1,5/50 Sonnar review

[Roger Hicks]

Dear Roger,

yes, the comparison with the lens baby was hyperbolic 🙂

Wide open, the Sonnar has much higher resolution in the center than
in the corners. For far distance shots (say 10m or so) that becomes very visible.
At f2.8, the corners start to "sharpen up". A lens optimized for 2.8 will
give you highest resolution at the widest aperture that it makes sense.
I think of it as giving me similar results for infinity shot than a 50/2.8
Elmar wide open. A lens optimized for 1.5 will give me lower performance,
for these typical "street" situations.

For me a lens optimized for 1.5 would therefore be much less useful.

I keep mentioning the M3 since using half the lower DOF mark (the f5.6 one)
of the M3 RF patch allows you to correct for wide open shift easily, without
taking the eye off the finder. So in a way, on older Ms, you can predict
the shift in the finder. That's how I use my Sonnar anyways. See

http://ferider.smugmug.com/gallery/2867043

The Sonnar has a 120 degree focus throw. When you look at its
RF cam it consists of 3 segments. Dr. Nasse confirmed to me that when
changing the Sonnar from 2.8 to 1.5 optimization, (1) the cam is rotated
by 180 degrees, (2) additional QA is performed.

Best,

Roland.

Dear Roland,

Sure, the Sonnar is soft at the edges at full aperture, but I can't imagine many circumstances in which I'd be shooting at infinity or even 10m at full aperture anyway -- and if I did (e.g. at night) I'd be unlikely to worry about soft edges.

In any case, coupling optimization isn't important at infinity: the infinity stop remains the same, so it won't matter if the lens is coupled at f/2.8 or f/1.5, or indeed uncoupled. D-o-f is such that in real-world pictures you can't really detect the focus shift beyond 3-5m anyway, even at f/1.5, i.e. this only matters with extreme close-ups.

Are you sure it is rotated 180 degrees? I just checked my Sonnar and it appears to have 3 cams, so 180 would place it in the middle of another cam. 120 degrees would seem more likely. Next time I see Dr. Nasse I'll ask; I suspect that those cams may be f/1.5, f/2.8 and f/5.6.

Cheers,

R.

 

[Roger Hicks]

Canon released the EF 50/1.2L a year ago, and the focus shift is driving photogs barmy. Of course, one can stop down and focus manually. 🙂

Well, now I do know...

Thanks for the information.

Cheers,

R.

 

[Huck Finn]

Can someone please clarify what is meant by "focus shift"? I thought it referred to the fact that if you focused the lens at one f-stop, and then changed the f-stop alone, the focal point would shift when it shouldn't. If that is indeed what we are referring to, Why does it matter? If I'm going to shoot at f1.5 I set that f-stop, focus and shoot. There is no focus shift. Unless we are talking about something else.

/T

Tolumne, you are correct in your definition of focus shift.

 

[Roger Hicks]

Tolumne, you are correct in your definition of focus shift.

Yes, if you are focusing through the lens at the shooting aperture -- which is not the case with a coupled rangefinder.

Cheers,

R.

 

[ferider]

Are you sure it is rotated 180 degrees? I just checked my Sonnar and it appears to have 3 cams, so 180 would place it in the middle of another cam. 120 degrees would seem more likely. Next time I see Dr. Nasse I'll ask; I suspect that those cams may be f/1.5, f/2.8 and f/5.6.

Yes, the 180 in my last sentence was a spelling, 120 degrees, of course.
And the third possible rotation, I would guess like you, is for f5.6,
like the old Canon 50/1.5 lenses.

Best,

Roland.

 

[Quercus]

Focus shift

Focus shift

All optics whatever their design exhibit focus shift - compound lenses are not just made up of bi convex lenses but are a mix of bi convex, plano convex/concave and even just plano elements. Therefore the instant that diameter of the elements effectively changes then so does the refractive index of the entire compound optic alters ever so slightly -this results in the point of focus (not the focal point) either moving ever so slightly forwards or backwards -

A single element lens does not exhibit this but then it exhibits so many aberrations that the relatively miniscule problem of focus shift for most lenses is just that - minimal

Off the top of my head i - i am adding up as i go from what i think i can remember the glass arrangement is - the 50 sonnar has 12 radii curves 6 thicknesses 5 airspaces and 6 glass types (glass is not a single constant material even schott) that is around 29 variables - sort of like a 29D bubble matrix each with different and differing refractive indexes that are not totally constant

So focus shift is really only to be expected - and as roger rightly say the cooke triplet not a computer generated mathematical matrix it is a beautifully realized artisan design that works

 

[Tom A]

Yesterday the weather was atrocious here so I decided to drag out some 50's and shoot a highly unscientific test with them. Flickr files are not good enough to judge lenses from (unless they are REALLY crappy), but I had fun and if you go to our Flickr site you will see 16 variations, ranging from 50f1.4 Asph, C Sonnar, Planar, DR Summicron, 75f2 Apo, 75/1,4, Heliar f2, Noktonf1.5, Color Skopar 50f2.5 and an old Summarit 50f1.5.
Film was an aodd choice, double XX in HC 110 (another experiment and if you like grain, it loos good).

 

[Tuolumne]

All optics whatever their design exhibit focus shift - compound lenses are not just made up of bi convex lenses but are a mix of bi convex, plano convex/concave and even just plano elements. Therefore the instant that diameter of the elements effectively changes then so does the refractive index of the entire compound optic alters ever so slightly -this results in the point of focus (not the focal point) either moving ever so slightly forwards or backwards -

A single element lens does not exhibit this but then it exhibits so many aberrations that the relatively miniscule problem of focus shift for most lenses is just that - minimal

Off the top of my head i - i am adding up as i go from what i think i can remember the glass arrangement is - the 50 sonnar has 12 radii curves 6 thicknesses 5 airspaces and 6 glass types (glass is not a single constant material even schott) that is around 29 variables - sort of like a 29D bubble matrix each with different and differing refractive indexes that are not totally constant

So focus shift is really only to be expected - and as roger rightly say the cooke triplet not a computer generated mathematical matrix it is a beautifully realized artisan design that works

Wow, this would make great marketing text for Zeiss. Just from that description alone I want to go out and buy one!

/T

 

[Dan States]

What bothered me about my Sonnar was not the performance at .7 meters, but at 3-5 meters, where the shift in focus clearly left the intended subject out of focus. Interesting that everyone is testing their Sonnar at minimum distance but not at typical shooting ranges. It seemed like the amount of focus shift was not a constant distance, but was a ratio of the focus distance, so that at longer distances the misfocus was up to a meter.

I'm happily using my 1950's Sonnar on my Contax IIA and have never had a problem nailing focus at any aperture. Guess I'll just keep hoping for one of those M adapters to land in my stocking!

Best wishes
Dan

 

[elude]

What bothered me about my Sonnar was not the performance at .7 meters, but at 3-5 meters, where the shift in focus clearly left the intended subject out of focus. Interesting that everyone is testing their Sonnar at minimum distance but not at typical shooting ranges

0.9meters ;-)
It's a portrait lens don't be astonished!

 

[summaron]

Tom A
I enjoyed your rainy day tour of lenses. What I notice again and again about the Sonnar--as opposed to symmetrical designs--is that the Sonnar's out of focus shapes are always sculptural or volumetric. The out of focus background shapes with the Planar, in comparison, are smooth and flat, sometimes a little bland. However, sometimes the Planar can make the background tonal variations look like those in a solarized print, which is pleasing.

The Sonnar and the rigid Summicron--a little less romantic than the Sonnar but nicely three dimensional--I'd have to say are my favorites.

I'll add to Quercus' and Roger's comments that according to Kingslake, the Goertz Dagor famously had focus shift, but was quite popular despite this. Also when the front element of the Tessar moved as it was focused, certain aberrations increased, so it had to be over corrected for this at one end of its movement or the other.

Kingslake talks about "degrees of freedom" a lens designer has--only four or five into which he has to fit in his whole laundry list of design wishes.

James

 

[larmarv916]

Fabulous 50's...Creative Results from Optical Designs

Fabulous 50's...Creative Results from Optical Designs

Yesterday the weather was atrocious here so I decided to drag out some 50's and shoot a highly unscientific test with them. Flickr files are not good enough to judge lenses from (unless they are REALLY crappy), but I had fun and if you go to our Flickr site you will see 16 variations, ranging from 50f1.4 Asph, C Sonnar, Planar, DR Summicron, 75f2 Apo, 75/1,4, Heliar f2, Noktonf1.5, Color Skopar 50f2.5 and an old Summarit 50f1.5.
Film was an aodd choice, double XX in HC 110 (another experiment and if you like grain, it loos good).

Thanks Tom for the great demonstration of creative differences that result from the pure technical design differences of each of these lenses. Over in the Rangfinder " Optical & Technical" forum I tried to get people to think and look at how each computation of a so called " normal" lens really offers many creative benefits for RF Shooters.

You discussion really helps drive home proof of what results when deeper understanding of a lenses "stylistic" differences are put to work. Thanks. So many people do not seem to see each lens...regardless of age as paint brushes of imagery. Also the major creative choices in both Color and B/W that delivers when matched with a particular film and developer.

The inclusion of the 75m lenses was a good idea as well.

 

[Quercus]

I rather tend to agree with Tom that for me the best all round "normal" currently available for M RFs is the Planar - like its predecessors in whatever guise(my personal favourite being the Rollei QMB HFT 1.8 4pin version - the colour rendition with the right reversal is beautiful) it is a lens that just works. In most lowish light conditions I rarely find I need the one extra stop of of a 1.4/1.5 - but again it depends upon what film is in there at the time - if its colour neg definitely no problems - 400 plus bw again ok - just reversal that can be tricky at 1/30 at f2, but then I rarely use reversal at low light its a waste of its wide gamut.

Also the price difference at about 300 euros is well a lot of film

Planars are very special lenses don't under-rate them

 

[dpetrzelka]

I would really like to try the 2/50 Planar and the 1,5/50 Sonnar ZM -
anyone looking for a 2,8/28 Biogon?

 

[Huck Finn]

Yes, if you are focusing through the lens at the shooting aperture -- which is not the case with a coupled rangefinder.

Cheers,

R.

. . . Huh?

 

[Roger Hicks]

. . . Huh?

You don't focus through the lens with a rangefinder camera. The rangefinder is a mechanical contrivance coupled to the focusing mount. If there is focus shift, the RF can only be set to correspond to the actual focused distance at one aperture.

Or were you asking something else? 'Huh?' is not very informative...

Cheers,

R.

 

[kshapero]

some new pics with C Sonnar

some new pics with C Sonnar

 

[larmarv916]

Thanks Tom for the great demonstration of creative differences that result from the pure technical design differences of each of these lenses. Over in the Rangfinder " Optical & Technical" forum I tried to get people to think and look at how each computation of a so called " normal" lens really offers many creative benefits for RF Shooters.

You discussion really helps drive home proof of what results when deeper understanding of a lenses "stylistic" differences are put to work. Thanks. So many people do not seem to see each lens...regardless of age as paint brushes of imagery. Also the major creative choices in both Color and B/W that delivers when matched with a particular film and developer.

The inclusion of the 75m lenses was a good idea as well.

Here is another example from the Koln Roman Museum, shot wide open with the zeiss sonnar 50mm at F1.5. Shot throught the glass of the display case and hand held at 1/125th of a second. Where all we wanted was the detail on the nearside of the small head. My old summilux 50mm would not have delivered this level of detail at the same close focal distance.

 

[Huck Finn]

You don't focus through the lens with a rangefinder camera. The rangefinder is a mechanical contrivance coupled to the focusing mount. If there is focus shift, the RF can only be set to correspond to the actual focused distance at one aperture.

Or were you asking something else? 'Huh?' is not very informative...

Cheers,

R.

I completely understand that, but what does that have to do with the definition of focus shift that Tolumne stated?

 

[fuwen]

I completely understand that, but what does that have to do with the definition of focus shift that Tolumne stated?

In practical sense for SLR u can confirm the focus shift by refocusing with the lens aperture stops down. However for range finder u will not be able to do this final focus shift check.