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!
New SK 35mm - any news?
- Thread starter f8orbust
- Start date
As an aside, and not meaning to hijack the thread, if new lenses are announced how will Schneider / Rodenstock get around the fact that Copal shutters are no longer made? Is it expected that we'll all need to use on FPS cams or with electronic shutter units?
Back on topic: Is the old 35mm SK a bit long in the tooth? Is it mainly light falloff and / or colour shift with movements?
Back on topic: Is the old 35mm SK a bit long in the tooth? Is it mainly light falloff and / or colour shift with movements?
My guess is that Schneider thinks that the wides need a bit more correction and be a bit more retrofocus to be shiftable on recent sensors. Hopefully they won't end up with Rodenstock 32mm-like design (a zillion glass elements and very high cost) but something more economical and light.
As far as I understand when you go really wide, say 35mm and wider, you also need more complex designs to deal with field curvature. The current 35mm has quite a bit of field curvature so the image circle with really high quality is more like 75mm rather than 90mm. I'd like to see a new 35 with a bit better correction, but not bring it too far.
It's a bit unclear to me for how long new lenses can be sold with Copal shutters. Hopefully they exist in stock for a while more. After that focal plane shutter cameras will probably be the main choice for field use, electronic shutters are still designed as studio products. Hopefully the new DHW shutter will change this but I'm skeptical. iPhone interfaces etc may look cool but will suck for tough field usage. I'm guessing that focal plane shutters will require that flange distance is longer than the current 35mm has. As a view camera user (Linhof Techno) I'm not particularly excited by the current FPS cameras, but theoretically Linhof could make a back to their camera which has an FPS, it would significantly increase flange distance though.
One issue for me concerning electronic shutters is that they cost ~€1000+VAT, which is a lot more than the Copal shutters. Therefore I think I will because of economical reasons look for a focal plane shutter solution, but there are still a few years before that will be a must.
As far as I understand when you go really wide, say 35mm and wider, you also need more complex designs to deal with field curvature. The current 35mm has quite a bit of field curvature so the image circle with really high quality is more like 75mm rather than 90mm. I'd like to see a new 35 with a bit better correction, but not bring it too far.
It's a bit unclear to me for how long new lenses can be sold with Copal shutters. Hopefully they exist in stock for a while more. After that focal plane shutter cameras will probably be the main choice for field use, electronic shutters are still designed as studio products. Hopefully the new DHW shutter will change this but I'm skeptical. iPhone interfaces etc may look cool but will suck for tough field usage. I'm guessing that focal plane shutters will require that flange distance is longer than the current 35mm has. As a view camera user (Linhof Techno) I'm not particularly excited by the current FPS cameras, but theoretically Linhof could make a back to their camera which has an FPS, it would significantly increase flange distance though.
One issue for me concerning electronic shutters is that they cost ~€1000+VAT, which is a lot more than the Copal shutters. Therefore I think I will because of economical reasons look for a focal plane shutter solution, but there are still a few years before that will be a must.
It should be possible to make something inbetween the Digaron 32 and Digitar 35, and I think that is the most likely scenario. It seems to be in line with the strategy of the whole Digitar lens lineup.
Hopefully they'll optimize it for the current baseline which seems to be 54x41mm 60 megapixels, rather than "future proof" it for like 200 megapixel sensors. As far as I understand the thing with wides is that they get really really complex (lots of lens elements) if they should resolve very high resolution over the whole image circle. Longer lenses can do with much simpler designs. So when it comes to wides it's better to not make them "too good" to avoid monster designs. I hope and think they'll find a balance.
Interesting. I can see the 24 being revamped due to it's age and small (60mm?) image circle (IC). Still a nice lens unshifted on a 37 x 49 sensor (or smaller) it has to be said.
The 28 is being updated (I guess) to allow for full shifting on IQ180/280 style sensors. This is a mind-bogglingly great lens already, with an amazing (for its focal length) 90mm IC. I've seen captures with it on the IQ160 and P45 - and both yield superb performance, even with large shifts (especially on the P45). Every new iteration of C1 seems to be able to deal with LCCs on wide angle tech camera lenses better than the last, but I guess the IQ180/280 sensors are just out of reach (for large shifts).
The 35 is my favourite lens - again I would imagine this is being updated so that it can cope with large shifts on IQ180/280 style sensors and, as torger points out, to alleviate field curvature and make the performance of the 90mm IC more consistent from centre to edge.
Looking forward to seeing them I must say. We already use the 28 - wouldn't look to upgrade that - but will look at the 35 assuming it's not the size of a tank (a la Rodie 32 - which should really be in a Copal 1 shutter given its size and weight methinks). Hopefully (as torger again points out) they are not going to try and design these new lenses for 200MP sensors - I think (hope) we're hitting the ceiling (of what 99% of people want) at 60/80MP and that future DBs will concentrate on other aspects of image capture beyond simply the number of pixels (CMOS, dynamic range, colour fidelity, high ISO noise etc.)
The 28 is being updated (I guess) to allow for full shifting on IQ180/280 style sensors. This is a mind-bogglingly great lens already, with an amazing (for its focal length) 90mm IC. I've seen captures with it on the IQ160 and P45 - and both yield superb performance, even with large shifts (especially on the P45). Every new iteration of C1 seems to be able to deal with LCCs on wide angle tech camera lenses better than the last, but I guess the IQ180/280 sensors are just out of reach (for large shifts).
The 35 is my favourite lens - again I would imagine this is being updated so that it can cope with large shifts on IQ180/280 style sensors and, as torger points out, to alleviate field curvature and make the performance of the 90mm IC more consistent from centre to edge.
Looking forward to seeing them I must say. We already use the 28 - wouldn't look to upgrade that - but will look at the 35 assuming it's not the size of a tank (a la Rodie 32 - which should really be in a Copal 1 shutter given its size and weight methinks). Hopefully (as torger again points out) they are not going to try and design these new lenses for 200MP sensors - I think (hope) we're hitting the ceiling (of what 99% of people want) at 60/80MP and that future DBs will concentrate on other aspects of image capture beyond simply the number of pixels (CMOS, dynamic range, colour fidelity, high ISO noise etc.)
Last edited:
With the IQ2 series only one had a new sensor, which is the IQ260, ie new development effort was put into 60 rather than 80 or more megapixels which I think was a great sign that what the market desires a pixel count around there. IQ180 won many users because the attractive upgrade plan was steered by Phase One P45+ => P65+ => IQ180, but I would guess that if the tech cam users could choose upgrade path more freely and knew about color cast issues (not so known when IQ180 was new) many more would have gone from P65+ to IQ160 rather than IQ180.
To make (ultra)wides work well with the IQ180/280 I'm afraid that we're forced into Rodie 32-like designs as the sensor simply has too much color cast, so I hope they leave that space for Rodenstock and focus on getting great performance with the 6um sensors (ie IQ260).
I think the color cast issue is a quite big thing concerning tech camera future. Sensors with low color cast allows for unique tech camera wide angle lens designs that make the system stand out from the rest. With high color cast one is forced into making lens designs more similar to standard camera systems thus losing some of the uniqueness. I'm not sure how much Phase One considers tech cameras when they design sensors together with Dalsa. I'd like to see a little bit of sensor design that adapts to optical designs rather than the other way around. I really don't like the IQ180 as I feel it has steered tech camera lens designs in the wrong direction. Ever since the P25 it's been the newer the sensor the worse color cast, IQ260 for the first time breaks that sequence (great!).
Therefore it will be very interesting to see how the new Schneider wides will be designed.
To make (ultra)wides work well with the IQ180/280 I'm afraid that we're forced into Rodie 32-like designs as the sensor simply has too much color cast, so I hope they leave that space for Rodenstock and focus on getting great performance with the 6um sensors (ie IQ260).
I think the color cast issue is a quite big thing concerning tech camera future. Sensors with low color cast allows for unique tech camera wide angle lens designs that make the system stand out from the rest. With high color cast one is forced into making lens designs more similar to standard camera systems thus losing some of the uniqueness. I'm not sure how much Phase One considers tech cameras when they design sensors together with Dalsa. I'd like to see a little bit of sensor design that adapts to optical designs rather than the other way around. I really don't like the IQ180 as I feel it has steered tech camera lens designs in the wrong direction. Ever since the P25 it's been the newer the sensor the worse color cast, IQ260 for the first time breaks that sequence (great!).
Therefore it will be very interesting to see how the new Schneider wides will be designed.
I do wonder what drives MFDB design. I suspect that a lot of (most?) buyers come from the industrial, government, academic/scientific etc. sectors - meaning that what we want as photographers is probably not a deciding factor. When NASA buys 50 x IQ180s, do they care about large shifts with symmetrical design wide-angle lenses ? Probably not.
ondebanks
Member
The military might - NASA would not. And it's not just because of their slashed budgets. There is little use for Bayer sensors in quantitative scientific applications. The IQ260 Achro would be potentially more useful, if it could only do long exposures like its colour brother.
Ray
It would be nice if the next generation of digital backs could incorporate 'digital shutters' - i.e. they could (if selected) switch on and capture data for 1/1000 sec, 1/500 sec or whatever. All we'd need then is a stand alone aperture on each lens, which shouldn't be too challenging to engineer.
Yes it was field curvature, which means that the image plane is bent, so when you focus in the center the sides are slightly out of focus. With an ALPA and RM3Di you can more easily make a focusing compromise, focus slightly past infinity to get more even center/corner sharpness. With view camera focusing it's more difficult. If I shift it more than a little I stop down to f/16 to compensate some.
There's no huge softening we're talking about, but pixel-peep noticable even lower megapixel backs. I'm sure a new 35 version will be better in this aspect.
It will however likely be worse concerning distortion. The current 35 has near zero distortion thanks to symmetrical design, but a new one will most likely be slightly retrofocus and thus have some distortion, just like the Rodenstock Digarons.
There's no huge softening we're talking about, but pixel-peep noticable even lower megapixel backs. I'm sure a new 35 version will be better in this aspect.
It will however likely be worse concerning distortion. The current 35 has near zero distortion thanks to symmetrical design, but a new one will most likely be slightly retrofocus and thus have some distortion, just like the Rodenstock Digarons.
jotloob
Subscriber Member
To build a retrofocus lens in the size and weight of an almost symetric lens like the SK 36mm , APO-SIRONAR-DIGITAL or CZ BIOGON you must reinvent optics .
First drawing is an APO-SIRONAR-DIGITAL (rather symetric)
View attachment 76035
This second drawing is a HR-DIGARON (retrofocus)
View attachment 76036
I have tried HH shooting with my HR 28mm lens . It is possible but a pain and you must be much more careful in handling the camera .
I do not believe in a smaller and lighter new design .
For example : the APO-SIRONAR-DIGITAL 35mm has 220 grams while the corresponding DIGARON-S 35mm has 480 grams .
First drawing is an APO-SIRONAR-DIGITAL (rather symetric)
View attachment 76035
This second drawing is a HR-DIGARON (retrofocus)
View attachment 76036
I have tried HH shooting with my HR 28mm lens . It is possible but a pain and you must be much more careful in handling the camera .
I do not believe in a smaller and lighter new design .
For example : the APO-SIRONAR-DIGITAL 35mm has 220 grams while the corresponding DIGARON-S 35mm has 480 grams .
whoops. Sorry about that. Question - years ago, there was a wonderful Kodak manual on lens design, and if remembered correctly, outlined that lenses focus on a radial basis, and that only some had flat field focus, mostly for reproduction. Is that still the case, and is the 35 even more so?
As far as I know all lenses have curvature issues to some extent, but they can be more or less corrected, and with longer lenses (narrower the field of view) the less of a problem is curvature. If you look at MTF diagrams you can often see that compensation. MTF shows the contrast at the flat image plane, and without correction the contrast becomes lower farther away from center as the plane of focus is bent away from the image plane. With a corrected lens you can see that when going out from the center it first becomes less contrasty, then a bit more contrasty again and then less contrasty (the general trend is less contrasty further out though), the more going back-and-forth the higher order of correction there is. In other words you cannot make the lens project a completely flat plane, but you can correct the inward curvature by superpositioning outward curvature on top. If you keep the curvature error small enough it will not be detectable.
This MTF curve from the Digaron-S 35mm has this effect quite visible in the 100 lpmm curve:
http://www.josephholmes.com/news/mf/35_HR_MTF.jpg
This MTF curve from the Digaron-S 35mm has this effect quite visible in the 100 lpmm curve:
http://www.josephholmes.com/news/mf/35_HR_MTF.jpg
I think the Rodenstock Digarons are large and heavy not only due to retrofocus design, but also due to larger aperture (f/4 instead of f/5.6), and more correction to provide better performance at say f/8.
I think a well-balanced lens for "large format digital" should be optimized for f/11, and largest aperture don't need to be larger than f/5.6. If they stay with that design target I think they can succeed making a smaller and lighter lens than the Digarons even if retrofocus.