The GetDPI Photography Forum

Great to see you here. Join our insightful photographic forum today and start tapping into a huge wealth of photographic knowledge. Completing our simple registration process will allow you to gain access to exclusive content, add your own topics and posts, share your work and connect with other members through your own private inbox! And don’t forget to say hi!

Technical Camera Images

mristuccia

Well-known member
Marco, I was just about to press post on my reply and saw that Warren has already answered your question.

Here is my version:

Yes, the adjustment to the infinity stop position is straightforward.

On the SK focus helicoid, there are three small Phillips-head screws located at 120° around the focus collar, just forward of the rubber grip.

Loosen the three screws, then grip the focus grip firmly and turn it against the infinity stop, so that the infinity symbol is pushed a small amount, such as 5 to 10mm, beyond its previous position at the infinity stop. That amount will provide you with the ability to find the focus for a subject at infinity by rocking the focus ring back and forth and viewing the image on the live view.

Finally, secure that adjustment by retightening the three small screws.

Two explanations are even better than one. Thank you Rod!
 

mristuccia

Well-known member
You can shift 13mm with the 3,5/60 lens? 😎.

I may remember wrong as I did not take a written note, but according to the following table it can do +/- 10mm on a 55x55mm "film" sensor. So, think it can do 13mm on a 44x33mm sensor. Its IC diameter must be greater than 75mm. And I think I even reached 15mm shift once, along the long side.

Screenshot 2025-07-14 at 23.51.05.jpg
 
Last edited:

diggles

Well-known member
I may remember wrong as I did not take a written note, but according to the following table it can do +/- 10mm on a 55x55 "film" sensor. So, think it can do 13mm on a 44x33mm sensor. Its IC diameter is 75mm. And I think I even reached 15mm shift once, along the long side.

View attachment 222259

The distortion seems to be well controlled. Did you have to apply any software corrections? Or is that way straight out of the camera?
 

mristuccia

Well-known member

Actually, I'm surprised myself.
According to the documentation there should be from 1% to 2% of distortion. At my shift level it should have reached 1.5-1.6%.
Maybe the dark areas surrounding the subject are hiding the distortion.

Screenshot 2025-07-15 at 00.10.00.jpg
 
Last edited:

mristuccia

Well-known member
This is the Distagon 3.5/60 CF at 17mm horizontal shift and 7mm rise, an image that I discarded of course, but which I'm putting here as an example of its capabilities.
You can clearly see the vignetting on the right due to reaching the IC limit.

20250713_BERLIN_AroundWedding_B_4272_v1_CL_GetDPI.jpg
Cambo WRS-1600 | CFV-100c | Distagon 3.5/60 CF | 17mm horizontal shift, 7mm lens rise

And this is a 100% crop of the upper-right area:

20250713_BERLIN_AroundWedding_B_4272_v1_CL_GetDPI_100percc.jpg

I think it still keeps up incredibly well at its maximum horizontal shift level for an old design. No additional sharpening added in this crop, just Phocus default. I've only enabled the "Adaptive" chromatic aberration correction which automatically removes some purple fringing. Kudos to Zeiss!

I need to try shifting on the other side, sometimes performance are not equal.

Edit: I forgot to say that this is at f11
 
Last edited:

diggles

Well-known member
Actually, I'm surprised myself.
According to the documentation there should be from 1% to 2% of distortion. At my shift level it should have reached 1.5-1.6%.
Maybe the dark areas surrounding the subject are hiding the distortion.

View attachment 222260

I’d seen that same chart, which is why I asked. Even in the second example you posted, the distortion is impressively well controlled and the corners look excellent—kudos to Zeiss indeed! It would be interesting to see the opposite side as well, though that would really just tell us about your specific copy.
 

mristuccia

Well-known member
I’d seen that same chart, which is why I asked. Even in the second example you posted, the distortion is impressively well controlled and the corners look excellent—kudos to Zeiss indeed! It would be interesting to see the opposite side as well, though that would really just tell us about your specific copy.

Curiously, I was not happy with the performances of this lens when I used it on my 500CM and CFV-50c Mk I years ago. It must have been an issue with the photographer rather than the lens. Unless lenses get better over time like wine. :)
 

rdeloe

Well-known member
My APO-Digitar 35/5.6 does not fit on my MAB Camera, so I went back into my lens drawer and fished out the Leica (Schneider-Kreuznach) PC Super-Angulon 28/2.8 that I set up a couple years ago. It's widely known that it's not great and has a tiny image circle. But there's really nothing else a bit wider than 28mm that is remotely as good. (If I liked really wide lenses I'd hack one of the Laowa 20mm tilt-shifts, but that's way too wide for me.)

Long story short, what a pleasant surprise. Yes, it has a tiny image circle, and yes, it has to be used at f/11 to get best results. But f/8 is decent if I'm not shifting, and f/11 is pretty good with a bit of shift. This is the Guelph Civic Museum, shot with ~3mm of rise at f/11. It's ground zero for summer camps for children, hence the chalk drawings on the pavement.

R. de Loe GFXE9224.jpg

The lens got dramatically better once I started exploring the position it needed on the rail of my camera. In theory, you set it at the lens' flange distance, and focus by rail. That's what I did previously, and it turns out that's why results were mediocre. I discovered that setting it closer than it technically should be produces much better results. This is what infinity looks like on the focus helicoid when the lens is locked at the optimum position on the rail. Bizarre. The only explanation I can think of is it's the thick cover glass on GFX cameras yet again messing with a wide angle lens, but this one is a retrofocus design so that really shouldn't be the reason.

Infinity.jpg

I'm sufficiently intrigued that I'm going to try a Schneider-Kreuznach WA-Digitar 28/2.8. Supposedly that's the same optics in a different format. I'm curious to see if careful shimming can squeeze out a better performance than the Leica PC Super-Angulon version is giving. It would be nice to have what I'm getting at f/11 at f/8. I'm also interested to see how that optical formula does without a close focusing helicoid system (which the PC Super-Angulon has).
 

corvus

Active member
I had the dealer show me both versions as I had it in the back of my mind as an option if I stayed with full frame. The sharpness was ok - basically similar. He said you can still get something out of it, depending on the specimen. However, what jumped out at me again at first glance when shifting was the moustache distortion. I am not able to correct this with moderate effort. In the picture you posted, Rob, this does not come into play, so it is certainly not an exclusion criterion.
 

rdeloe

Well-known member
I had the dealer show me both versions as I had it in the back of my mind as an option if I stayed with full frame. The sharpness was ok - basically similar. He said you can still get something out of it, depending on the specimen. However, what jumped out at me again at first glance when shifting was the moustache distortion. I am not able to correct this with moderate effort. In the picture you posted, Rob, this does not come into play, so it is certainly not an exclusion criterion.

In the centre, the sharpness of this lens is very good in my view. The problem is the edges. They improve dramatically across the frame for the radial lines, but are still bad for tangential. Interestingly, the chart for the vaunted and rare 60 XL looks crappy too, in my view, but people who have it sing its praises. As always, you just have to see for yourself!

MTF.jpg

Regarding the moustache distortion, it can get quite nasty and hard to correct. It's not bad in this image. I have some others from the same session that don't look as good.

Honestly and truly, this is a terrible lens for architecture! The image circle is small, and the distortion is strong. However, for my purposes, which almost never involve architecture, it's growing on me. I don't think I'll carry it with my main kit (F-Universalis) because I'm fine with the 35mm APO-Digitar for my widest wide. But on my MAB Camera, 50mm as the widest was a bit constraining. The nice thing about a 100 MP sensor is I can use 28mm with the intention of cropping in a bit; that takes care of issues on the edges.

Mind you, if I close it down to f/11 (or even f/9.5), it's good right across the frame. This was just a test of swing performance. It's completely fine. Sharpness is good everywhere it needs to be. There was a tiny bit of purple fringing on those dormers at the top left, but it tidied up with a click. Considering that there is, for me, no alternative, this will do.

R. de Loe GFXE9188.jpg
 

Whisp3r

Well-known member
0001-P0007071-Edit-2-Edit-2-Melvinkobe-Photography.jpg
0002-P0007081-Edit-Edit-Melvinkobe-Photography.jpg

The Tondeur Diffusion Tower, located in Anderlecht, Belgium. Designed by architect François Roos, built in 1965-1967.
First image: frontal view, Arca-Swiss RM3di, IQ4 150, and SK 60XL at almost 30mm rise.
Second image: side view, Arca-Swiss RM3di, IQ4 150, and Sinaron 5.6/90 at 30mm rise.
 

rdeloe

Well-known member
I wanted a wider lens for my MAB Camera outfit. The standards don't come together close enough to let me use my Schneider-Kreuznach APO-Digitar 35/5.6, so I went back to a Leica PC Super-Angulon 28/2.8 that I rebuilt in 2022. I was never particularly happy with this lens. It's terrific in the middle, but it falls apart quickly as you move away from centre, and it only shifts ~3-4mm on a GFX camera before hard vignetting appears. But needs must as they say; this was the only option.

After a lot of messing around and testing, I discovered that better performance is available if I put it closer on the rail than it should be based on its actual flange focal length. It doesn't make sense that it's the GFX cover glass issue again because this is not a symmetrical lens. I don't understand, but here we are.

This is the kind of lens that if you only judge it against test charts, you will never be happy. However, I've learned that the only way to really know if a lens is going to be good enough for your purposes is to use it for your purposes! So I went to downtown Guelph, near the River Run Centre (a performing arts facility). It's a fun area because in one scene you can find remnants of the historic buildings, a janky looking car wash, and a very fancy condo building. The car wash is a leftover from the 1950s from the looks of it. I'm convinced the owners painted it that garish yellow to annoy the fancy condo people.

R. de Loe GFXE9493.jpg

Long story short, if you respect its limits, you can squeeze a good performance out of the lens. This is processed of course, but with 3mm of rise, some swing, and f/13, I was able to get everything in focus. If you had the full resolution version, you'd see that the top left corner is a bit mushy, but not so much that it can't be tidied up in Lightroom.

I liked the lens well enough to give its cousin the Schneider-Kreuznach WA-Digitar 28/2.8 a chance. Assuming Canada Post doesn't go on strike again, it should land next week. Schneider-Kreuznach built the PC Super-Angulon 28/2.8 for Leica, and the optical formula is, apparently, nearly the same. The key difference is the PC Super-Angulon has a floating element design, whereas the WA-Digitar is a simple technical camera lens. I'm hopeful that with careful attention to cell spacing, I may be able to eke out as good or better from the WA-Digitar, in which case I'll be quite happy because it allows me to focus by rail, whereas the PC Super-Angulon is focused by the helicoid.

A couple more examples, both with ~ 3mm rise. The second one also has a bit of swing. Both of these are f/11.

R. de Loe GFXE9477.jpg

R. de Loe GFXE9467.jpg
 
Top