Reading mode: Light Dark

Leica rumored new MF camera this year?

  • Thread starter Thread starter Chris Dodkin
  • Start date Start date
Go to last post
GXAlan said:
Not sure about what has been said but I find the focus on the S006 to be excellent. Improvements in AF were noted going from S2 to 006 to 007, so I am not sure where the focus error lies. That said, you are dealing with 6 micron pixels so the AF may only be set to that level.

It is interesting that the latest SL lenses have very high performance for the MTF charts compared to the S lenses, but in actual galleries, it seems as if the S lenses have a more GFX-like bite.

At some point, maybe someone will come out with an electronic S to GFX adapter. Apparently even the Fotodiox-no-aperture-control version is miscalculated where infinity focus is not allowed.

Someone near Leica Store Miami May be able to test the Thalia cinema versions of the Leica S lenses on a GFX...

At least for a lens rendering standpoint, the first Hollywood production to use Thalia lenses will be released in Nov:

Fantastic Beasts: The Crimes of Grindelwal
Click to expand...
Hi,

Comparing MTF curves, you need to keep in mind that the Leica S images need less magnification. So if you for instance check the 20 lp/mm line for the Leica-SL (24x36 mm) lens, you would need to compare it to a 16 lp/mm line for the Leica S.

Another factor may be that the Leica S could have a different micro lens design from the Leica SL or the Leica SL could have an OLP filter.

A third factor is that Lightroom/ACR always has a baseline sharpening that is specific for each system. That is active even with sharpening set to zero.

Regarding focusing issues on the Leica S/S2. Much of the info is coming from Lloyd Chambers. He reported the issue in two articles. The issue he reported affected at least four different cameras. Lloyd Chambers's site is a pay site, unfortunately.

https://diglloyd.com/prem/s/MF/LeicaS/autofocus-granite.html

https://diglloyd.com/prem/prot/MF/LeicaS/autofocus-alpine.html

There were also a couple of Leica S users who reported focusing problems with Leica S on the GetDPI forum.

Update: here is the link: https://www.getdpi.com/forum/medium...60441-leica-007-warning-potential-buyers.html

With the guy on GetDPI, his camera was three times to Germany for repairs, but was considered within specs.

Honestly, I was a bit baffled.

With regard to Diglloyd, I did subscribe to some of his sites for a long time. He has found many issues with many systems over time. I would suggest that in many cases, AF-systems are not really accurate enough for critical work. If you shoot stopped down to f/8, it will cover up a lot of focusing errors. Stopping down to f/16 firmly puts any sensor in the diffraction affected territory.

If you shoot at open aperture and focus accurately using magnified live view, that will show what the system is really capable of.

It is quite possible that AF will deliver acceptable sharpness. But, if you pay 85k$AUS for a system with excellent lenses, you want to have a focusing system that actually makes use of all the quality. Which means that only perfect is good enough.

The best solution is to focus manually, using maximum magnification in live view at shooting aperture.

It is quite possible that Diglloyd gets quite a few bad samples. It has been noted that very often he gets his stuff on loan from B&H. It may be that those lenses he loans from B&H are open box returns. It could be that those lenses were returned to B&H for reason.

The really bad thing with "Justalexanders" experience is that he got no help. His gear has been to Leica three times and they neither found or acknowledged an error.

Best regards

Erik
 
Last edited:
Spizzi said:
Dear god in heaven, please have this be a new system departure from the S body where Leica's past 3 years was a parallel development strategy to use the SL as an incubator for the future of Leica's higher end product set. Not saying the S goes away but maybe a product family focusing solely on Sony's 50MP sensor for MF. Where this new product family has learned and enhanced an SL-like body based on how the current SL is doing in the market today. Let this new line be another incubator for Leica's MF future.

Hopefully August will present more teasers. Hopefully dreams do come true.
Click to expand...
Not much chance, I’m afraid. The raison d’etre for the S has been to provide a DSLR experience in a medium format body. In my view, that precludes the use of the existing Sony chip with its 4:3 aspect ratio and 300ms readout time. They will almost certainly keep the form factor including mirror box and optical viewfinder. They will likely keep the 3:2 aspect ratio. I certainly don’t see them going with the existing Sony chip, and I don’t think even the replacement 100mp chip as very likely. They need to be distinguished from the Fuji, Hasselblad, and even the Pentax offering,and I think that means an evolutionary change not a big one. If nothing else, Leica has a huge R&D investment in the S glass so they will need to keep the same, long flange distance.
 
Hi Jared,

Your analysis is basically sound.

You are probably quite right about the Sony chip not being optimized for EVF applications.

Thanks for sharing!

Best regards

Erik
Jared Willson said:
Spizzi said:
Dear god in heaven, please have this be a new system departure from the S body where Leica's past 3 years was a parallel development strategy to use the SL as an incubator for the future of Leica's higher end product set. Not saying the S goes away but maybe a product family focusing solely on Sony's 50MP sensor for MF. Where this new product family has learned and enhanced an SL-like body based on how the current SL is doing in the market today. Let this new line be another incubator for Leica's MF future.

Hopefully August will present more teasers. Hopefully dreams do come true.
Click to expand...
Not much chance, I’m afraid. The raison d’etre for the S has been to provide a DSLR experience in a medium format body. In my view, that precludes the use of the existing Sony chip with its 4:3 aspect ratio and 300ms readout time. They will almost certainly keep the form factor including mirror box and optical viewfinder. They will likely keep the 3:2 aspect ratio. I certainly don’t see them going with the existing Sony chip, and I don’t think even the replacement 100mp chip as very likely. They need to be distinguished from the Fuji, Hasselblad, and even the Pentax offering,and I think that means an evolutionary change not a big one. If nothing else, Leica has a huge R&D investment in the S glass so they will need to keep the same, long flange distance.
Click to expand...
 
Jared Willson said:
Spizzi said:
Dear god in heaven, please have this be a new system departure from the S body where Leica's past 3 years was a parallel development strategy to use the SL as an incubator for the future of Leica's higher end product set. Not saying the S goes away but maybe a product family focusing solely on Sony's 50MP sensor for MF. Where this new product family has learned and enhanced an SL-like body based on how the current SL is doing in the market today. Let this new line be another incubator for Leica's MF future.

Hopefully August will present more teasers. Hopefully dreams do come true.
Click to expand...
Not much chance, I’m afraid. The raison d’etre for the S has been to provide a DSLR experience in a medium format body. In my view, that precludes the use of the existing Sony chip with its 4:3 aspect ratio and 300ms readout time. They will almost certainly keep the form factor including mirror box and optical viewfinder. They will likely keep the 3:2 aspect ratio. I certainly don’t see them going with the existing Sony chip, and I don’t think even the replacement 100mp chip as very likely. They need to be distinguished from the Fuji, Hasselblad, and even the Pentax offering,and I think that means an evolutionary change not a big one. If nothing else, Leica has a huge R&D investment in the S glass so they will need to keep the same, long flange distance.
Click to expand...
Agree. Possibly they can do a 30mm x 45mm sensor with 2x number of pixels (DALSA?).

I belive (?) their glass will be good enough (75 megapixels).🤔

They would make a big splash if the managed to include a true "global shutter".😀

It would be very useful for some of their target audiences.
 
Last edited:
Erik Kaffehr said:
Hi Jared,

Your analysis is basically sound.

You are probably quite right about the Sony chip not being optimized for EVF applications.
Click to expand...
Where did Jared say that? I think the current Sony chip does a good job with the EVFs in the H1D and GFX.

Jim

Erik Kaffehr said:
Jared Willson said:
Spizzi said:
Dear god in heaven, please have this be a new system departure from the S body where Leica's past 3 years was a parallel development strategy to use the SL as an incubator for the future of Leica's higher end product set. Not saying the S goes away but maybe a product family focusing solely on Sony's 50MP sensor for MF. Where this new product family has learned and enhanced an SL-like body based on how the current SL is doing in the market today. Let this new line be another incubator for Leica's MF future.

Hopefully August will present more teasers. Hopefully dreams do come true.
Click to expand...
Not much chance, I’m afraid. The raison d’etre for the S has been to provide a DSLR experience in a medium format body. In my view, that precludes the use of the existing Sony chip with its 4:3 aspect ratio and 300ms readout time. They will almost certainly keep the form factor including mirror box and optical viewfinder. They will likely keep the 3:2 aspect ratio. I certainly don’t see them going with the existing Sony chip, and I don’t think even the replacement 100mp chip as very likely. They need to be distinguished from the Fuji, Hasselblad, and even the Pentax offering,and I think that means an evolutionary change not a big one. If nothing else, Leica has a huge R&D investment in the S glass so they will need to keep the same, long flange distance.
Click to expand...
Click to expand...
 
Hi,

True global shutter is probably not very attractive with today's technology as it would mean increasing noise by around 41%.

Future technology may be different, but future technology is sort of 'future'.

Best regards

Erik
KWNJr said:
Jared Willson said:
Spizzi said:
Dear god in heaven, please have this be a new system departure from the S body where Leica's past 3 years was a parallel development strategy to use the SL as an incubator for the future of Leica's higher end product set. Not saying the S goes away but maybe a product family focusing solely on Sony's 50MP sensor for MF. Where this new product family has learned and enhanced an SL-like body based on how the current SL is doing in the market today. Let this new line be another incubator for Leica's MF future.

Hopefully August will present more teasers. Hopefully dreams do come true.
Click to expand...
Not much chance, I’m afraid. The raison d’etre for the S has been to provide a DSLR experience in a medium format body. In my view, that precludes the use of the existing Sony chip with its 4:3 aspect ratio and 300ms readout time. They will almost certainly keep the form factor including mirror box and optical viewfinder. They will likely keep the 3:2 aspect ratio. I certainly don’t see them going with the existing Sony chip, and I don’t think even the replacement 100mp chip as very likely. They need to be distinguished from the Fuji, Hasselblad, and even the Pentax offering,and I think that means an evolutionary change not a big one. If nothing else, Leica has a huge R&D investment in the S glass so they will need to keep the same, long flange distance.
Click to expand...
Agree. Possibly they can do a 30mm x 45mm sensor with 2x number of pixels (DALSA?).

I belive (?) their glass will be good enough (75 megapixels).🤔

They would make a big splash if the managed to include a true "global shutter".😀
Click to expand...
 
KWNJr said:
Jared Willson said:
Spizzi said:
Dear god in heaven, please have this be a new system departure from the S body where Leica's past 3 years was a parallel development strategy to use the SL as an incubator for the future of Leica's higher end product set. Not saying the S goes away but maybe a product family focusing solely on Sony's 50MP sensor for MF. Where this new product family has learned and enhanced an SL-like body based on how the current SL is doing in the market today. Let this new line be another incubator for Leica's MF future.

Hopefully August will present more teasers. Hopefully dreams do come true.
Click to expand...
Not much chance, I’m afraid. The raison d’etre for the S has been to provide a DSLR experience in a medium format body. In my view, that precludes the use of the existing Sony chip with its 4:3 aspect ratio and 300ms readout time. They will almost certainly keep the form factor including mirror box and optical viewfinder. They will likely keep the 3:2 aspect ratio. I certainly don’t see them going with the existing Sony chip, and I don’t think even the replacement 100mp chip as very likely. They need to be distinguished from the Fuji, Hasselblad, and even the Pentax offering,and I think that means an evolutionary change not a big one. If nothing else, Leica has a huge R&D investment in the S glass so they will need to keep the same, long flange distance.
Click to expand...
Agree. Possibly they can do a 30mm x 45mm sensor with 2x number of pixels (DALSA?).

I belive (?) their glass will be good enough (75 megapixels).🤔

They would make a big splash if the managed to include a true "global shutter".😀

It would be very useful for some of their target audiences.
Click to expand...
The main point of global shutter in my mind is freedom from shutter vibration. Not sure it's much of a win in a DSLR.

Jim
 
Hi Jim,

Jared did not use that wording, but he mentioned 300 ms readout time.

I am basing my statement on something written by Ming Thein, who is director of strategy or something like that at Hasselblad. He stated that the current 44x33 mm sensor is optimized for DSLRs rather than EVF.

Chris claimed, I think, that Sony does some modification for FUJI GFX with faster readout. That may be reasonable.

The Sony 44x33 mm sensor has been around some time and I would guess that they have a lot of feed back from camera makers on features and priorities.

My guess would be that the new sensors that are to be released this year will be more optimised for EVF systems than DSLRs.

But, all of this is just a guess...

Best regards

Erik

JimKasson said:
Erik Kaffehr said:
Hi Jared,

Your analysis is basically sound.

You are probably quite right about the Sony chip not being optimized for EVF applications.
Click to expand...
Where did Jared say that? I think the current Sony chip does a good job with the EVFs in the H1D and GFX.

Jim
JimKasson said:
Jared Willson said:
Spizzi said:
Dear god in heaven, please have this be a new system departure from the S body where Leica's past 3 years was a parallel development strategy to use the SL as an incubator for the future of Leica's higher end product set. Not saying the S goes away but maybe a product family focusing solely on Sony's 50MP sensor for MF. Where this new product family has learned and enhanced an SL-like body based on how the current SL is doing in the market today. Let this new line be another incubator for Leica's MF future.

Hopefully August will present more teasers. Hopefully dreams do come true.
Click to expand...
Not much chance, I’m afraid. The raison d’etre for the S has been to provide a DSLR experience in a medium format body. In my view, that precludes the use of the existing Sony chip with its 4:3 aspect ratio and 300ms readout time. They will almost certainly keep the form factor including mirror box and optical viewfinder. They will likely keep the 3:2 aspect ratio. I certainly don’t see them going with the existing Sony chip, and I don’t think even the replacement 100mp chip as very likely. They need to be distinguished from the Fuji, Hasselblad, and even the Pentax offering,and I think that means an evolutionary change not a big one. If nothing else, Leica has a huge R&D investment in the S glass so they will need to keep the same, long flange distance.
Click to expand...
Click to expand...
Click to expand...
 
Hi Jim,

Could also be about ultra fast flash X-sync times. I don't think it important for the wast majority of photographers.

Best regards

Erik
JimKasson said:
KWNJr said:
Jared Willson said:
Spizzi said:
Dear god in heaven, please have this be a new system departure from the S body where Leica's past 3 years was a parallel development strategy to use the SL as an incubator for the future of Leica's higher end product set. Not saying the S goes away but maybe a product family focusing solely on Sony's 50MP sensor for MF. Where this new product family has learned and enhanced an SL-like body based on how the current SL is doing in the market today. Let this new line be another incubator for Leica's MF future.

Hopefully August will present more teasers. Hopefully dreams do come true.
Click to expand...
Not much chance, I’m afraid. The raison d’etre for the S has been to provide a DSLR experience in a medium format body. In my view, that precludes the use of the existing Sony chip with its 4:3 aspect ratio and 300ms readout time. They will almost certainly keep the form factor including mirror box and optical viewfinder. They will likely keep the 3:2 aspect ratio. I certainly don’t see them going with the existing Sony chip, and I don’t think even the replacement 100mp chip as very likely. They need to be distinguished from the Fuji, Hasselblad, and even the Pentax offering,and I think that means an evolutionary change not a big one. If nothing else, Leica has a huge R&D investment in the S glass so they will need to keep the same, long flange distance.
Click to expand...
Agree. Possibly they can do a 30mm x 45mm sensor with 2x number of pixels (DALSA?).

I belive (?) their glass will be good enough (75 megapixels).🤔

They would make a big splash if the managed to include a true "global shutter".😀

It would be very useful for some of their target audiences.
Click to expand...
The main point of global shutter in my mind is freedom from shutter vibration. Not sure it's much of a win in a DSLR.

Jim
Click to expand...
 
Erik Kaffehr said:
Comparing MTF curves, you need to keep in mind that the Leica S images need less magnification. So if you for instance check the 20 lp/mm line for the Leica-SL (24x36 mm) lens, you would need to compare it to a 16 lp/mm line for the Leica S.
Click to expand...
True for printing but at 1:1 pixel peeping, I find a S006 to beat any SL. Likewise, just broadly there appears to be more astigmatism as well as general waviness. :)

The S006 had offset microlenses which may do something and the S006 has higher spec for the hue accuracy relative to other Kodak sensors of that vintage.

I have seen your posts on other forums with the disappointment you have with your P45 — the S may be diff.
Erik Kaffehr said:
Another factor may be that the Leica S could have a different micro lens design from the Leica SL or the Leica SL could have an OLP filter.
Click to expand...
Definitely different microlenses for sure. Neither of them have an OLP.
Erik Kaffehr said:
A third factor is that Lightroom/ACR always has a baseline sharpening that is specific for each system. That is active even with sharpening set to zero.
Click to expand...
Possibly although it seems in the latest version you have specific sharpening that is tailored for the system.
Erik Kaffehr said:
Regarding focusing issues on the Leica S/S2. Much of the info is coming from Lloyd Chambers. He reported the issue in two articles. The issue he reported affected at least four different cameras.
Click to expand...
This is where firmware plays a huge role where they introduced “a completely reprogrammed autofocus algorithm that improves AF performance and enables significantly faster and more precise focusing, the update also includes optimised AF drive control for all S-Lenses.”

Loading…

en.leica-camera.com

Erik Kaffehr said:
With the guy on GetDPI, his camera was three times to Germany for repairs, but was considered within specs.

Honestly, I was a bit baffled.
Click to expand...
Maybe user error? The S really works reliably in back button focus but the half press shutter approach from Canon and Nikon fails to work reliably since the Leica is a bit spongy. Could be something as silly as a dirty AF sensor.

To be frank, for a working pro (in the US) it is pretty hard to choose something other than Canon given the superior CPS service over Sony and Fuji. I don’t have first hand experience with Nikon.
Erik Kaffehr said:
With regard to Diglloyd, I did subscribe to some of his sites for a long time. He has found many issues with many systems over time. I would suggest that in many cases, AF-systems are not really accurate enough for critical work. If you shoot stopped down to f/8, it will cover up a lot of focusing errors. Stopping down to f/16 firmly puts any sensor in the diffraction affected territory.
Click to expand...
I think your comment about using rental equipment is an issue. Outside of Lensrentals.com, no one actually tests to make sure that the gear going out meets spec. I don’t think the Leica S 006 AF is great. It is slow and in low light it just won’t try. But at the metric of 6 micron pixels (which is easier than a higher density sensor), I rarely have a misfocus with a static subject where the camera says I am in focus and it is not. I do get the “I am not able to find anything” state in low light.
Erik Kaffehr said:
If you shoot at open aperture and focus accurately using magnified live view, that will show what the system is really capable of.
Click to expand...
Agree.
Erik Kaffehr said:
It is quite possible that AF will deliver acceptable sharpness. But, if you pay 85k$AUS for a system with excellent lenses, you want to have a focusing system that actually makes use of all the quality. Which means that only perfect is good enough.
Click to expand...
Except that the S lenses are easily acquired used at 50% discount right now... which may change if the S008 is good.
Erik Kaffehr said:
The really bad thing with "Justalexanders" experience is that he got no help. His gear has been to Leica three times and they neither found or acknowledged an error.

Best regards

Erik
Click to expand...
Agree. Leica remains a better camera for hobbyists than professionals. Any professional shooting a Leica should hope to be within driving distance to a Leica Store. In the US, the San Francisco store (and Walnut Creek and Rancho Mirage CameraWest sister stores) and Miami store have a good track record. If an S 007 had that kind of problem then those US stores would likely have just taken care of the problem with a new camera.

The lack of consistency in service and lack of reliability is why Canon is the choice of so many pros despite poorer dynamic range than competitors and in some cases less sharp glass than competitors.

The SL may bring Panasonic levels of reliability. I haven’t heard a lot of problems like I have with the S.
 
Hi,

Some comments
GXAlan said:
This is where firmware plays a huge role where they introduced “a completely reprogrammed autofocus algorithm that improves AF performance and enables significantly faster and more precise focusing, the update also includes optimised AF drive control for all S-Lenses.”

https://en.leica-camera.com/Company...CA-S2-S2-P-FW-1.4.0.0.-and-all-Leica-S-Lenses
GXAlan said:
With the guy on GetDPI, his camera was three times to Germany for repairs, but was considered within specs.

Honestly, I was a bit baffled.
Click to expand...
Click to expand...
That update was from 2014, but the user experience was reported 2016.

The camera was sent three times to Wetzlar and was found to be within specs.

I would expect that Leica would upgrade camera firmware and clean the AF-sensor if a camera would be sent in for repair with AF-issues.

The image below indicates some of the problems he had:



Image at top is focused in live view, camera indicates focus distance 8.02 m real distance is 500 m.OK, Leica responded that lens barrel marking and camera display should not be used for focusing. But why have an indication with three digits if all of them are wrong?Leica suggested that he uses LV for focusing, and that is good advice on the S (typ 007), but it would not help S2 or (typ 006) users.

Image at top is focused in live view, camera indicates focus distance 8.02 m real distance is 500 m.OK, Leica responded that lens barrel marking and camera display should not be used for focusing. But why have an indication with three digits if all of them are wrong?Leica suggested that he uses LV for focusing, and that is good advice on the S (typ 007), but it would not help S2 or (typ 006) users.
GXAlan said:
Maybe user error? The S really works reliably in back button focus but the half press shutter approach from Canon and Nikon fails to work reliably since the Leica is a bit spongy. Could be something as silly as a dirty AF sensor.
Click to expand...
I don't think so. He spent a lot of effort on this.
GXAlan said:
GXAlan said:
With regard to Diglloyd, I did subscribe to some of his sites for a long time. He has found many issues with many systems over time. I would suggest that in many cases, AF-systems are not really accurate enough for critical work. If you shoot stopped down to f/8, it will cover up a lot of focusing errors. Stopping down to f/16 firmly puts any sensor in the diffraction affected territory.
Click to expand...
Click to expand...
GXAlan said:
I think your comment about using rental equipment is an issue. Outside of Lensrentals.com, no one actually tests to make sure that the gear going out meets spec. I don’t think the Leica S 006 AF is great. It is slow and in low light it just won’t try. But at the metric of 6 micron pixels (which is easier than a higher density sensor), I rarely have a misfocus with a static subject where the camera says I am in focus and it is not. I do get the “I am not able to find anything” state in low light.
Click to expand...
I would make the point that Diglloyd has issues with almost any camera with regard to AF.

Lensrentals looked into AF accuracy a few years ago. They looked at AF on Canon and Nikon, and neither was even close to manual AF in live view.

They found out that the latest generation of Canon lenses on latest generation Canon DSLR bodies had PDAF as good as manual AF in live view.

GXAlan said:
GXAlan said:
It is quite possible that AF will deliver acceptable sharpness. But, if you pay 85k$AUS for a system with excellent lenses, you want to have a focusing system that actually makes use of all the quality. Which means that only perfect is good enough.
Click to expand...
Except that the S lenses are easily acquired used at 50% discount right now... which may change if the S008 is good.
Click to expand...
Well, that was the amount he spent on his gear.
GXAlan said:
GXAlan said:
The really bad thing with "Justalexanders" experience is that he got no help. His gear has been to Leica three times and they neither found or acknowledged an error.

Best regards

Erik
Click to expand...


The lack of consistency in service and lack of reliability is why Canon is the choice of so many pros despite poorer dynamic range than competitors and in some cases less sharp glass than competitors.
Click to expand...
As far as I know, Canon has extremely good lenses. Telephotos were always excellent. The ultra wide zooms were not so good, but that has been resolved with the 16-35/4L and the 16-35/2.8LIII.

The latest generation Canons have pretty decent DR.



892643cdef6947f98b33b868d1ce9146.jpg.png


The reason I bought into Hasselblad / P45+ was pretty much that many statements about MFD did not make any sense to me and wanted to find out on my own. So it is not that I am unhappy with it. It performs as I expected.

It did have an advantage in resolution over my 24 MP camera, at least with some of the lenses that was the reason I was using it for 3-4 years.

But, I feel that buying into an old system just to get into MFD makes poor sense.

Best regards

Erik





--
Erik Kaffehr
Website: http://echophoto.dnsalias.net
Magic uses to disappear in controlled experiments…
Gallery: http://echophoto.smugmug.com
Articles: http://echophoto.dnsalias.net/ekr/index.php/photoarticles
 
Erik Kaffehr said:
Hi,

Some comments
Erik Kaffehr said:
This is where firmware plays a huge role where they introduced “a completely reprogrammed autofocus algorithm that improves AF performance and enables significantly faster and more precise focusing, the update also includes optimised AF drive control for all S-Lenses.”

https://en.leica-camera.com/Company...CA-S2-S2-P-FW-1.4.0.0.-and-all-Leica-S-Lenses
Erik Kaffehr said:
With the guy on GetDPI, his camera was three times to Germany for repairs, but was considered within specs.

Honestly, I was a bit baffled.
Click to expand...
Click to expand...
That update was from 2014, but the user experience was reported 2016.

The camera was sent three times to Wetzlar and was found to be within specs.

I would expect that Leica would upgrade camera firmware and clean the AF-sensor if a camera would be sent in for repair with AF-issues.

The image below indicates some of the problems he had:

Image at top is focused in live view, camera indicates focus distance 8.02 m real distance is 500 m.OK, Leica responded that lens barrel marking and camera display should not be used for focusing. But why have an indication with three digits if all of them are wrong?Leica suggested that he uses LV for focusing, and that is good advice on the S (typ 007), but it would not help S2 or (typ 006) users.

Image at top is focused in live view, camera indicates focus distance 8.02 m real distance is 500 m.OK, Leica responded that lens barrel marking and camera display should not be used for focusing. But why have an indication with three digits if all of them are wrong?Leica suggested that he uses LV for focusing, and that is good advice on the S (typ 007), but it would not help S2 or (typ 006) users.
Erik Kaffehr said:
Maybe user error? The S really works reliably in back button focus but the half press shutter approach from Canon and Nikon fails to work reliably since the Leica is a bit spongy. Could be something as silly as a dirty AF sensor.
Click to expand...
I don't think so. He spent a lot of effort on this.
Erik Kaffehr said:
Erik Kaffehr said:
With regard to Diglloyd, I did subscribe to some of his sites for a long time. He has found many issues with many systems over time. I would suggest that in many cases, AF-systems are not really accurate enough for critical work. If you shoot stopped down to f/8, it will cover up a lot of focusing errors. Stopping down to f/16 firmly puts any sensor in the diffraction affected territory.
Click to expand...
I think your comment about using rental equipment is an issue. Outside of Lensrentals.com, no one actually tests to make sure that the gear going out meets spec. I don’t think the Leica S 006 AF is great. It is slow and in low light it just won’t try. But at the metric of 6 micron pixels (which is easier than a higher density sensor), I rarely have a misfocus with a static subject where the camera says I am in focus and it is not. I do get the “I am not able to find anything” state in low light.
Click to expand...
I would make the point that Diglloyd has issues with almost any camera with regard to AF.

Lensrentals looked into AF accuracy a few years ago. They looked at AF on Canon and Nikon, and neither was even close to manual AF in live view.

They found out that the latest generation of Canon lenses on latest generation Canon DSLR bodies had PDAF as good as manual AF in live view.
Click to expand...
What is "manual AF"?
Erik Kaffehr said:
Erik Kaffehr said:
Erik Kaffehr said:
It is quite possible that AF will deliver acceptable sharpness. But, if you pay 85k$AUS for a system with excellent lenses, you want to have a focusing system that actually makes use of all the quality. Which means that only perfect is good enough.
Click to expand...
Except that the S lenses are easily acquired used at 50% discount right now... which may change if the S008 is good.
Click to expand...
Well, that was the amount he spent on his gear.
Erik Kaffehr said:
Erik Kaffehr said:
The really bad thing with "Justalexanders" experience is that he got no help. His gear has been to Leica three times and they neither found or acknowledged an error.

Best regards

Erik
Click to expand...
The lack of consistency in service and lack of reliability is why Canon is the choice of so many pros despite poorer dynamic range than competitors and in some cases less sharp glass than competitors.
Click to expand...
As far as I know, Canon has extremely good lenses. Telephotos were always excellent. The ultra wide zooms were not so good, but that has been resolved with the 16-35/4L and the 16-35/2.8LIII.

The latest generation Canons have pretty decent DR.

892643cdef6947f98b33b868d1ce9146.jpg.png


The reason I bought into Hasselblad / P45+ was pretty much that many statements about MFD did not make any sense to me and wanted to find out on my own. So it is not that I am unhappy with it. It performs as I expected.

It did have an advantage in resolution over my 24 MP camera, at least with some of the lenses that was the reason I was using it for 3-4 years.

But, I feel that buying into an old system just to get into MFD makes poor sense.

Best regards

Erik

--
Erik Kaffehr
Website: http://echophoto.dnsalias.net
Magic uses to disappear in controlled experiments…
Gallery: http://echophoto.smugmug.com
Articles: http://echophoto.dnsalias.net/ekr/index.php/photoarticles
Click to expand...
 
GXAlan said:
The S006 had offset microlenses which may do something and the S006 has higher spec for the hue accuracy relative to other Kodak sensors of that vintage.

Click to expand...
What, pray tell, would a spec for "hue accuracy" look like?

Jim
 
Erik Kaffehr said:
Hi Jim,

Jared did not use that wording, but he mentioned 300 ms readout time.
Click to expand...
That's the readout time for the entire raw image. The readout time for a pixel subset sufficient for refreshing an EVF is considerably faster. I believe the GFX refreshes the display at 50 Hz.
Erik Kaffehr said:
I am basing my statement on something written by Ming Thein, who is director of strategy or something like that at Hasselblad. He stated that the current 44x33 mm sensor is optimized for DSLRs rather than EVF.
Click to expand...
That would be ironic, since it seems that most of the production of that chip are going into MILC cameras. Are you sure he's not talking about CDAF only?
Erik Kaffehr said:
Chris claimed, I think, that Sony does some modification for FUJI GFX with faster readout. That may be reasonable.
Click to expand...
I have not seen that statement from an authoritative source.
Erik Kaffehr said:
The Sony 44x33 mm sensor has been around some time and I would guess that they have a lot of feed back from camera makers on features and priorities.

My guess would be that the new sensors that are to be released this year will be more optimised for EVF systems than DSLRs.
Click to expand...
 
JimKasson said:
GXAlan said:
The S006 had offset microlenses which may do something and the S006 has higher spec for the hue accuracy relative to other Kodak sensors of that vintage.
Click to expand...
What, pray tell, would a spec for "hue accuracy" look like?

Jim

--
http://blog.kasson.com
Click to expand...
Maybe hue precision is better term than hue accuracy.

———

Loading…

pdf1.alldatasheet.com

Hope the link works. Kodak reported the nominal red green hue shift as a nominal 1.2% and blue green hue shift as a nominal 2.6% for the KAF-37500

With the footnote

“Gradual variations in hue (red with respect to green pixels and blue with respect to green pixels) in regions of interest (168×168 blocks)
within the sensor.”

the M9 sensor 1.8%

Loading…

www.onsemi.com

but the KAF-40000 (Pentax 645D) is listed as 5%

Loading…

www.onsemi.com

And the KAF-50100 (Hasselblad) also at 5%

Loading…

www.onsemi.com

———-

you then see some reciprocal changes. A higher hue accuracy is claimed for the Leica S006 and S2 CCD sensors and thus it must give something up. We see LOWER quantum efficiency specifications than the Pentax 645D sensor.

the KAF37500 is claimed 23/39/35% in RGB

the KAF40000 is claimed 42/44/38% in RGB

Alan
 
GXAlan said:
JimKasson said:
GXAlan said:
The S006 had offset microlenses which may do something and the S006 has higher spec for the hue accuracy relative to other Kodak sensors of that vintage.
Click to expand...
What, pray tell, would a spec for "hue accuracy" look like?

Jim

--
http://blog.kasson.com
Click to expand...
Maybe hue precision is better term than hue accuracy.

———

https://pdf1.alldatasheet.com/datas...37500/+WW_Q-JVLLS.xKdxGlhCG.zC+/datasheet.pdf

Hope the link works.
Click to expand...
Doesn't work for me.



3bc5884a01e74d648d0e505ddf6811c2.jpg.png


GXAlan said:
Kodak reported the nominal red green hue shift as a nominal 1.2% and blue green hue shift as a nominal 2.6% for the KAF-37500

With the footnote

“Gradual variations in hue (red with respect to green pixels and blue with respect to green pixels) in regions of interest (168×168 blocks)
within the sensor.”

the M9 sensor 1.8%

http://www.onsemi.com/pub/Collateral/KAF-18500-D.PDF

but the KAF-40000 (Pentax 645D) is listed as 5%

http://www.onsemi.com/pub/Collateral/KAF-40000-D.PDF

And the KAF-50100 (Hasselblad) also at 5%

http://www.onsemi.com/pub/Collateral/KAF-50100-D.PDF
Click to expand...
I don't see those numbers. Where should I be looking?
GXAlan said:
———-

you then see some reciprocal changes. A higher hue accuracy is claimed for the Leica S006 and S2 CCD sensors and thus it must give something up. We see LOWER quantum efficiency specifications than the Pentax 645D sensor.

the KAF37500 is claimed 23/39/35% in RGB

the KAF40000 is claimed 42/44/38% in RGB
Click to expand...
Are you saying that differences in QE across CFA channels have anything at all to do with hue accuracy in the image after it's passed through the compromise matrix? I'll have to see some evidence of that. maybe I could understand you better if that top link worked for me.

Please explain your metrics.

Jim



--

the last word the last word - Photography meets digital computer technology. Photography wins -- most of the time.

Photography meets digital computer technology. Photography wins -- most of the time.
blog.kasson.com blog.kasson.com
 
JimKasson said:
Top link
Click to expand...
1) just google KAF-37500 PDF data sheet

2) numbers are a little less than halfway through the documents where the full list of specifications are listed right before the show the QE curves. It is a different page number I think for each of the docs but around pg 8.

3) You are probably much better equipped to do this testing with your gear and experience with Matlab. Unless we can find a Kodak/OnSemi engineer to give us more than the marketing description of what was being measured, we are out of luck.

The argument is that weak CFAs have increased low light sensitivity until you get to the point where the CFA is clear. The argument is that the maximally strong CFA can only be accurate for a given light spectra at the expense of noise.

The Leica-Kodak CCDs were always spec’d to a better nominal hue shift value (even though they said up to 12% is ok). Whether pure marketing number or not, it really does suggest that something different was done for those sensors in particular.

Original M9 CCD press release talked about new red pigment in the filter...
 
Last edited:
GXAlan said:
GXAlan said:
Top link
Click to expand...
1) just google KAF-37500 PDF data sheet
Click to expand...
Thanks. I've now got the data sheet.
GXAlan said:
2) numbers are a little less than halfway through the documents where the full list of specifications are listed right before the show the QE curves. It is a different page number I think for each of the docs but around pg 8.
Click to expand...
I see it now. I had no idea what it means, but I did find this cryptic explanation in a Kodak data sheet:

Gradual variations in hue (red with respect to green pixels and blue with respect to green pixels) in regions of interest (146 x 146 blocks) within the sensor.

I can't tell how they are measuring hue, or what the spatial filtering is, but it looks to me that they are talking about the consistency of the CFA filters, not their absolute spectral characteristics.
GXAlan said:
3) You are probably much better equipped to do this testing with your gear and experience with Matlab. Unless we can find a Kodak/OnSemi engineer to give us more than the marketing description of what was being measured, we are out of luck.

The argument is that weak CFAs have increased low light sensitivity until you get to the point where the CFA is clear. The argument is that the maximally strong CFA can only be accurate for a given light spectra at the expense of noise.

The Leica-Kodak CCDs were always spec’d to a better nominal hue shift value (even though they said up to 12% is ok). Whether pure marketing number or not, it really does suggest that something different was done for those sensors in particular.

Original M9 CCD press release talked about new red pigment in the filter...
Click to expand...
 
JimKasson said:
GXAlan said:
JimKasson said:
Top link
Click to expand...
1) just google KAF-37500 PDF data sheet
Click to expand...
Thanks. I've now got the data sheet.
JimKasson said:
2) numbers are a little less than halfway through the documents where the full list of specifications are listed right before the show the QE curves. It is a different page number I think for each of the docs but around pg 8.
Click to expand...
I see it now. I had no idea what it means, but I did find this cryptic explanation in a Kodak data sheet:

Gradual variations in hue (red with respect to green pixels and blue with respect to green pixels) in regions of interest (146 x 146 blocks) within the sensor.

I can't tell how they are measuring hue, or what the spatial filtering is, but it looks to me that they are talking about the consistency of the CFA filters, not their absolute spectral characteristics.
JimKasson said:
3) You are probably much better equipped to do this testing with your gear and experience with Matlab. Unless we can find a Kodak/OnSemi engineer to give us more than the marketing description of what was being measured, we are out of luck.

The argument is that weak CFAs have increased low light sensitivity until you get to the point where the CFA is clear. The argument is that the maximally strong CFA can only be accurate for a given light spectra at the expense of noise.

The Leica-Kodak CCDs were always spec’d to a better nominal hue shift value (even though they said up to 12% is ok). Whether pure marketing number or not, it really does suggest that something different was done for those sensors in particular.

Original M9 CCD press release talked about new red pigment in the filter...
Click to expand...
Click to expand...
 
left eye said:
JimKasson said:
GXAlan said:
left eye said:
Top link
Click to expand...
1) just google KAF-37500 PDF data sheet
Click to expand...
Thanks. I've now got the data sheet.
left eye said:
2) numbers are a little less than halfway through the documents where the full list of specifications are listed right before the show the QE curves. It is a different page number I think for each of the docs but around pg 8.
Click to expand...
I see it now. I had no idea what it means, but I did find this cryptic explanation in a Kodak data sheet:

Gradual variations in hue (red with respect to green pixels and blue with respect to green pixels) in regions of interest (146 x 146 blocks) within the sensor.

I can't tell how they are measuring hue, or what the spatial filtering is, but it looks to me that they are talking about the consistency of the CFA filters, not their absolute spectral characteristics.
left eye said:
3) You are probably much better equipped to do this testing with your gear and experience with Matlab. Unless we can find a Kodak/OnSemi engineer to give us more than the marketing description of what was being measured, we are out of luck.

The argument is that weak CFAs have increased low light sensitivity until you get to the point where the CFA is clear. The argument is that the maximally strong CFA can only be accurate for a given light spectra at the expense of noise.

The Leica-Kodak CCDs were always spec’d to a better nominal hue shift value (even though they said up to 12% is ok). Whether pure marketing number or not, it really does suggest that something different was done for those sensors in particular.

Original M9 CCD press release talked about new red pigment in the filter...
Click to expand...
--
http://blog.kasson.com
Click to expand...
There was the emerald green which I remember back in the day...

https://en.wikipedia.org/wiki/RGBE_filter

Just as inkjet printers use different hues of their CMYK inks (i.e. vivid magenta), etc, I would expect the likes of Fuji to be requesting certain purities/wavelength of the RGB filters from Sony, I really don't see why not.
Click to expand...
Fuji makes colorants:

http://www.fujifilmusa.com/products/semiconductor_materials/image-sensor-color-mosaic/index.html

It is possible that the X1D has Fuji colorants.
left eye said:
Also I don't see why they would be stating those exact frequencies in any marketing blurb - or even making it easy for other manufactures to find out.
Click to expand...
Anybody with a monochromator can figure that out. People producing camera profiles commercially often use those instruments. Not specifying the CFA characteristics doesn't hide anything from the competition, just from the customers.
left eye said:
Certainly the (now classic) RGB filters of Kodak-made CCD sensors would not have same filteration wavelengths and purity as their contemporary Sony sensors. The colour differences between cameras with either Kodak or Sony sensors at the time were much more noticeable then, than now with the current ubiquitous use of Sony sensors.

Canon in a way has retained some of this lost Kodak-CCD colour magic by using their own RGB filters on their sensors - though even with Canon this colour magic/quality seems to be waning in some of their current bodies.

I've never liked the 'video-camera' Sony look, and still don't, though there's many ways to push raw colour profiles that mostly mitigate the issue entirely.

I suppose it comes down to whether you think all bayer RGB filters have exactly the same wavelengths and purity, or not.
Click to expand...
Please define "purity" in this context. I don't know what you mean by that. Are you referring to the distance between the half-power points?

Certainly, all Bayer CFAs don't have the same spectral characteristics.
left eye said:
I don't need to see the technical evidence to know the colour quality I saw when I used Kodak CCD - I still miss it.
Click to expand...
Jim

--
http://blog.kasson.com
 
Last edited:
You must log in or register to reply here.

Keyboard shortcuts

f
Forum
m
My threads
Mobile site
About
Editorial content
Cameras & Lenses
Community
All content, design, and layout are Copyright © 1998–2025 Digital Photography Review All Rights Reserved.
Reproduction in whole or part in any form or medium without specific written permission is prohibited.
When you use DPReview links to buy products, the site may earn a commission.
©GPS Media - Guides, Products, Services.
Back
Top