p.13 #1First Sighting: A7rii Thin filter in action
uhoh7 wrote: MAubrey wroteI would expect that Steve is talking about the optical measurement of the glass. Physically measuring it is one thing, optically measuring its thickness is another thing.
From Roger at Lensrentals: Earlier quoteOur database for sensor-stack thickness remains incomplete (although I’m hoping for more contributions soon). Also remember that physical measurement of thickness and optical thickness may be different. If the glass used has a high index of refraction, it would have an optical effect greater than what is measured physically. For example, we might measure a physical thickness of 2mm on two different cameras, but if one has low-refractile glass and the other high-refractile glass, the optical measurement made might be 2mm and 2.5mm.https://www.lensrentals.com/blog/2014/07/sensor-stack-thickness-part-iii-the-summary/The variables are small. If you can identify the glass the numbers for calculating the precise optical depth fro any given thickness are right there. What is the biggest factor in optical depth? The thickness of the glass.
If you don't know the glass, you can easily determine the lowest probable Refractive Index. If the Sony had a hugely lower RI than Canikon it would be a big consideration.
Except the problem is higher RI's. Not lower ones.
p.13 #2First Sighting: A7rii Thin filter in action
uhoh7 wrote:
Our database for sensor-stack thickness remains incomplete (although I’m hoping for more contributions soon). Also remember that physical measurement of thickness and optical thickness may be different. If the glass used has a high index of refraction, it would have an optical effect greater than what is measured physically. For example, we might measure a physical thickness of 2mm on two different cameras, but if one has low-refractile glass and the other high-refractile glass, the optical measurement made might be 2mm and 2.5mm.
The variables are small. If you can identify the glass the numbers for calculating the precise optical depth fro any given thickness are right there. What is the biggest factor in optical depth? The thickness of the glass.
If you don't know the glass, you can easily determine the lowest probable Refractive Index. If the Sony had a hugely lower RI than Canikon it would be a big consideration. It's probably worse than Canikon. But even with the benefit of the doubt and a superior RI the thickness difference is overwhelming. So the whole "it's really the optical thickness and we can't know that" thing belies the facts on the ground which are obvious: we very likely already know all we need to know, and if somebody suspects otherwise, they should find out and report.
The deeply curious can pull the filter stack and use a refractometer. In fact I'm sure there are astro people who would have access to one. Or you could send out the filter stack. We might be too lazy to do so as it's a lot of work to verify the highly likely, but if somebody thinks it's telling a secret story, have at it. I'd love to see the Canikon and Sony filter stacks tested. I know where I'd put my money.
I don't say that to bash Sony. I just know how cost conscious Sony is; and for that matter, Leica, who choose a very cheap IR cut for the M8, while Canikon had spent more, especially on the big top of the line DSLRs, which are reputed to have very thin and expensive IR cuts, but of course not on the cover glass, as Leica does, alone. The young designers of the A7 were very honest about their confidence in processor corrections related to the thickness of the filter stack, let alone the cover glass. The same would hold true for RI. No need to spend a fortune on super low RI if the processor is going to do alot of work, or so the thinking goes.
The RI for the UT is lower than I expected, less than v2 or M9 either IR cut, if I am not mistaken, and combined with it's thinness....well it would be very fun to just see one as a coverglass The Optical depth is of greater difference than the physical thickness vs either the V2, or a Stock camera, simply by the nature of the math.
For optical depth there are only two factors: RI and thickness.
So if somebody feels this is a factor which somehow trumps the physical thickness with typical RI around 1.5, there are many people to correspond with, and contirbute something more than vague questions as a way to cast doubt on the meaning of physical thickness in this context. The actual results point to what is the simplest explanation: Sony is not using some secret glass with ultra low RI to reach a thin optical depth in comparison to Canikon. If anything their RI is very likely higher. It's not in their interest to be otherwise.
...........
. ...Show more →
Nice reading. I couldn't help casting a "Like" to your post, although the devil's advocate inside me tells me I would have liked it even more if it were a bit shorter..
p.13 #3First Sighting: A7rii Thin filter in action
MAubrey wrote:
I would expect that Steve is talking about the optical measurement of the glass. Physically measuring it is one thing, optically measuring its thickness is another thing.
From Roger at Lensrentals: Earlier quoteOur database for sensor-stack thickness remains incomplete (although I’m hoping for more contributions soon). Also remember that physical measurement of thickness and optical thickness may be different. If the glass used has a high index of refraction, it would have an optical effect greater than what is measured physically. For example, we might measure a physical thickness of 2mm on two different cameras, but if one has low-refractile glass and the other high-refractile glass, the optical measurement made might be 2mm and 2.5mm.https://www.lensrentals.com/blog/2014/07/sensor-stack-thickness-part-iii-the-summary/
Incomplete and wrong are two very different things.
The quoted Nikon glass thicknesses, for example, do not take in to account the sensor cover glass and the dust shaker.
It is a pity that this factually wrong piece is considered as a basis for many discussions.
p.13 #4First Sighting: A7rii Thin filter in action
Thanks for all the information and testing done in this thread! I really appreciate the work done here!
I’m considering doing the new Kolari UT mod on my A7rii. But the decision will be based on have much improvement I will get on these two lenses.
Leica Summicron-M 1:2/28 ASPH
Leica Summicron-M 1:2/35 ASPH
I Have not read the hole thread, but I am wondering if anyone here has any experience with these two Leica Summicron lenses on a A7rii Kolari UT mod? Our on one of the other thin filter mods?
I have already seen very good results on Zeiss 35 1.4 ZM and Zeiss 50mm 1.5 ZM (that I also have), so that’s very encouraging.
p.13 #5First Sighting: A7rii Thin filter in action
realVivek wrote: MAubrey wroteI would expect that Steve is talking about the optical measurement of the glass. Physically measuring it is one thing, optically measuring its thickness is another thing.
From Roger at Lensrentals: Earlier quoteOur database for sensor-stack thickness remains incomplete (although I’m hoping for more contributions soon). Also remember that physical measurement of thickness and optical thickness may be different. If the glass used has a high index of refraction, it would have an optical effect greater than what is measured physically. For example, we might measure a physical thickness of 2mm on two different cameras, but if one has low-refractile glass and the other high-refractile glass, the optical measurement made might be 2mm and 2.5mm.https://www.lensrentals.com/blog/2014/07/sensor-stack-thickness-part-iii-the-summary/Incomplete and wrong are two very different things.
The quoted Nikon glass thicknesses, for example, do not take in to account the sensor cover glass and the dust shaker.
It is a pity that this factually wrong piece is considered as a basis for many discussions.
I am sure that Roger would be delighted if you could provide him with better information
p.13 #6First Sighting: A7rii Thin filter in action
PersonCyborg wrote:
Thanks for all the information and testing done in this thread! I really appreciate the work done here!
I’m considering doing the new Kolari UT mod on my A7rii. But the decision will be based on have much improvement I will get on these two lenses.
Leica Summicron-M 1:2/28 ASPH
Leica Summicron-M 1:2/35 ASPH
I Have not read the hole thread, but I am wondering if anyone here has any experience with these two Leica Summicron lenses on a A7rii Kolari UT mod? Our on one of the other thin filter mods?
I have already seen very good results on Zeiss 35 1.4 ZM and Zeiss 50mm 1.5 ZM (that I also have), so that’s very encouraging.
Jan there are literally only a handful of cameras with the UT mod. In fact, there may only be the two talked about here. Sebboh (Derek) has one and Uhoh7 (Charlie) has one. It is a brand new mod. The two talked about here are both A7 mods, so there may not even be an A7rII UT mod that has been done yet. Charlie has the Leica M 28 f/2 ASPH and I think he has some shots with it and the UT earlier in the thread and perhaps he can comment more. The Leica M 35 f/2 ASPH is one of the worst performers on the stock Sony camera. It would be interesting to see its performance with the ultra thin mod, but I haven't seen that yet. I would expect the UT mod to help quite a bit, but it is always nice to see the actual evidence that it does. I don't know if Derek or Charlie actually have access to one, but they might not. It really is good that they have taken the time to provide the testing they have and it may be awhile before they have time to do any more. They both have busy full time work.
p.13 #7First Sighting: A7rii Thin filter in action
realVivek wrote:
Charlie, Have you tried shooting in IR with the UT variant? Looks like (from wild variations in WB) it would work for near IR.
Hi Vivek,
Sorry I need to catch up. How would you suggest I try it? I'm all ears
mdemeyer wrote:
I find the AWB in the A7 to be somehow sensitive to aperture, for reasons I can not come up with a theory for. This is true unmodified and with the v1,2, and 3 thin filter mods. Any ideas why?
Also, after a better technical brief from Ilija at Kolari I will have a camera switched to the UT mod and share some comparisons.
Michael
Michael, I'm very excited you will try the UT!
What I think may help in this idea, is to get beyond the obvious and consider what exactly Sony has sandwiched in those thick stacks. Clues again, come from the Astro people, who talk about the various layers in cameras they are taking apart, in some cases down past the CFA. I find references to color filters, independent of the IR cut.
The M9 has a pretty radical WB shift with 28 cron when you go WO. This may be related, as the Leicas are the simplest in terms of the "mix" of filters over the sensor.
In addition, I have also recently discovered the glued coverglass in Canikon is not always clear.
But I do feel the V1 and V2 WB is more "stable" than the UT. However that is not stay in the end a better result may be had from the UT, just that AWB is more varied. I see much more saturation with the UT. It tones down easy enough.
sebboh wrote:
i tried to test charlie's theory on slr wides improving with the thin filter today on the only suitable lens i had lying around. not sure that i see an obvious conclusion from these... caveat is the lens has a floating element and i have to use different adapters for each camera.
Hi Derek, I can't tell anything there, but the range is not the same. At some point you be around a stock A7 I'm sure and we can get a better look with any SLR lens you have.
I'll be shocked if there is no difference.
"What about lenses designed for little or no sensor stack? Actually, it’s already been shown they don’t do well on camera with significant sensor stacks. Panavision has made premium lenses for their film cameras for many years. Recently they’ve released their Primo V series of lenses, which are their Primo lenses modified, according to their website to ” eliminate coma, astigmatism, and other aberrations caused by the extra layers of glass in digital cameras.” U. S. Patent application 14/024,578 describes adding additional optics to the existing lenses to correct for the glass in the imaging pathway, that is between the rear of the lens and the camera sensor."
Roger at Lens Rentals.
My seat of the pants feelings certainly may be wrong, but often they are right. What they tell me:
FD 35/2 thorium, and nFD 24/2 are different on the v2 compared to stock cameras.
One thing we already know: the stock cameras vary widely on how they work with particular film lenses, even between each other. The best example is Charles K and the 75 Lux. He liked it fine on A7s and other early A7 models. Shooting it on the R2, he sold the lens. Charles has a scientific background and a very keen eye.
We are pulling all sorts of things out from in front of the sensor besides the plain properties of optical glass, which are certainly significant themselves. Now, maybe some of those things are better left in. I can see a day when other filters are added to an IR cut in a mod which starts with new glued coverglass. Anyway, my point is: better or worse in various respects, but being the "same" is not likely.
One of my heroes at LUF is Adan, and he has a hilarious series of posts about what is the base ISO of the M10. At first he convinces most readers it's 200, and he's quite emphatic. Then he changes his mind altogether, and proceeds to show it's 100 after all, as the onlookers gape, aghast. That thread is active now. Adan is more agile technically than me, but I hope I have the same attitude.
artur5 wrote:
Do we know for a fact that Canon DSLRs have a thinner sensor stack than Sony A7 series ?
In the context of FF, A7 vs Canikon, yes we know for a fact the Sony is much thicker, actually and "optically". In our context, I believe Roger's idea of "optical thickness" is simply RI times actual thickness. Since the RI in all the glasses varies between 1.4ish to 1.6ish, and the actual thickness varies between 1.1 (D810) and 1.9 (A7), you can easily see which is the primary factor. The bigger Canikon FFs are thinner yet. You must add at least .5mm to that for the base coverglass in those cameras and then compare to Leica, which has NO filter stack, only coverglass which is the IR cut. So the Sony is literally 1.9mm thicker, give or take .3mm. No variable of RI viable for Sony will make the difference anything less than huge.
The whole term of "optical thickness" should be thrown out of this thread. First, it has a bunch of meanings, so it is extremely confusing to research. Second, those "optical thickness" numbers we have now from Lens Rentals are totally suspect. Roger did not make them. We don't know how they were calculated, and if other variables were introduced. Yes, there are thin films which increase optical depth in other contexts dramatically, But not the other way around. So the Sony soup in that thick stack has the potential to make things extra thick, but not extra thin. On the other hand, the Leica is dead simple. My M9 has a sheet of .8mm S8612 over the sensor. Optical properties are well documented and you can look them up. Same for Sam Lee and NK.
The physical thickness as a comparison in this context is conservative. It's easy to verify. It's what the techs are talking about and the DIY astro people. Optical thickness is multi-context, and will only amplify what the physical measurements are giving us, considering the materials we are working with. For us, discussing optical thickness will not help in making a better mod, but just confuse things to no end.
That said, what would be very nice to know: more about what's in the soup of the stock cameras' cover glass and filter stacks.. That might inform a super-mod someday
People who are members at the paid Nikon and Canon technical forums could really help us if they learn more about the glass Canikon is using at various levels above the sensor in FF cameras. I know there are people in those forums who know alot about it. I've seen some of them scoff at Leica and note more sophisticated and expensive filters over the sensor developed by Canikon, especially for the big DSLRs.
p.13 #8First Sighting: A7rii Thin filter in action
PersonCyborg wrote:
Thanks for all the information and testing done in this thread! I really appreciate the work done here!
I’m considering doing the new Kolari UT mod on my A7rii. But the decision will be based on have much improvement I will get on these two lenses.
Leica Summicron-M 1:2/28 ASPH
Leica Summicron-M 1:2/35 ASPH
I Have not read the hole thread, but I am wondering if anyone here has any experience with these two Leica Summicron lenses on a A7rii Kolari UT mod? Our on one of the other thin filter mods?
I have already seen very good results on Zeiss 35 1.4 ZM and Zeiss 50mm 1.5 ZM (that I also have), so that’s very encouraging.
The 28 Cron is a bit better on the UT but it is quite usable on the V2. The ultimate might be a NK mod with the STC at .85mm as IR cut. You would need a Chinese speaker to help in ordering, but there are quite a number who watch this thread, so anyone determined can surmount that obstacle I'm sure. NK has a face book page, and you can pm Sam Lee, also.
I'd like to see Derek post more general shots in this thread, so people do not need to rely only on my color editing to see what they will get with the UT. But if you are comfortable with samples here in terms of WB (not all, but you know what I mean) I think the UT will do very well with those lenses. The 28 cron should not show much more FC than with the 240 on the UT, if any. I'm not sure about the 35 cron, but it should be close if not the same also to the 240 (without profile)
Whatever you get, I hope you will post some shots here and let us know how it's going
p.13 #9First Sighting: A7rii Thin filter in action
uhoh7 wrote:
I'd like to see Derek post more general shots in this thread, so people do not need to rely only on my color editing to see what they will get with the UT.
I'm afraid my general shots lately aren't terribly useful for people trying to understand color performance. here's a few with the UT and various lenses.
p.13 #10First Sighting: A7rii Thin filter in action
Phillip Reeve wrote: realVivek wroteMAubrey wroteI would expect that Steve is talking about the optical measurement of the glass. Physically measuring it is one thing, optically measuring its thickness is another thing.
From Roger at Lensrentals: Earlier quoteOur database for sensor-stack thickness remains incomplete (although I’m hoping for more contributions soon). Also remember that physical measurement of thickness and optical thickness may be different. If the glass used has a high index of refraction, it would have an optical effect greater than what is measured physically. For example, we might measure a physical thickness of 2mm on two different cameras, but if one has low-refractile glass and the other high-refractile glass, the optical measurement made might be 2mm and 2.5mm.https://www.lensrentals.com/blog/2014/07/sensor-stack-thickness-part-iii-the-summary/Incomplete and wrong are two very different things.
The quoted Nikon glass thicknesses, for example, do not take in to account the sensor cover glass and the dust shaker.
It is a pity that this factually wrong piece is considered as a basis for many discussions. I am sure that Roger would be delighted if you could provide him with better information
Suggestion to correct incorrect information fell on deaf ears.
p.13 #11First Sighting: A7rii Thin filter in action
realVivek wrote: Phillip Reeve wroterealVivek wroteMAubrey wroteI would expect that Steve is talking about the optical measurement of the glass. Physically measuring it is one thing, optically measuring its thickness is another thing.
From Roger at Lensrentals: Earlier quoteOur database for sensor-stack thickness remains incomplete (although I’m hoping for more contributions soon). Also remember that physical measurement of thickness and optical thickness may be different. If the glass used has a high index of refraction, it would have an optical effect greater than what is measured physically. For example, we might measure a physical thickness of 2mm on two different cameras, but if one has low-refractile glass and the other high-refractile glass, the optical measurement made might be 2mm and 2.5mm.https://www.lensrentals.com/blog/2014/07/sensor-stack-thickness-part-iii-the-summary/Incomplete and wrong are two very different things.
The quoted Nikon glass thicknesses, for example, do not take in to account the sensor cover glass and the dust shaker.
It is a pity that this factually wrong piece is considered as a basis for many discussions. I am sure that Roger would be delighted if you could provide him with better information Suggestion to correct incorrect information fell on deaf ears.
Which evidence did you provide?
p.13 #12First Sighting: A7rii Thin filter in action
Phillip Reeve wrote: realVivek wrotePhillip Reeve wroterealVivek wroteMAubrey wroteI would expect that Steve is talking about the optical measurement of the glass. Physically measuring it is one thing, optically measuring its thickness is another thing.
From Roger at Lensrentals: Earlier quoteOur database for sensor-stack thickness remains incomplete (although I’m hoping for more contributions soon). Also remember that physical measurement of thickness and optical thickness may be different. If the glass used has a high index of refraction, it would have an optical effect greater than what is measured physically. For example, we might measure a physical thickness of 2mm on two different cameras, but if one has low-refractile glass and the other high-refractile glass, the optical measurement made might be 2mm and 2.5mm.https://www.lensrentals.com/blog/2014/07/sensor-stack-thickness-part-iii-the-summary/Incomplete and wrong are two very different things.
The quoted Nikon glass thicknesses, for example, do not take in to account the sensor cover glass and the dust shaker.
It is a pity that this factually wrong piece is considered as a basis for many discussions. I am sure that Roger would be delighted if you could provide him with better information Suggestion to correct incorrect information fell on deaf ears. Which evidence did you provide?
Panasonic m43rds sensor glass thickness (all of it = total glass).
Sony glass thickness (NEX-5N, A7) (all of it = total glass).
I have measured myself and provided them to Brian Caldwell (one of the authors in the quoted articles). When the article appeared I did post in the Q&A/comments (all those are now gone).
p.13 #13First Sighting: A7rii Thin filter in action
realVivek wrote: Phillip Reeve wroterealVivek wrotePhillip Reeve wroterealVivek wroteMAubrey wroteI would expect that Steve is talking about the optical measurement of the glass. Physically measuring it is one thing, optically measuring its thickness is another thing.
The quoted Nikon glass thicknesses, for example, do not take in to account the sensor cover glass and the dust shaker.
It is a pity that this factually wrong piece is considered as a basis for many discussions. I am sure that Roger would be delighted if you could provide him with better information Suggestion to correct incorrect information fell on deaf ears. Which evidence did you provide?Panasonic m43rds sensor glass thickness (all of it = total glass).
Sony glass thickness (NEX-5N, A7) (all of it = total glass).
I have measured myself and provided them to Brian Caldwell (one of the authors in the quoted articles). When the article appeared I did post in the Q&A/comments (all those are now gone).
p.13 #14First Sighting: A7rii Thin filter in action
Phillip Reeve wrote: realVivek wrotePhillip Reeve wroterealVivek wrotePhillip Reeve wroterealVivek wroteIncomplete and wrong are two very different things.
The quoted Nikon glass thicknesses, for example, do not take in to account the sensor cover glass and the dust shaker.
It is a pity that this factually wrong piece is considered as a basis for many discussions. I am sure that Roger would be delighted if you could provide him with better information Suggestion to correct incorrect information fell on deaf ears. Which evidence did you provide?Panasonic m43rds sensor glass thickness (all of it = total glass).
Sony glass thickness (NEX-5N, A7) (all of it = total glass).
I have measured myself and provided them to Brian Caldwell (one of the authors in the quoted articles). When the article appeared I did post in the Q&A/comments (all those are now gone).
So, the numbers in the articles on physical thickness is are random or incorrect.Do you have any pictures or other documentation on how exactly you performed the measurements?
Your link seems to be beyond a paywall.
https://www.flickr.com/photos/vivek-iyer/albums/72157600032172823 shows a few pics of dismantled sensors (Nikon, Panasonic, Sony) . These, I did for conversion to IR or UV or Monochrome (debayering the sensors). I have here a box full of stuff with all the glass. Measurements were done using a Vernier. Not exactly a "rocket science".
My purpose was different but the information (Sony/Panasonic communicated via email) gained from it was useful to Brian Caldwell for one of his projects. Just to be clear- while his theory (its origins are in a post in the Leica section at photo.net) on the sensor glass thickness and aberrations and the conclusions of the articles posted on the lensrentals blog are correct, the numbers are wrong. While it does not affect the overall theory there it causes problems when discussions such as the one here tend to quote the values posted in the articles.
p.13 #15First Sighting: A7rii Thin filter in action
realVivek wrote: Phillip Reeve wroterealVivek wrotePhillip Reeve wroterealVivek wrotePhillip Reeve wroteI am sure that Roger would be delighted if you could provide him with better information Suggestion to correct incorrect information fell on deaf ears. Which evidence did you provide?Panasonic m43rds sensor glass thickness (all of it = total glass).
Sony glass thickness (NEX-5N, A7) (all of it = total glass).
I have measured myself and provided them to Brian Caldwell (one of the authors in the quoted articles). When the article appeared I did post in the Q&A/comments (all those are now gone).
So, the numbers in the articles on physical thickness is are random or incorrect.Do you have any pictures or other documentation on how exactly you performed the measurements?
Your link seems to be beyond a paywall.https://www.flickr.com/photos/vivek-iyer/albums/72157600032172823 shows a few pics of dismantled sensors (Nikon, Panasonic, Sony) . These, I did for conversion to IR or UV or Monochrome (debayering the sensors). I have here a box full of stuff with all the glass. Measurements were done using a Vernier. Not exactly a "rocket science".
My purpose was different but the information (Sony/Panasonic communicated via email) gained from it was useful to Brian Caldwell for one of his projects. Just to be clear- while his theory (its origins are in a post in the Leica section at photo.net) on the sensor glass thickness and aberrations and the conclusions of the articles posted on the lensrentals blog are correct, the numbers are wrong. While it does not affect the overall theory there it causes problems when discussions such as the one here tend to quote the values posted in the articles.
p.13 #16First Sighting: A7rii Thin filter in action
Just wanted to say thank you all for the information and research provided in this thread. It led me to getting the A7 UT mod done and has given new life to my old wide angles, especially my Voigtlander 28mm f3.5. Much appreciated!
Thanks to Paul Roark for the in-depth and un-biased review!
He also shoots the r2 and Loxia 21.
His thoughts on cornerfix are especially interesting. Leica provides quite a number of closely tuned profiles for a wide range of lenses. Quite a few lenses can get away without that even on the V2, like the 28 cron, but in the future I would like to see some "profile-like" choices for the Sony mods, shared, in either CF, LR FF plugin, or even better the Sony in-camera lens correction app.
p.13 #19First Sighting: A7rii Thin filter in action
Nice indeed. Thanks for the link
What the reviewer doesn't mention is that what he calls 'color artifacts' i.e. cyan/purplish corners that are a real issue with some RF glass on the A7r, are less so with the A7/A7II and become practically irrelevant with the BSI sensor of the A7r2 (and future cameras fitted this kind of sensor). Therefore, even "trouble-makers" like the ZM 21/4.5 might become very usable in UT moded Sonys. No more need of added hassle in postprocess with Cornerfix or Lightroom.