Technorati: F and Z mounts and microlenses

bclaff · Sep 19, 2018

Marianne Oelund said:
bclaff said:

Marianne Oelund said:

I think we've converged to the same point.

Click to expand...

Perhaps.

But, unless I misunderstand, you are implying that the red and green lines in the following are at the same angle:

That's my chief issue with what I thought you said.

Click to expand...

No, that's not what I'm referring to. The red line is the physical stop location. The apparent exit pupil is what you see through the lens elements.

I've measured this for a number of lenses, since it's a parameter that relates to the PDAF system. For some lenses, especially rear focus types like the DC lenses, the position changes greatly with focus.

The red line is not the aperture stop, it is the exit pupil (close in this case).

The edge virtual image of the exit pupil is the green line.
Note that it comes from the last lens element.

--
Bill ( Your trusted source for independent sensor data at PhotonsToPhotos )

Astrophotographer 10 · Sep 19, 2018

rhlpetrus said:
I have read that corners of sensors require different microlenses to compensate for light reduction due to angle of incidence of light coming from camera lens. Is that true, for example, for F mount lenses and Nikon dslrs?

Z mount uses shorter flange distance and larger diameter, so that typical angle of incidence of light reaching sensor corners is different, possibly requiring less microlens compensation. Also, is that true?

But then, F mount lenses used with FTZ adapter would show, possibly, more vignetting, if answers to above questions were both “yes”.

Could our tech experts, Iliah, Jim, Bobn2, Horshack and others, clarify the issue?

Thanks.

That was true on the Sony A7r but I haven't heard of that being done on any other camera.

Greg.

Marianne Oelund · Sep 19, 2018

bclaff said:
Marianne Oelund said:

bclaff said:

Marianne Oelund said:

I think we've converged to the same point.

Click to expand...

Perhaps.

But, unless I misunderstand, you are implying that the red and green lines in the following are at the same angle:

That's my chief issue with what I thought you said.

Click to expand...

No, that's not what I'm referring to. The red line is the physical stop location. The apparent exit pupil is what you see through the lens elements.

I've measured this for a number of lenses, since it's a parameter that relates to the PDAF system. For some lenses, especially rear focus types like the DC lenses, the position changes greatly with focus.

Click to expand...

The red line is not the aperture stop, it is the exit pupil (close in this case).

I'm not that picky. The stop is always placed close to the exit pupil, otherwise it will be an additional source of vignetting.

bclaff said:
The edge virtual image of the exit pupil is the green line.

OK, but again, that's not the guiding parameter for designing the microlens array.

bclaff said:
Note that it comes from the last lens element.

That isn't how I think about it. It comes from all of the lens elements. You can't ignore the other contributions to the design. Most of the rear elements together determine the angle of the green line, not just the rearmost element. The back element could be made larger, then its edge wouldn't define the angle; its matching size is just a cost consideration.

I prefer to use the concept of apparent exit pupil distance to find the chief ray angle. It's easy to measure, you don't need to know the details of the lens design and you don't need to carry out laborious calculations. It's an "engineer thing," a way to find answers quickly to sufficient accuracy, just like Thevenin and Norton equivalents in electrical circuit analysis.

--
Source credit: Prov 2:6
- Marianne

bclaff · Sep 19, 2018

Marianne Oelund said:
bclaff said:

Marianne Oelund said:

bclaff said:

Marianne Oelund said:

I think we've converged to the same point.

Click to expand...

Perhaps.

But, unless I misunderstand, you are implying that the red and green lines in the following are at the same angle:

That's my chief issue with what I thought you said.

Click to expand...

No, that's not what I'm referring to. The red line is the physical stop location. The apparent exit pupil is what you see through the lens elements.

I've measured this for a number of lenses, since it's a parameter that relates to the PDAF system. For some lenses, especially rear focus types like the DC lenses, the position changes greatly with focus.

Click to expand...

The red line is not the aperture stop, it is the exit pupil (close in this case).

Click to expand...

I'm not that picky. The stop is always placed close to the exit pupil, otherwise it will be an additional source of vignetting.

I suppose that depends on what you call "close", especially for wide-angle or fish-eye designs.
Here are two very different 18mm designs:

8131dae655f84b06bb1b3665bcc303f5.jpg.png

6fd4624650874bc0914bdfcc6eeaec9e.jpg.png

The black tick marks are the Aperture Stop (wide open) and the red line labelled P' is the Exit Pupil.
In both cases the Exit Pupil is displaced from the Aperture Stop.
One is closer to the sensor and the other further away.
Hard to make a generalization; I think.

Marianne Oelund said:
Marianne Oelund said:

The edge virtual image of the exit pupil is the green line.

Click to expand...

OK, but again, that's not the guiding parameter for designing the microlens array.

Agreed; but I'm pretty sure neither of us knows the guiding parameter

I'm pretty sure it's a compromise in any case.

Marianne Oelund said:
Marianne Oelund said:

Note that it comes from the last lens element.

Click to expand...

That isn't how I think about it. It comes from all of the lens elements. You can't ignore the other contributions to the design. Most of the rear elements together determine the angle of the green line, not just the rearmost element. The back element could be made larger, then its edge wouldn't define the angle; its matching size is just a cost consideration.

I prefer to use the concept of apparent exit pupil distance to find the chief ray angle. It's easy to measure, you don't need to know the details of the lens design and you don't need to carry out laborious calculations. It's an "engineer thing," a way to find answers quickly to sufficient accuracy, just like Thevenin and Norton equivalents in electrical circuit analysis.

Well, the Chief Ray always points at the center of the Aperture Stop, so that leave the Exit Pupil out entirely.

ff69e7126b2b41409bd7dd2dee408316.jpg.png

The green line is the Chief Ray.

FWIW, I'm pretty sure we aren't having a big disagreement, just fine-tuning our communication ;-)

Regards

--
Bill ( Your trusted source for independent sensor data at PhotonsToPhotos )

Marianne Oelund · Sep 19, 2018

bclaff said:
bclaff said:

OK, but again, that's not the guiding parameter for designing the microlens array.

Click to expand...

Agreed; but I'm pretty sure neither of us knows the guiding parameter
I'm pretty sure it's a compromise in any case.

Oh, it's a compromise, alright - a huge one. Three degrees isn't anywhere close to what would work best with most lenses.

It'll be interesting to see what they actually did for the Z bodies. Eventually, I'll have one because I can't resist the performance of the new S line lenses, but I might wait for the "pro" model to be released.

Also, on an earlier point of confusion - you are equating the exit pupil to the rear principal plane, but the definition I use for exit pupil (ref. Hecht) is the image of the aperture stop. http://hyperphysics.phy-astr.gsu.edu/hbase/geoopt/stop.html

Iliah Borg · Sep 19, 2018

JimKasson said:
While the ray tracing is sparse, this illustration from a Canon white paper shows a tantalizing look at what the combination of a wide throat and a short mount might allow:

Jim

7.53 mm back focal distance Biogon 2,8/35

34.85 mm FFD in Contax mount

f02c375e44a8459ebd8212491acf84c3.jpg.png

--

About LibRaw | LibRaw

www.libraw.org

JimKasson · Sep 19, 2018

Iliah Borg said:
JimKasson said:

While the ray tracing is sparse, this illustration from a Canon white paper shows a tantalizing look at what the combination of a wide throat and a short mount might allow:

Jim

Click to expand...

7.53 mm back focal distance Biogon 2,8/35

34.85 mm FFD in Contax mount

Thanks for that. Nothing new under the sun, huh?

--
Posted as a regular forum member.

the last word the last word - Photography meets digital computer technology. Photography wins -- most of the time.

Photography meets digital computer technology. Photography wins -- most of the time.

blog.kasson.com

Iliah Borg · Sep 19, 2018

JimKasson said:
Iliah Borg said:

JimKasson said:

While the ray tracing is sparse, this illustration from a Canon white paper shows a tantalizing look at what the combination of a wide throat and a short mount might allow:

Jim

Click to expand...

7.53 mm back focal distance Biogon 2,8/35

34.85 mm FFD in Contax mount

Click to expand...

Thanks for that. Nothing new under the sun, huh?

Sometimes SLRs use an accessory viewfinder, like for 2.1cm f/4-16 Nikkor-O

--

About LibRaw | LibRaw

www.libraw.org

rhlpetrus · Sep 20, 2018

Iliah Borg said:
JimKasson said:

Iliah Borg said:

JimKasson said:

While the ray tracing is sparse, this illustration from a Canon white paper shows a tantalizing look at what the combination of a wide throat and a short mount might allow:

Jim

Click to expand...

7.53 mm back focal distance Biogon 2,8/35

34.85 mm FFD in Contax mount

Click to expand...

Thanks for that. Nothing new under the sun, huh?

Click to expand...

Sometimes SLRs use an accessory viewfinder, like for 2.1cm f/4-16 Nikkor-O

Pretty kit. Why the need for the VF? Too dark TTL?

--
Renato.

Renato Pedrosa

Explore Renato Pedrosa’s 937 photos on Flickr!

www.flickr.com

OnExposure member

http://www.onexposure.net/

Good shooting and good luck
(after Ed Murrow)

Iliah Borg · Sep 20, 2018

rhlpetrus said:
Iliah Borg said:

Sometimes SLRs use an accessory viewfinder, like for 2.1cm f/4-16 Nikkor-O

Click to expand...

Pretty kit. Why the need for the VF? Too dark TTL?

Lens protrudes into the mirror chamber (that black cylinder on the rear of the lens goes far into the chamber), so to mount the lens the mirror needs to be locked in the upper position before mounting. Thus the TTL viewfinder is blocked.

--
http://www.libraw.org/

JimKasson · Sep 20, 2018

Iliah Borg said:
rhlpetrus said:

Iliah Borg said:

Sometimes SLRs use an accessory viewfinder, like for 2.1cm f/4-16 Nikkor-O

Click to expand...

Pretty kit. Why the need for the VF? Too dark TTL?

Click to expand...

Lens protrudes into the mirror chamber (that black cylinder on the rear of the lens goes far into the chamber), so to mount the lens the mirror needs to be locked in the upper position before mounting. Thus the TTL viewfinder is blocked.

If i remember right, with the Nikon F and the early 8mm fisheye, there was no interlock, and you could damage the mirror if you forgot to lock it up before mounting the lens.

--
Posted as a regular forum member.

the last word the last word - Photography meets digital computer technology. Photography wins -- most of the time.

Photography meets digital computer technology. Photography wins -- most of the time.

blog.kasson.com

Iliah Borg · Sep 20, 2018

JimKasson said:
Iliah Borg said:

rhlpetrus said:

Iliah Borg said:

Sometimes SLRs use an accessory viewfinder, like for 2.1cm f/4-16 Nikkor-O

Click to expand...

Pretty kit. Why the need for the VF? Too dark TTL?

Click to expand...

Lens protrudes into the mirror chamber (that black cylinder on the rear of the lens goes far into the chamber), so to mount the lens the mirror needs to be locked in the upper position before mounting. Thus the TTL viewfinder is blocked.

Click to expand...

If i remember right, with the Nikon F and the early 8mm fisheye, there was no interlock, and you could damage the mirror if you forgot to lock it up before mounting the lens.

Yes, Fish-eye-NIKKOR 1:8 f=8mm and Fish-eye-NIKKOR 1:5.6 f=7.5mm, both protruded very far.

Fisheye-NIKKOR Auto 1:2.8 f=8mm was much less protruding, still needed mirror out of the way to mount it properly (we used it to shoot welding seems inside pipes for documentation purposes).

Some images of the lenses here

http://www.destoutz.ch/lenses_fisheye_8mm.html

Another protruding lens is Fisheye-NIKKOR 1:5.6 f=6mm, it has 11.3mm back focus distance, but the actual clearance is even less, because of the metal tube on the back, 6.2mm. Coverage is 21.6mm circle.

A "modest" Fisheye-Nikkor 1:5.6 f=10mm OP (orthographic projection "fisheye for scientists"), also protruding. It is used to measure brightness across the field, as it has no vignetting. It is the only Nikkor that I know to contain a plastic lens element.

--
http://www.libraw.org/

Andre Yew · Sep 20, 2018

The Zeiss Hologon is another extreme example with a 4.5mm back-plane distance for the M-mount 15mm f/8.

Iliah Borg · Sep 20, 2018

Andre Yew said:
The Zeiss Hologon is another extreme example with a 4.5mm back-plane distance for the M-mount 15mm f/8.

Also Voigtlander SL 12/5.6 and SL 15/4.5, should work with Z-mount + adapter.

More than a few older lenses are deeply recessed non-retrofocus, which normally require mirror lockup. The photographic value of such lenses on a high resolution digital mirrorless camera should be an interesting study.

JimKasson · Sep 20, 2018

Iliah Borg said:
Andre Yew said:

The Zeiss Hologon is another extreme example with a 4.5mm back-plane distance for the M-mount 15mm f/8.

Click to expand...

Also Voigtlander SL 12/5.6 and SL 15/4.5, should work with Z-mount + adapter.

More than a few older lenses are deeply recessed non-retrofocus, which normally require mirror lockup. The photographic value of such lenses on a high resolution digital mirrorless camera should be an interesting study.

BSI will help with CFA crosstalk, but the sensor stack may cause corner smearing, as with many short M-mount lenses.

Jim

just Tony · Sep 20, 2018

JimKasson said:
Iliah Borg said:

Andre Yew said:

The Zeiss Hologon is another extreme example with a 4.5mm back-plane distance for the M-mount 15mm f/8.

Click to expand...

Also Voigtlander SL 12/5.6 and SL 15/4.5, should work with Z-mount + adapter.

More than a few older lenses are deeply recessed non-retrofocus, which normally require mirror lockup. The photographic value of such lenses on a high resolution digital mirrorless camera should be an interesting study.

Click to expand...

BSI will help with CFA crosstalk, but the sensor stack may cause corner smearing, as with many short M-mount lenses.

And as a practical matter, the mirror lockups on the F, F2, and F3 were purely mechanical. I have concerns that most of the subsequent reflex cameras would be at risk of a collision upon interruption of the electrical power source.

But mirrorless is another story, as long as the BFD exceeds the thickness of the sensor stack + shutter anyway.

Iliah Borg · Sep 20, 2018

JimKasson said:
Iliah Borg said:

Andre Yew said:

The Zeiss Hologon is another extreme example with a 4.5mm back-plane distance for the M-mount 15mm f/8.

Click to expand...

Also Voigtlander SL 12/5.6 and SL 15/4.5, should work with Z-mount + adapter.

More than a few older lenses are deeply recessed non-retrofocus, which normally require mirror lockup. The photographic value of such lenses on a high resolution digital mirrorless camera should be an interesting study.

Click to expand...

BSI will help with CFA crosstalk, but the sensor stack may cause corner smearing, as with many short M-mount lenses.

Jim

I'm thinking that, but also comparing performance to retrofocus designs for the same focal lengths.

Technorati: F and Z mounts and microlenses

Veteran Member

Veteran Member

Veteran Member

Veteran Member

Veteran Member

Veteran Member

Community Leader

Veteran Member

Veteran Member

Veteran Member

Community Leader

Veteran Member

Veteran Member

Veteran Member

Community Leader

Veteran Member

Veteran Member

Keyboard shortcuts