Fast Sonnar Clone Shootout - Canon, Jupiter, Nikkor

[furcafe]

Maybe improved coatings accounts for the revival of the Ernostar (e.g., 90/2.8 "Sonnar" in G mount), too.

Just find it interesting that in the 50s, glass was used in the middle of the front triplet, to avoid air/glass surfaces (and achieve the above mentioned flare resistance), while with today's coating technology, the C-Sonnar doesn't need this.

Roland.

 

[Roger Hicks]

When the light goes down today I'm going to try to shoot my next phase; open aperture focused at infinity. I always hear this or that lens is "optimized for [close up, infinity]" and I need to see it.

Do people mean, simply, that the lens is calibrated to be most accurately in focus on a rangefinders mechanism either near, or far? I just don't get it.

Some lenses deliver maximum resolution close up; some at infinity; and some at (for example) 3-5 metres. Even with floating elements it's difficult to optimize for more than one distance.

Rangefinder coupling and focus shift are separate issues from the above.

Cheers,

R.

 

[Mark C]

I agree Roger. But that is the question. Which are people talking about? My impression is that in these sonnar discussions people are talking almost entirely about focus accuracy.

 

[Roger Hicks]

I agree Roger. But that is the question. Which are people talking about? My impression is that in these sonnar discussions people are talking almost entirely about focus accuracy.

Mine too: sorry for 'explaining' what you already knew, but you can never be sure on line what someone else knows.

Cheers,

R.

 

[Heru Anggono]

All LTM Nikkor 50/1.4 copies are Sonnars.

Nikkor 50/1.5:

Nikkor 50/1.4:

The Olympic RF Nikkor 50/1.4 (all black, S mount only) was a double Gauss (as is the modern S-mount re-incarnation).

Nikon stretched the Sonnar design in 1950 to open it up to 1.4 and claim the fastest RF 50 record ....

Which I believe is where the flare comes from Garrett. If you close the Nikkor down less than half a stop, the veiling flare disappears (vignetting at infinity on full frame, too), and contrast and resolution pick up dramatically.

When you zoom in, say, f1.7 and up, you should see that the Nikkor has noticeably higher center resolution than the other two lenses.

Also, regarding a previous question, the Canon is typically optimized for middle apertures (f5.6 or so), while the Nikkor is optimized for f1.4. Of course, on an M body, you can control this via the use of a different LTM adapter. Then again the contrast changes so much when opening up further than f2, that focus shift is hardly noticeable for any of those lenses.

Cheers,

Roland.

While we're on the topic. Is there any perceptible visual difference between traditional Sonnar design (3 groups) compared to modern Sonnar design? Modern Sonnar such as:

- Zeiss 50mm f/1.5 ZM (6 elements, 4 group)
- Konica M-Hex 90mm f/2.8 (5 elements, 4 group)
- Contax G 90mm f/2.8 (5 elements, 4 group)
- Zeiss C/Y 85mm f/2.8 (5 elements, 4 group)

Perhaps moving from traditional 3 groups design to 4 groups removes the characteristic/signature of a Sonnar lens?

 

[goamules]

That's a good question, but I only have the classic versions of Sonnars.

 

[sevo]

Perhaps moving from traditional 3 groups design to 4 groups removes the characteristic/signature of a Sonnar lens?

The middle lens in the front group was supposed to be a anti-flare measure in pre coating days, and probably can be replaced by an air space with impunity, now that we multicoat.

The other design changes that they made will have more impact - even more so as optical designers have a more ugly name for "Sonnar characteristics": Design flaw.

 

[kermaier]

The M-Hex 90/2.8 is an Ernostar, not a Sonnar.

::Ari

 

[furcafe]

Hard to say definitively, since the newer lenses have different glass & are all multicoated.

Also, per kermaier, I think the Konica & CZ 90/2.8 are actually updated Ernostars (ancestor of the Sonnar, by the same designer, Bertele) &, as sevo indicated, are better corrected than the classic Sonnar. I have the CZ 90/2.8 in G mount, but haven't shot enough w/it over the years to form an opinion as to its Sonnar-like qualities. I have used the modern C Sonnar (in Nikon RF mount) extensively & it does seem to provide an updated (less flare, better-corrected) version of the Sonnar look.

While we're on the topic. Is there any perceptible visual difference between traditional Sonnar design (3 groups) compared to modern Sonnar design? Modern Sonnar such as:

- Zeiss 50mm f/1.5 ZM (6 elements, 4 group)
- Konica M-Hex 90mm f/2.8 (5 elements, 4 group)
- Contax G 90mm f/2.8 (5 elements, 4 group)
- Zeiss C/Y 85mm f/2.8 (6 elements, 4 group)

Perhaps moving from traditional 3 groups design to 4 groups removes the characteristic/signature of a Sonnar lens?

 

[thompsonks]

Could you please go back to JSU's early question and fill in the parts we don't know yet?

"It would help to know what camera, and if digital, are these out of the camera jpegs or RAW conversions, which would then prompt the question which RAW convertor is used?"

We know a 4:3 camera has given us images of center resolution, tone, and contrast. But I have old/new Sonnars and Nikkors, and their images resolve less and have gentler bokeh at wide apertures. Did some sharpening occur in making jpgs or in processing RAW files – or in jpg conversion to 'Save for Web'? ?

Anyhow thank you for comparing them. My personal preference so far is for the Jupiter and Nikkor. IMO the Canon looks too much like a modern lens that I'd use if I didn't like old Sonnars!

I appreciated the comment above, that the best evidence would come not from a cropped sensor, but from a chance to compare center with corners and with the whole image. My 1938 Sonnar has amazing center resolution, but has some difficulty in other respects.

 

[goamules]

I used a Panasonic G1, shot as JPGs, then posted on Flickr, which creates several sized JPGs. One size I linked to. I don't know what Flickr does when it creates them. If you go to my signature line link, you can find the original, larger images.

 

[thompsonks]

They look great in the large versions.

But again, and simply as JSU asked: "It would help to know ... are these out of the camera jpegs or RAW conversions, which would then prompt the question which RAW convertor is used?"

 

[Heru Anggono]

It would be great if we could include late version of Zeiss (Opton) 50mm f/1.5 Sonnar in the evaluation.

Just out of curiosity I double check on Zeiss Contax C/Y SLR lens:

- 85mm f/2.8 Sonnar
- 105mm f/3.5 Sonnar
- 135mm f/2.8 Sonnar

They're all 5 elements, 4 group design, so I guess most likely it is Ernostar as mentioned earlier, not actually a traditional Sonnar (3 group design).

 

[Lightshow]

I believe the Helios 103 is also a Sonnar.
Thanks for shooting the comparison, I'm looking for a Canon 50/1.5(&28/2.8)to go with my set(35/2.8, 50/1.8, 100/3.5)

 

[jim_buchanan]

Great thread. Thanks to OP. and, yes it is difficult to interpret a comment when the word "optimized" is used.

I have a later Sonnar Opton in Sony NEX mount that is very interesting, but it is not full frame ready. Unless, Sony releases a full frame NEX.

I know it's not a Sonnar design, but I'm finishing up a Nikon Millenium Leica M mount build and am looking forward to using it on a NEX and hopefully a M9 soon. You don't hear too much about this lens, I guess due to the S mount. If it even comes close to the performance of the ultra expensive Summilux 50 ASPH, I will be happy.

Sorry, for veering slightly off Sonnar topic.

Some lenses deliver maximum resolution close up; some at infinity; and some at (for example) 3-5 metres. Even with floating elements it's difficult to optimize for more than one distance.

Rangefinder coupling and focus shift are separate issues from the above.

Cheers,

R.

 

[raid]

The old 50mm testing that I did years ago still provides useful side by side comparisons (done with Roland). It includes the Opton 50/1.5 that someone here asked about.



http://ferider.smugmug.com/Technical/Raids-50mm-Tests

 

[gb hill]

It was Raid's test that made me buy a J3. Best of the lot in both examples. Properly shimmed (like mine is) to my Bessa R, thanks to Brian S. It's as good as the higher priced leitz lenses of the same era. Not a Sonnar but the I 26m is a killer lens too.

 

[raid]

Such "tests" are maybe not free of human {focus) error and/or lenses not being optimized, but they may be helpful for an overall feeling how some lenses perform. There was no digital camera used. Each lens had replicate test shots. Results were posted here, and people chimed in with their own experiences.

 

[gb hill]

Such "tests" are maybe not free of human {focus) error and/or lenses not being optimized, but they may be helpful for an overall feeling how some lenses perform. There was no digital camera used. Each lens had replicate test shots. Results were posted here, and people chimed in with their own experiences.

I agree. It really is individuals choice of preference. I'm quite pleased with my lenses. I prefer test to be made out in the real world.

 

[raid]

This is also how I feel about it, Greg.