rdeloe
Well-known member
Bingo. If it has moustache distortion, then the case for me putting that much money down to replace my perfectly good Pentax-A 645 35mm just got a lot weaker.
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!
APO SILVETAR f5,6 / 35mm
- Thread starter diggles
- Start date
The killer feature for me of the Contax 645 35mm is that Capture One has a distortion profile - and the corrections work shifted. It's kind of a huge lens though.
4x5Australian
Well-known member
Paula at Linhof & Studio Ltd has uploaded additional material, supplied by Silvestri, relating to the Apo-Silvetar 5.6/35mm lens:
(a) a set of six images from a Fujifilm GFX 50 camera, in both raw and TIFF formats.
(b) a PDF of technical data comprising a graph for relative illumination and two graphs for MTF.
The downloads are available from the order page here: Silvestri Apo Silvetar 5.6/35 mm (linhofstudio.com)
The individual images can be quickly previewed on the Google Drive page by clicking on the first filename, and thereafter advancing to the next image using the arrow button on the right.
My first impressions are:
So far I've looked at the preview images only to judge the amount of distortion. The first image shows surprisingly little distortion in the vertical columns. The second image shows moderate barrel distortion in the horizontals. Subsequent images are obliques and do not help much in evaluating distortion.
According to the graph of relative illumination, the illumination falls to 60% at the midway point and to 17% at the edge of the field radius of 37.5mm. The mid-point value of 60% is typical. The edge value of 17% is more than 2 stops, but similar to that seen in the Digaron-S 35mm. A centre filter may be beneficial. Note: The aperture used in the test is not specified.
Two graphs of MTF are given. Unfortunately, neither graph is labelled with the lens aperture used, so until we are told, we will have to guesstimate. One graph specifies the field radius in mm and the other does it in degrees, but that difference is inconsequential.
The first MTF graph, on page 2, is new to us. It shows very high centre performance, similar to the Digaron-S 35mm lens at f/5.6, and much weaker performance in the mid-field and edge. Accordingly, my guess is that this graph shows the MTF of the Apo-Silvetar 5.6/35mm lens wide open at f/5.6.
The second MTF graph, on page 3, is the one we've already seen in the lens brochure. It shows slightly lower, though still very high, centre performance that extends well into the mid-field, similar to the Digaron-S 35mm f/4 lens at f/8, with very good performance at the edge. Accordingly, my guess is that this graph shows the MTF of the Apo-Silvetar 5.6/35mm lens at a similar f/stop, likely either f/8 or f/11.
Whatever the latter aperture is, this excellent resolving power - at an affordable price - is what this lens is all about.
I've sent a request for some further details to Silvestri.
Rod
(a) a set of six images from a Fujifilm GFX 50 camera, in both raw and TIFF formats.
(b) a PDF of technical data comprising a graph for relative illumination and two graphs for MTF.
The downloads are available from the order page here: Silvestri Apo Silvetar 5.6/35 mm (linhofstudio.com)
The individual images can be quickly previewed on the Google Drive page by clicking on the first filename, and thereafter advancing to the next image using the arrow button on the right.
My first impressions are:
So far I've looked at the preview images only to judge the amount of distortion. The first image shows surprisingly little distortion in the vertical columns. The second image shows moderate barrel distortion in the horizontals. Subsequent images are obliques and do not help much in evaluating distortion.
According to the graph of relative illumination, the illumination falls to 60% at the midway point and to 17% at the edge of the field radius of 37.5mm. The mid-point value of 60% is typical. The edge value of 17% is more than 2 stops, but similar to that seen in the Digaron-S 35mm. A centre filter may be beneficial. Note: The aperture used in the test is not specified.
Two graphs of MTF are given. Unfortunately, neither graph is labelled with the lens aperture used, so until we are told, we will have to guesstimate. One graph specifies the field radius in mm and the other does it in degrees, but that difference is inconsequential.
The first MTF graph, on page 2, is new to us. It shows very high centre performance, similar to the Digaron-S 35mm lens at f/5.6, and much weaker performance in the mid-field and edge. Accordingly, my guess is that this graph shows the MTF of the Apo-Silvetar 5.6/35mm lens wide open at f/5.6.
The second MTF graph, on page 3, is the one we've already seen in the lens brochure. It shows slightly lower, though still very high, centre performance that extends well into the mid-field, similar to the Digaron-S 35mm f/4 lens at f/8, with very good performance at the edge. Accordingly, my guess is that this graph shows the MTF of the Apo-Silvetar 5.6/35mm lens at a similar f/stop, likely either f/8 or f/11.
Whatever the latter aperture is, this excellent resolving power - at an affordable price - is what this lens is all about.
I've sent a request for some further details to Silvestri.
Rod
Steve Hendrix
Well-known member
I downloaded the raw files. From what I can tell - as much as I was intrigued by this lens, I don't see anything in these images that encourages me. I do see some distortion (pretty easily corrected). But what I don't see is anything that I could describe as bitingly sharp. And a lot that I could describe as very meh. I am wondering if these sample files were shot at f/22, because that is what they look like. If so, that could explain a lot and perhaps the lens is capable of actual positivity at f/11 - f/16. Look at the exposure specs ISO 200 (why ?) at 1/20th - 1/30th of a second for a daytime shot. There's no aperture information so I have to assume f/22. Keep in mind this is on a 44mm x 33mm sensor at 50mp. I hope they produce some more raw files at f/11 - f/16, and it would also be positive if they indicated the degree of shift (if there even is any) for each shot.
Steve Hendrix/CI
Steve Hendrix/CI
rdeloe
Well-known member
I downloaded the samples (RAF). The samples don't suggest it's a brilliant lens, but distortion seems quite modest and correctable. Where I'm seeing issues, I can't tell if it's operator trouble, a small aperture, and/or the effects of movement. Overall, I think I could do my job with this lens. The problem for Silvestri is that for a customer like me, this lens needed to leave the Pentax-A 645 35/3.5 for dead, and I don't think it does. I look forward to seeing more raw samples, with aperture and movement information.
4x5Australian
Well-known member
I've examined the TIFF files whilst going back and forth to my test images of the Zeiss Contax 645 3.5/35mm lens, which was pre-set to f/11 and mounted with a Kipon shift adapter on a GFX 50R.
As Steve pointed out, the file properties show that the Silvetar 35 images were made at ISO 200 with shutter speeds of 1/20s to 1/30s, implying the aperture used was f/22 or thereabouts, so the resolution was likely degraded by diffraction.
My general impression is that the level of detail in these images, made with a GFX 50SII, seems similar to that from the Contax 645 35mm on a GFX 50R.
However, the images are the result of a combination of lens + sensor, and I suspect that the 50Mpx sensor is the limiting factor in this combination. If, instead, the lens was mounted in front of a 100Mpx sensor or an IQ4-150, perhaps our impressions of detail and sharpness would be better.
Let alone what using f/8 or f/11 might do for it.
The images show little or no colour fringing, despite the abundant white window frames. For example, the white perforated metal mesh on the left side of 4527 looks very clean.
The distortion in these images seems less than the Contax 645 35mm, the HD Pentax 645 D-FA 35mm and the Digaron-S 35mm.
These test shots are also compromised, I think, by questionable camera alignment and focus, and the dull overcast lighting.
I hope that Silvestri will make a sample lens available to someone who can better demonstrate its capabilities.
Rod
As Steve pointed out, the file properties show that the Silvetar 35 images were made at ISO 200 with shutter speeds of 1/20s to 1/30s, implying the aperture used was f/22 or thereabouts, so the resolution was likely degraded by diffraction.
My general impression is that the level of detail in these images, made with a GFX 50SII, seems similar to that from the Contax 645 35mm on a GFX 50R.
However, the images are the result of a combination of lens + sensor, and I suspect that the 50Mpx sensor is the limiting factor in this combination. If, instead, the lens was mounted in front of a 100Mpx sensor or an IQ4-150, perhaps our impressions of detail and sharpness would be better.
Let alone what using f/8 or f/11 might do for it.
The images show little or no colour fringing, despite the abundant white window frames. For example, the white perforated metal mesh on the left side of 4527 looks very clean.
The distortion in these images seems less than the Contax 645 35mm, the HD Pentax 645 D-FA 35mm and the Digaron-S 35mm.
These test shots are also compromised, I think, by questionable camera alignment and focus, and the dull overcast lighting.
I hope that Silvestri will make a sample lens available to someone who can better demonstrate its capabilities.
Rod
Last edited:
diggles
Well-known member
Good call on checking ISO and shutter speed.
The weather at my home today is overcast, similar to what is seen in the sample Silvetar images. I set my camera to ISO 200 at f/22 to see what the camera would suggest to set the shutter speed for proper exposure. It came up with 1/20th. In the same range as the sample images. To me this supports the assumption that the Silvetar lens was set at f/22.
The raw file DSCF4525 is a the closest image to a one point perspective. Based on the perspective it looks like the center of the lens is about 1/3 from the left hand side. So there is definitely some shift. I would imagine there is some camera fall as well, but hard to tell just how much.
Distortion and CA look very good, easily correctable.
Image circle-wise, I agree about the "Meh". Looking at image 4525, you can see a clear difference in sagittal/tangential resolution in the upper right corner. (Compare the sharpness of the vertical sides of the window recesses with the converging horizontal lines of the shutters.) Maybe a different compromise could have been achieved with some focus tweaking? If this is shot at f/22, corners are not going to be better at wider apertures. This is shown in the MTF plot as the difference between the dashed & solid lines toward the right.
It does look like the camera is pretty square to the background building - no obvious horizontal convergence there. Based on that assumption, looks like this frame had the lens shifted about 7mm right and 2.5mm up, putting the top right corner ~70mm from the optical axis. In this image I've added CA correction and +20 light falloff in Capture One, but no distortion correction.
Image circle-wise, I agree about the "Meh". Looking at image 4525, you can see a clear difference in sagittal/tangential resolution in the upper right corner. (Compare the sharpness of the vertical sides of the window recesses with the converging horizontal lines of the shutters.) Maybe a different compromise could have been achieved with some focus tweaking? If this is shot at f/22, corners are not going to be better at wider apertures. This is shown in the MTF plot as the difference between the dashed & solid lines toward the right.
It does look like the camera is pretty square to the background building - no obvious horizontal convergence there. Based on that assumption, looks like this frame had the lens shifted about 7mm right and 2.5mm up, putting the top right corner ~70mm from the optical axis. In this image I've added CA correction and +20 light falloff in Capture One, but no distortion correction.
Last edited:
Vivseries1
Member
The Silvestri website says this lens is available in stock. So has anyone had a chance to actually use it? Results? Thanks
IMHO, very clever from Silvestri to design and offer such a lens. Love the touch of “small rear group size … allows ... movements inside the bayonets of mirrorless cameras” (Emphasis added).
Lack of a useable wide angle lens (i.e. short focal length) when a cameras was used, was always a deficient of the Silvestri. (Not an issue when using digital back since the DB's sensor is closer to the lens.)
In case Silvestri is open for of another lens suggestion that would fill another gap: a lightweight Tele (tele-design), say, 160mm!.
Lack of a useable wide angle lens (i.e. short focal length) when a cameras was used, was always a deficient of the Silvestri. (Not an issue when using digital back since the DB's sensor is closer to the lens.)
In case Silvestri is open for of another lens suggestion that would fill another gap: a lightweight Tele (tele-design), say, 160mm!.