L-Mount alliance: cameras with different sensor glass thickness

[chrismuc]

Advertisement (gone after registration)

Leica, Sigma and Panasonic are the initial members of the L Mount alliance. But Leica uses a thinner sensor glass while Sigma and Panasonic use a thicker sensor glass. That basically makes it impossible that one particular (wide angle) lens design can achieve the same, excellent corner sharpness with all three cameras.

1. Is there any knowledge if Leica or Sigma or Panasonic wide angle prime or zoom lenses perform different on Leica SL/SL2/SL3 vs. on S1R or FP-L?
2. When re-batching for example Sigma L 14-24f2.8 or Panasonic L 35f2 lenses, did Leica adjust the optical formula to optimise the lenses for the thinner sensor glass? Did Peter Karbe ever comment on that topic?

[Advertisement]

The Leica Q3 is available for pre-order at B&H Photo Video and Adorama

[hdmesa]

Re #2: My understanding is that thinner cover glass only helps sharpness, but I could be wrong.

[jankap]

vor 5 Stunden schrieb chrismuc:

Leica, Sigma and Panasonic are the initial members of the L Mount alliance. But Leica uses a thinner sensor glass while Sigma and Panasonic use a thicker sensor glass. That basically makes it impossible that one particular (wide angle) lens design can achieve the same, excellent corner sharpness with all three cameras.

1. Is there any knowledge if Leica or Sigma or Panasonic wide angle prime or zoom lenses perform different on Leica SL/SL2/SL3 vs. on S1R or FP-L?
2. When re-batching for example Sigma L 14-24f2.8 or Panasonic L 35f2 lenses, did Leica adjust the optical formula to optimise the lenses for the thinner sensor glass? Did Peter Karbe ever comment on that topic?

I believe, that the fp and the fp-l differ in this respect.?

[jaapv]

5 hours ago, hdmesa said:

Re #2: My understanding is that thinner cover glass only helps sharpness, but I could be wrong.

The thicker glass will also show more refraction towards the corners which will have a negative effect on lenses with a steep incidence angle resulting in smearing and colour shifts. 

[chrismuc]

The glass in front of the sensor (either glued to the sensor like most cameras or in a certain distance like GFX) basically acts like a lens, a lens with plane parallel surfaces. So it can be added to the lens design, that means the corners of wide angle lenses can be perfectly sharp, if that glass thickness was considered in the lens design. My M 24f3.8 and M 35f2.4 (which were designed for no glass or thin glass) are quite much sharper towards the corners on my M9 (thin glass) than on my S1R (thick glass). I am just wondering if Leica considered that matter when releasing their L 14-24f2.8 and L 35f2 versions of the respective Sigma and Panasonic lenses, otherwise these lenses might perform a bit worse on Leica SL cameras than on Panasonic and Sigma cameras, they were designed for. 

[jaapv]

That depends on the telecentricity oc the design. Why don’t you ask Sigma? 

[BernardC]

Advertisement (gone after registration)

19 hours ago, chrismuc said:

Leica, Sigma and Panasonic are the initial members of the L Mount alliance. But Leica uses a thinner sensor glass while Sigma and Panasonic use a thicker sensor glass. That basically makes it impossible that one particular (wide angle) lens design can achieve the same, excellent corner sharpness with all three cameras.

Telecentric lenses don't care how thick the sensor glass is. Nearly-telecentric lenses like we get for digital cameras are barely affected, because all of the light hits the sensor glass at nearly the same angle. The net effect of the sensor glass is to move the optical focal point back by 1/3 of the thickness of the glass. In other words, there is no reason to believe that a particular L-Mount lens will be "better" on Leica vs. Lumix or Sigma.

Sensor glass thickness is only significant for legacy film-era lenses, especially certain rangefinder wide angle designs. Leica is the only L-mount partner that explicitly supports these types of lenses (M-mount lenses in this case). Blackmagic, DJI, Panasonic, Sigma, etc., do not. M-mount lenses won't be optimized with these non-Leica cameras, although they'll usually perform better than they would on non-L-mount cameras made by Sony and Nikon.

Olympus was the first brand to make a big deal of this, over 20 years ago, but telecentric lenses have other use cases and were common 50 years before that. For instance, lenses designed to be used in front of a beam splitter (video, Technicolor) were telecentric.

People talk a lot about "sensor glass thickness," but that doesn't mean it's the most critical factor. You can replace the sensor glass on a Sony camera (or send it out to have the glass replaced), and it will be slightly better than before (with M lenses), but still not nearly as good as a Leica M or SL.

 

 

[Dr. G]

19 minutes ago, BernardC said:

Telecentric lenses don't care how thick the sensor glass is. Nearly-telecentric lenses like we get for digital cameras are barely affected, because all of the light hits the sensor glass at nearly the same angle. The net effect of the sensor glass is to move the optical focal point back by 1/3 of the thickness of the glass. In other words, there is no reason to believe that a particular L-Mount lens will be "better" on Leica vs. Lumix or Sigma.

Sensor glass thickness is only significant for legacy film-era lenses, especially certain rangefinder wide angle designs. Leica is the only L-mount partner that explicitly supports these types of lenses (M-mount lenses in this case). Blackmagic, DJI, Panasonic, Sigma, etc., do not. M-mount lenses won't be optimized with these non-Leica cameras, although they'll usually perform better than they would on non-L-mount cameras made by Sony and Nikon.

Olympus was the first brand to make a big deal of this, over 20 years ago, but telecentric lenses have other use cases and were common 50 years before that. For instance, lenses designed to be used in front of a beam splitter (video, Technicolor) were telecentric.

People talk a lot about "sensor glass thickness," but that doesn't mean it's the most critical factor. You can replace the sensor glass on a Sony camera (or send it out to have the glass replaced), and it will be slightly better than before (with M lenses), but still not nearly as good as a Leica M or SL.

 

 

Correct me if I'm wrong, but doesn't the FP-L have an anti aliasing filter?  I would think that would have more effect on the image than slightly thicker glass.  

[BernardC]

1 minute ago, Dr. G said:

Correct me if I'm wrong, but doesn't the FP-L have an anti aliasing filter?  I would think that would have more effect on the image than slightly thicker glass.  

You are correct. Sensor glass thickness isn't the most relevant factor. The fp-l has an anti-aliasing filter because it has advantages for video (where resolution is limited). It may be slightly less sharp than a camera without such a filter, when used for stills, but probably not to a great extent. Maybe someone here has an fp-l and an SL3, and can run some tests.

Lensrentals addressed the issue of sensor glass thickness 10 years ago. Their general conclusion was that it's only an issue with certain film-era lenses, especially rangefinder designs (like M lenses). All L-mount lenses were designed for digital sensors, so they shouldn't be materially affected by small differences in sensor glass thickness.

 

[Dr. G]

14 minutes ago, BernardC said:

You are correct. Sensor glass thickness isn't the most relevant factor. The fp-l has an anti-aliasing filter because it has advantages for video (where resolution is limited). It may be slightly less sharp than a camera without such a filter, when used for stills, but probably not to a great extent. Maybe someone here has an fp-l and an SL3, and can run some tests.

Lensrentals addressed the issue of sensor glass thickness 10 years ago. Their general conclusion was that it's only an issue with certain film-era lenses, especially rangefinder designs (like M lenses). All L-mount lenses were designed for digital sensors, so they shouldn't be materially affected by small differences in sensor glass thickness.

 

I wonder why they used an anti aliasing filter on the Nikon Zf when it's primarily a stills camera.  I was thinking about trying the Zf with some of the Z Mount Voigtlander APO lenses, as I still find manual focusing difficult on the SL3.  I thought that the manual focus aids on the Zf with the Voigtlanders would be a great comibination.  I was told by a few people that if I was used to the images from the SL platform with the APO Summicron-SL lenses I wouldn't be happy with the Zf images because of the AA filter.  

[BernardC]

3 minutes ago, Dr. G said:

I wonder why they used an anti aliasing filter on the Nikon Zf when it's primarily a stills camera.

It's part of the "special sauce" that each brand brings to the table. The 3 big Japanese brands have traditionally favoured anti-aliasing, at the expense of ultimate sharpness, whereas Leica has gone the other way (starting with the M8). It probably doesn't make a huge difference with very high-resolution sensors, unless you are pixel-peeping.

[LBJ2]

6 hours ago, chrismuc said:

The glass in front of the sensor (either glued to the sensor like most cameras or in a certain distance like GFX) basically acts like a lens, a lens with plane parallel surfaces. So it can be added to the lens design, that means the corners of wide angle lenses can be perfectly sharp, if that glass thickness was considered in the lens design. My M 24f3.8 and M 35f2.4 (which were designed for no glass or thin glass) are quite much sharper towards the corners on my M9 (thin glass) than on my S1R (thick glass). I am just wondering if Leica considered that matter when releasing their L 14-24f2.8 and L 35f2 versions of the respective Sigma and Panasonic lenses, otherwise these lenses might perform a bit worse on Leica SL cameras than on Panasonic and Sigma cameras, they were designed for. 

I don't remember hearing Peter Karbe address this topic regarding L-Mount in any of the articles or videos e.g., Youtube videos where he was interviewed. If someone knows otherwise, I would be interested to know as well. 

From what I understand, Leica digital cameras include thinner sensor stack and customized microlenses. As has already been mentioned, this should benefit adapting Leica M/rangefinder lenses made for film particularly the wider lenses. *FWIW, I've noticed my Carl Zeiss Contax G 28 made for film lens the corners still smear when adapted to the SL2, albeit a little less corner smear than when I adapt the Contax G 28  to my thicker sensor stack Sony camera. 

Edited December 30, 2024 by LBJ2

[Photoworks]

from the different talk of Leica designers, the SL camera has an evolution in the micro prism glass.

Leica has made improvements from SL1 to SL2 with thinner glass with the goal of reproducing the same quality images with all the M lenses.

The SL3 increased the distance of the cover glass to the sensor in order to minimize sensor dust.

The SL cameras are all a bit sharper with any lens than the Panasonic counterparts.
I have noticed that the SL2-3 renders sharper and less smearing with adapted lenses; the same lenses used on a Sony body are much softer in the corners.

[FlashGordonPhotography]

It matters for M glass. Not so much for SL lenses.

I did extensive testing between the S1R and SL2 when they were both new. I did most of my testing with the 50 Lux and 90 SL APO and the 24-90 and 90-280. I really couldn’t see any functional differences with the zooms. There was a TINY difference with the prime lenses.And it didn’t always show. I was at 2-400% trying to see subtle differences. Sometimes the Leica showed a bit more *bite*.

I concluded that for my uses any differences were irrelevant. Once the file went to print any differences basically evaporated anyway. I stopped worrying about it and moved on.

Gordon

[jaapv]

6 hours ago, Dr. G said:

I wonder why they used an anti aliasing filter on the Nikon Zf when it's primarily a stills camera.  I was thinking about trying the Zf with some of the Z Mount Voigtlander APO lenses, as I still find manual focusing difficult on the SL3.  I thought that the manual focus aids on the Zf with the Voigtlanders would be a great comibination.  I was told by a few people that if I was used to the images from the SL platform with the APO Summicron-SL lenses I wouldn't be happy with the Zf images because of the AA filter.  

Try bird photography… I have learnt to cope with aliasing in post, but it is a real problem. 

[laowai_]

On 12/30/2024 at 7:43 AM, Dr. G said:

I wonder why they used an anti aliasing filter on the Nikon Zf when it's primarily a stills camera.  I was thinking about trying the Zf with some of the Z Mount Voigtlander APO lenses, as I still find manual focusing difficult on the SL3.  I thought that the manual focus aids on the Zf with the Voigtlanders would be a great comibination.  I was told by a few people that if I was used to the images from the SL platform with the APO Summicron-SL lenses I wouldn't be happy with the Zf images because of the AA filter.  

As sensor pixel count increases, the Nyquist frequency, i.e. the spatial frequency at which aliasing happens, gets pushed to higher and higher values. I guess that’s why some vendors chose to remove the AA filter above a certain pixel count and for certain target customer segments.
The upside is an increase in perceived ‘sharpness’ of the image, the downside is that you might see aliasing on subjects with high spatial frequencies such as feathers of birds.
Engineering is all about trade offs, and marketing is about positioning. Leica and Nikon are doing a fantastic job at both.

 

[hdmesa]

2 hours ago, laowai_ said:

As sensor pixel count increases, the Nyquist frequency, i.e. the spatial frequency at which aliasing happens, gets pushed to higher and higher values. I guess that’s why some vendors chose to remove the AA filter above a certain pixel count and for certain target customer segments.
The upside is an increase in perceived ‘sharpness’ of the image, the downside is that you might see aliasing on subjects with high spatial frequencies such as feathers of birds.
Engineering is all about trade offs, and marketing is about positioning. Leica and Nikon are doing a fantastic job at both.

It does, but we still need a much higher pixel density to avoid aliasing at f/8 and wider. I still saw image-ruining aliasing in woven fabrics with the SL3 and GFX 100S (which share base silicon and pixel density), even with older lens designs adapted to each platform. If I recall correctly, @JimKasson has said we need 200-250mp on the GFX as an ideal resolution.

Canon has the best AA filters out there for retaining sharpness at the pixel level while still avoiding aliasing.

[Jon Warwick]

5 hours ago, hdmesa said:

It does, but we still need a much higher pixel density to avoid aliasing at f/8 and wider. I still saw image-ruining aliasing in woven fabrics with the SL3 and GFX 100S (which share base silicon and pixel density).

Yes, I often see aliasing on the color M11 (and SL3 that I tested), but it’s a bit less common on the GFX100 (not sure why - the GF lenses are sharp, but maybe less sharp than some of the best of the best Leica and Voigtlander APOs?). “Enhance” in ACR seems to help to clean up some of the artefacts though.

In comparison, the M11 Monochrom is magnificent in its purity of capture (ie, no aliasing) when head to head with all those color filter array cameras, and due to the lack of aliasing in particular, it seems to record more fine detail as a result (remarkably, to my eyes, I see more fine detail with the M11M than I get off the 100mp medium format).

It’s aliasing and getting rid of it at capture is why I’d like to see the SL3 with a multi shot mode. Will Leica ever produce pixel shift for the SL3 in a firmware update?!

Edited January 2 by Jon Warwick

[Advertisement]

Hallo,

Look here: Leica L-Mount

[Dr. G]

19 minutes ago, Jon Warwick said:

Yes, I often see aliasing on the color M11 (and SL3 that I tested), but it’s a bit less common on the GFX100 (not sure why - the GF lenses are sharp, but maybe less sharp than some of the best of the best Leica and Voigtlander APOs?). “Enhance” in ACR seems to help to clean up some of the artefacts though.

In comparison, the M11 Monochrom is magnificent in its purity of capture (ie, no aliasing) when head to head with all those color filter array cameras, and due to the lack of aliasing in particular, it seems to record more fine detail as a result (remarkably, to my eyes, I see more fine detail with the M11M than I get off the 100mp medium format).

It’s aliasing and getting rid of it at capture is why I’d like to see the SL3 with a multi shot mode. Will Leica ever produce pixel shift for the SL3 in a firmware update?!

Remember that a color sensor with 100MP probably has less effective and perceived resolution than a 60MP monochrome sensor due to the presence of a Bayer (or X-Trans) filter.  

Edited January 2 by Dr. G

[hdmesa]

1 hour ago, Dr. G said:

Remember that a color sensor with 100MP probably has less effective and perceived resolution than a 60MP monochrome sensor due to the presence of a Bayer (or X-Trans) filter.  

Not really, they're about the same in practice, though lens choice makes a difference. I never tested them side-by-side when I owned them, but I also never thought the M11M punched above the GFX 100S, though I did think it was more or less its equal.

Edited January 3 by hdmesa