p.2 #1 · The Glass in the Path: Sensor Stacks and Adapted Lenses
Interesting.
The other day I tested some of my wide angles on an A7 and an A7r - mainly with regards to flare and ghosting when pointed into the sun. Flare was the same, but the A7 had significantly worse ghosting characteristics than the A7r.
I was surprised by this and I suppose it's due to the type of cover glass/AA filter (or lack thereof) between the two cameras. I was initially very disappointed with the A7, which I tested first, but after testing the A7r was excited again. The A7r ghosting characteristics were nearly identical to my D800 with the same lenses.
Also - I had not known that m43 sensors have such a thick covering on the sensor - why is that?
I've not been very pleased with the ghosting characteristics of my m43 gear and always thought is was the lenses or the interaction of the lens rear element and the glass surface of the sensor cover. But it could be due to the really thick sensor cover showing a lot of reflections between the two surfaces.
Love these kinds of articles!
BTW - anyone know the sensor cover glass thickness of the A7, A7r, and D800?
p.2 #3 · The Glass in the Path: Sensor Stacks and Adapted Lenses
Not only the glass thickness but its distance from the sensor is important. Don't know about the A7/r but on the A900 the glass was further away from the sensor to allow for dust shake off and to minimize dust visibility in photos.
p.2 #4 · The Glass in the Path: Sensor Stacks and Adapted Lenses
Roger hinted a while back at optical solutions for correcting wide angle rangefinder glass, I'm not sure if this is relevant to the discussion here but I'd be interested to know if this was actually happening.
p.2 #5 · The Glass in the Path: Sensor Stacks and Adapted Lenses
I think glass cover thickness is only part of the corner issue with A7R and WA rangefinders. Else, many "older" wides as well as more recent wides designed for little or no glass cover should work, and they don't.
p.2 #6 · The Glass in the Path: Sensor Stacks and Adapted Lenses
I disassembled an A7 and removed the filter stack for better performance with M lenses - especially wider angle lenses. The A7 filter stack was measured at 1.8mm with an outside micrometer.
This doesn't include the first-birefringent layer epoxied to the ceramic carrier which I didn't remove.
Generally the birefringent layers are thicknessed according to the pixel size - eg. smaller pixel wells, need thinner glass due to the way the low pass filter works.
This explains why the A7r disassembled by Roger measured a slightly different value to my A7.
Incidentally, I got an M-mount adapter re-machined to get infinity focus back, and am using M lenses fairly successfully with improved corner performance.
p.2 #7 · The Glass in the Path: Sensor Stacks and Adapted Lenses
philber wrote:
I think glass cover thickness is only part of the corner issue with A7R and WA rangefinders. Else, many "older" wides as well as more ecet wides designed for little or no glass cover should work, and they don't.
Probably, but also don't underestimate the difference with the M sensor cover which I believe must be around 1mm (0.8mm according to some sources).
p.2 #9 · The Glass in the Path: Sensor Stacks and Adapted Lenses
My hope is that Sony designs the FE lenses to accommodate with little problems a wide range of sensor cover thickness. At least that would my goal, if I were a Sony engineer, in order to cope with future leaps in sensor technology ( self filtered photosites not needing IR filters,; Foven-like pixel arrays not needing AA filters, etc.. )
p.2 #10 · The Glass in the Path: Sensor Stacks and Adapted Lenses
artur5 wrote:
My hope is that Sony designs the FE lenses to accommodate with little problems a wide range of sensor cover thickness. At least that would my goal, if I were a Sony engineer, in order to cope with future leaps in sensor technology
That is not going to happen, because the required telecentricity implies relatively large lens designs which violate the philosophy of a small, compact, and light system.
p.2 #11 · The Glass in the Path: Sensor Stacks and Adapted Lenses
_julian_ wrote:
I disassembled an A7 and removed the filter stack for better performance with M lenses - especially wider angle lenses. The A7 filter stack was measured at 1.8mm with an outside micrometer.
This doesn't include the first-birefringent layer epoxied to the ceramic carrier which I didn't remove.
Generally the birefringent layers are thicknessed according to the pixel size - eg. smaller pixel wells, need thinner glass due to the way the low pass filter works.
This explains why the A7r disassembled by Roger measured a slightly different value to my A7.
Incidentally, I got an M-mount adapter re-machined to get infinity focus back, and am using M lenses fairly successfully with improved corner performance.
that is great, I have done the same on my nex-7 and I am considering it to my a7r too. how is corner performance with wide angle lenses? do you have any sample image? how do you cope with ir/uv contamination?
p.2 #12 · The Glass in the Path: Sensor Stacks and Adapted Lenses
Toothwalker wrote:
That is not going to happen, because the required telecentricity implies relatively large lens designs which violate the philosophy of a small, compact, and light system.
True. And even a perfectly telecentric design wouldn't help with filter-induced spherical aberration at large apertures, which is the problem Roger and I initially encountered with the boosted Otus without placing the necessary glass plate in the converging beam.
p.2 #13 · The Glass in the Path: Sensor Stacks and Adapted Lenses
edwardkaraa wrote:
Not only the glass thickness but its distance from the sensor is important. Don't know about the A7/r but on the A900 the glass was further away from the sensor to allow for dust shake off and to minimize dust visibility in photos.
The distance from the sensor does not have any immediate effect on spherical aberration or CA, but it has a very large effect on stuff like internal system scattering and reflection based aberrations.
Most manufacturers do not use any kind of anti-reflection coatings on the plates - at all. Unless you count the hotmirror surface as an AR, which it in a kind of way is... Unfortunately that's only one of the four normally included surfaces. In some Sony's I've even seen the plates split into three groups. They might have some very good reason to do this (I hope) - because otherwise it's pure madness from an image quality PoV.
p.2 #14 · The Glass in the Path: Sensor Stacks and Adapted Lenses
theSuede wrote:
........ In some Sony's I've even seen the plates split into three groups. They might have some very good reason to do this (I hope) - because otherwise it's pure madness from an image quality PoV.
My cynical guess is that they do it on purpose, just to prevent customers from using third party lenses, specially RF wides..
p.2 #15 · The Glass in the Path: Sensor Stacks and Adapted Lenses
artur5 wrote:
My cynical guess is that they do it on purpose, just to prevent customers from using third party lenses, specially RF wides..
Likely not at this point yet, but Sony might decide to stick to it instead of facilitating the use of rangefinder glass with a thinner glass to promote their own Sony/Zeiss branded lenses. I might be wrong, but I don't expect Sony to change this array to make it easier with third party lenses. In fact, they might make it even worse somehow (I certainly don't hope so...).
p.2 #16 · The Glass in the Path: Sensor Stacks and Adapted Lenses
I should have a table up on a few sensor thickness measurements and some exit pupil distances tomorrow. Anyone know of any accurate exit pupil distances for Leica lenses? I thought they used to put that on their lens specification pdfs but it seems they don't anymore.
p.2 #17 · The Glass in the Path: Sensor Stacks and Adapted Lenses
retrofocus wrote:
Likely not at this point yet, but Sony might decide to stick to it instead of facilitating the use of rangefinder glass with a thinner glass to promote their own Sony/Zeiss branded lenses. I might be wrong, but I don't expect Sony to change this array to make it easier with third party lenses. In fact, they might make it even worse somehow (I certainly don't hope so...).
If the rumored RX1 replacement does indeed use a curved sensor, Sony will at some point use that on their FF E-Mount and be assured of a proprietary lens system which will allow smaller lenses with better performance (while, of course, at the same time, completely killing any alt lens use).
p.2 #18 · The Glass in the Path: Sensor Stacks and Adapted Lenses
Tariq Gibran wrote:
If the rumored RX1 replacement does indeed use a curved sensor, Sony will at some point use that on their FF E-Mount and be assured of a proprietary lens system which will allow smaller lenses with better performance (while, of course, at the same time, completely killing any alt lens use).
Yep, I am afraid you are right. My only hope is that Sony recognizes that a big group of their mirrorless camera users comes with alt lenses. That's why I personally wait whenever Sony releases an A7R successor to see what is found especially about the alt/rangefinder lens use on such new future camera. I will definitely hold on to my current A7R in case people claim a worse alt lens performance of such camera then.
p.2 #19 · The Glass in the Path: Sensor Stacks and Adapted Lenses
But it would also make their own entire collection of existing A, E and FE lenses incompatible... unless there was a special intermediate optical flat field to curved field converter..
p.2 #20 · The Glass in the Path: Sensor Stacks and Adapted Lenses
rscheffler wrote:
But it would also make their own entire collection of existing A, E and FE lenses incompatible... unless there was a special intermediate optical flat field to curved field converter..
Their entire, fully native arsenal of FE FF lenses? Perhaps there is a reason why Sony is not investing heavily in the development of native glass at the moment (and will let Zeiss do that for them). The optical adapter is an interesting idea.
Perhaps there is a reason why Sony is not investing heavily in the development of native glass at the moment (and will let Zeiss do that for them). The optical adapter is an interesting idea.