp as any modern 6/4 Gaussian design but the lens had to be slightly refocused (towards infinity) in order to obtain maximum sharpness even though rf focus seemed spot on at infinity with my camera. I later learned from Erwin Puts review of the ZM lenses (seen here: http://www.imx.nl/photosite/comments/c016.html under MECHANICAL QUALITIES) that ZM lens design intentionally focus' more forward from the lens plane than do Leica lenses due to Zeiss' philosophy on outward bulging of film. So, if that's the case, I would assume this might also help account for complaints of front focus with the Sonnar.
Click to expand...

Alan, thanks for the link. I have read this article many times before but glossed over the section you referenced. Very helpful.

Out of curiosity, I decided to review the previous Sonnar threads to see what cameras were being used by those who were complaining about the front focus issue. In every case but one - including the test report from Luminous Landscape - the camera being used was something other than a ZI. Cameras used included the following: R-D1, M8, MP, M6, & R2. The R2 should be similar to the ZI in terms of the issue described by Puts, but the only R2 used was by the Luminous Landscape review & that only as an auxiliary to the M8. I don't think that the results with the R2 got most of their attention. The one case in which a Zeiss Ikon owner found front focus issues was in the case of the test by FanMan, but while he documented the front focus in his test, it didn't seem to be an issue for him prior to the test & he just didn't seem to be complaining about in such a strong way as were the others.

So, I think you may be onto something here. Puts says it's only the difference of a few human hairs but that's at the lens/film plane itself; ithe difference will be magnified in a photo.

It was interesting to skim these threads. I realized that htis issue had been raised before, but I didn't pay a lot of attention to it. Your report of your test with the Noctilux was most interesting as was your reference to the Pop Photo article. Thanks for those contributions.

Huck
 
You must log in or register to reply here.
Back
Top Bottom