Sensor Stacks and Adapted Lenses

Cruzan80

Mu-43 All-Pro
Joined
Aug 23, 2012
Messages
1,107
Location
Denver, Co
Real Name
Sean Rastsmith
I wonder what difference the sensors have from body to body. Does m4/3 all have a thick sensor, or just the one they happened to pull from the GX1?
 

barry13

Mu-43.com Editor
Joined
Mar 7, 2014
Messages
9,192
Location
Southern California
Real Name
Barry
Hi, I read recently that many new sensors, including the 16mp Sonys, are cutting back on filters to gain effective sensitivity.
Afaict, that explains the purple fringing problems with some Pana lenses on Oly bodies.
Oly lenses reportedly make up for it in their coatings.

Barry


Sent from my iPad using Mu-43
 

MajorMagee

Mu-43 All-Pro
Joined
Feb 24, 2011
Messages
1,630
Location
Dayton, OH
Wow, that was interesting. It makes me wonder about future camera bodies, and if they'll need to keep the same overall glass thickness, even if they drop things like the AA filter, just to avoid distorting the optics of their entire lens lineup.

Is it part of the m43 design standard?

I appreciate that Metabones is sophisticated enough to have designed this effect into their speed boosters. I would be surprised if the cheaper alternatives even have a clue about the need to correct for this.
 

HarryS

Mu-43 Top Veteran
Joined
Jun 23, 2012
Messages
988
Location
Midwest, USA
I would be certain that Samyang uses CAD tools to design lenses and even a Zhongyi (lens turbo) probably has the software. Lens design is high level work no matter where you do it. From the article: Brian’s design like that of most lenses) has an optical formula that plans on light rays leaving the lens passing through such a stack before reaching the sensor... I wonder if Metabones uses the same converter lens for its Sony and M43 products, or is the difference big enough to need two different lenses..

It does suggest that lenses put into multiple mounts (like Tamron, Sigma, Rokinon, et al) might NOT work the same on different cameras.


Thanks to OP for the link. An interesting article.
 

fortwodriver

Mu-43 Top Veteran
Joined
Nov 15, 2013
Messages
993
Location
Toronto, Ontario, Canada
Real Name
Frank
The m43 camera I've seen has a filter stack that's about that thickness.
That would be the E-M1, G1, G2. I haven't had a chance to look at any others.
If one didn't, you'd have serious problems with lenses on that camera. It would essentially remove compatibility by eliminating something from the optical path. Incidentally, this relatively thick stack allows for much better accommodation of sensor dust. A dust particle three times farther from the sensor plate is going to blur much more than one that's closer. So, even if the piezo shaker doesn't shed all of the dust particles, you'd have to work with very tiny apertures to see some of the smaller particles stuck to that filter stack.

I think even the early Olympus E-10 and E-20 had a thick cover glass over their 2/3 sensors.
 

MajorMagee

Mu-43 All-Pro
Joined
Feb 24, 2011
Messages
1,630
Location
Dayton, OH
I was thinking about the problems that Fotodiox had with their first focal length reducing adapters.
 

fortwodriver

Mu-43 Top Veteran
Joined
Nov 15, 2013
Messages
993
Location
Toronto, Ontario, Canada
Real Name
Frank
I was thinking about the problems that Fotodiox had with their first focal length reducing adapters.
I would guess that in Fotodiox's case, it was whoever they were contracting out to do the optical work for early versions. They were not paying for the tightest standards. In my opinion, Fotodiox products aren't particularly precise although they have become better over the years.
 

fortwodriver

Mu-43 Top Veteran
Joined
Nov 15, 2013
Messages
993
Location
Toronto, Ontario, Canada
Real Name
Frank
Interesting indeed and it would be nice to see what cameras have what thickness of glass in front of their sensors!
I remember the Nikon D100 and D1 had one of the thinnest stacks I'd seen... They used to advertise that they were made with Lithium Niobate. I remember taking apart a D1 sensor that was dead and finding that the cover glass wasn't much thicker than your typical microscope cover glass - Maybe a mm or two thick. It was so easy to damage those.
 

tjdean01

Mu-43 Top Veteran
Joined
Feb 20, 2013
Messages
862
1) If there's all this glass in front of our sensors, why are we worried about cleaning them ourselves? NEX and Pentax Q clearly has glass. I thought my PM2 had nothing but the sensor. So, I CAN touch it!

2) I'm not understanding how this will affect anything. We here have tried hundreds of adapted lenses with very good, predictable results. What the article and you guys are saying makes sense, however, I'm not seeing any lack of quality.

3) Speed booster: the glass simply makes the image circle of the 35mm lens smaller and thus faster and wider. Since you're using a higher percentage of the intended image circle, sure, I can see how this might be better. But I don't see how the sensor stack could be making anything better or worse. I'm confused.

So, even if the piezo shaker doesn't shed all of the dust particles, you'd have to work with very tiny apertures to see some of the smaller particles stuck to that filter stack..
Again, confused. Regardless of the aperture the image circle would be the same size, right?



So, in short, why is finding out the thickness of the sensor stack making those who use speed boosters happier? And, if a flange distance of a certain mount is X, it will be X regardless. You don't even need to know the thickness of the sensor stack, right?
 

OzRay

Mu-43 Hall of Famer
Joined
Jan 29, 2010
Messages
4,991
Location
South Gippsland, Australia
Real Name
Ray, not Oz
All lenses are designed to focus on a specific image plane (distance from the exit pupil to film/sensor plane); with film it used to be pretty much a constant across the board. With digital, the plane of focus depends on what is in front of the sensor, which becomes part of the optical path, post-lens, and every digital camera's OEM lens design takes this into account. While lenses designed for say a Nikon can work on a m4/3 body very well, there may be things happening that are not ideal, but may not matter greatly in the long run regarding image quality. With the lens speed boosters, it's a different matter altogether, as you are now introducing another optical layer into the equation, which complicates things significantly. I think the article was specifically discussing the speed booster concept, but it can apply to other types of adapters as well.
 

biomed

Mu-43 Hall of Famer
Joined
Aug 22, 2013
Messages
2,669
Location
Seattle area
Real Name
Mike
From the article:

Testing Implications

We may need to make corrections on our optical bench when testing lenses designed to have a certain thickness of glass between their rear element and the sensor. Obviously, we’ll be going back to doing more testing there, too. I suspect, for example, that the numbers I posted in last week’s 50mm article might actually be a tiny bit lower than reality for the Sigma Art and Zeiss Otus lenses.

Or perhaps not. This is a new area and we’ll have to run lots of copies on the bench, and correlate them with Imatest or other complete-systems measurements before we know for sure.

Of course, it’s possible that sensor stack thickness might end up being no big deal. But hey, if it’s important enough for Panavision, it’s important enough for me.


What is the "real world" implication? I have been using both M and F mount lenses on my u4/3 bodies and have been getting excellent results. In my opinion some of the adapted lenses I use are better on a u4/3 body than on the original film body for which they were designed.
 

MajorMagee

Mu-43 All-Pro
Joined
Feb 24, 2011
Messages
1,630
Location
Dayton, OH
Every transition from one material to another (air to glass, glass to plastic, etc.) causes a change in direction of the light rays. The thickness and incidence angle will determine what comes out on the other side. Now add in the complication that the response is frequency dependent (red versus green versus blue light) and you can start to see how it takes a PhD and some pretty complicated optics software to design these to work as intended without degrading the image.
 

fortwodriver

Mu-43 Top Veteran
Joined
Nov 15, 2013
Messages
993
Location
Toronto, Ontario, Canada
Real Name
Frank
1)
Again, confused. Regardless of the aperture the image circle would be the same size, right?
It's got nothing to do with that. It's got to do with how the rays of light strike any dirt on the cover glass. That's why you can see dust particles in your photos when they are taken at small apertures. Large apertures let more oblique rays through the lens system effectively bending around the dust particle. When you close the aperture down (with the exception of the rays that bend around the blades (i.e.: diffraction) the light is centralized causing a deeper shadow behind the dust particle.

Also, these stacks are not simple glass. They are filter stacks. They often contain colour filter surfaces, polarization surfaces and can't simply be considered "flat, plane-perfect" glass. They are essentially specialized filters.
 

RnR

Mu-43 All-Pro
Joined
Sep 25, 2011
Messages
1,238
Location
Brisbane, Australia
Real Name
Hasse
1) If there's all this glass in front of our sensors, why are we worried about cleaning them ourselves? NEX and Pentax Q clearly has glass. I thought my PM2 had nothing but the sensor. So, I CAN touch it!
I think the danger is possible damage to the image stablisation system.

2) I'm not understanding how this will affect anything. We here have tried hundreds of adapted lenses with very good, predictable results. What the article and you guys are saying makes sense, however, I'm not seeing any lack of quality.
Focal reducers really give a 'whoa' result. Its kinda amazing the first time you are in your image editor looking over results. Actually I could see something wonderful was up through the evf.

3) Speed booster: the glass simply makes the image circle of the 35mm lens smaller and thus faster and wider. Since you're using a higher percentage of the intended image circle, sure, I can see how this might be better. But I don't see how the sensor stack could be making anything better or worse. I'm confused.
Have a look at the Metabones Speedbooster white paper on page 5 - http://www.metabones.com/assets/a/stories/Speed Booster White Paper.pdf. The first picture quite clearly shows the effect a piece of flat glass has on the light rays. It also suggests to me that the Metabones Speedbooster for the m43 mount takes the ~4mm filter into account. Also remember that some of the long tele lenses with drop in filters recommends to have simple uv/skylight filter in place while not using any other filter, because a filter in that location is part of the optical design.

Again, confused. Regardless of the aperture the image circle would be the same size, right?
Ignore fortwodriver, it has nothing to do with dirt.

So, in short, why is finding out the thickness of the sensor stack making those who use speed boosters happier? And, if a flange distance of a certain mount is X, it will be X regardless. You don't even need to know the thickness of the sensor stack, right?
I'm happy because there is choice in the market place for focal reducers. Roxsen is selling a decent one. Metabones is selling their Speedbooster which was designed by Brian Caldwell, the gent in the article. From the pdf linked above and Brian's comments in the article, it seems very likely that the Speedbooster design is taking the sensor stack into account giving very good results. Its not clear if the Roxsen does. When I bought mine I took a chance on Brian's reputation on his previous designs, and feel my purchase decision have been right for me.

Edit: confirmed that the Speedbooster corrects for the sensor/filter stack/pack... from the pdf linked above (bottom of page 12);

A dirty little secret in digital photography is that the filter pack (low-pass filter + infraredfilter + sensor cover glass) found in virtually all digital cameras can contribute significant aberration to the image when a lens designed for film is used. This problem is particularly troublesome for very fast lenses. The Speed Booster corrects this defect, and automatically corrects for filter pack aberrations regard less of the objective lens used.
As he says, its troublesome for fast lenses. Perhaps the effect only becomes visible for f2.8 and faster. But it does explain why a great many threads on this subforum is about getting film era fast 50's working well on m43 sensors. Given the comment above, it could be a bit of a fools errant unless you adapt with a focal reducer. I have a Konica 57mm f1.2... I'm kinda looking forward to a hypothetical Metabones Konica adapter... *crosses fingers*.
 

Reflector

Mu-43 Hall of Famer
Joined
Aug 31, 2013
Messages
2,114
So, in short, why is finding out the thickness of the sensor stack making those who use speed boosters happier? And, if a flange distance of a certain mount is X, it will be X regardless. You don't even need to know the thickness of the sensor stack, right?
Here's an implication: You might actually lose out by straight adapting over a lens with an adapter to correct for flange as the lens optics are not designed for a different filter stack. (Ex: Canon might have a different thickness than Nikon, they both have a different thickness than m43s.) While the properly designed telecompressors will compensate for this so that you get the MTF gain when you adapt a lens from one system to another wherein the filter stacks are significantly different.
 
Links on this page may be to our affiliates. Sales through affiliate links may benefit this site.
Mu-43 is a fan site and not associated with Olympus, Panasonic, or other manufacturers mentioned on this site.
Forum post reactions by Twemoji: https://github.com/twitter/twemoji
Copyright © 2009-2019 Amin Forums, LLC
Top Bottom