fredmiranda.com

Moderated by: Fred Miranda

New fredmiranda.com Mobile Site

New Feature: SMS Notification alert

New Feature: Buy & Sell Watchlist

FM Forums | Sony Forum | Join Upload & Sell

1 2 3 4

6 … 53 54 end

Archive 2015 · A thinner sensor stack may be possible after all!

charles.K
Offline
• • • • • •
Upload & Sell: Off

p.5 #1 · A thinner sensor stack may be possible after all!

I agree with all the perspectives and potential issues discussed here

Presently I still have my M glass, WATE, 24 Lux Asph, 35 Lux FLE, 50 Lux Asph, 50 Nocti f/1.0, 50 Cron DR, 75 Lux, 75 Cron AA, 90 Cron AA. I just recently sold my 21 SEM and 28 Cron Asph on the premise there would not be an option to use these on the A7 series.

If this "thin filter" performs close to the test results presented so far, I will be very happy. I suspect that even the Loxia 35 may perform better on the edges too.

The issue of corrosion is not really an issue if it lasts at for least 12 to 18 months, and is covered under warranty. With the continued development, I am sure there will be improved new thin glass covers, and I would be pleased to upgrade at that time.

I do not think Sony will alter from their R&D/product development goals, so this option most likely will be after market choice.

I had discussed these options of using a "thin filter" in Australia with Camera Clinic about 8 months ago, as they do a lot of filter mods, and I was about to test out this option. I was unsure what if any improvements would be seen. But fortunately we do have some definitive test results now, so really it a small investment. I do hope this resolves the problematic issues with my remaining M lenses.

Also it is worthwhile to note that in closely reviewing Lloyd's test results for the 18 SEM, 24 Elmar and 50 Cron AA, that the magenta shift is improved in every f stop comparison between the stock and mod A7r. I don't think the ZM 21 f/4.5 Biogon is a player at all, even though there are improvements.

Jan 20, 2015 at 08:07 PM

uhoh7
Offline
• • • •
Upload & Sell: Off

p.5 #2 · A thinner sensor stack may be possible after all!

mdemeyer wrote:
Lloyd published results with the 50mm APO Summicron and Zeiss ZM 21mm Biogon-C f4.5 tonight. While the improvements in the Biogon are expected (and it still has huge color shift issues, also as expected) the improvements with the 50mm APO Summicron show that, with a good enough lens, the problem is not limited to wides and ultra-wide lenses if one is looking for the ultimate in performance - which is what having 36MP is about, yes?

You hit the nail on the head here. My view is we have had muddy results with everything to some degree, except the native E mounts which use the processor to make up for the crazy dust cover. The design team alluded to this in Japanese interviews a year ago. The A7r drove me insane with SLR glass on close inspection.

Steve Spencer wrote:
I also have one other question that I hope the early adopters can address, what about sensor reflections? I worry that because the glass used for the mod in uncoated that it might be prone to sensor reflections.

How would you set up a test for this, Steve?

charles.K wrote:
I feel the results for the ZM 21 Biogon C f/4.5 is not in the same league as the 18 SEM and 24 Elmar. As Lloyd mentioned it could be a lens/adapter/camera issue. The other issue with or without the modification there is a very strong magenta cast with this lens. I feel this would be difficult to fully compensate with a profile for a given f stop.

So, Charles, you think the SEM 21 may really become usable at infinity with A7? My test shot above shows colorshift on A7 with 21SEM is not so bad.

As to the speculation about Sony dealing with the sensor stack themselves, and the reasoning, this is like predicting which pond a brontosaurus will stomp into next and why.

The thick stack has been a mistake, and the camera would have many more converts without it. But it's success in spite of this masks the bad judgment.

Bottomline: perhaps, for the first time, there will be a FF digital M serious alternative possible for about 1300USD or a bit less. How do I come up with that figure? A used A7 for 900 and the mod for 400.

This would be less than half the cost of an M9 and give the user a camera with additional strengths, SLR and Long glass, AF etc.

In fact I really expect to see better results with the Sony 70400 via LAEA4. But then again, I thought the A7r would be a Leica killer out of the box

BTW: my A7 has arrived safely in NJ

Jan 20, 2015 at 08:27 PM

charles.K
Offline
• • • • • •
Upload & Sell: Off

p.5 #3 · A thinner sensor stack may be possible after all!

uhoh7 wrote:
So, Charles, you think the SEM 21 may really become usable at infinity with A7? My test shot above shows colorshift on A7 with 21SEM is not so bad.

As to the speculation about Sony dealing with the sensor stack themselves, and the reasoning, this is like predicting which pond a brontosaurus will stomp into next and why.

The thick stack has been a mistake, and the camera would have many more converts without it. But it's success in spite of this masks the bad judgment.

Bottomline: perhaps, for the first time, there will be a FF digital M serious alternative possible for...Show more →

I feel the 21 SEM will work well with the mod. I would not have expected the 18 SEM to work as well with Lloyd's tests, but I would be happy to use on the basis of comparisons.

What ever the reasoning for the thicker glass filter, this no doubt has been an issue for all the lens designers, so much so that Zeiss acknowledged that the Loxia design had to compensate for the thick filter. I think with the Loxia design, there is wishful thinking that the filter will be reduced in thickness, as it appears that Zeiss have taken the middle ground in their design. Maybe this will incorporated in the new sensor designs of A7rII/A9.

Good luck with your A7! My A7r is in California due for delivery tomorrow morning

Jan 20, 2015 at 09:05 PM

naturephoto1
Offline
• • • • • •
Upload & Sell: Off

p.5 #4 · A thinner sensor stack may be possible after all!

I will be interested in Charlie's and Charles' observations and experiences with the thinner sensor stacks on their A7 and A7r cameras respectively.

I doubt that I will make the modification to my A7r due to the fact that I also use the camera heavily with longer Leica R lenses including my 100mm f2.8 Apo-Macro-Elmarit, my 280mm f2.8 Apo Telyt (naked and with the Leica R 1.4X Apo and Leica R 2X Apo extenders), Leica R 180 f3.4 Apo Telyt, Leica R 28mm f2.8 Elmarit V2, and many other R lenses. I am concerned about the loss of performance of all of my lenses that are not the short M RF lenses of which I only have the Leica M WATE (and my Minolta CLE MC 40mm f2 M-Rokkor) lens (es).

Rich

Jan 20, 2015 at 09:18 PM

sebboh
Offline
• • • • • •
Upload & Sell: Off

p.5 #5 · A thinner sensor stack may be possible after all!

naturephoto1 wrote:
I will be interested in Charlie's and Charles' observations and experiences with the thinner sensor stacks on their A7 and A7r cameras respectively.

I doubt that I will make the modification to my A7r due to the fact that I also use the camera heavily with longer Leica R lenses including my 100mm f2.8 Apo-Macro-Elmarit, my 280mm f2.8 Apo Telyt (naked and with the Leica R 1.4X Apo and Leica R 2X Apo extenders), Leica R 180 f3.4 Apo Telyt, Leica R 28mm f2.8 Elmarit V2, and many other R lenses. I am concerned about the loss of performance of all...Show more →

rich � since those R lenses were all designed for film they should also perform as good or better with the thin sensor stack as it is closer in geometry to film.

it's only if you are using lenses designed for a thicker sensor stack where you will see a drop in performance (e.g. FE lenses).

Jan 20, 2015 at 09:22 PM

naturephoto1
Offline
• • • • • •
Upload & Sell: Off

p.5 #6 · A thinner sensor stack may be possible after all!

sebboh wrote:
rich � since those R lenses were all designed for film they should also perform as good or better with the thin sensor stack as it is closer in geometry to film.

it's only if you are using lenses designed for a thicker sensor stack where you will see a drop in performance (e.g. FE lenses).

Derek,

OK. I will reconsider once I hear the observations of those taking the plunge.

Rich

Jan 20, 2015 at 09:23 PM

mdemeyer
Offline
• •
Upload & Sell: Off

p.5 #7 · A thinner sensor stack may be possible after all!

Keep in mind that, while a "quick test", it is interesting that Roger's tests with the Native lenses didn't show a performance loss. I don't view that as definitive (it was a quick test), but it is promising.

It raises a question - when we have lenses "optimized" for thick stacks, how much of that is just making them more retrofocal to minimize the effects, vs. actually designing the lens to be at its best with the glass in the path? I don't know the answer to that and am not throwing stones. I think Roger speculated on that question, too. After all, some lines of cameras (example Canon and Nikon DSLR) use different thnkness filter stacks on different models. (See the LensRentals database for details.) in that case, what would the lens be optimized for?

Jan 20, 2015 at 09:39 PM

Steve Spencer
Offline
• • • • • • •
Upload & Sell: On

p.5 #8 · A thinner sensor stack may be possible after all!

uhoh7 wrote:
You hit the nail on the head here. My view is we have had muddy results with everything to some degree, except the native E mounts which use the processor to make up for the crazy dust cover. The design team alluded to this in Japanese interviews a year ago. The A7r drove me insane with SLR glass on close inspection.

How would you set up a test for this, Steve?

So, Charles, you think the SEM 21 may really become usable at infinity with A7? My test shot above shows colorshift on A7 with 21SEM is not so bad.

As to the...Show more →

Charlie, if you just take some night shots that have street lights in them that will give us a decent idea of whether sensor flare is an issue. The A7 has this issue, but the A7r, and A7 II, and A7s it is a much smaller issue. I worry that with a uncoated cover glass this sensor flare might be fairly strong. This probably won't be an issue for A7 owners and they already had the problem, but A7r owners it might really be a bummer to fix one problem only to get another. How serious of an issue it is probably depends on what you shoot, however. Personally, I almost never shoot in conditions that would show sensor flare, but I am happy to have the A7 II which lower the concern even further.

Jan 20, 2015 at 09:59 PM

Steve Spencer
Offline
• • • • • • •
Upload & Sell: On

p.5 #9 · A thinner sensor stack may be possible after all!

mdemeyer wrote:
Keep in mind that, while a "quick test", it is interesting that Roger's tests with the Native lenses didn't show a performance loss. I don't view that as definitive (it was a quick test), but it is promising.

It raises a question - when we have lenses "optimized" for thick stacks, how much of that is just making them more retrofocal to minimize the effects, vs. actually designing the lens to be at its best with the glass in the path? I don't know the answer to that and am not throwing stones. I think Roger speculated on that question,...Show more →

Roger's tests were all at close distance and showed very mild interference with performance on FE lenses. If problems arise for the FE lenses they are probably most likely at infinity focus, however, so it is way too early, IMO, to conclude their won't be problems for FE lenses with the modified cover glass.

Jan 20, 2015 at 10:01 PM

hiepphotog
Offline
• • • •
Upload & Sell: Off

p.5 #10 · A thinner sensor stack may be possible after all!

mdemeyer wrote:
Keep in mind that, while a "quick test", it is interesting that Roger's tests with the Native lenses didn't show a performance loss. I don't view that as definitive (it was a quick test), but it is promising.

It raises a question - when we have lenses "optimized" for thick stacks, how much of that is just making them more retrofocal to minimize the effects, vs. actually designing the lens to be at its best with the glass in the path? I don't know the answer to that and am not throwing stones. I think Roger speculated on that question,...Show more →

Personally, I think up till now the degradation of SLR film lens on DSLR is negligible due to the longer exit pupil distance and because of the lower resolution on these early DSLRs. However, if you notice the optical bench result of the Otus from Lensrental, you see significant degradation without a piece of glass. So I think all the optical makers knew about it all along but they had been mum on the subject. They just silently upgraded their optics over the years.

Jan 20, 2015 at 10:10 PM

Tariq Gibran
Offline
• • • • • •
Upload & Sell: Off

p.5 #11 · A thinner sensor stack may be possible after all!

uhoh7 wrote:
The thick stack has been a mistake, and the camera would have many more converts without it. But it's success in spite of this masks the bad judgment.

Leica really had no other option but go with a thinner cover glass given their existing rangefinder lens designs. Everything is a trade off here though so it's not like Leica's choice is perfect with regard to IQ with no negatives. Even the m240 still suffers IR contamination for instance, as have all previous digital M's (the M8's being terrible in this respect of course). Then, of course, we have that "corrosion" issue with the M9, ME and MM that curiously is also an issue with this conversion. Thus one could just as easily argue that Sony, starting from a clean slate, took the best option for overall IQ with also a thought towards limiting consumer frustration over sensor issues as compared to the route Leica was forced to take. Arguing that Sony made a mistake and used bad judgement going with a thicker stack versus a thin stack with it's own IQ and reliability issues doesn't really hold much water imo.

Has it been mentioned yet if this mod affects IR contamination with the A7?

Jan 20, 2015 at 11:01 PM

mdemeyer
Offline
• •
Upload & Sell: Off

p.5 #12 · A thinner sensor stack may be possible after all!

I understand from KolariVision that the IR characteristics of the new filter are (essentially) the same as the stock filters. Regarding humidity resistance, Leica used S8612 in the M9, etc. which is more prone to humidity damage (according to the Schott technical information, although they are not as transparent on the subject as one might wish) than the BG39 used here.

It's a tricky balance - IR rejection, matching the visible band transmission curve to minimize white balance changes, availability of glass in small quantities in the needed thickness, adjusting sensor position to maintain proper optical path length (focus), etc. Having been involved in the project since last fall (I'm the Michael mentioned on the KolariVision web site. I initiated the project - although I have no commercial interest in the project or KolariVision) I have seen personally how difficult it is to come up with a modification like this that actually works and can be implemented at a reasonable cost IMHO for what it delivers.

Perfect? I'm sure not. Could it be done better by a manufacturer in volume? Almost certainly. But so far none but Leica seem to have taken on the challenge and I can't justify putting that much into a rapidly depreciating digital asset. I love my RF and other glass acquired over the years and was frustrated by years of people talking about the problem and not solving it. I'm delighted that Ilija at KolariVision was willing to take it on and develop something that, for many of us, will breath life into the glass we treasure for whatever reason. For those who think it's an imperfect or un-needed solution, that's fine. We each have our priorities and values and can choose to participate or not.

At least we have another option now. I, for one, am happy to have that choice and to share it with others of like mind.

Michael

Jan 21, 2015 at 12:06 AM

charles.K
Offline
• • • • • •
Upload & Sell: Off

p.5 #13 · A thinner sensor stack may be possible after all!

Thank you very much Michael for initiating the project! I have been in discussion with Ilija from KolariVision, and I have been very impressed so far.

I do hope there will be a reasonable volume of thin filter mods, so the process will mature and continue to evolve as an option for those with M glass.

I am interested in your thoughts, as you must already have the A7r modified with a number M mount lenses. I would not ask for test results per se as this is a pain, but what improvements you have seen?

Jan 21, 2015 at 12:26 AM

ZoneV
Offline
• • •
Upload & Sell: Off

p.5 #14 · A thinner sensor stack may be possible after all!

Steve Spencer wrote:
..I worry that because the glass used for the mod in uncoated that it might be prone to sensor reflections.

For sure it has more reflection issues.
For normal incident coated glass has less than 1% reflection on each surface - probably only 0.1%. The uncoated BG 39 about 4%.
Especially with short symmetrical rangefinder lenses most reflections are reflected away into the mount. With long focal lenses or image side telecentric lenses the reflection are reflected to the lens system - and back to the sensor again...

...
It raises a question - when we have lenses "optimized" for thick stacks, how much of that is just making them more retrofocal to minimize the effects, vs. actually designing the lens to be at its best with the glass in the path?...

From my basic lens design knowledge I would say a more retrofocal / telecentric design (image side) would help. But fast lenses would still suffer from a different sensor stack thickness - they have wideopen always some very oblique rays that hit the sensor.

hiepphotog wrote:
... So I think all the optical makers knew about it all along but they had been mum on the subject. They just silently upgraded their optics over the years.

I am not sure about that. I don�t think manfacturers would change much on the optical design of existing lenses without give it a different name.
Pure lens manufacturers work with the optical bench as measurment tool, for me it seems sometimes they forget the later use on cameras.
How long was the Canon EF 50/1.0L produced? The 50mm/1.2L is likely designed with the sensor stack in mind.

mdemeyer wrote:
I understand from KolariVision that the IR characteristics of the new filter are (essentially) the same as the stock filters. Regarding humidity resistance, Leica used S8612 in the M9, etc. which is more prone to humidity damage (according to the Schott technical information, although they are not as transparent on the subject as one might wish) than the BG39 used here. ...

Does Leica uses the S8612 glass uncoated?

Jan 21, 2015 at 01:54 AM

uhoh7
Offline
• • • •
Upload & Sell: Off

p.5 #15 · A thinner sensor stack may be possible after all!

Tariq Gibran wrote:
Thus one could just as easily argue that Sony, starting from a clean slate, took the best option for overall IQ with also a thought towards limiting consumer frustration over sensor issues as compared to the route Leica was forced to take. Arguing that Sony made a mistake and used bad judgement going with a thicker stack versus a thin stack with it's own IQ and reliability issues doesn't really hold much water imo.

Hi Tariq
Alway enjoy your posts, and hope we are friendly, despite some fun skirmishes over time

Re: the Clean slate. This one doesn't hold water for me, since there were only two lenses to start with, and Sony went about widely touting the use of ALT glass to make up for it. Worse, many of their new FE lenses are quite large, e.g. the 70200 zoom and the 1635. The thick stack is driving what had the potential to be a great compact system, right into DSLR territory.

You infer that a good thin stack is impossible because the M9 has some issues etc Really, If at first we don't succeed, forget it, sell all your great glass, and start over with a big fat cover that needs special lenses, which BTW, are horrible to MF?

The entire success of the whole Nex series, the E-Mount "brand", comes from it's wide adaptability. Throwing that out the window with a picky thick sensor stack strikes me as utterly oblivious behavior.

mdemeyer wrote:
I understand from KolariVision that the IR characteristics of the new filter are (essentially) the same as the stock filters. Regarding humidity resistance, Leica used S8612 in the M9, etc. which is more prone to humidity damage (according to the Schott technical information, although they are not as transparent on the subject as one might wish) than the BG39 used here.

It's a tricky balance - IR rejection, matching the visible band transmission curve to minimize white balance changes, availability of glass in small quantities in the needed thickness, adjusting sensor position to maintain proper optical path length (focus), etc....Show more →

Great post Michael. I could kiss you!!! For many of us you may have saved the whole system.

Jan 21, 2015 at 03:57 PM

naturephoto1
Offline
• • • • • •
Upload & Sell: Off

p.5 #16 · A thinner sensor stack may be possible after all!

uhoh7 wrote:
Hi Tariq
Alway enjoy your posts, and hope we are friendly, despite some fun skirmishes over time

Re: the Clean slate. This one doesn't hold water for me, since there were only two lenses to start with, and Sony went about widely touting the use of ALT glass to make up for it. Worse, many of their new FE lenses are quite large, e.g. the 70200 zoom and the 1635. The thick stack is driving what had the potential to be a great compact system, right into DSLR territory.

You infer that a good thin stack is impossible because the M9 has...Show more →

Hi Charlie,

You can not argue with Sony that the M RF lenses didn't work on the Sony A7 series cameras. Sony never stated that the WA RF M lenses worked well into the corners, but they did work on the cameras and to this date, they are the only FF camera besides Leica on which these lenses can be used and mounted. It was hopeful thinking on the part of all those with WA RF M lenses or those that wanted to use these lenses on the cameras that they would work well.

Rich

Edited on Jan 21, 2015 at 05:08 PM · View previous versions

Jan 21, 2015 at 04:01 PM

MAubrey
Offline
• • •
Upload & Sell: On

p.5 #17 · A thinner sensor stack may be possible after all!

hiepphotog wrote:
Personally, I think up till now the degradation of SLR film lens on DSLR is negligible due to the longer exit pupil distance and because of the lower resolution on these early DSLRs. However, if you notice the optical bench result of the Otus from Lensrental, you see significant degradation without a piece of glass. So I think all the optical makers knew about it all along but they had been mum on the subject. They just silently upgraded their optics over the years.

Brian Cadwell, the speedbooster designer, said on the comment thread of that particular lensrentals blog post that it was in the late 1990's that Canon & Nikon started giving their lenses the longer exit pupils that cause fewer problems for digital. Whether that was an anticipation of the sensor stack or some other design reason, I don't think we can guess.

Anyway, the reason for the degradation of the Otus was less about the Otus and more about the fact that the micro four thirds speedbooster they were mounting the Otus on was designed for a 4mm sensor stack.

Jan 21, 2015 at 04:23 PM

Tariq Gibran
Offline
• • • • • •
Upload & Sell: Off

p.5 #18 · A thinner sensor stack may be possible after all!

uhoh7 wrote:
Alway enjoy your posts, and hope we are friendly, despite some fun skirmishes over time

My main point is that we have not seen the "no issue/ perfect thin stack" used on FF as of yet...and Leica has been working on it for about 8 years thus far. As mentioned, even the current M240 suffers IR contamination making it a non-starter for some uses without additional filtration. Ideally, Leica would like to be using an even thinner stack but what they have now is still a compromise (which they would be first to admit). Sony, on the other hand, was not in the same position of having to find a solution for a digitally unfriendly existing lens line. One can argue all day long that a little IR contamination and possible sensor glass durability issues are worth dealing with for the huge benefit of smaller lenses but that's just not going to fly with the majority of users/ larger market (beyond a small niche)...and I suspect Sony fully understood that from the outset. I think if it were technically possible at the time of development, Sony would have done it (and Leica would have successfully done it by now as well). All that said, I'm very happy we have the additional option of this mod now with our Sony's. I certainly could see having this done on a dedicated body myself. For some, it will be a complete solution.

Jan 21, 2015 at 04:34 PM

pdmphoto
Offline
• • • •
Upload & Sell: Off

p.5 #19 · A thinner sensor stack may be possible after all!

Tariq Gibran wrote:
Well, I hope my post was not taken as confrontational at all. Sorry if that's how it came off as I certainly meant it in the spirit of a friendly debate.

My main point is that we have not seen the "no issue/ perfect thin stack" used on FF as of yet...and Leica has been working on it for about 8 years thus far. As mentioned, even the current M240 suffers IR contamination making it a non-starter for some uses without additional filtration. Ideally, Leica would like to be using an even thinner stack but what they have now is...Show more →

Being that the A7 series was designed to be a small system with high performance, Sony likely could have done more to minimize the thickness of the cover glass. I get the feeling that Sony don't really care. They want to sell cameras and lenses, not necessarily cater to alts. They have done a good job of enticing them though. The cover glass issue seems much like the lossy RAW file issue. Sony could have very likely made the change, but aren't all that interested. If they change it, it will be touted as a new feature on their latest high end model

Jan 21, 2015 at 04:45 PM

artur5
Offline
• • •
Upload & Sell: Off

p.5 #20 · A thinner sensor stack may be possible after all!

Tariq Gibran wrote:
.....
Thus one could just as easily argue that Sony, starting from a clean slate, took the best option for overall IQ with also a thought towards limiting consumer frustration over sensor issues as compared to the route Leica was forced to take. Arguing that Sony made a mistake and used bad judgement going with a thicker stack versus a thin stack with it's own IQ and reliability issues doesn't really hold much water imo.
....

Tariq, I always enjoy reading your posts, but this time I have to disagree somehow with you on the infamous filter-stack matter.
I don't think it's because of technical problems that Sony choose the safe 'superthick' way. It's rather a sort of 'safety belt' for protecting their, still scarce, lineup of FE lenses
The theory that a thin filter has nasty drawbacks as increased fragility or IR contamination clashes with the fact that the Leica M240 doesn't suffer from corrosion or cracked sensors and, if it has some IR contamination, it must be very mild for I haven't seen significant complaints from Leica's demanding users base.
Last, but non least, let's not forget that Ricoh's A12 M mount GXR module, which is tuned for M glass, has no IR contamination at all ( I can testify that) and I haven't heard of a single occurrence of corrosion or cracked filters, which must be very thin because there's no smearing to speak of, even with 15 or 18mm. focal lenses. Agreed, it's an APS-C sensor, but I reckon that the angles of incidence of a 15-16mm. lens at the corners of a cropped sensor may be as steep or more than with a 24-28mm. lens on a FF sensor.
We know that the Biogons ZM 25 and ZM 28 have serious problems with the A7 series while lenses like the CV15 /4.5 or ZM18, are terrific on a Ricoh A12.
Maybe the people of Kolari should ask Ricoh which glass they used for the filter of the A12 module. Or perhaps they should ask Sony directly, cause it's very probable that Sony made that Ricoh sensor.

Jan 21, 2015 at 05:10 PM

1 2 3 4 5 6 … 53 54 end
FM Forums \| Sony Forum \| Join Upload & Sell 1 2 3 4 5 6 … 53 54 end