A7r-II BSI sensor on tech wides like SK35?

[torger]

I have not followed the tech development well during the past 6-8 months as I've spent so much time coding camera profiling software.

Hence I did not discover until now that the new A7r-II actually has BSI technology, the sensor technology that move the photo diode to the surface. My hopes is that it would drastically improve the color cast issues / crosstalk issues seen in the old A7r and indeed the IQ250 et al. It's probably only a matter of time until a BSI sensor arrives in a MFD back. Therefore I think it would be extremely interesting to know how this performs on tech wide angles.

Has someone tested the new A7r-II with say a SK35 in strongly shifted positions, to see if the BSI actually makes a difference in practice?

I've seen examples with the A7r + SK35 posted here in the past, and while it worked it suffered strongly from color fidelity loss and also had some mazing due to crosstalk. It would be fantastic news if the A7r-II despite the higher resolution can handle it without those issues.

[Chris Barrett]

Torger, have you heard about the Kolari mod?  A7r thin filter mod

They replace the sensor cover glass on the Sony with a much thinner piece.  This has proven to improve the corner performance of Leica-M glass on the Sony's.  My A7r should be returning from them this week and if I like the results, I'll send my A7r2.

I sold my wide Schneiders and am using all retro-focus lenses these days to avoid LCCs altogether, but I do expect this would improve smearing and crosstalk on the A7 series.

CB

[torger]

I haven't heard of the mod, thanks for the link.

The problem they fix there is however as far as I understand only image smearing (which is important too of course). Anyway, the color filter array sits closest to the sensor surface, under both cover glass and microlenses.

The problem I'd like to see solved is crosstalk, the issue when light comes in through one color filter on the sensor surface (say a red one) and then inside the chip jumps over to a neighboring pixel (say a green) and then mixes up colors. This can be seen as desaturation and in severe cases mazing. A thinner cover glass won't help for that. In the past sensors this happened a lot as the photo diodes sits deep down in the chip with open space between neighbors where light could freely pass over to the neighbor if coming in at an angle. Kodak CCDs had put in walls (light shields) between pixels, but those where not available in Dalsa CCDs and not in Sony CMOS.

In theory on a BSI sensor the photo diode sits directly on the surface of the chip, virtually sandwiched to the color filter. This means that light has no chance to skip over to the next pixel. If practice match theory there would be virtually no crosstalk with this sensor. Color cast is probably still there though as I think part of that is due to changed color filter response when light comes in at an angle. Color cast is correctable though, while crosstalk really isn't.

[vjbelle]

Torger, have you heard about the Kolari mod?  A7r thin filter mod

They replace the sensor cover glass on the Sony with a much thinner piece.  This has proven to improve the corner performance of Leica-M glass on the Sony's.  My A7r should be returning from them this week and if I like the results, I'll send my A7r2.

I sold my wide Schneiders and am using all retro-focus lenses these days to avoid LCCs altogether, but I do expect this would improve smearing and crosstalk on the A7 series.

CB

Chris.... according to Chambers, and shown with examples, the modification you are referring to is only a gain for lenses designed for the Leica sensor..... such as all of the wide and even normal M lenses and Zeiss lenses for that specific platform.  The modification is actually detrimental for non-M lenses, again shown with examples.

Victor

[BernardLanguillier]

Chris.... according to Chambers, and shown with examples, the modification you are referring to is only a gain for lenses designed for the Leica sensor..... such as all of the wide and even normal M lenses and Zeiss lenses for that specific platform.  The modification is actually detrimental for non-M lenses, again shown with examples.

So you need 2 of them, right?

Cheers,
Bernard

[vjbelle]

So you need 2 of them, right?

Cheers,
Bernard

Yes....... according to Chambers and certainly looking at his examples.

Victor

[narikin]

I sold my wide Schneiders and am using all retro-focus lenses these days to avoid LCCs altogether, but I do expect this would improve smearing and crosstalk on the A7 series.

To be honest Chris, I think it's a mistake to avoid LCCs altogether. Even straight shooting, I find it advantageous to apply an LCC, where you can clearly see angular color errors cleaned up. If you can't be bothered to shoot them regularly, then just take a couple of hours one day to build a library you can apply. You'll get better results in the end!

[mgrayson]

LCCs are great for dust removal, too!

[ErikKaffehr]

Hi,

I wouldn't agree. The Kolari modification would be helpful with any lens having large beam angles and being designed for film. Under these conditions any filter in the optical path will cause astigmatism.

With lenses designed for digital systems, I would expect that some optical glass is taken into account. Most DSLRs use around 2 mm it seems ,while 4/3 has around 4 mm. Leica has 0.8 mm AFAIK.

Lensrentals noted in their lab testing that the new Sony lenses worked best with around 2 mm. With DSLR lenses, the cover glass usually did not matter as they used to have reasonable beam angles.

So, my guess is that the Kolari modification is beneficial with old lenses made for film, but not so with modern lenses optimised for digital sensors having a cover glass, which they all do.

Best regards
Erik

Chris.... according to Chambers, and shown with examples, the modification you are referring to is only a gain for lenses designed for the Leica sensor..... such as all of the wide and even normal M lenses and Zeiss lenses for that specific platform.  The modification is actually detrimental for non-M lenses, again shown with examples.

Victor

[alatreille]

So, my guess is that the Kolari modification is beneficial with old lenses made for film, but not so with modern lenses optimised for digital sensors having a cover glass, which they all do.

Best regards
Erik

Or if applying movements to the lens and increasing that angle!

[Chris Barrett]

Or if applying movements to the lens and increasing that angle!

Well, yeah... that was my thinking.

[vjbelle]

Hi,

I wouldn't agree. The Kolari modification would be helpful with any lens having large beam angles and being designed for film. Under these conditions any filter in the optical path will cause astigmatism.

With lenses designed for digital systems, I would expect that some optical glass is taken into account. Most DSLRs use around 2 mm it seems ,while 4/3 has around 4 mm. Leica has 0.8 mm AFAIK.

Lensrentals noted in their lab testing that the new Sony lenses worked best with around 2 mm. With DSLR lenses, the cover glass usually did not matter as they used to have reasonable beam angles.

So, my guess is that the Kolari modification is beneficial with old lenses made for film, but not so with modern lenses optimised for digital sensors having a cover glass, which they all do.

Best regards
Erik

Let me clarify..... The modification is detrimental for lenses designed specifically for the A7R/Rll.  If one were wanting to shoot the Sony with both native lenses and Leica M type lenses then it would be best, according to Chambers, to have two cameras. His examples are very informative. 

Victor

Victor

[Chris Barrett]

Curious, the resolution tests at LensRentals showed nearly identical performance with Sony lenses on the modded vs non-modded.

I'll do some real world testing when my Mk1 comes back from Kolari.

[ErikKaffehr]

Hi,

Lensrentals has found that the Sony lenses they tested performed with 2.0 mm optical glass in front. An earlier test indicated only a small loss of sharpness on the Kolari-modied body, but the earlier tests were done using Imatest and not on the optical bench.

Best regards
Erik

Let me clarify..... The modification is detrimental for lenses designed specifically for the A7R/Rll.  If one were wanting to shoot the Sony with both native lenses and Leica M type lenses then it would be best, according to Chambers, to have two cameras. His examples are very informative. 

Victor

Victor

[Chris Barrett]

I don't understand how you could test the effect of the cover glass on an optical bench, since only the lens and not the camera body goes on the bench.

[Bart_van_der_Wolf]

I don't understand how you could test the effect of the cover glass on an optical bench, since only the lens and not the camera body goes on the bench.

Hi Chris,

Here is the explanation from LensRentals. Modern lenses are designed with an assumption of a filter stack in the optical path. Taking away that filter stack, results in porrer optical quality. Old lens designs were not calculated/designed for digital use, and would be better if they had been redesigned. Obviously, glass technology and computer capacity has also progressed, so later designs (especially the better/specialized lenses) tend to perform better that early designs, even those with the same design name.

How much of that lens quality transfers to the sensor it will be used on, is another matter, but a better lens design will always improve the system performance, the combined MTF of the optics and the MTF of the sensor (size/aperture/microlenses/cross-talk/etc.) .

Cheers,
Bart

[vjbelle]

Hi,

Lensrentals has found that the Sony lenses they tested performed with 2.0 mm optical glass in front. An earlier test indicated only a small loss of sharpness on the Kolari-modied body, but the earlier tests were done using Imatest and not on the optical bench.

Best regards
Erik

I can only report on what I have seen.  Look for yourself..... the images don't lie.  I would not invest in the Kolari modified body if I intended to use lenses designed for the Sony. 

Victor

[alan_b]

Has someone tested the new A7r-II with say a SK35 in strongly shifted positions, to see if the BSI actually makes a difference in practice?

Bump for the question in the OP.  Any experience/examples of a SK35 shifted on A7rII?

[ErikKaffehr]

Hi,

If the optical thickness of the cover glass is know , you just put the corresponding thickness of optical glass in front of the focal plane. In this case Lensrentals tested with 1, 2 and 3 mm, and 2 mm gave the best results with the Sony lenses. As it happens, the filter stack on the Sony cameras seems to be around 2 mm.

Best regards
Erik

I don't understand how you could test the effect of the cover glass on an optical bench, since only the lens and not the camera body goes on the bench.

[torger]

Bump for the question in the OP.  Any experience/examples of a SK35 shifted on A7rII?

Bump indeed... it seems noone's tested.

Even better check if it works on the SK28. A coming CMOS back with BSI will most likely be 44x33mm (at least one in a sane price range) and then I would need the 28 to get wide enough view.

I've heard that Schneider Digitar is leaving the tech lens game though so the interest in this lens range is declining. Tests on the Rodenstock Digarons are relevant too though.

BSI coming to MFD or not, and if it's really wide angular response like theory has it can be if not a game changer a welcome boost to the usability of tech cams. What if it's the other way around and the trend continues with narrower and narrower angular response, leaving also the Rodenstocks unusable? A reversal of this trend would be fantastic news for us tech cam users.

So... I think it's a super-interesting experiment. I think those selling 135-compatible tech cams (Cambo, Arca-Swiss) should have tested it but they haven't been good at it historically so I guess we have to wait for some user testing this time around too.