The PC Super-Angulon 28mm f/2.8 by Schneider-Kreuznach (and Leica): a Deep Dive

[rdeloe]

I couldn't find a lot of information about this lens when I was thinking about getting one. I'm glad I did because it exceeds my expectations. No, it's not an HR Digaron-S 28/4.5! But I can't use one of those on my outfit so this is the next best thing.

For anyone else who is looking at this lens, here are some user impressions I posted at the other medium format forum. https://www.dpreview.com/forums/thread/4676611

I'm happy to answer any questions, or address any errors or glaring omissions.

Rob

 

[Alkibiades]

Schneider PC 28 mm was the best shift lens for 35 mm cameras, but it was develope for film.
I used this lens very often with my Canon 5D- the lens was much better as canon TS-e 24 mm verion I.
What i disliked was the strong distortion and the poor edges at max movents - bus still better than canon.
The things changed the new version of the 24 mm: Canon TS-e 24 mm verion II, that is a real modern, for digital use designed wide angle tilt-shift lens.
is has no visible distortion, is extramly sharp, the performance at the max edges is the best of that kind of lenses- but still not perfect.
I use the 24 II on Fuji GFX100S and if you dont shift to extreme the perfomance is really good.
But If you look for even better lens just take canon ts-e 50 mm. This is my favorite lens with performance like Digaron-W 50 mm ( but not such huge image circle).
also the other new TS-e lenses 90 mm and 135 mm are superior on GFX.

 

[MGrayson]

Art Director: Why is there only half of the building?
Photographer: Well, the 50mm lens was sharper than the 24mm...

(Disclaimer: I have a lot of glass that was purchased because it was good, not because I needed or wanted the focal length.)

 

[rdeloe]

I haven't used any of the Canon t/s lenses, except the old FD 35, which was amazingly good for its time. But I've heard good things about the Mark II 24mm, and reports about the new 50mm confirm that it's superb on GFX. Unfortunately, because Canon dropped the aperture ring and made the EF lenses electronic, I can't use these on my current MAB Cam. When my F-Universalis shows up I could use EF lenses if I was willing to pay over USD$1,000 for the lens board (ouch). But that's a lot of money for a focal length I don't use a lot (24mm).

Of course the other major design flaw of the 24mm Canon is that it's not 28mm. I needed a 28mm lens, not a 24mm lens. hence the Leica PC Super-Angulon 28/2.8.

 

[Alkibiades]

I haven't used any of the Canon t/s lenses, except the old FD 35, which was amazingly good for its time. But I've heard good things about the Mark II 24mm, and reports about the new 50mm confirm that it's superb on GFX. Unfortunately, because Canon dropped the aperture ring and made the EF lenses electronic, I can't use these on my current MAB Cam. When my F-Universalis shows up I could use EF lenses if I was willing to pay over USD$1,000 for the lens board (ouch). But that's a lot of money for a focal length I don't use a lot (24mm).

Of course the other major design flaw of the 24mm Canon is that it's not 28mm. I needed a 28mm lens, not a 24mm lens. hence the Leica PC Super-Angulon 28/2.8.

I prefer personally also not too wide wide angle- to get more naturalistic look. so for me 28-35 mm would be better than 24 mm. I hope some day Fuji 30 mm shift ..
I use canon TS-E lenses on Fuji with easy cheap adapter.

 

[Alan]

Did you consider the old 28mm/2.8 Digitar since you seem to be OK with the distortion? It would eliminate the need for hacking - or maybe that's half the fun.

 

[Alkibiades]

Did you consider the old 28mm/2.8 Digitar since you seem to be OK with the distortion? It would eliminate the need for hacking - or maybe that's half the fun.

The old 28mm/2.8 Digitar is the Schneider PC 28 . these are the same lenses.

 

[rdeloe]

Did you consider the old 28mm/2.8 Digitar since you seem to be OK with the distortion? It would eliminate the need for hacking - or maybe that's half the fun.

I did consider it Alan. However, the PC Super-Angulon version has close focusing elements that really make a difference. Those are not present in the Digitar version.

 

[rdeloe]

The old 28mm/2.8 Digitar is the Schneider PC 28 . these are the same lenses.

I thought that too. However, I no longer believe that is true because the PC Super-Angulon has a close focusing element, and it makes a huge difference. I tested how important it is by using the PC S-A as a unit focusing lens; it's terrible at the edges and corners if you use it like the Digitar, which is a simple unit focusing design.

 

[Alkibiades]

the lens - the optical part- was simply puted in different mount.
If you look for the last version of the 28 mm PC-TS that a fully different lens. it based on mamiya -phase one 28mm, that was also not the best lens, but with bigger image circle.
Schneider improved a little the mamiya version. I talk with Schneider technician personally about these lenses...

 

[rdeloe]

This is the lens block diagram for the PC Super-Angulon. Note the floating element. It moves when you turn the focusing ring.


This is the lens block diagram for the Digitar. It is similar, but not the same.




How would that floating element move in the Digitar version given that the front cell and the back cell are screwed into the Copal shutter, and the entire lens is unit focused? Did the Schneider person you talked to explain that?

I should emphasize again that I can empirically prove that the floating element makes a big difference... On my camera I can focus the lens as a unit, or using the focusing helicoid. It makes a huge difference. If the Digitar does not take advantage of the floating element, then it is not the same lens even if the cells are similar.

 

[Alkibiades]

I see the same lensdesign.
both lenses have also the same performance.
If the one version have a floting element so this can improve image quality in other distance, so here in closer distance. but in the classic use as a lens for architecture, landscape optimized
on longer distance i cant see a use of floting elements.
Its fine when it makes a differance in your work.

 

[Makten]

I have the lens in C/Y mount and I agree that it's pretty good. However, my main gripe with it is the quite pronounced cyan color cast towards the periphery.
If you use the lens unshifted, it's easy to get rid of with an LCC profile. But for shifted images you'd have to make a new profile for virtually every image.

Edit: I didn't know it had a floating element, so I might want to check the infinity calibration with my adapter. If it's off, there is probably some improvement possible.

 

[rdeloe]

I have the lens in C/Y mount and I agree that it's pretty good. However, my main gripe with it is the quite pronounced cyan color cast towards the periphery.
If you use the lens unshifted, it's easy to get rid of with an LCC profile. But for shifted images you'd have to make a new profile for virtually every image.

Edit: I didn't know it had a floating element, so I might want to check the infinity calibration with my adapter. If it's off, there is probably some improvement possible.

I have not noticed a pronounced cyan cast on my copy, but I'll keep an eye out for it. I wonder if your copy has taken a knock and something is out of alignment. Let me know if you want to see some 50R RAFs from mine so you can compare.

It's extremely sensitive to being at the right flange distance if you're working at close distances, where the floating element is engaged. I haven't figured out the exact threshold yet, but distances of 1 m definitely wall within it. I suppose that if someone only focuses at long distances and doesn't care about what's happening in the corners in the foreground, it will matter less. But that's not me.

 

[Makten]

I have not noticed a pronounced cyan cast on my copy, but I'll keep an eye out for it. I wonder if your copy has taken a knock and something is out of alignment. Let me know if you want to see some 50R RAFs from mine so you can compare.

You can see it in the photo of the maple leaf, and the brick wall shots. It's not always apparent, and more visible in certain types of light. It took me a while to notice it in my own photos, but once you see it, it's difficult to unsee.
I really haven't used mine since I got the GF 30. Which I don't use much either, it's still too wide for me.

It's extremely sensitive to being at the right flange distance if you're working at close distances, where the floating element is engaged. I haven't figured out the exact threshold yet, but distances of 1 m definitely wall within it. I suppose that if someone only focuses at long distances and doesn't care about what's happening in the corners in the foreground, it will matter less. But that's not me.

If it makes a difference at close range, it should at long range too. At least if it's not in the proper position.
Since most adapters are too short on purpose, mine is probably too. I'll have a look when there is light and I have time.

 

[rdeloe]

You can see it in the photo of the maple leaf, and the brick wall shots. It's not always apparent, and more visible in certain types of light. It took me a while to notice it in my own photos, but once you see it, it's difficult to unsee.
I really haven't used mine since I got the GF 30. Which I don't use much either, it's still too wide for me.

Ah yes, I recall you really see it.

I don't know if we're talking about the same thing, but what I have noticed is a bit of chromatic aberration at high magnification in some textures. It does not clean up with Lightroom automatically, and there's not enough for me to sort it out with the manual dropper. I don't find it bothersome, but some might.

If it makes a difference at close range, it should at long range too. At least if it's not in the proper position.

I've tested at long range, and it seems to work like the close focus element in my Pentax-A 645 35/3.5, in other words, it's only activated once you get to roughly the 1m point on the focus ring. To see this in action, you need to be able to focus the lens both ways: (1) by putting it at the correct flange distance and focusing with the helicoid, and (2) by setting it to infinity and focusing with the rail. I've done that test, and it's clear to me that at longer distances, the close focusing element is not brought into play.

This is the point where Alkibiades and I were disagreeing. He's arguing that the Digitar and this version are the same lens even though in the Digitar the close focusing element is fixed but it's mobile in the PC Super-Angulon 28/2.8. I argue that the differences is important enough to make them different lenses even though the rest of the optics are the same. That we disagree does not matter. Where we agree is that the close focusing element only does it's thing up close. I just don't know where exactly the cutoff is, i.e., is it 1 m, 1.5 m, 2 m? I'll find out.

Since most adapters are too short on purpose, mine is probably too. I'll have a look when there is light and I have time.

I predict you'll discover it's fine. I have used the lens with its native shift mount and a Fotodiox R to GFX adapter, and on my current digital view camera, which requires me to guestimate the flange distance empirically. Image quality at close distances was slightly better on the Fotodiox adapter.

 

[Makten]

Ah, for some reason I assumed that the movement of the CRC element was linear, but of course it could "stay put" for most of the travel and then start to move when approaching MFD. That is surely the case in many modern lenses with plastic helicoids, but it might be in some older lenses too.

I've been a bit bothered by lateral CA, but at least it's possible to clean up in C1, when the lens is on axis.

 

[John Black]

I think Makten and I share some of the sensitivities when it comes to labeling lens performance I've had two of the Leica 28mm F2.8 PC's. They are okay up to a point - which is 5 to 7mm of shift. Beyond that things can get really messy - like the CA performance. But it also depends on the image. If the picture was of tree with a bunch of leaves, it would be a mess. But if it's buildings and sky, it's not great, but it's not as offensive as the tree leaves likely would be. Unshifted the result on the Leica M11 (60 MP) is okay. It doesn't have the crisp sharpness of my Leica M ASPH lenses (28, 35, etc.)

One niggle I have with the lens is its barrel distortion. That's always a tug-of-war to try to get a pleasing fix in C1 because it's a mash up of distortion and movement corrections to get something that is aesthetically correct. I've given up going for "technically" correct. Aesthetically, its sun-stars suck.

To use filters on it, buy the B+W 67-77mm step up ring. It's the only one I've found that fits. From there I use a 77-105mm step up ring and 105mm and can do full shifts without any mechanical vignetting (2 filters w/ B+W WA 105mm metal lens hood). Leica M10-M w/ polarizer + dark red filter + lens hood -



These were shot a week or two apart -

28mm PC / Leica M11 -


Rodenstock / IQ3 -


This isn't too demanding of a shot - I think it takes ~7mm of shift depend on high / low the tripod is set-up. In terms of the geometry and the distortion, the Rodenstock runs circles around the Leica 28mm PC - and it damn well should given the cost. On the flip side, I can execute a shot with the Leica M11 so, so much faster. Another curiosity in the mix is shooting the 28 Lux on the smaller sensor, thus is has more DOF whereas with the Rodenstock I have to play with tilt.

 

[rdeloe]

Thanks for the feedback and the tips on attachment options John. That's very helpful.

I can say that without a doubt the Leica PC Super-Angulon 28/2.8 is the best 28mm lens that allows for tilt, swing and a bit of shift on my camera... Bar none. Unfortunately, that doesn't mean it's an excellent lens. It has major limitations, which you've summarized in your post, and which I go into into considerable detail in my "deep dive" at DPReview.

Alas, there are no other good options I can use with a GFX camera and the digital view camera setup I use. Even if I could afford an HR Digaron-S 28/4.5, that lens does not work on my outfit.

Assuming the price isn't ridiculous, I could get a Fuji GF 30/5.6 tilt-shift lens when they become available in 2023. But I don't want to because it won't be small, it won't be light, it's not the focal length I wanted (but it is very close, I grant), and most importantly it doesn't mount on my current or future digital view camera setup. Were I an architectural photographer who needed a lot of shift, the GF 30/5.6 t-s would be a no-brainer. But that's not what I do, so here we are.

Thankfully, my need for 28mm is occasional. I'm just back from a full day of shooting and I needed the 28mm for two scenes, but everything else was 35mm, 50mm, 90mm and 150mm today.

Nothing to write home about... but for this project, I needed this picture with this perspective. Warts and all, the PC Super-Angulon 28/2.8 came through after a bit of rise, some swing, and some tilt.

 

[John Black]

Cough, cough... about the best 28mm with shift, etc... perhaps not a totally fair statement. If you want to search for a needle in haystack there is a Schneider 28mm PC-TS. From what I've read, only 100 were made before being discontinued. Their price was crazy, $6000 or $8000. No wonder they didn't sell, right? But if you can find one, you will get "better".

It's kind of a crazy lens to use (I had the 90mm). It only shifts in one direction. It only tilts in one direction. So lots of rotating to get things in the mix needed for the shot. It was available in different mounts. Here's a link I found with a quick search - https://www.bhphotovideo.com/c/prod...6_1075956_pc_ts_super_angulon_28mm_f_4_5.html

Another lens worth searching out (and MUCH easier to find) is the Contax 35mm F2.8 Distagon PC. I don't think there is as much value in a 35mm shift lens vs 28mm, but it's a better lens vs the Schneider. Still quite compact.

If you're contemplating adapting to the Fuji, I'd be more inclined to go with the Contax. It'll end up as a 28mm FOV (assuming .8 crop factor is applied for some quick math).