[stevebrot]

Another excellent article from "The Online Photographer", Mike Johnston with embedded "guest", Bruno Masset...

The Online Photographer: Two Reasons...

Significantly, the article explains why exit pupil distance is important as well as the cause of potential issues associated with a deeper stack.


Steve

[SunnyG.]

Excellent idea. It the "guy" is Roger Cicala, he is usually very thorough and very readable.

To save you the trouble...

The 2014 three-part series:

Lens Rentals | Blog: The Glass in the Path: Sensor Stacks and Adapted Lenses

Lens Rentals | Blog: Sensor Stack Thickness: When Does It Matter?

Lens Rentals | Blog: Sensor Stack Thickness Part III: The Summary

The part III summary is the most useful, but it helps to read the other two. I am familiar with the series based on the earlier discussion here a few years back.

The 2015 followup:

Lens Rentals | Blog: A Thinner Sensor Stack

The last is quite telling in regards to adapted rangefinder lenses.

We should all be on the same page now.


Steve

Thanks for the all in link! Other than the 3rd part I read all of the articles together!

---------- Post added 09-02-20 at 11:43 PM ----------

OK...this is the part I am not seeing in any of the four Cicala articles. In fact, he appears to avoid drawing any lines in the sand. Going forward, all Pentax SLR lenses will have exit pupil distance (at infinity) 45.46mm or longer, simply to allow for the mirror box. According to the Part II article (LINK), the sole Pentax lens measured (Super Tak 50mm f/1.4) measured at 70.5mm. This does not surprise me since traditional Pentax normals are mildly retrofocus.

Going to the Part III article (LINK), his comments regarding SLR lenses begin with a simple statement:


In regards to Pentax physical stack thickness, the Part II article has the K10D at 1.6mm. Removing the AA filter, as with all current model Pentax dSLRs, I would expect the same or thinner on those. It is probably enough that Cicala basically lumped dSLRs together as having about 2mm filter stacks. I take Cicala's Summary from Part III as a reasonable guide for practical matters:


As for shooting designed-for-film SLR lenses on any recent model Pentax dSLRs, the simple answer is that the results we get are the results we get. Whether poor results are a result of sensor stack depth is hard to say, though the more usual culprits are missed focus and camera movement.


Steve

I agree it's not mentioned in the article! But after throughly reading the article, that's what I inferred. Anything in and around 70 mm aren't at their peak performance. That's where things start to fall off.

---------- Post added 09-02-20 at 11:49 PM ----------

Another excellent article from "The Online Photographer", Mike Johnston with embedded "guest", Bruno Masset...

The Online Photographer: Two Reasons...

Significantly, the article explains why exit pupil distance is important as well as the cause of potential issues associated with a deeper stack.


Steve

Thank you for this article as well!!! I really liked this one too. Now my mind is at peace 😌

[pinholecam]

What appears to be the case is that this is mostly a non issue when using adapted SLR or DSLR lenses on Sony. It is more of an issue with Leica M lenses on m4/3. Basically if the sensor glass thickness is significantly different, there can be an optical effect.

Sensor Stack Thickness: When Does It Matter?

Oh and the closer the exit pupil to the sensor, the more impact this has. Slr wide angles with their retrofocal designs are typically less impacted than rangefinder wide angles.

Depending on how picky one is with edge sharpness, it can/does affect even SLR lenses (at least the Pentax K mount ones).
I have compared the results shot on my K24/2.8 on a modded and non-modded A7, and there was a slight difference.
In comparison, this same K24/2.8 does perform better on the K1 than the modded A7 when the file is reduced to 24mp.
I've also seen samples passed to be from the A7r with the K24/2.8 (also a 36mp sensor) and they were outright horrible on the edges (unlike the K1)




The effect for old RF wides are significantly worse, thats for sure.
That said, the newer RF lenses designed to cater for this (and out for many years already ) does take care of the issue.
Some examples being the latest versions of the Voigtlander 10mm,15mm, 21mm.

[SunnyG.]

Depending on how picky one is with edge sharpness, it can/does affect even SLR lenses (at least the Pentax K mount ones).
I have compared the results shot on my K24/2.8 on a modded and non-modded A7, and there was a slight difference.
In comparison, this same K24/2.8 does perform better on the K1 than the modded A7 when the file is reduced to 24mp.
I've also seen samples passed to be from the A7r with the K24/2.8 (also a 36mp sensor) and they were outright horrible on the edges (unlike the K1)




The effect for old RF wides are significantly worse, thats for sure.
That said, the newer RF lenses designed to cater for this (and out for many years already ) does take care of the issue.
Some examples being the latest versions of the Voigtlander 10mm,15mm, 21mm.

Still the newer RF lenses have a problem, because the stack thickness of Leica cameras are between 0.8mm to 1mm. Where as sony is 2mm++ as far as I have read.

The only place where perform better is canon, since canon has stack thickness of 2mm or less

[stevebrot]

I've also seen samples passed to be from the A7r with the K24/2.8 (also a 36mp sensor) and they were outright horrible on the edges (unlike the K1)

Might this be related to Sony's special slanted microlens treatment at the edges?


Steve

[PJ1]

So does this mean that when we read about cleaning a sensor, what is really being cleaned is the stack in front of the sensor?

[RGlasel]

So does this mean that when we read about cleaning a sensor, what is really being cleaned is the stack in front of the sensor?

Yes. The cleaning device/solution doesn't contact the photosensitive layer of the semiconductor (which is exposed to light and nothing else).

[Ian Stuart Forsyth]

So does this mean that when we read about cleaning a sensor, what is really being cleaned is the stack in front of the sensor?

Yes here is the stack removed from the surface of the pentax K7 sensor


Ignore the black circle on the lower part of the stack, it was just felt marker where I ran a blade across it to see what it would take to scratch its surface

This is the K7 sensor without the stack in front of it

Here is the pentax K5 closeup of the sensor without the stack

[pinholecam]

Might this be related to Sony's special slanted microlens treatment at the edges?


Steve

No idea.
Whatever it is, Sony seemed to have resolved most of it on their later "r" designated cameras and I've not seen complaints

---------- Post added 09-04-2020 at 12:06 PM ----------

Was going thru my Flickr and found some sample shots I shot with the CV35/1.4 on a stock A7 and a thin filter modded A7.




A7-kolari modded



A7-stock