jpaulmoore
Active member
I just read where Lloyd Chambers (Digital Lloyd) is seeing some pronounced Posterization with some A7r ll files. Anybody seeing evidence of this?
J. Paul
J. Paul
The GetDPI Photography Forum
Great to see you here. Join our insightful photographic forum today and start tapping into a huge wealth of photographic knowledge. Completing our simple registration process will allow you to gain access to exclusive content, add your own topics and posts, share your work and connect with other members through your own private inbox! And don’t forget to say hi!
Seeing any Posterization issues with A7r ll?
- Thread starter jpaulmoore
- Start date
eleanorbrown
New member
No I haven't...not so far anyway (I read Lloyds article on this)....but I'm converting my RAWs in C1, not LR so that may (or may not?) make a difference. Will be curious to see what others have to say. Eleanor
Guy Mancuso
Administrator, Instructor
Nothing from C1
tashley
Subscriber Member
It would be interesting to drop that file that Lloyd has shown (his daughter by the blue lake) into C1 because it is really, really nasty as output from ACR. I have emailed him to ask if he could try it. But I have to say that the underlying unease I expressed at the outset about the files from the A7RII is not going away.
Last edited:
k-hawinkler
Well-known member
Is the raw file available for others to try? <--- Probably a stupid question.
In the end, maybe Mr. Chambers does all of us a favor by forcing Sony to support 14 bit lossless compression in their cameras?
In the end, maybe Mr. Chambers does all of us a favor by forcing Sony to support 14 bit lossless compression in their cameras?
Last edited:
Amin
Active member
Doesn't seem to be. Seems very odd that it isn't.
I have seen it on the web with files PP with Adobe but not with mine or others using C1.
I think Adobe rushed their update because they knew C1 already released their update and have shown some issues like LE hot pixels, shadow noise, and posterization that are nonexistent using C1's basic conversion. When I first saw early images I did not know if the effect was a result of increased DR or that the OP's were just moving the shadow and highlight sliders too far. Posterization defines what I saw pretty well with Adobe conversions.
I think Adobe rushed their update because they knew C1 already released their update and have shown some issues like LE hot pixels, shadow noise, and posterization that are nonexistent using C1's basic conversion. When I first saw early images I did not know if the effect was a result of increased DR or that the OP's were just moving the shadow and highlight sliders too far. Posterization defines what I saw pretty well with Adobe conversions.
Amin
Active member
Here is a RAW file I took with my A7RII that has some water ripples with blue and green tones:
https://drive.google.com/file/d/0B93HF6w-u-VXRk1Ga3huaUFsVkk/view?usp=sharing
I'm away from home and don't have ACR on this computer. Can someone process it there and tell me if you see posterization? I don't see any in the C1 conversion.
https://drive.google.com/file/d/0B93HF6w-u-VXRk1Ga3huaUFsVkk/view?usp=sharing
I'm away from home and don't have ACR on this computer. Can someone process it there and tell me if you see posterization? I don't see any in the C1 conversion.
ShooterSteve
New member
Just opened it in Photoshop and it looks fine to me. Is there something I can push in the adjustments to bring out any issues?
ErikKaffehr
Well-known member
Hi,
This example may be caused by the camera profile. I have seen some odd banding/posterisation artefacts cuased by home made DNG-profiles and also Capture One on files from my Phase One P45+ back using what they call 16-bit files. With some DNG profiles and the Capture One profile there was horrible banding, with Adobe Standard none at all. Profiles are complicated, far to complicated I would say.
Sony's raw compression can cause artefacts, though, but nothing like what Lloyd shows. The best example I have seen have been shown by Lloyd and it is a star tracks image:
This image has been used by the RawDigger folks who explain how it works: RawDigger: detecting posterization in SONY cRAW/ARW2 files | RawDigger
Personally, I would expect that the "delta" compression could cause artefacts but Lloyd seems to be obsessed with density curve compression which would not give any posterisation.
Best regards
Erik
This example may be caused by the camera profile. I have seen some odd banding/posterisation artefacts cuased by home made DNG-profiles and also Capture One on files from my Phase One P45+ back using what they call 16-bit files. With some DNG profiles and the Capture One profile there was horrible banding, with Adobe Standard none at all. Profiles are complicated, far to complicated I would say.
Sony's raw compression can cause artefacts, though, but nothing like what Lloyd shows. The best example I have seen have been shown by Lloyd and it is a star tracks image:
This image has been used by the RawDigger folks who explain how it works: RawDigger: detecting posterization in SONY cRAW/ARW2 files | RawDigger
Personally, I would expect that the "delta" compression could cause artefacts but Lloyd seems to be obsessed with density curve compression which would not give any posterisation.
Best regards
Erik
Last edited:
ErikKaffehr
Well-known member
Hi,
I don't see any artefacts with LR6. With artefacts on rippled water I would expects some problems with colour aliasing, but not at f/11 on the Sony. Anyway I cannot see anything I would called posterisation.
There can be a lot of causes to posterisation, screen profiles and colour profiles can have discontinuous gradients.
Best regards
Erik
I don't see any artefacts with LR6. With artefacts on rippled water I would expects some problems with colour aliasing, but not at f/11 on the Sony. Anyway I cannot see anything I would called posterisation.
There can be a lot of causes to posterisation, screen profiles and colour profiles can have discontinuous gradients.
Best regards
Erik
tashley
Subscriber Member
Nope, it's an ARW file. I have written and asked him to try it in C1 or to let me have the original file via dropbox so I can try it.
The RAWDigger histograms of LLoyd's file (inside sub area) are quite worrisome. In each color channel that each have about 10M value (R, 2 x G & B channels as sensor is Bayer filtered) there is huge "piling" of luminance values into same values, near half million same values (in a full 14 bit RAW there would be potentially 16 000 different values) and about 50% of each channels contains value selected from less than 20-30 different values.
If the file was from a Nikon D810 for example there would be more even distribution, where Sony has half million same values Nikon woiuld have 250K etc. ==> Way easier task for demosaic algorithm to put out clean output.
If the file was from a Nikon D810 for example there would be more even distribution, where Sony has half million same values Nikon woiuld have 250K etc. ==> Way easier task for demosaic algorithm to put out clean output.
Guy Mancuso
Administrator, Instructor
Sorry not buying that argument.