Home

Please support mflenses.com if you need any graphic related work order it from us, click on above banner to order!
SearchSearch MemberlistMemberlist RegisterRegister ProfileProfile Log in to check your private messagesLog in to check your private messages Log inLog in

Leica-Minolta collaboration
Post new topic   Reply to topic
Home   Manual Focus Lenses
View previous topic :: View next topic  


PostPosted: Fri Jul 21, 2023 12:39 pm    Post subject: Reply with quote
cbass wrote:
All the mediocre Leica lenses were designs they purchased from Minolta. Just kidding.


Talking about SLR lenses in the 1970s Leica simply didn't have the knowledge to produce good wideangles and superwides. They were purchased from Zeiss/Pentax (3.5/15mm), Schneider (4/21mm) and Minolta (2.8/24mm). Leica didn't produce their own zooms but bought them from Angéenieux and Minolta, and the 500mm mirror was Minolta as well.

cbass wrote:
Leica made some very special lenses that were ahead of their time.
The Summilux 50 in particular. Even today it's very difficult to correct for coma on fast 50's and Leica did an exceptional job here and they calculated that lens in the 1960s. There also was no other lens that could match its performance at f/1.4 in the center for decades. There were lenses that could do better in the corners because Leica allowed massive curvature in the lens to get that performance. It would fail stevemark's typical test as the extreme corners didn't get crisp until f/8 due to the curvature. That lens was designed for handheld night photography where it had no serious competition for a long time. If you wanted a lens with less curvature where the corner performance was much better at wider apertures, then Leica had the summicron 50. The Summicron 50 would perform very strongly in stevemark's typical test.


Thanks for this information! I am accutely aware that my tests are dealing with a tiny part of porperties of vintage lenses, and I really appreciate this additional information. It would be good, if you could substantiate your claims with additional images and/or crops, e. g. comparing a Leica R Summilux 1.4/50mm with a contemporary Minolta MC/MD 1.4/50 or a Canon FD 1.4/50. I (and probably others as wee) would really appreciate it: To see whether those differences are relevant / irrelevant for my kind of photograohy.

While I usually refrain from buying Leica R lenses, I have e. g. a Elmarit-R 2.8/135mm (2nd computation). Mechanically it clearly is the nicest vintage 135mm lens I own, but optically ...?? Quite low performance, at least for landscapes. What was Leitz thinking when developing that lens?

Combining the Elmarit 2.8/135mm with the dedicated Leitz 2x converter it becomes really bad. Combining my cheap Minolta MD-III 2.8/135mm with the Minolta 2x Converter S gives much better results. Again - what was Leitz doing back then ?!?

S

Last edited by stevemark on Fri Jul 21, 2023 1:50 pm; edited 1 time in total


PostPosted: Fri Jul 21, 2023 1:26 pm    Post subject: Reply with quote
stevemark wrote:
Talking about SLR lenses in the 1970s Leica simply didn't have the knowledge to produce good wideangles and superwides. They were purchased from Zeiss/Pentax (3.5/15mm), Schneider (4/21mm) and Minolta (2.8/24mm). Leica didn't produce their own zooms but bought them from Angéenieux and Minolta, and the 500mm mirror was Minolta as well.


If memory serves me right (I could be wrong), Leica also lacked the experience in mechatronics required to design the more advanced electronic exposure systems that were emerging at the time, and impressed with Minolta's demonstrable prowess in that area collaborated on the development of camera bodies (Minolta CL & Leica CL, Minolta XE & Leica R3, Minolta XD7/XD11 & Leica R4).


PostPosted: Fri Jul 21, 2023 2:28 pm    Post subject: Reply with quote
RokkorDoctor wrote:


If memory serves me right (I could be wrong), Leica also lacked the experience in mechatronics required to design the more advanced electronic exposure systems that were emerging at the time, and impressed with Minolta's demonstrable prowess in that area collaborated on the development of camera bodies (Minolta CL & Leica CL, Minolta XE & Leica R3, Minolta XD7/XD11 & Leica R4).


Yes-and no. When the R3 was developed, for example, the electronics were outsourced to the British Ferranti factory. ( https://en.wikipedia.org/wiki/Ferranti ). They added spot metering and completely revised the electronics. In my opinion, if that was not the case, leaving the original would have been a much better idea.

(source: my memory from an in-deep article on an italian magazine of the time)


PostPosted: Fri Jul 21, 2023 11:58 pm    Post subject: Reply with quote
stevemark wrote:


Thanks for this information! I am accutely aware that my tests are dealing with a tiny part of porperties of vintage lenses, and I really appreciate this additional information. It would be good, if you could substantiate your claims with additional images and/or crops, e. g. comparing a Leica R Summilux 1.4/50mm with a contemporary Minolta MC/MD 1.4/50 or a Canon FD 1.4/50. I (and probably others as wee) would really appreciate it: To see whether those differences are relevant / irrelevant for my kind of photograohy.

While I usually refrain from buying Leica R lenses, I have e. g. a Elmarit-R 2.8/135mm (2nd computation). Mechanically it clearly is the nicest vintage 135mm lens I own, but optically ...?? Quite low performance, at least for landscapes. What was Leitz thinking when developing that lens?

Combining the Elmarit 2.8/135mm with the dedicated Leitz 2x converter it becomes really bad. Combining my cheap Minolta MD-III 2.8/135mm with the Minolta 2x Converter S gives much better results. Again - what was Leitz doing back then ?!?

S

My comment about Minolta was tongue in cheek. They made some exceptional lenses.

It's very possible that a more modern Minolta MD 50 f/1.4 would surpass a Summilux. The Summilux was designed as a handheld night photography lens and compromises were made. The most impressive performance characteristic is the coma correction. Leica also knew toward the end that the design of the Summilux was aging, and that competition surpassed it and they would need a replacement. The Summilux-R E60 was the successor that was a much better lens stopped down largely due to the correction to curvature. Leica added another element to correct some of the curvature and astigmatism and even further corrected coma. The Summilux-R E60 is very expensive and IMO not worth the price anybody wants for it today. Even going back to the original lenses, the Summilux-M 50 was better than the Summilux-R. The issue is that many M lenses don't adapt well to mirrorless. After the Summilux-R E60 Leica made the Summilux-M 50 Asph. That also had an additional lens element to correct for curvature and astigmatism but this time it was aspherical and also improved performance significantly at f/1.4. In addition, it has floating elements in order to correct for close focus. All previous Summilux performance dropped at close range and is pretty mediocre until about f/2.8. This wasn't such a huge issue as it cleaned up pretty quick around 1.5-2m. The Asph Summilux was the best corrected f/1.4 50 until the Otus came out.

Leica lenses are a poor choice for your needs. Both the M lenses and the R lenses allowed for a lot of curvature. Often this curvature was not surpassed by the DOF until f/5.6 or f/8. Zeiss focused more on planar designs and those are better suited for your testing and needs. Although Zeiss had some bad curvature on the older designs as well especially the wide angles like the 28mm lenses. Zeiss would lean toward overall performance while Leica would lean toward the best central performance at the widest aperture at the expense of other factors. Leica also does not focus solely on in focus areas, but also takes into consideration how the out of focus areas look. I am not sure if this was a design consideration at Zeiss or not.

The modern SL lenses are a departure from this, and they have very planar fields and are extremely high performance. They also are not traditional Leica with compact size, but huge lenses that rival medium format lenses in size. They also spread correction over several lens elements instead of using as little as possible like previous designs.

One area where Leica is mediocre is veiling flare. This isn't an issue on every lens and some designs have strong flare performance. However, many of their designs including modern designs are behind the competition when it comes to veiling flare. Many SL lenses and even the over $8K APO-Summicron 50 does worse when it comes to veiling flare compared to the competition.

On the other hand, when it comes to low contrast or dull light there is where Leica excels. Other lenses can be dull in these lighting conditions while Leica maintains contrast and saturation. I know that's a bold claim, but I have an idea for a test that I think can prove it.

Last edited by cbass on Sat Jul 22, 2023 5:26 am; edited 1 time in total


PostPosted: Sat Jul 22, 2023 5:00 am    Post subject: Reply with quote
stevemark wrote:

It would be good, if you could substantiate your claims with additional images and/or crops, e. g. comparing a Leica R Summilux 1.4/50mm with a contemporary Minolta MC/MD 1.4/50 or a Canon FD 1.4/50. I (and probably others as wee) would really appreciate it: To see whether those differences are relevant / irrelevant for my kind of photograohy.



Steve,

It depends on which claims you are interested in. Coma correction or corner performance?

I currently don't own a Minolta MD 50 f/1.4 or Canon FD 50 f/1.4. I can obtain one if you are really interested. I also own just an APS-C and MF camera. The APS-C will cut off the worst of the coma and corners, but I can use the MF camera in FF crop mode.

There is enough information of many previous tests, however, especially when it comes to coma.

https://lens.ws/minolta-md-50mm-f1-4/

There is a coma test there for the Minolta MD 50 f/1.4 and as you can see it is in line with the nasty pronounced coma of most fast 50's.

Leica's official literature talks about coma correction and resistance to bright sources of light. You can see manufacturer claims inlcuded here:

https://lens-db.com/leitz-wetzlar-leitz-canada-summilux-r-50mm-f14-i-type-1-1970/

Here you can see a comparison between Summilux-M versions:

https://www.fredmiranda.com/forum/topic/1812141/1#16267704

Summilux-M V1 that was produced very briefly had your typical nasty wing coma as most fast 50's. I think V1 was not a Leica design and in production only a few years. Starting from V2 the coma correction is very good and exceptional if you consider that it came out in the 1960s. V3 was the same optical formula as V2, but with the ability to focus closer. The ASPH version improved upon the coma even further and is still one of the best 50's. Today people say the Otus is better, but you can make an argument otherwise, but I won't get into that.

The Summilux-R was Leica porting the rangefinder design to a SLR with some differences but maintaining the same design goals and characteristics with coma correction and bright point sources of light. The E60 Summilux-R was an evolution of the design before the Aspherical version of the Summilux-M both improving upon coma correction.

Here is a night test from a blog that includes the Summilux-R. You can compare coma with other fast 50's. I believe it is testing that Rolf has done which used to be or still is a member on here. There is also a link to download all this test images for those interested.

https://theothersideofbokeh.wordpress.com/2022/05/01/finest-fifties-at-night/

This was more an issue in the film days where high-speed film options were limited especially in the 1960s-1970s. Today's digital cameras with high ISO performance and IBIS where you can stop down to f/2.8 and hand hold instead of having to use f/1.4 change things. However, there are still special needs where the fast aperture and coma correction are still relevant for certain astrophotography applications and some lab experiments. There are few options here and very few vintage options.

There was the Noctilux 50 f/1.2 that used two aspherical hand ground surfaces and was very difficult to produce. This was a very expensive lens and then due to rarity became obnoxiously expensive. Leica decided to produce this lens again recently, but it isn't cheap today either.

Nikon had the Noct 58 f/1.2, which again had to use a hand ground aspherical element and again was very expensive and difficult to produce and today also sells for a high price. The thing with the Noct it the contrast at f/1.2 was extremely poor when compared to the Noctilux f/1.2 and the Summilux. It had good coma correction, but in all other optical criteria it wasn't anything close to Leica's offerings.

Finally, there was/is the Summilux 50. The earlier versions only used spherical lens elements unlike the faster options that used aspherical. Except for the most recent version that also uses an aspherical lens element.

Does any of this matter? Probably not to most people. If you can use a tripod, then just stop the lens down where coma is no longer an issue. You can also stop down and take advantage of IBIS if possible. But it not only was special at the time of production, but still is special today.

For a more affordable option the Voigtlander 50 APO-Lanthar has the best coma correction I know of today and is much cheaper than even an older Summilux. It is a f/2 lens, however, and a f/2 lens is not an f/1.4 lens, but with digital most should be able to get away with f/2. I also think the Loxia 50 is also pretty good, but I am not 100% confident about that. Again, it's f/2 not f/1.4 so not the same.

https://www.lenstip.com/613.7-Lens_review-Voigtlander_Apo_Lanthar_50_mm_f_2_Aspherical_Coma__astigmatism_and_bokeh.html


PostPosted: Sat Jul 22, 2023 10:12 pm    Post subject: Reply with quote
cbass wrote:
stevemark wrote:

It would be good, if you could substantiate your claims with additional images and/or crops, e. g. comparing a Leica R Summilux 1.4/50mm with a contemporary Minolta MC/MD 1.4/50 or a Canon FD 1.4/50. I (and probably others as wee) would really appreciate it: To see whether those differences are relevant / irrelevant for my kind of photograohy.



Steve,

It depends on which claims you are interested in. Coma correction or corner performance?


For my personal purposes I'm usually into "corner-to-corner performance", since I do use my lenses e. g. for calendar images.

However, I really would love to see a direct comparison of a Summilux R 1.4/50 image with another image taken with a "classical" 1.4/50mm lens (Canon, Konica, Minolta, Nikon, Zeiss ... you name it) - showing a situation where conditions are favourable to the Summilux (night time images? Portraits?)

cbass wrote:
I currently don't own a Minolta MD 50 f/1.4 or Canon FD 50 f/1.4. I can obtain one if you are really interested. I also own just an APS-C and MF camera. The APS-C will cut off the worst of the coma and corners, but I can use the MF camera in FF crop mode.


No problem - I have access to many of the famous Leica R lenses if I really need them, thanks to a collector friend. I really should try e. g. the R 1.4/50 and the R 1.4/80.

Thanks a lot for all the additional information and links - I'll go through it!

S


PostPosted: Sun Jul 23, 2023 5:37 am    Post subject: Reply with quote
stevemark wrote:


For my personal purposes I'm usually into "corner-to-corner performance", since I do use my lenses e. g. for calendar images.

However, I really would love to see a direct comparison of a Summilux R 1.4/50 image with another image taken with a "classical" 1.4/50mm lens (Canon, Konica, Minolta, Nikon, Zeiss ... you name it) - showing a situation where conditions are favourable to the Summilux (night time images? Portraits?)

cbass wrote:
I currently don't own a Minolta MD 50 f/1.4 or Canon FD 50 f/1.4. I can obtain one if you are really interested. I also own just an APS-C and MF camera. The APS-C will cut off the worst of the coma and corners, but I can use the MF camera in FF crop mode.


No problem - I have access to many of the famous Leica R lenses if I really need them, thanks to a collector friend. I really should try e. g. the R 1.4/50 and the R 1.4/80.

Thanks a lot for all the additional information and links - I'll go through it!

S


Rolf's blog that I linked has many 50s in controlled night testing and there is a link for all the images. He doesn't have a Minolta 50 f/1.4 in the test, but he does test 8 50mm lenses including the Summilux and Summicron.

https://ibb.co/album/GsbYdV

Your use case is more complex than just corner to corner performance as you as looking for it to peak as the widest aperture possible. If you needed it at f/8 it would be different. The Zeiss Contax 35-70/3.4 zoom would be a strong candidate. The best performance is from f/8 to f/11, but the 35mm range beats most primes when it comes to corners even today. Just nowhere close to wide open. At 50mm and f/8 it's about as good as anything out there made historically. The 70mm end is also very strong, but again at f/8. Stopped down to f/8-f/11 that zoom is pretty much three strong primes.

The best 50 for your use case is the Voigtlander APO-Lanther 50mm f/2. There is no other 50 as well corrected at the price point.

https://www.flickr.com/photos/royprasad/50078214387

https://phillipreeve.net/blog/review-voigtlander-50mm-f2-apo-lanthar/

Not as cheap as a vintage prime, but also no vintage prime will touch the performance at wider apertures. It's also not a huge lens. There is an M mount and a Sony E mount. Buy the mount for the right body. Buying M mount plus adding adapter for E mount is not the same. Corner performance will not be the same.

I don't think you will like the Leica R 80/1.4. It's competent when stopped down, but full of aberrations at the wider apertures. The love of the lens does not come from how well corrected it is, but how it draws and renders. It is loved by portrait photographers and less so by landscape photographers. The love is for the out of focus areas more than in focus areas.

The Summicron-R 50 second version is also a good lens, but it really starts to become excellent at f/4. The lens was designed to be a budget lens not requiring any special glass types with strong performance, but this is a 1970s design and it shows its age. f/2 it has the Leica glow. The Leica glow is pretty much mostly spherical aberration. It's called a glow instead of spherical aberration with other makers because the lenses retain strong contrast and still render a lot of detail although SA is present. This is a unique compared to most other manufacturers as typically contrast and details fall. I am not a fan of the glow. I just see SA. Most legendary 50/1.4 lenses will do better than the Summicron-R 50 at f/2 and f/2.8. The Zeiss Contax 50/1.4 being one of them.

The place the Summilux-R 50/1.4 will shine is when comparing coma at f/1.4. Also, when you look at infinity performance at f/1.4. It has the least SA of any vintage lens I compared it against. It is not absent, but impressively minimal for its age.


PostPosted: Sun Jul 23, 2023 6:38 am    Post subject: Reply with quote
[quote="cbass"]
stevemark wrote:


Buying M mount plus adding adapter for E mount is not the same. Corner performance will not be the same.



But why?

Question


PostPosted: Sun Jul 23, 2023 11:32 am    Post subject: Reply with quote
LittleAlex wrote:
cbass wrote:


Buying M mount plus adding adapter for E mount is not the same. Corner performance will not be the same.



But why?

Question


Leica M cameras have a very thin AA-filter stack in front of the sensor (<1mm); Sony E cameras have a 3 mm thick filter stack in front of the sensor (check e. g. the patents for Sony E/EF lenses). The thick filter stack must be incorporated in the lens design, otherwise it will degrade the lens performance (especially for fast non-retrofocus lenses <50mm).

Check Roger Cicalas excellent blog entries about sensor stack thickness and lens performance / degradation.

S


PostPosted: Sun Jul 23, 2023 2:37 pm    Post subject: Reply with quote
LittleAlex wrote:
cbass wrote:


Buying M mount plus adding adapter for E mount is not the same. Corner performance will not be the same.



But why?

Question


This is beyond my understanding, but it is something that has been discovered by others that shoot both Leica and Sony that thought they could just buy the M mount and also use on their Sony via adapter. Unfortunately, the lens lost the most impressive feature of corner performance at wide apertures. There is a discussion here with samples and links to further discussion.

https://www.fredmiranda.com/forum/topic/1736389/0

The sensor cover glass thickness is often listed as the culprit and many articles have been written about it as stevemark pointed out. It is the main culprit attributed to differences in performance.

Digital rangefinders were hell for Leica to design. They had to get help from Kodak. Film was more tolerant of steep ray angles than digital. Leica took advantage of that to design their M lenses especially wide angles since a rangefinder did not have a mirror box and they could use a shorter flange focal distance because they didn't need room for a mirror box. Sadly, most M lenses especially wide angles do not perform the same especially in corners on mirrorless cameras. You either need a modified mirrorless camera with a thin sensor glass or you need to spend the money on a digital M camera. Telephotos are typically fine. Normal lenses around 50mm can be hit or miss, but I think most are affected just not as badly as wide angles.

https://diglloyd.com/blog/2015/20150808_1515-SonyA7R_II-LeicaM-sensorCoverGlass.html


PostPosted: Sun Jul 23, 2023 8:15 pm    Post subject: Reply with quote
stevemark wrote:
LittleAlex wrote:
cbass wrote:


Buying M mount plus adding adapter for E mount is not the same. Corner performance will not be the same.



But why?

Question


Leica M cameras have a very thin AA-filter stack in front of the sensor (<1mm); Sony E cameras have a 3 mm thick filter stack in front of the sensor (check e. g. the patents for Sony E/EF lenses). The thick filter stack must be incorporated in the lens design, otherwise it will degrade the lens performance (especially for fast non-retrofocus lenses <50mm).

Check Roger Cicalas excellent blog entries about sensor stack thickness and lens performance / degradation.

S


Indeed. When using legacy non-native lenses designed for film photography, the thick sensor filter stack on the SONY E cameras affects not only corner performance but also centre performance at large apertures (starts to become noticeable at f/2, obvious at f/1.4). Micro 4/3 is even worse at ca. 4mm stack thickness.

When these lenses were designed it was not anticipated that a 3mm thick sheet of glass would be inserted between the lens and the "film". Spherical aberration is perhaps the most obvious to understand of the various aberrations introduced by the thick planar sheet of filter glass (at those larger apertures).

Conversely, it is also the reason that lenses designed for use with a rear glass filter (super tele's, some extreme wide-angles) should not be used without one of those filters fitted, even if just a clear glass filter; it forms part of the optical calculation incl. e.g. the correction for spherical aberration.


PostPosted: Sun Jul 23, 2023 8:27 pm    Post subject: Reply with quote
cbass wrote:
LittleAlex wrote:
cbass wrote:


Buying M mount plus adding adapter for E mount is not the same. Corner performance will not be the same.



But why?

Question


This is beyond my understanding,


As I said before - the 3mm thick sensor stack ("glass") in front of the CMOS sensors in Sony E/EF cameras changes the way of light rays in a pretty dramatic way if they are coming in in a flat angle (e. g. in symmetrical wideangles). And if these wideangle are even fast, then the incoming corner rays arrive with varying angles (since the aperture opening is pretty wide). That leads to unsharp corners.

Removing the sensor stack completely results in proper corner resolution, also with Sony A7 series cameras. I know this since I had access to a Kolari "UT" modified Sony A7S. Sensor stack is 0.2 mm (much lesss than any digital Leica M), and the results are breathtaking even though the cam only has 12 MP. Lenses such as the Zeiss Biogon C 4.5/21mm ZM or the rare Zeiss Distagon 2.8/15mm ZM (which are neraly unusable on a ordinary A7) now were extremely sharp, even wide open, and even in the extreme corners.

Removing the sensor stack (or replacing it by a Kolari UT 0.2 mm) has a big disadvantage though - much more infrared light is getting to sensor, resulting in a strong orange or reddish color cast. Since IR radiation now is contributing to the red image dots on the sensor, you can't simply adjust white balance to get correct colors. Partial corrction is possible, but the resulting image will always look strange. Due to the increased IR transmission, the Kolari UT modification would be perfect for high res b/w landscape fotos, though (at least as long as you don't use native Sony/Zeiss lenses ... !!)

S


PostPosted: Mon Jul 24, 2023 4:15 am    Post subject: Reply with quote
Now some of this stuff is coming back to me. I remember reading some time ago that an APS-C sensor camera with a film lens on a speed booster could perform better than the lens adapted to FF sensor camera because the speed booster optics could correct for the sensor glass thickness aberrations. I didn't know if anything ever came out of this or not.

Fun additional reading:

https://www.lensrentals.com/blog/2014/06/sensor-stack-thickness-when-does-it-matter/

Actually, I found what I was referencing.

Quote:
9) It’s too Good to be True! – Part 6, Filter Effect is Cured
A dirty little secret in digital photography is that the filter pack (low-pass filter + infrared
filter + sensor cover glass) found in virtually all digital cameras can contribute significant
aberration to the image when a lens designed for film is used. This problem is particularly
troublesome for very fast lenses. The Speed Booster corrects this defect, and automatically
corrects for filter pack aberrations regardless of the objective lens used.


Quote:
Cool It’s too Good to be True! – Part 5, Telecentricity is Improved
Another nice benefit of the Speed Booster is that it moves the exit pupil further from the
image plane for any objective lens. In other words, it actually improves the telecentricity
of the lens system. This helps to improve corner illumination and reduce sensor artifacts
such as color variations caused by pixel crosstalk.
Figure 11 illustrates how telecentricity is improved by the Speed Booster. To find the exit
pupil location you need to extend a chief ray (i.e., the ray that passes through the center of
the aperture stop and the corner of the image) back through the lens until it crosses the
optical axis. The amount by which the exit pupil location is increased by the Speed
Booster changes depending on the degree of telecentricity of the objective lens, but the
general trend is the same for all existing SLR objective lenses.


Full information:

https://www.lensrentals.com/product-assets/e98e3c31-1515-4bac-875f-1a9d5e51eb6d/Speed%20Booster%20White%20Paper.pdf
Post new topic   Reply to topic
Home   Manual Focus Lenses
Goto page Previous  1, 2, 3, 4
Page 4 of 4