S lens MTF charts

[Dsauro]

Advertisement (gone after registration)

Looking for a bit of education regarding the MTF lens charts.

i think it’s great that Leica posts the MTF charts for their s lens via their website or through lens description downloads.  In reading various literature on how to interpret MTF charts, certain discussions suggest that reviewing MTF charts should be taken with a grain of salt.  The articles suggest that lens makers publish MTF charts in combination with the use of their camera sensor.  Meaning, canon lens MTF charts are optimized, or perhaps skewed, based on the performance of the lens when complimented by a canon sensor.

Can anyone help me understand if Leica basis their lens MTF performance in a similar fashion.  Or are Leica MTF charts “agnostic” and based solely on lens performance (without sensor optimization)?

 

 

[Jeff S]

Lensrentals is in the process of publishing their own real (not computed... avg of multiple copies) from various manufacturers....

https://www.lensrentals.com/blog/2019/03/just-mtf-charts-canon-prime-lenses/

Jeff

[Stuart Richardson]

In general, I find Leica's MTF pretty accurate, but I am not an optical scientist. Leica is one of the few optical manufacturers to provide measured MTF data, rather than theoretical. So they are taking actual lenses they have made and posting that data. One thing to be aware of, however, is that MTF are the measure of the contrast in a flat field. Sometimes manufacturers choose to make a lens with a more curved field, particularly with wider lenses. The 70mm Summarit exhibits this behavior. In the real world, sometimes this is something that people prefer, as three dimensional compositions often feature detail closer to the camera at the sides, with the central point a bit further away. Personally, I prefer flat field lenses, but they are not the best for all situations. The MTF of curved field lenses often look very sharp on center, but then drop rather sharply towards the edges. They might still be able of rendering really sharp detail, just not at the same focused distance as the center. In general, MTF can be dangerous because a lot of people use them as the sole judge of a lenses performance, but they are just one aspect.  One very important thing to note is quality control and sample variation. For example, I shoot a lot of 4x5 and lots of photographers rave about the 150mm APO Sironar S. It is one of the most recent LF lenses, made with the latest and greatest technology (well, 90s anyway). I even printed a show for a friend shot on one that went up in the National Museum here...12 prints at 108x135cm or so (a bit bigger than 40x50 inches). The images were superb. I have bought three of the damn things and they ranged from ok to terrible. (well, the terrible one was actually a frankenlens...someone screwed the rear element from a 150mm symmar into the front of a 150mm apo sironar s and sold it like that). One was superb, and 2 more were just ok. Leica seems a bit better about this than most, but it is quite common. If you get a good copy of a lens that you like, think twice about selling it, no matter what the MTF of a latest and greatest says. 

Edited June 12, 2019 by Stuart Richardson

[Jeff S]

Lensrentals (Roger) has written and tested extensively regarding sample variation for many lenses from many manufacturers.  That’s why the MTF article I linked above describes their use of multiple lens copies for each chart.

Jeff

[BernardC]

One thing to note, even when MTF charts are based on real lenses (as Leica, Zeiss and Lensrentals do), they should only be compared with other MTF charts from the same source.

In other words, it's OK to compare two Leica MTF charts, but it's tricky to compare a Leica chart with a Zeiss chart. The general shapes can tell you a lot, but the absolute values won't. They may be based on different distances, light spectrum, etc. They may not even use the same scales (30 lp/mm or 40 lp/mm?).

There's also a lot MTF won't tell you, like flare, distortion, the types of aberrations, focus feel, etc. They also won't tell you how the lens performs away from center at optimal focus, which is important for off-center subjects.

[albertknappmd]

On 6/12/2019 at 7:35 PM, Stuart Richardson said:

In general, I find Leica's MTF pretty accurate, but I am not an optical scientist. Leica is one of the few optical manufacturers to provide measured MTF data, rather than theoretical. So they are taking actual lenses they have made and posting that data. One thing to be aware of, however, is that MTF are the measure of the contrast in a flat field. Sometimes manufacturers choose to make a lens with a more curved field, particularly with wider lenses. The 70mm Summarit exhibits this behavior. In the real world, sometimes this is something that people prefer, as three dimensional compositions often feature detail closer to the camera at the sides, with the central point a bit further away. Personally, I prefer flat field lenses, but they are not the best for all situations. The MTF of curved field lenses often look very sharp on center, but then drop rather sharply towards the edges. They might still be able of rendering really sharp detail, just not at the same focused distance as the center. In general, MTF can be dangerous because a lot of people use them as the sole judge of a lenses performance, but they are just one aspect.  One very important thing to note is quality control and sample variation. For example, I shoot a lot of 4x5 and lots of photographers rave about the 150mm APO Sironar S. It is one of the most recent LF lenses, made with the latest and greatest technology (well, 90s anyway). I even printed a show for a friend shot on one that went up in the National Museum here...12 prints at 108x135cm or so (a bit bigger than 40x50 inches). The images were superb. I have bought three of the damn things and they ranged from ok to terrible. (well, the terrible one was actually a frankenlens...someone screwed the rear element from a 150mm symmar into the front of a 150mm apo sironar s and sold it like that). One was superb, and 2 more were just ok. Leica seems a bit better about this than most, but it is quite common. If you get a good copy of a lens that you like, think twice about selling it, no matter what the MTF of a latest and greatest says. 

this has been my experience with the 30-90 zoomlens.. It took 4 tries but I have aral winner.. have kept it for past eight years and will never let it go!

Albert

[albertknappmd]

Advertisement (gone after registration)

oops, it took 4 tries but I have a real winner...

[ZHNL]

Leica's MTF is also calculated per my memory. However, I have found their MTF match real world results closely that I trust their curve. 

PS, I also have no complain about 30-90 zoom performance. It resolve 37M with no problem at landscape aperture cross range. There must be either strong copy variation problem or I think most likely is the AF accuracy problem. Especially many users use it with S2/S006 without LV implementation. 

Edited June 19, 2019 by ZHNL