Green/blue color cast when re-developing a RAW from older ACDSee version

[Trevor]

(Originally posted as a reply to another thread but since it looks like a different issue , now moved to a new topic here:

I have a very annoying, green/blue cast problem

I developed several RAW images (Nikon NEF's from a Z7) on computer "A" with Ultimate 2021
on computer "A" they look just fine to me after developing.
I then transferred those processed RAW images, together with their side car .xmp files and the their "preview" jpg file (contained in the hidden folder called "[Developed]" ) to Computer "B"


On Computer "B" I'm using Ultimate 2023. Here, the the thumbnails look fine and the image still looks fine in VIEW mode. (I think it's the preview jpg from the "[developed]" folder that is actually displayed in VIEW mode?)

So far so good,...
BUT, when I start to Develop the same image again on computer B, it opens and, after about 4 seconds, the displayed image will change its "good colors" so that a definite green/blue cast beomes visible :-(
I think what I see now on the monitor is being decoded from RAW (with my copied sidecar settings).

The green/blue cast I see is as if the white balance has somehow been changed to a more green setting of the "Tint" and a somewhat bluer "temperature colour" . ( or maybe it is some of the other colour setings that are changed?)

I've made a drop box link with an example of such an image, if you'd like to take a look

https://www.dropbox.com/scl/fo/vfdmnoxwlgxqzbaqfzdyp/h?rlkey=1cwaxwgrbicelzaj9q1pnq6h1&dl=0

It would be interesting if you can see and confirm the color change to more blue/green when going from VIEW to DEVELOP Mode using Ultimate 2023?
Please make sure you copy also the hidden file.
.... perhaps you have some tips ?

Many thanks
Trevor

By the way I (think that I ) used the ulimate 2021 on Computer A to copy the images to computer B , (computers were connected via my network) so that all the hidden folders were created too

[Greyfox]

Trevor

Unfortunately the zip file only contains a JPG file, and as the issue is all about the development of a Nikon Z7 Raw file, I would need to have the original RAW file, together with any sidecar xmp files produced in the two versions of ACDSee to be able to make any informed comment. In addition to that, as this is a color issue, details of ACDSee's Color management, and the color profile selected for your monitor would help

In terms of ACDSee's View mode, when you have a RAW file selected there is a toggle in the lower tool bar to toggle between the embedded preview, and the decoded RAW image. You should NOT expect the two to be identical.

[Trevor]

Trevor

Unfortunately the zip file only contains a JPG file, and as the issue is all about the development of a Nikon Z7 Raw file, I would need to have the original RAW file, together with any sidecar xmp files produced in the two versions of ACDSee to be able to make any informed comment. In addition to that, as this is a color issue, details of ACDSee's Color management, and the color profile selected for your monitor would help

In terms of ACDSee's View mode, when you have a RAW file selected there is a toggle in the lower tool bar to toggle between the embedded preview, and the decoded RAW image. You should NOT expect the two to be identical.

Hi Greyfox,
Thanks for taking a look
First to the drop box link, ...
Please try again. Sorry, it seems an extra, unwanted, icon appeared when I copy/pasted my original post to a new thread. Maybe you clicked on that extra icon?
My original post is edited now to show only what I originally posted, and it works OK for me.

When I click on my link, I get this:
... two files the NEF and the .xmp and a folder called "[developed]" containing the jpg preview file calledDSC_6643.NEF.jpg

and clicking on that "Download" button gives me a zip with all 3 files


My colour settings on Computer "B" using ACDSee U 2023 are like in this screeshot. The selections on Computer "A" are the same.

I assume ACDSee will use the same colour profile for my monitor on Computer "B", when it is in VIEW mode and DEVELOP mode ?

In terms of ACDSee's View mode, when you have a RAW file selected there is a toggle in the lower tool bar to toggle between the embedded preview, and the decoded RAW image. You should NOT expect the two to be identical.

Yes, agree, I'm aware of that.
This toggle only appears for untouched RAW files. That toggle will disappear as soon as an image has been developed by ACDSee, For a developed image, the thumbnail and the image in view mode are based on the jpg stored extra in the hidden folder called "[Developed]"
So in my example because it was developed already on computer "A", the toggle is not there on computer "A" or "B".

So, my problem is: on computer "B", first I see the good hidden jpg which was created originally by ACDSee U2021 on computer "A",
And then, when I press "develop", it uses my RAW file and will decode anew with the .xmp that was also made by U2021 on computer "A" and the result on Computer "B" looks like it has a green/blue cast

Any clues?

[Greyfox]

Trevor

Got them, thanks.

OK, starting from the basics, The following was done in ACDSee Ultimate 2023, with the Color Profile settings as per your post. My Monitors profile are sRGB

The first two screen shots are taken in view mode with the ex camera NEF raw without any xmp file


As expected the RAW decoded image is slightly "flatter" than the embedded preview. (The embedded preview has generally had the same post processing done in camera as would be the case for an ex camera JPG. With RAW, that post processing is expected to be done by the user). But importantly, to me there is no green cast.

OK, then next screen shot shows the same NEF RAW file opened in Develop Mode, again with no xmp sidecar file in play. The shot was taken after ACDSee's basic development had completed (so on the default program settings).



Again, I don't see any green cast.

For the next screen shot however, the NEF RAW file was paired with the xmp side car file you provided.



So now the green cast is quite evident. You can see from the blue indicators that you have applied changes to the General Settings, White balance, Light EQ, Color EQ, and to the Post Crop Vignette, as well as some brush work.

In the next screen shot, I have simply turned off the White balance adjustment (so basically restoring it to default).The rest of the adjustments are as you had them.



And the green cast is gone. So it appears in this case to have been introduced by your White Balance adjustment.

Hope that helps, but if that White Balance adjustment looked OK on the PC when the adjustment was made, but is clearly not right on both your other PC, and my PC here, it does raise the question of as to what is wrong with the color management on the original PC..




.

[Trevor]

Many thanks Greyfox for that detailed analysis, I added some notes in blue between your text below

Trevor

Got them, thanks.

OK, starting from the basics, The following was done in ACDSee Ultimate 2023, with the Color Profile settings as per your post. My Monitors profile are sRGB

The first two screen shots are taken in view mode with the ex camera NEF raw without any xmp file


As expected the RAW decoded image is slightly "flatter" than the embedded preview. (The embedded preview has generally had the same post processing done in camera as would be the case for an ex camera JPG. With RAW, that post processing is expected to be done by the user). But importantly, to me there is no green cast.

OK, then next screen shot shows the same NEF RAW file opened in Develop Mode, again with no xmp sidecar file in play. The shot was taken after ACDSee's basic development had completed (so on the default program settings).



Again, I don't see any green cast.

This is good, it confirms that my original RAW file (NEF) was OK from the camera. so far so good .

For the next screen shot however, the NEF RAW file was paired with the xmp side car file you provided.



So now the green cast is quite evident. You can see from the blue indicators that you have applied changes to the General Settings, White balance, Light EQ, Color EQ, and to the Post Crop Vignette, as well as some brush work.

Thanks, that is very similar to what I see on my computer "B" and confirms the problem I have , now we need the cause... ;-)

In the next screen shot, I have simply turned off the White balance adjustment (so basically restoring it to default).The rest of the adjustments are as you had them.



And the green cast is gone. So it appears in this case to have been introduced by your White Balance adjustment.

yes, probably the problem is with the white balance. I might have also made WB corrections in the brushed areas too, I'll have to check that,...
Generally, when processing my RAW's I will make some adjustements to the temperature and tint sliders (as needed/desired)

... But what caused this WB problem ?,...

Hope that helps, but if that White Balance adjustment looked OK on the PC when the adjustment was made, but is clearly not right on both your other PC, and my PC here, it does raise the question of as to what is wrong with the color management on the original PC..



.

Yes it does help, thanks :-)
Did you also take a look at the "preview" jpg file, that I put in the "hidden" folder called [Developed]?
this file was created with computer "A" and looks OK to me on both computer "A" & "B"
I'm thinking, that if I had a problem with the colour profile on the old computer "A", then this preview would also show some colour cast on the new computer "B", but it doesn't.
So I assume that the colour profile used on computer A is also more or less OK.
By the way, I calibrated both computers with the "Color Monkey "

Is there anything (in the colour part of the .xmp file) that needs to be converted when going from U2021 to U2023?
I just copied the .xmp file (using ACDSee), I didn't "convert" it in any way. Should I have done so?



Many thanks again
Trevor

[Trevor]

I have just noticed one strange difference between :
-U2021 on computer “A” and
-U2023 on computer “B”
But I don’t know quite how it is relevant?

I put only the untouched NEF file on a separate folder (without the .xmp file and without the hidden jpg preview file) and clicked on DEVELOP
then looked at the “As shot” values for WB
I assumed these would be the same for both BUT I see completely different values ??

U2021 on computer “A” it shows “as shot” = 4706k , Temp = -4, Tint = -7
U2023 on computer “B” it shows “as shot” = 5144K, Temp = 11, Tint = 20


On computer “A” I had originally developed and manually modified the WB settings to a warmer setting of Temp =19, Tint = -9
And the image was as I desired.

However, with the same settings on U2023 the settings 19,-9 give that green cast :-(

Mmmmh, So why could the “As shot” values differ so much?

[Greyfox]

..
U2021 on computer “A” it shows “as shot” = 4706k , Temp = -4, Tint = -7
U2023 on computer “B” it shows “as shot” = 5144K, Temp = 11, Tint = 20

Good pickup Trevor.

As it happens, I have U2021, U2022 and U2023 installed on the same PC. All using the same color management settings, same hardware, same monitor profiles.
I can confirm that opening the original DSC_6642.NEF Raw image (no xmp file) in Develop mode shows

U2021 - White Balance As shot” = 4706k , Temp = -4, Tint = -7
U2022 - White Balance As shot” = 4706k , Temp = -4, Tint = -7
U2023 - White Balance As shot" = 5144K, Temp = 11, Tint = 20

Here is the comparison screen shots when each version "develops" the image using its "As shot values"


So yes the RAW development process in ACDSee Ultimate 2023 does give different "As Shot" values both white balance temperature and tint to those in U2021 and U2022.

As the above screen shots show, the "developed" result is much the same, providing the values relate to the version. The issue is that the actual values used are what is recorded in the xmp file, so for your Nikon Z7 (V3.4 software) raw images at least, that is not going to be correct if a U23021 or U2022 xmp file is used in U2023

I've just run the same check on RAW images from my Sony SLT-A57 camera and the "as shot" values are the same across.the three versions
I've also run the same check on RAW images from a Nikon D7100 and again the "as shot" values are the same across.the three versions

Clearly this is something ACDSee need to address. Can you raise the issue with Customer Support at https://www.acdsee.com/en/support/form/

Edit: I've just run a check using a couple of NEF raw images from a Nikon Z7 MKII (V1.0 software).
The "As Shot" values are identical in U2021, U2022 and U2023

[Trevor]

Good pickup Trevor.

As it happens, I have U2021, U2022 and U2023 installed on the same PC. All using the same color management settings, same hardware, same monitor profiles.
I can confirm that opening the original DSC_6642.NEF Raw image (no xmp file) in Develop mode shows

U2021 - White Balance As shot” = 4706k , Temp = -4, Tint = -7
U2022 - White Balance As shot” = 4706k , Temp = -4, Tint = -7
U2023 - White Balance As shot" = 5144K, Temp = 11, Tint = 20

Here is the comparison screen shots when each version "develops" the image using its "As shot values"


So yes the RAW development process in ACDSee Ultimate 2023 does give different "As Shot" values both white balance temperature and tint to those in U2021 and U2022.

As the above screen shots show, the "developed" result is much the same, providing the values relate to the version. The issue is that the actual values used are what is recorded in the xmp file, so for your Nikon Z7 (V3.4 software) raw images at least, that is not going to be correct if a U23021 or U2022 xmp file is used in U2023

I've just run the same check on RAW images from my Sony SLT-A57 camera and the "as shot" values are the same across.the three versions
I've also run the same check on RAW images from a Nikon D7100 and again the "as shot" values are the same across.the three versions

Clearly this is something ACDSee need to address. Can you raise the issue with Customer Support at https://www.acdsee.com/en/support/form/

Edit: I've just run a check using a couple of NEF raw images from a Nikon Z7 MKII (V1.0 software).
The "As Shot" values are identical in U2021, U2022 and U2023

Thanks for the confirmation.
I started a customer service case.

I wonder if anyone can try the same comparison with another Z7 RAW NEF file, just to eliminate the possibilty that something is wrong with my particular files
It would also be good to know what the "as shot" WB values are for a Z7 NEF file in copmparison between any of these
... U2021, U2022, U2023 & especially U2024 !

Thanks!

[Greyfox]

..I wonder if anyone can try the same comparison with another Z7 RAW NEF file, just to eliminate the possibilty that something is wrong with my particular files
It would also be good to know what the "as shot" WB values are for a Z7 NEF file in copmparison between any of these
... U2021, U2022, U2023 & especially U2024

I've downloaded some sample Z7 (V1.00 software) Raw Images which all exhibited the same problem.
For example in U2021 and U2022 one image has "As Shot" Temp= 8, Tint = -8, where as in U2023 the "As Shot" values show as Temp +21, Tint +21.
The White Balance mode from Exif metadata was Auto1 which is the default for that camera.;

It appears only to be an issue for the Z7 model, and I've now seen it with both V1.0 and V3.3 camera software. It doesn't appear to occur with the Z7 MKII images.

I don't know about U2024 as it hasn't been released yet. But worth asking Customer Support.

[Trevor]

I've downloaded some sample Z7 (V1.00 software) Raw Images which all exhibited the same problem.
For example in U2021 and U2022 one image has "As Shot" Temp= 8, Tint = -8, where as in U2023 the "As Shot" values show as Temp +21, Tint +21.
The White Balance mode from Exif metadata was Auto1 which is the default for that camera.;

It appears only to be an issue for the Z7 model, and I've now seen it with both V1.0 and V3.3 camera software. It doesn't appear to occur with the Z7 MKII images.

I don't know about U2024 as it hasn't been released yet. But worth asking Customer Support.

Thanks once again!
I pointed customer support to this thread so they should see the concern about U2024 as well

This issue makes upgrading to a new ultimate version troublesome, for a Z7 owner (and any other camera that might be affetced) because every (previously developed) image would get the green/blue cast "baked in" as soon as it is opened in the U2023 developer, ...because If an image is opened in the U2023 developer it is usually saved automatically with these changes !

Maybe this also affects images that were processed in the EDIT mode (I only tested develop mode)?

[Greyfox]

..Maybe this also affects images that were processed in the EDIT mode (I only tested develop mode)?

(1) My tests show that issue only occurs if the image has been processed in Develop mode in U2021 or U2022 and the changes saved to an xmp file, and then ONLY if the Whte Balance settings were manually changed in the process..

(2) In regards to Edit mode. Settings done in edit Mode filters do not get written into sidecar xmp files, and only affect what ever format image that is saved or exported from Edit mode. So changes made in Edit Mode in U2021 or U2022 are not carried into U2023 via an xmp file.

That said, if the White balance has been manually altered in U2021 or U2022 Develop mode and saved into the xmp file, then in U2023 if the image is opened in Edit mode the cast will show.

The sliders in both the Edit Mode White balance Filter, and the Adjustment Layer White balance control open at the default setting Temperature = 0, Tint =0, Strength=100 regardless of what is in the xmp file.

[Trevor]

(1) My tests show that issue only occurs if the image has been processed in Develop mode in U2021 or U2022 and the changes saved to an xmp file, and then ONLY if the Whte Balance settings were manually changed in the process..

(2) In regards to Edit mode. Settings done in edit Mode filters do not get written into sidecar xmp files, and only affect what ever format image that is saved or exported from Edit mode. So changes made in Edit Mode in U2021 or U2022 are not carried into U2023 via an xmp file.

That said, if the White balance has been manually altered in U2021 or U2022 Develop mode and saved into the xmp file, then in U2023 if the image is opened in Edit mode the cast will show.

The sliders in both the Edit Mode White balance Filter, and the Adjustment Layer White balance control open at the default setting Temperature = 0, Tint =0, Strength=100 regardless of what is in the xmp file.

Thanks for that detailed analysis!

I got a confirmation from James W. at ACDSee support, that this problem exists as described here.
He has passed it on to the quality department for the appropriate remedy.

This problem has affected nearly all of my many images. I nearly always use Develop mode and I'll adjust the WB manually in practically every shot. It's one of the advantages of using RAW that I can set the WB after the shot. Sometimes adjusting locally as well as globally with brushed areas or gradients.
I'd need a way to quickly and automatically adjust each image.
Let's see what ACDSee come up with,...

[Greyfox]

...This problem has affected nearly all of my many images. I nearly always use Develop mode and I'll adjust the WB manually in practically every shot. It's one of the advantages of using RAW that I can set the WB after the shot. Sometimes adjusting locally as well as globally with brushed areas or gradients.

I can understand your concern.
For what it is worth, it also appears to me that the problem has also carried over into the new Ultimate 2024 version, so even an upgrade to that isn't going to help.

Just on the offchance I tried changing the ID_DCRaw.api in U2024 to the one from U2021, but the cast still occurs.

One would hope ACDSee gets on top of this very quickly.