Sensor stack - affect of glass thickness and lens

[GaryLH]

http://www.lensrentals.com/blog/2014/06/the-glass-in-the-path-sensor-stacks-and-adapted-lenses

Another pretty interesting article from Roger. It turns out that the lenses are designed to account for the thickness of the glass (aa/ir filter glass). Leica uses the thinnest and m43 the thickest. Roger shows the effect of thickness of the glass in front of sensor vs the mtf info that is produced by his new lens testing rig.

Gary

 

[Paulbe]

Roger's posts are always excellent. His analysis deflates a LOT of internet "wisdom."

Paul

 

[krötenblender]

The article was also reposted on Petapixel. As interesting and enlightening as I found the article, as amazed I was at some of the comments below...

 

[Rob-F]

When the M8 came out, it was said that the reason Leica minimized the filtration in front of the sensor was to avoid the degrading effects they have on the image. Hence the need for we M8 owners to put UV/IR filters on our lenses. This was evidently to get peak IQ from our existing lenses.

But then when Leica came out with the M9 and subsequent, they evidently put back some of the filtration they took out. For instance, you don't need the UV-IR filters on the lens with the M9. That suggests they must have increased the thickness of the filter stack. And that seems to imply poorer performance with our existing lenses, not designed for that filter stack.

Comments in the recent thread comparing the M8 and M9 don't, as I recall, say with any consistency, that the M9 has any better IQ than the M8, its full size sensor and higher pixel count not withstanding. At least one poster said he thought the M8 IQ was just a little better. So, I'm suspecting that IQ degradation caused by the M9 filter stack might be just about offset by the larger sensor and higher pixel count; with the result that the two cameras come out about even.

Agree or disagree?

 

[Chris101]

When the M8 came out, it was said that the reason Leica minimized the filtration in front of the sensor was to avoid the degrading effects they have on the image. Hence the need for we M8 owners to put UV/IR filters on our lenses. This was evidently to get peak IQ from our existing lenses.

But then when Leica came out with the M9 and subsequent, they evidently put back some of the filtration they took out. For instance, you don't need the UV-IR filters on the lens with the M9. That suggests they must have increased the thickness of the filter stack. And that seems to imply poorer performance with our existing lenses, not designed for that filter stack.

Comments in the recent thread comparing the M8 and M9 don't, as I recall, say with any consistency, that the M9 has any better IQ than the M8, its full size sensor and higher pixel count not withstanding. At least one poster said he thought the M8 IQ was just a little better. So, I'm suspecting that IQ degradation caused by the M9 filter stack might be just about offset by the larger sensor and higher pixel count; with the result that the two cameras come out about even.

Agree or disagree?

Did they increase the thickness, or just the optical density of the IR absorbing filter? Filtration is a function of both the thickness, AND the concentration of the absorbing material in the glass.

 

[Rob-F]

Did they increase the thickness, or just the optical density of the IR absorbing filter? Filtration is a function of both the thickness, AND the concentration of the absorbing material in the glass.

I don't know. Hopefully we can get some info contributed about that. But if they can increase the density, why didn't they do that in the first place?

 

[willie_901]

Rob,

If I remember correctly, alledgedly IR materials with the proper optical qualities and sufficient density were not yet commercially available.

 

[ColSebastianMoran]

http://www.lensrentals.com/blog/2014/06/the-glass-in-the-path-sensor-stacks-and-adapted-lenses

Another pretty interesting article from Roger. It turns out that the lenses are designed to account for the thickness of the glass (aa/ir filter glass). … snip ...

Thanks for this link. The article by Roger Cicala is very interesting, and not just theory, he has measurements.

The gist of the article, as I read it:
- glass filter pack in front of the sensor has optical properties (e.g. chromatic aberration)
- lens designed for a specific camera is probably optimized for that filter pack
- to some extent, taking a lens designed for one filter stack and using it on a camera with different filter stack reduces quality
- we are de-optimizing our legacy lenses, no matter how good, when we use them on digicams with a filter stack (since legacy lenses were designed for no glass at the focal plane)
- most dramatic in Roger's article is micro 4/3, since it has a thick filter pack
- and, the question is how big is this effect? Real world effect?

Roger promises further articles on these questions.

 

[icebear]

LOL - as Roger was mentioned, I assumed the OP was talking about "our" Roger. No it's R. Cicala

I am also not sure about real world implications about effects that you can detect with bench testing.
Optimum focus and avoidance of movement/camera shake will be more important. Just my $0.02 😉

 

[bwidjaja]

Based on my very limited real life testing is that Leica 35mm Summicron ASPH did not perform well on my X-E1. On the other hand, CV 28mm f3.5 Color Skopar is amazing on the same camera with the same adapter.

 

[GaryLH]

Another great follow up article by Roger

Another great follow up article by Roger

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

It starts to show that most rf designed lenses, the closer the exit pupil is to the sensor.. The more that legacy lens will have issues and never perform as well as on the comera body it was intended. Why Leica drf bodies have the thinnest stack thickness.

Gary

Ps.. Since the gxr module was designed w/ m lenses in mind, I wish that it was included in the thickness database 🙁

 

[Spanik]

Read it also yesterday. What surprisd me the most was that Roger of all persons was so surprise about it. It is long known that for instance those mirror lenses with filters at the back need to have at least a clear one to function. Everything with a refraction ndew in the light path will have an influence.