Discussion:
[darktable-user] A better starting-point for Nikon RAW-files
Vidar Hoel
2016-07-26 16:44:52 UTC
Permalink
Hi,

some time ago I noticed that the JPGs from my Nikon were better then
what I could achieve from processing the RAW-file (NEF) in Darktable.
After some digging on the Internet, I figured out I could use an
alternative to the default base-curves in Darktable - and that made me
get a better starting-point when processing my Nikon RAW-files (and I
was able to achieve a better image than the JPGs from my camera!)

Today I finished a step-by-step video of what I did, so anyone can see,
learn and perhaps do it themselves.

Some notes before viewing:
- I am not really sure behind the technology behind this. I am not sure
if a Nikon RAW-file has a profile embedded, but it seems so.
- I am not sure if there is an other way to achieve this is Darktable?
- As explained in the video, you will need Windows, or a friends laptop
with Windows, once, to make this work.
- My native language is not English :-)



Enjoy :-)

Best regards,
Vidar Hoel
Colin Adams
2016-07-26 17:09:37 UTC
Permalink
That looks very interesting.
But I have a question. The image showed much less noise. But can you also
use profiled denoise with this approach, or is there an inherent conflict?
Post by Vidar Hoel
Hi,
some time ago I noticed that the JPGs from my Nikon were better then
what I could achieve from processing the RAW-file (NEF) in Darktable.
After some digging on the Internet, I figured out I could use an
alternative to the default base-curves in Darktable - and that made me
get a better starting-point when processing my Nikon RAW-files (and I
was able to achieve a better image than the JPGs from my camera!)
Today I finished a step-by-step video of what I did, so anyone can see,
learn and perhaps do it themselves.
- I am not really sure behind the technology behind this. I am not sure
if a Nikon RAW-file has a profile embedded, but it seems so.
- I am not sure if there is an other way to achieve this is Darktable?
- As explained in the video, you will need Windows, or a friends laptop
with Windows, once, to make this work.
- My native language is not English :-)
http://youtu.be/O1nS3URVSYA
Enjoy :-)
Best regards,
Vidar Hoel
____________________________________________________________________________
darktable user mailing list
to unsubscribe send a mail to
Vidar Hoel
2016-07-26 18:35:57 UTC
Permalink
Hi,

No, I use the "denoise (profiled)" as before, and it works the same way,
regardless of what modules (base curve or input color profile) I have
applied before. It sure seems like it, anyway.

Regards,
Vidar Hoel
Post by Colin Adams
That looks very interesting.
But I have a question. The image showed much less noise. But can you
also use profiled denoise with this approach, or is there an inherent
conflict?
Hi,
some time ago I noticed that the JPGs from my Nikon were better then
what I could achieve from processing the RAW-file (NEF) in Darktable.
After some digging on the Internet, I figured out I could use an
alternative to the default base-curves in Darktable - and that made me
get a better starting-point when processing my Nikon RAW-files (and I
was able to achieve a better image than the JPGs from my camera!)
Today I finished a step-by-step video of what I did, so anyone can see,
learn and perhaps do it themselves.
- I am not really sure behind the technology behind this. I am not sure
if a Nikon RAW-file has a profile embedded, but it seems so.
- I am not sure if there is an other way to achieve this is Darktable?
- As explained in the video, you will need Windows, or a friends laptop
with Windows, once, to make this work.
- My native language is not English :-)
http://youtu.be/O1nS3URVSYA
Enjoy :-)
Best regards,
Vidar Hoel
____________________________________________________________________________
darktable user mailing list
to unsubscribe send a mail to
Roman Lebedev
2016-07-26 18:40:36 UTC
Permalink
Hi.

Base curve and input color profile are located *after* the denoise
(profiled) in pipe,
so they can not interfere with denoise (profiled) in any way...

https://www.darktable.org/usermanual/ch03s02.html.php

Roman.
Post by Vidar Hoel
Hi,
No, I use the "denoise (profiled)" as before, and it works the same way,
regardless of what modules (base curve or input color profile) I have
applied before. It sure seems like it, anyway.
Regards,
Vidar Hoel
Post by Colin Adams
That looks very interesting.
But I have a question. The image showed much less noise. But can you
also use profiled denoise with this approach, or is there an inherent
conflict?
Hi,
some time ago I noticed that the JPGs from my Nikon were better then
what I could achieve from processing the RAW-file (NEF) in Darktable.
After some digging on the Internet, I figured out I could use an
alternative to the default base-curves in Darktable - and that made me
get a better starting-point when processing my Nikon RAW-files (and I
was able to achieve a better image than the JPGs from my camera!)
Today I finished a step-by-step video of what I did, so anyone can see,
learn and perhaps do it themselves.
- I am not really sure behind the technology behind this. I am not sure
if a Nikon RAW-file has a profile embedded, but it seems so.
- I am not sure if there is an other way to achieve this is Darktable?
- As explained in the video, you will need Windows, or a friends laptop
with Windows, once, to make this work.
- My native language is not English :-)
http://youtu.be/O1nS3URVSYA
Enjoy :-)
Best regards,
Vidar Hoel
____________________________________________________________________________
darktable user mailing list
to unsubscribe send a mail to
____________________________________________________________________________
darktable user mailing list
Jean-Luc CECCOLI
2016-07-26 18:16:01 UTC
Permalink
Hello,

 

Just one thing that puzzles me : Nikon ViewNX2 generates a profile for each picture it processes, so each
profile is based upon the parameters of this picture.

That would mean that, for each picture, one would have to use the corresponding imported profile, else one
might get even worse results than using the DT dedicated basecurve, isn't it ?

That said, I think that that is the way profiles work...

 

Regards,

 

J.-Luc

 

 
Message du 26/07/16 18:45
De : "Vidar Hoel"
Objet : [darktable-user] A better starting-point for Nikon RAW-files
Hi,
some time ago I noticed that the JPGs from my Nikon were better then
what I could achieve from processing the RAW-file (NEF) in Darktable.
After some digging on the Internet, I figured out I could use an
alternative to the default base-curves in Darktable - and that made me
get a better starting-point when processing my Nikon RAW-files (and I
was able to achieve a better image than the JPGs from my camera!)
Today I finished a step-by-step video of what I did, so anyone can see,
learn and perhaps do it themselves.
- I am not really sure behind the technology behind this. I am not sure
if a Nikon RAW-file has a profile embedded, but it seems so.
- I am not sure if there is an other way to achieve this is Darktable?
- As explained in the video, you will need Windows, or a friends laptop
with Windows, once, to make this work.
- My native language is not English :-)
http://youtu.be/O1nS3URVSYA
Enjoy :-)
Best regards,
Vidar Hoel
____________________________________________________________________________
darktable user mailing list
Vidar Hoel
2016-07-26 18:41:35 UTC
Permalink
Post by Jean-Luc CECCOLI
Hello,
Hi!
Post by Jean-Luc CECCOLI
Just one thing that puzzles me : Nikon ViewNX2 generates a profile for
each picture it processes, so each
profile is based upon the parameters of this picture.
Almost. It generates a different profile for each camera setting (it
seems like). So of mine 98 pictures, I got 10 different profiles.
Post by Jean-Luc CECCOLI
That would mean that, for each picture, one would have to use the
corresponding imported profile, else one
might get even worse results than using the DT dedicated basecurve, isn't it ?
After applying all 10 different profiles to the same image, I discovered
it was only minor differences, and so small I can neglect them. I have
never seen them make a worse result than the dedicated base curves. But
this is a subjective observation!

Regards,
Vidar Hoel
Post by Jean-Luc CECCOLI
That said, I think that that is the way profiles work...
Regards,
J.-Luc
Message du 26/07/16 18:45
Objet : [darktable-user] A better starting-point for Nikon RAW-files
Hi,
some time ago I noticed that the JPGs from my Nikon were better then
what I could achieve from processing the RAW-file (NEF) in Darktable.
After some digging on the Internet, I figured out I could use an
alternative to the default base-curves in Darktable - and that
made me
get a better starting-point when processing my Nikon RAW-files (and I
was able to achieve a better image than the JPGs from my camera!)
Today I finished a step-by-step video of what I did, so anyone can
see,
learn and perhaps do it themselves.
- I am not really sure behind the technology behind this. I am not
sure
if a Nikon RAW-file has a profile embedded, but it seems so.
- I am not sure if there is an other way to achieve this is Darktable?
- As explained in the video, you will need Windows, or a friends
laptop
with Windows, once, to make this work.
- My native language is not English :-)
http://youtu.be/O1nS3URVSYA
Enjoy :-)
Best regards,
Vidar Hoel
____________________________________________________________________________
darktable user mailing list
to unsubscribe send a mail to
I. Ivanov
2016-07-26 23:50:44 UTC
Permalink
Is this behavior "Nikon specific"? Does anybody know if Canon behaves
the same?

Regards,

B
Post by Vidar Hoel
Post by Jean-Luc CECCOLI
Hello,
Hi!
Post by Jean-Luc CECCOLI
Just one thing that puzzles me : Nikon ViewNX2 generates a profile for
each picture it processes, so each
profile is based upon the parameters of this picture.
Almost. It generates a different profile for each camera setting (it
seems like). So of mine 98 pictures, I got 10 different profiles.
Post by Jean-Luc CECCOLI
That would mean that, for each picture, one would have to use the
corresponding imported profile, else one
might get even worse results than using the DT dedicated basecurve, isn't it ?
After applying all 10 different profiles to the same image, I
discovered it was only minor differences, and so small I can neglect
them. I have never seen them make a worse result than the dedicated
base curves. But this is a subjective observation!
Regards,
Vidar Hoel
Post by Jean-Luc CECCOLI
That said, I think that that is the way profiles work...
Regards,
J.-Luc
Message du 26/07/16 18:45
Objet : [darktable-user] A better starting-point for Nikon
RAW-files
Hi,
some time ago I noticed that the JPGs from my Nikon were better
then
what I could achieve from processing the RAW-file (NEF) in
Darktable.
After some digging on the Internet, I figured out I could use an
alternative to the default base-curves in Darktable - and that
made me
get a better starting-point when processing my Nikon RAW-files
(and I
was able to achieve a better image than the JPGs from my camera!)
Today I finished a step-by-step video of what I did, so anyone can
see,
learn and perhaps do it themselves.
- I am not really sure behind the technology behind this. I am not
sure
if a Nikon RAW-file has a profile embedded, but it seems so.
- I am not sure if there is an other way to achieve this is
Darktable?
- As explained in the video, you will need Windows, or a friends
laptop
with Windows, once, to make this work.
- My native language is not English :-)
http://youtu.be/O1nS3URVSYA
Enjoy :-)
Best regards,
Vidar Hoel
____________________________________________________________________________
darktable user mailing list
to unsubscribe send a mail to
____________________________________________________________________________
darktable user mailing list
to unsubscribe send a mail to
Jean-Luc CECCOLI
2016-07-27 14:39:06 UTC
Permalink
Hello,

 

Very interesting.

Thanks for sharing.

 

Rgds,

 

J.-Luc

 

 
Message du 26/07/16 20:41
De : "Vidar Hoel"
Objet : Re: [darktable-user] A better starting-point for Nikon RAW-files
Post by Jean-Luc CECCOLI
Hello,
Hi!
Post by Jean-Luc CECCOLI
Just one thing that puzzles me : Nikon ViewNX2 generates a profile for
each picture it processes, so each
profile is based upon the parameters of this picture.
Almost. It generates a different profile for each camera setting (it
seems like). So of mine 98 pictures, I got 10 different profiles.
Post by Jean-Luc CECCOLI
That would mean that, for each picture, one would have to use the
corresponding imported profile, else one
might get even worse results than using the DT dedicated basecurve, isn't it ?
After applying all 10 different profiles to the same image, I discovered
it was only minor differences, and so small I can neglect them. I have
never seen them make a worse result than the dedicated base curves. But
this is a subjective observation!
Regards,
Vidar Hoel
Post by Jean-Luc CECCOLI
That said, I think that that is the way profiles work...
Regards,
J.-Luc
Message du 26/07/16 18:45
De : "Vidar Hoel"
Objet : [darktable-user] A better starting-point for Nikon RAW-files
Hi,
some time ago I noticed that the JPGs from my Nikon were better then
what I could achieve from processing the RAW-file (NEF) in Darktable.
After some digging on the Internet, I figured out I could use an
alternative to the default base-curves in Darktable - and that
made me
get a better starting-point when processing my Nikon RAW-files (and I
was able to achieve a better image than the JPGs from my camera!)
Today I finished a step-by-step video of what I did, so anyone can
see,
learn and perhaps do it themselves.
- I am not really sure behind the technology behind this. I am not
sure
if a Nikon RAW-file has a profile embedded, but it seems so.
- I am not sure if there is an other way to achieve this is Darktable?
- As explained in the video, you will need Windows, or a friends
laptop
with Windows, once, to make this work.
- My native language is not English :-)
http://youtu.be/O1nS3URVSYA
Enjoy :-)
Best regards,
Vidar Hoel
____________________________________________________________________________
darktable user mailing list
to unsubscribe send a mail to
____________________________________________________________________________
darktable user mailing list
Colin Adams
2016-07-31 11:18:10 UTC
Permalink
So I tried this (I have a Nikon D810).

When selecting any of the generated icc profiles except the last one, I get
a message:
"unsupported input profile has been replaced by linear Rec709 RGB!"
So I selected the last one, and set unbreak input profile to 0.0 as you
instruct.
The resulting image indeed looks to have slightly better quality, but it
has a green cast.
This is with darktable 2.0.5-1 rather than 2.0.4. I'm doubtful if that
should be making the difference though.
Post by Jean-Luc CECCOLI
Hello,
Very interesting.
Thanks for sharing.
Rgds,
J.-Luc
Message du 26/07/16 20:41
Objet : Re: [darktable-user] A better starting-point for Nikon RAW-files
Post by Jean-Luc CECCOLI
Hello,
Hi!
Post by Jean-Luc CECCOLI
Just one thing that puzzles me : Nikon ViewNX2 generates a profile for
each picture it processes, so each
profile is based upon the parameters of this picture.
Almost. It generates a different profile for each camera setting (it
seems like). So of mine 98 pictures, I got 10 different profiles.
Post by Jean-Luc CECCOLI
That would mean that, for each picture, one would have to use the
corresponding imported profile, else one
might get even worse results than using the DT dedicated basecurve, isn't it ?
After applying all 10 different profiles to the same image, I discovered
it was only minor differences, and so small I can neglect them. I have
never seen them make a worse result than the dedicated base curves. But
this is a subjective observation!
Regards,
Vidar Hoel
Post by Jean-Luc CECCOLI
That said, I think that that is the way profiles work...
Regards,
J.-Luc
Message du 26/07/16 18:45
Objet : [darktable-user] A better starting-point for Nikon RAW-files
Hi,
some time ago I noticed that the JPGs from my Nikon were better then
what I could achieve from processing the RAW-file (NEF) in Darktable.
After some digging on the Internet, I figured out I could use an
alternative to the default base-curves in Darktable - and that
made me
get a better starting-point when processing my Nikon RAW-files (and I
was able to achieve a better image than the JPGs from my camera!)
Today I finished a step-by-step video of what I did, so anyone can
see,
learn and perhaps do it themselves.
- I am not really sure behind the technology behind this. I am not
sure
if a Nikon RAW-file has a profile embedded, but it seems so.
- I am not sure if there is an other way to achieve this is
Darktable?
Post by Jean-Luc CECCOLI
- As explained in the video, you will need Windows, or a friends
laptop
with Windows, once, to make this work.
- My native language is not English :-)
http://youtu.be/O1nS3URVSYA
Enjoy :-)
Best regards,
Vidar Hoel
____________________________________________________________________________
Post by Jean-Luc CECCOLI
darktable user mailing list
to unsubscribe send a mail to
____________________________________________________________________________
darktable user mailing list
to unsubscribe send a mail to
____________________________________________________________________________
darktable user mailing list to unsubscribe send a mail to
Edgardo Hoszowski
2016-07-31 17:13:42 UTC
Permalink
Same here with a D5100, with a D7100 all went fine, I have 2 or 3 different
profiles that give different colors cast, but so far there's always one
closer to the camera jpg.
Using the D7100 profiles on a D5100 raw seems ok too, much closer to the
camera jpg.
Post by Colin Adams
So I tried this (I have a Nikon D810).
When selecting any of the generated icc profiles except the last one, I
"unsupported input profile has been replaced by linear Rec709 RGB!"
So I selected the last one, and set unbreak input profile to 0.0 as you
instruct.
The resulting image indeed looks to have slightly better quality, but it
has a green cast.
This is with darktable 2.0.5-1 rather than 2.0.4. I'm doubtful if that
should be making the difference though.
Post by Jean-Luc CECCOLI
Hello,
Very interesting.
Thanks for sharing.
Rgds,
J.-Luc
Message du 26/07/16 20:41
Objet : Re: [darktable-user] A better starting-point for Nikon RAW-files
Post by Jean-Luc CECCOLI
Hello,
Hi!
Post by Jean-Luc CECCOLI
Just one thing that puzzles me : Nikon ViewNX2 generates a profile for
each picture it processes, so each
profile is based upon the parameters of this picture.
Almost. It generates a different profile for each camera setting (it
seems like). So of mine 98 pictures, I got 10 different profiles.
Post by Jean-Luc CECCOLI
That would mean that, for each picture, one would have to use the
corresponding imported profile, else one
might get even worse results than using the DT dedicated basecurve, isn't it ?
After applying all 10 different profiles to the same image, I
discovered
it was only minor differences, and so small I can neglect them. I have
never seen them make a worse result than the dedicated base curves. But
this is a subjective observation!
Regards,
Vidar Hoel
Post by Jean-Luc CECCOLI
That said, I think that that is the way profiles work...
Regards,
J.-Luc
Message du 26/07/16 18:45
Objet : [darktable-user] A better starting-point for Nikon RAW-files
Hi,
some time ago I noticed that the JPGs from my Nikon were better then
what I could achieve from processing the RAW-file (NEF) in
Darktable.
Post by Jean-Luc CECCOLI
After some digging on the Internet, I figured out I could use an
alternative to the default base-curves in Darktable - and that
made me
get a better starting-point when processing my Nikon RAW-files (and
I
Post by Jean-Luc CECCOLI
was able to achieve a better image than the JPGs from my camera!)
Today I finished a step-by-step video of what I did, so anyone can
see,
learn and perhaps do it themselves.
- I am not really sure behind the technology behind this. I am not
sure
if a Nikon RAW-file has a profile embedded, but it seems so.
- I am not sure if there is an other way to achieve this is
Darktable?
Post by Jean-Luc CECCOLI
- As explained in the video, you will need Windows, or a friends
laptop
with Windows, once, to make this work.
- My native language is not English :-)
http://youtu.be/O1nS3URVSYA
Enjoy :-)
Best regards,
Vidar Hoel
____________________________________________________________________________
Post by Jean-Luc CECCOLI
darktable user mailing list
to unsubscribe send a mail to
____________________________________________________________________________
darktable user mailing list
to unsubscribe send a mail to
____________________________________________________________________________
darktable user mailing list to unsubscribe send a mail to
____________________________________________________________________________
darktable user mailing list to unsubscribe send a mail to
Edgardo Hoszowski
2016-08-01 18:10:22 UTC
Permalink
Sure, see attached
Hi Edgardo,
I wrote you related about 7100 icc profiles. I have a mac updated to
capitan and I couldn't install view NX 2 to generate the profiles.
Could you send me your 7100 generated profiles?
Thanks!
Post by Edgardo Hoszowski
Same here with a D5100, with a D7100 all went fine, I have 2 or 3
different profiles that give different colors cast, but so far there's
always one closer to the camera jpg.
Using the D7100 profiles on a D5100 raw seems ok too, much closer to the
camera jpg.
Post by Colin Adams
So I tried this (I have a Nikon D810).
When selecting any of the generated icc profiles except the last one, I
"unsupported input profile has been replaced by linear Rec709 RGB!"
So I selected the last one, and set unbreak input profile to 0.0 as you
instruct.
The resulting image indeed looks to have slightly better quality, but it
has a green cast.
This is with darktable 2.0.5-1 rather than 2.0.4. I'm doubtful if that
should be making the difference though.
On Wed, 27 Jul 2016 at 15:39 Jean-Luc CECCOLI <
Post by Jean-Luc CECCOLI
Hello,
Very interesting.
Thanks for sharing.
Rgds,
J.-Luc
Message du 26/07/16 20:41
Objet : Re: [darktable-user] A better starting-point for Nikon
RAW-files
Post by Jean-Luc CECCOLI
Hello,
Hi!
Post by Jean-Luc CECCOLI
Just one thing that puzzles me : Nikon ViewNX2 generates a profile
for
Post by Jean-Luc CECCOLI
each picture it processes, so each
profile is based upon the parameters of this picture.
Almost. It generates a different profile for each camera setting (it
seems like). So of mine 98 pictures, I got 10 different profiles.
Post by Jean-Luc CECCOLI
That would mean that, for each picture, one would have to use the
corresponding imported profile, else one
might get even worse results than using the DT dedicated basecurve,
isn't it ?
After applying all 10 different profiles to the same image, I
discovered
it was only minor differences, and so small I can neglect them. I
have
never seen them make a worse result than the dedicated base curves.
But
this is a subjective observation!
Regards,
Vidar Hoel
Post by Jean-Luc CECCOLI
That said, I think that that is the way profiles work...
Regards,
J.-Luc
Message du 26/07/16 18:45
Objet : [darktable-user] A better starting-point for Nikon
RAW-files
Post by Jean-Luc CECCOLI
Hi,
some time ago I noticed that the JPGs from my Nikon were better
then
Post by Jean-Luc CECCOLI
what I could achieve from processing the RAW-file (NEF) in
Darktable.
Post by Jean-Luc CECCOLI
After some digging on the Internet, I figured out I could use an
alternative to the default base-curves in Darktable - and that
made me
get a better starting-point when processing my Nikon RAW-files
(and I
Post by Jean-Luc CECCOLI
was able to achieve a better image than the JPGs from my camera!)
Today I finished a step-by-step video of what I did, so anyone can
see,
learn and perhaps do it themselves.
- I am not really sure behind the technology behind this. I am not
sure
if a Nikon RAW-file has a profile embedded, but it seems so.
- I am not sure if there is an other way to achieve this is
Darktable?
Post by Jean-Luc CECCOLI
- As explained in the video, you will need Windows, or a friends
laptop
with Windows, once, to make this work.
- My native language is not English :-)
http://youtu.be/O1nS3URVSYA
Enjoy :-)
Best regards,
Vidar Hoel
____________________________________________________________________________
Post by Jean-Luc CECCOLI
darktable user mailing list
to unsubscribe send a mail to
____________________________________________________________________________
darktable user mailing list
to unsubscribe send a mail to
____________________________________________________________________________
darktable user mailing list to unsubscribe send a mail to
____________________________________________________________________________
darktable user mailing list to unsubscribe send a mail to
____________________________________________________________________________
darktable user mailing list to unsubscribe send a mail to
Terry Duell
2016-08-01 22:20:12 UTC
Permalink
Hello Edgardo,

On Tue, 02 Aug 2016 04:10:22 +1000, Edgardo Hoszowski
Post by Edgardo Hoszowski
Sure, see attached
The 'attached' that arrived here was 6.5 MB.
I'm not sure what others think, but in view it is bad etiquette to attach
large files to posts to mailing lists, as everyone ends up getting a copy.
In my view the 'correct' approach is to provide a link to the file on
Dropbox or similar.

Cheers,
--
Regards,
Terry Duell
Edgardo Hoszowski
2016-08-01 22:26:42 UTC
Permalink
Sorry about that, I didn't notice the size of the file, I'll have it in
mind the next time.
Post by Terry Duell
Hello Edgardo,
On Tue, 02 Aug 2016 04:10:22 +1000, Edgardo Hoszowski <
Sure, see attached
The 'attached' that arrived here was 6.5 MB.
I'm not sure what others think, but in view it is bad etiquette to attach
large files to posts to mailing lists, as everyone ends up getting a copy.
In my view the 'correct' approach is to provide a link to the file on
Dropbox or similar.
Cheers,
--
Regards,
Terry Duell
____________________________________________________________________________
darktable user mailing list
to unsubscribe send a mail to
Saint Germain
2016-08-02 00:59:01 UTC
Permalink
Hello,

I've tried your profiles with my D7200, and I can confirm that the
results are much closer to the camera jpeg.
I can see a slight blue/green cast and the colors are less
saturated, but it's a much better starting point than the base curve.

I'll have to generate my own profiles to be sure (perhaps D7200 is a
bit different from D7100 ?).

Thanks to Edgardo and Vidar !

On Mon, 1 Aug 2016 15:10:22 -0300, Edgardo Hoszowski
Post by Edgardo Hoszowski
Sure, see attached
Hi Edgardo,
I wrote you related about 7100 icc profiles. I have a mac updated
to capitan and I couldn't install view NX 2 to generate the
profiles.
Could you send me your 7100 generated profiles?
Thanks!
2016-07-31 19:13 GMT+02:00 Edgardo Hoszowski
Post by Edgardo Hoszowski
Same here with a D5100, with a D7100 all went fine, I have 2 or 3
different profiles that give different colors cast, but so far
there's always one closer to the camera jpg.
Using the D7100 profiles on a D5100 raw seems ok too, much closer
to the camera jpg.
Post by Colin Adams
So I tried this (I have a Nikon D810).
When selecting any of the generated icc profiles except the last
"unsupported input profile has been replaced by linear Rec709
RGB!" So I selected the last one, and set unbreak input profile
to 0.0 as you instruct.
The resulting image indeed looks to have slightly better quality,
but it has a green cast.
This is with darktable 2.0.5-1 rather than 2.0.4. I'm doubtful if
that should be making the difference though.
On Wed, 27 Jul 2016 at 15:39 Jean-Luc CECCOLI <
Post by Jean-Luc CECCOLI
Hello,
Very interesting.
Thanks for sharing.
Rgds,
J.-Luc
Message du 26/07/16 20:41
Objet : Re: [darktable-user] A better starting-point for Nikon
RAW-files
Post by Jean-Luc CECCOLI
Hello,
Hi!
Post by Jean-Luc CECCOLI
Just one thing that puzzles me : Nikon ViewNX2 generates a profile
for
Post by Jean-Luc CECCOLI
each picture it processes, so each
profile is based upon the parameters of this picture.
Almost. It generates a different profile for each camera
setting (it seems like). So of mine 98 pictures, I got 10
different profiles.
Post by Jean-Luc CECCOLI
That would mean that, for each picture, one would have to
use the corresponding imported profile, else one
might get even worse results than using the DT dedicated
basecurve, isn't it ?
After applying all 10 different profiles to the same image, I
discovered
it was only minor differences, and so small I can neglect them. I
have
never seen them make a worse result than the dedicated base curves.
But
this is a subjective observation!
Regards,
Vidar Hoel
Post by Jean-Luc CECCOLI
That said, I think that that is the way profiles work...
Regards,
J.-Luc
Message du 26/07/16 18:45
Objet : [darktable-user] A better starting-point for Nikon
RAW-files
Post by Jean-Luc CECCOLI
Hi,
some time ago I noticed that the JPGs from my Nikon were better
then
Post by Jean-Luc CECCOLI
what I could achieve from processing the RAW-file (NEF) in
Darktable.
Post by Jean-Luc CECCOLI
After some digging on the Internet, I figured out I could
use an alternative to the default base-curves in Darktable
- and that
made me
get a better starting-point when processing my Nikon RAW-files
(and I
Post by Jean-Luc CECCOLI
was able to achieve a better image than the JPGs from my camera!)
Today I finished a step-by-step video of what I did, so anyone can
see,
learn and perhaps do it themselves.
- I am not really sure behind the technology behind this. I am not
sure
if a Nikon RAW-file has a profile embedded, but it seems so.
- I am not sure if there is an other way to achieve this is
Darktable?
Post by Jean-Luc CECCOLI
- As explained in the video, you will need Windows, or a friends
laptop
with Windows, once, to make this work.
- My native language is not English :-)
http://youtu.be/O1nS3URVSYA
Enjoy :-)
Best regards,
Vidar Hoel
____________________________________________________________________________
Post by Jean-Luc CECCOLI
darktable user mailing list
to unsubscribe send a mail to
____________________________________________________________________________
darktable user mailing list
to unsubscribe send a mail to
____________________________________________________________________________
darktable user mailing list to unsubscribe send a mail to
____________________________________________________________________________
darktable user mailing list to unsubscribe send a mail to
____________________________________________________________________________
darktable user mailing list to unsubscribe send a mail to
____________________________________________________________________________
darktable user mailing list
to unsubscribe send a mail to
André Felipe Carvalho
2016-08-02 19:16:29 UTC
Permalink
Hello,

I've following this thread with great interest, since I have a Nikon 7200
and I was even thinking about developing from jpegs instead of raws.

Could you please instruct me with what to do with those .icm files? e.g.,
where to put it in my system. I know it's for the D7100, but since Saint
Germain reccomended it, I would like to git it a try.

And about the generation of .icm files specifically for the D7200, if
someone could send them to me, I would greatly appreciate it.


Thank you in advance!
André Felipe
Post by Jean-Luc CECCOLI
Hello,
I've tried your profiles with my D7200, and I can confirm that the
results are much closer to the camera jpeg.
I can see a slight blue/green cast and the colors are less
saturated, but it's a much better starting point than the base curve.
I'll have to generate my own profiles to be sure (perhaps D7200 is a
bit different from D7100 ?).
Thanks to Edgardo and Vidar !
On Mon, 1 Aug 2016 15:10:22 -0300, Edgardo Hoszowski
Post by Edgardo Hoszowski
Sure, see attached
Hi Edgardo,
I wrote you related about 7100 icc profiles. I have a mac updated
to capitan and I couldn't install view NX 2 to generate the
profiles.
Could you send me your 7100 generated profiles?
Thanks!
2016-07-31 19:13 GMT+02:00 Edgardo Hoszowski
Post by Edgardo Hoszowski
Same here with a D5100, with a D7100 all went fine, I have 2 or 3
different profiles that give different colors cast, but so far
there's always one closer to the camera jpg.
Using the D7100 profiles on a D5100 raw seems ok too, much closer
to the camera jpg.
Post by Colin Adams
So I tried this (I have a Nikon D810).
When selecting any of the generated icc profiles except the last
"unsupported input profile has been replaced by linear Rec709
RGB!" So I selected the last one, and set unbreak input profile
to 0.0 as you instruct.
The resulting image indeed looks to have slightly better quality,
but it has a green cast.
This is with darktable 2.0.5-1 rather than 2.0.4. I'm doubtful if
that should be making the difference though.
On Wed, 27 Jul 2016 at 15:39 Jean-Luc CECCOLI <
Post by Jean-Luc CECCOLI
Hello,
Very interesting.
Thanks for sharing.
Rgds,
J.-Luc
Message du 26/07/16 20:41
Objet : Re: [darktable-user] A better starting-point for Nikon
RAW-files
Post by Jean-Luc CECCOLI
Hello,
Hi!
Post by Jean-Luc CECCOLI
Just one thing that puzzles me : Nikon ViewNX2 generates a profile
for
Post by Jean-Luc CECCOLI
each picture it processes, so each
profile is based upon the parameters of this picture.
Almost. It generates a different profile for each camera
setting (it seems like). So of mine 98 pictures, I got 10
different profiles.
Post by Jean-Luc CECCOLI
That would mean that, for each picture, one would have to
use the corresponding imported profile, else one
might get even worse results than using the DT dedicated
basecurve, isn't it ?
After applying all 10 different profiles to the same image, I
discovered
it was only minor differences, and so small I can neglect them. I
have
never seen them make a worse result than the dedicated base curves.
But
this is a subjective observation!
Regards,
Vidar Hoel
Post by Jean-Luc CECCOLI
That said, I think that that is the way profiles work...
Regards,
J.-Luc
Message du 26/07/16 18:45
Objet : [darktable-user] A better starting-point for Nikon
RAW-files
Post by Jean-Luc CECCOLI
Hi,
some time ago I noticed that the JPGs from my Nikon were better
then
Post by Jean-Luc CECCOLI
what I could achieve from processing the RAW-file (NEF) in
Darktable.
Post by Jean-Luc CECCOLI
After some digging on the Internet, I figured out I could
use an alternative to the default base-curves in Darktable
- and that
made me
get a better starting-point when processing my Nikon RAW-files
(and I
Post by Jean-Luc CECCOLI
was able to achieve a better image than the JPGs from my camera!)
Today I finished a step-by-step video of what I did, so
anyone can
see,
learn and perhaps do it themselves.
- I am not really sure behind the technology behind this.
I am not
sure
if a Nikon RAW-file has a profile embedded, but it seems so.
- I am not sure if there is an other way to achieve this is
Darktable?
Post by Jean-Luc CECCOLI
- As explained in the video, you will need Windows, or a friends
laptop
with Windows, once, to make this work.
- My native language is not English :-)
http://youtu.be/O1nS3URVSYA
Enjoy :-)
Best regards,
Vidar Hoel
____________________________________________________________________________
Post by Edgardo Hoszowski
Post by Edgardo Hoszowski
Post by Colin Adams
Post by Jean-Luc CECCOLI
Post by Jean-Luc CECCOLI
darktable user mailing list
to unsubscribe send a mail to
____________________________________________________________________________
Post by Edgardo Hoszowski
Post by Edgardo Hoszowski
Post by Colin Adams
Post by Jean-Luc CECCOLI
darktable user mailing list
to unsubscribe send a mail to
____________________________________________________________________________
Post by Edgardo Hoszowski
Post by Edgardo Hoszowski
Post by Colin Adams
Post by Jean-Luc CECCOLI
darktable user mailing list to unsubscribe send a mail to
____________________________________________________________________________
Post by Edgardo Hoszowski
Post by Edgardo Hoszowski
Post by Colin Adams
darktable user mailing list to unsubscribe send a mail to
____________________________________________________________________________
Post by Edgardo Hoszowski
Post by Edgardo Hoszowski
darktable user mailing list to unsubscribe send a mail to
____________________________________________________________________________
Post by Edgardo Hoszowski
darktable user mailing list
to unsubscribe send a mail to
____________________________________________________________________________
darktable user mailing list
to unsubscribe send a mail to
--
André Felipe

https://www.flickr.com/photos/andrefelipecarvalho/
Bernhard
2016-08-02 19:37:10 UTC
Permalink
I opened this in gnome color manager and viewed the profile.
• The white of the output medium is distorted by blobs of color
Nevertheless it's interesting to view the gamut and color curves of the
profile ...

DispCalGUI refuses to open the files.
--
regards
Bernhard

http://www.bilddateien.de
Hello,
I've following this thread with great interest, since I have a Nikon
7200 and I was even thinking about developing from jpegs instead of raws.
Could you please instruct me with what to do with those .icm files?
e.g., where to put it in my system. I know it's for the D7100, but
since Saint Germain reccomended it, I would like to git it a try.
And about the generation of .icm files specifically for the D7200, if
someone could send them to me, I would greatly appreciate it.
Thank you in advance!
André Felipe
Hello,
I've tried your profiles with my D7200, and I can confirm that the
results are much closer to the camera jpeg.
I can see a slight blue/green cast and the colors are less
saturated, but it's a much better starting point than the base curve.
I'll have to generate my own profiles to be sure (perhaps D7200 is a
bit different from D7100 ?).
Thanks to Edgardo and Vidar !
On Mon, 1 Aug 2016 15:10:22 -0300, Edgardo Hoszowski
Post by Edgardo Hoszowski
Sure, see attached
Hi Edgardo,
I wrote you related about 7100 icc profiles. I have a mac
updated
Post by Edgardo Hoszowski
to capitan and I couldn't install view NX 2 to generate the
profiles.
Could you send me your 7100 generated profiles?
Thanks!
2016-07-31 19:13 GMT+02:00 Edgardo Hoszowski
Post by Edgardo Hoszowski
Same here with a D5100, with a D7100 all went fine, I have 2 or 3
different profiles that give different colors cast, but so far
there's always one closer to the camera jpg.
Using the D7100 profiles on a D5100 raw seems ok too, much
closer
Post by Edgardo Hoszowski
Post by Edgardo Hoszowski
to the camera jpg.
2016-07-31 8:18 GMT-03:00 Colin Adams
Post by Colin Adams
So I tried this (I have a Nikon D810).
When selecting any of the generated icc profiles except the last
"unsupported input profile has been replaced by linear Rec709
RGB!" So I selected the last one, and set unbreak input profile
to 0.0 as you instruct.
The resulting image indeed looks to have slightly better
quality,
Post by Edgardo Hoszowski
Post by Edgardo Hoszowski
Post by Colin Adams
but it has a green cast.
This is with darktable 2.0.5-1 rather than 2.0.4. I'm
doubtful if
Post by Edgardo Hoszowski
Post by Edgardo Hoszowski
Post by Colin Adams
that should be making the difference though.
On Wed, 27 Jul 2016 at 15:39 Jean-Luc CECCOLI <
Post by Jean-Luc CECCOLI
Hello,
Very interesting.
Thanks for sharing.
Rgds,
J.-Luc
Message du 26/07/16 20:41
Objet : Re: [darktable-user] A better starting-point for
Nikon
Post by Edgardo Hoszowski
Post by Edgardo Hoszowski
Post by Colin Adams
Post by Jean-Luc CECCOLI
RAW-files
Post by Jean-Luc CECCOLI
Hello,
Hi!
Post by Jean-Luc CECCOLI
Just one thing that puzzles me : Nikon ViewNX2 generates a
profile
for
Post by Jean-Luc CECCOLI
each picture it processes, so each
profile is based upon the parameters of this picture.
Almost. It generates a different profile for each camera
setting (it seems like). So of mine 98 pictures, I got 10
different profiles.
Post by Jean-Luc CECCOLI
That would mean that, for each picture, one would have to
use the corresponding imported profile, else one
might get even worse results than using the DT dedicated
basecurve, isn't it ?
After applying all 10 different profiles to the same image, I
discovered
it was only minor differences, and so small I can neglect them. I
have
never seen them make a worse result than the dedicated base
curves.
But
this is a subjective observation!
Regards,
Vidar Hoel
Post by Jean-Luc CECCOLI
That said, I think that that is the way profiles work...
Regards,
J.-Luc
Message du 26/07/16 18:45
Objet : [darktable-user] A better starting-point for
Nikon
Post by Edgardo Hoszowski
Post by Edgardo Hoszowski
Post by Colin Adams
Post by Jean-Luc CECCOLI
RAW-files
Post by Jean-Luc CECCOLI
Hi,
some time ago I noticed that the JPGs from my Nikon were
better
then
Post by Jean-Luc CECCOLI
what I could achieve from processing the RAW-file
(NEF) in
Post by Edgardo Hoszowski
Post by Edgardo Hoszowski
Post by Colin Adams
Post by Jean-Luc CECCOLI
Darktable.
Post by Jean-Luc CECCOLI
After some digging on the Internet, I figured out I could
use an alternative to the default base-curves in
Darktable
Post by Edgardo Hoszowski
Post by Edgardo Hoszowski
Post by Colin Adams
Post by Jean-Luc CECCOLI
Post by Jean-Luc CECCOLI
- and that
made me
get a better starting-point when processing my Nikon
RAW-files
(and I
Post by Jean-Luc CECCOLI
was able to achieve a better image than the JPGs from my
camera!)
Today I finished a step-by-step video of what I did, so
anyone can
see,
learn and perhaps do it themselves.
- I am not really sure behind the technology behind this.
I am not
sure
if a Nikon RAW-file has a profile embedded, but it seems
so.
- I am not sure if there is an other way to achieve
this is
Post by Edgardo Hoszowski
Post by Edgardo Hoszowski
Post by Colin Adams
Post by Jean-Luc CECCOLI
Darktable?
Post by Jean-Luc CECCOLI
- As explained in the video, you will need Windows, or a
friends
laptop
with Windows, once, to make this work.
- My native language is not English :-)
http://youtu.be/O1nS3URVSYA
Enjoy :-)
Best regards,
Vidar Hoel
____________________________________________________________________________
Post by Edgardo Hoszowski
Post by Edgardo Hoszowski
Post by Colin Adams
Post by Jean-Luc CECCOLI
Post by Jean-Luc CECCOLI
darktable user mailing list
to unsubscribe send a mail to
____________________________________________________________________________
Post by Edgardo Hoszowski
Post by Edgardo Hoszowski
Post by Colin Adams
Post by Jean-Luc CECCOLI
darktable user mailing list
to unsubscribe send a mail to
____________________________________________________________________________
Post by Edgardo Hoszowski
Post by Edgardo Hoszowski
Post by Colin Adams
Post by Jean-Luc CECCOLI
darktable user mailing list to unsubscribe send a mail to
____________________________________________________________________________
Post by Edgardo Hoszowski
Post by Edgardo Hoszowski
Post by Colin Adams
darktable user mailing list to unsubscribe send a mail to
____________________________________________________________________________
Post by Edgardo Hoszowski
Post by Edgardo Hoszowski
darktable user mailing list to unsubscribe send a mail to
____________________________________________________________________________
Post by Edgardo Hoszowski
darktable user mailing list
to unsubscribe send a mail to
____________________________________________________________________________
darktable user mailing list
to unsubscribe send a mail to
--
André Felipe
https://www.flickr.com/photos/andrefelipecarvalho/
____________________________________________________________________________
darktable user mailing list to unsubscribe send a mail to
--
regards
Bernhard

http://www.bilddateien.de
Colin Adams
2016-08-02 19:54:17 UTC
Permalink
What's the command to open a .icm file in gnome-color-manager? I'd like to
look at one of my Nikon D810 files.
Post by Bernhard
I opened this in gnome color manager and viewed the profile.
• The white of the output medium is distorted by blobs of color
Nevertheless it's interesting to view the gamut and color curves of the
profile ...
DispCalGUI refuses to open the files.
--
regards
Bernhard
http://www.bilddateien.de
Hello,
I've following this thread with great interest, since I have a Nikon 7200
and I was even thinking about developing from jpegs instead of raws.
Could you please instruct me with what to do with those .icm files? e.g.,
where to put it in my system. I know it's for the D7100, but since Saint
Germain reccomended it, I would like to git it a try.
And about the generation of .icm files specifically for the D7200, if
someone could send them to me, I would greatly appreciate it.
Thank you in advance!
André Felipe
Post by Jean-Luc CECCOLI
Hello,
I've tried your profiles with my D7200, and I can confirm that the
results are much closer to the camera jpeg.
I can see a slight blue/green cast and the colors are less
saturated, but it's a much better starting point than the base curve.
I'll have to generate my own profiles to be sure (perhaps D7200 is a
bit different from D7100 ?).
Thanks to Edgardo and Vidar !
On Mon, 1 Aug 2016 15:10:22 -0300, Edgardo Hoszowski
Post by Edgardo Hoszowski
Sure, see attached
Hi Edgardo,
I wrote you related about 7100 icc profiles. I have a mac updated
to capitan and I couldn't install view NX 2 to generate the
profiles.
Could you send me your 7100 generated profiles?
Thanks!
2016-07-31 19:13 GMT+02:00 Edgardo Hoszowski
Post by Edgardo Hoszowski
Same here with a D5100, with a D7100 all went fine, I have 2 or 3
different profiles that give different colors cast, but so far
there's always one closer to the camera jpg.
Using the D7100 profiles on a D5100 raw seems ok too, much closer
to the camera jpg.
Post by Colin Adams
So I tried this (I have a Nikon D810).
When selecting any of the generated icc profiles except the last
"unsupported input profile has been replaced by linear Rec709
RGB!" So I selected the last one, and set unbreak input profile
to 0.0 as you instruct.
The resulting image indeed looks to have slightly better quality,
but it has a green cast.
This is with darktable 2.0.5-1 rather than 2.0.4. I'm doubtful if
that should be making the difference though.
On Wed, 27 Jul 2016 at 15:39 Jean-Luc CECCOLI <
Post by Jean-Luc CECCOLI
Hello,
Very interesting.
Thanks for sharing.
Rgds,
J.-Luc
Message du 26/07/16 20:41
Objet : Re: [darktable-user] A better starting-point for Nikon
RAW-files
Post by Jean-Luc CECCOLI
Hello,
Hi!
Post by Jean-Luc CECCOLI
Just one thing that puzzles me : Nikon ViewNX2 generates a profile
for
Post by Jean-Luc CECCOLI
each picture it processes, so each
profile is based upon the parameters of this picture.
Almost. It generates a different profile for each camera
setting (it seems like). So of mine 98 pictures, I got 10
different profiles.
Post by Jean-Luc CECCOLI
That would mean that, for each picture, one would have to
use the corresponding imported profile, else one
might get even worse results than using the DT dedicated
basecurve, isn't it ?
After applying all 10 different profiles to the same image, I
discovered
it was only minor differences, and so small I can neglect them. I
have
never seen them make a worse result than the dedicated base curves.
But
this is a subjective observation!
Regards,
Vidar Hoel
Post by Jean-Luc CECCOLI
That said, I think that that is the way profiles work...
Regards,
J.-Luc
Message du 26/07/16 18:45
Objet : [darktable-user] A better starting-point for Nikon
RAW-files
Post by Jean-Luc CECCOLI
Hi,
some time ago I noticed that the JPGs from my Nikon were better
then
Post by Jean-Luc CECCOLI
what I could achieve from processing the RAW-file (NEF) in
Darktable.
Post by Jean-Luc CECCOLI
After some digging on the Internet, I figured out I could
use an alternative to the default base-curves in Darktable
- and that
made me
get a better starting-point when processing my Nikon RAW-files
(and I
Post by Jean-Luc CECCOLI
was able to achieve a better image than the JPGs from my
camera!)
Today I finished a step-by-step video of what I did, so
anyone can
see,
learn and perhaps do it themselves.
- I am not really sure behind the technology behind this.
I am not
sure
if a Nikon RAW-file has a profile embedded, but it seems so.
- I am not sure if there is an other way to achieve this is
Darktable?
Post by Jean-Luc CECCOLI
- As explained in the video, you will need Windows, or a
friends
laptop
with Windows, once, to make this work.
- My native language is not English :-)
http://youtu.be/O1nS3URVSYA
Enjoy :-)
Best regards,
Vidar Hoel
____________________________________________________________________________
Post by Edgardo Hoszowski
Post by Edgardo Hoszowski
Post by Colin Adams
Post by Jean-Luc CECCOLI
Post by Jean-Luc CECCOLI
darktable user mailing list
to unsubscribe send a mail to
____________________________________________________________________________
Post by Edgardo Hoszowski
Post by Edgardo Hoszowski
Post by Colin Adams
Post by Jean-Luc CECCOLI
darktable user mailing list
to unsubscribe send a mail to
____________________________________________________________________________
Post by Edgardo Hoszowski
Post by Edgardo Hoszowski
Post by Colin Adams
Post by Jean-Luc CECCOLI
darktable user mailing list to unsubscribe send a mail to
____________________________________________________________________________
Post by Edgardo Hoszowski
Post by Edgardo Hoszowski
Post by Colin Adams
darktable user mailing list to unsubscribe send a mail to
____________________________________________________________________________
Post by Edgardo Hoszowski
Post by Edgardo Hoszowski
darktable user mailing list to unsubscribe send a mail to
____________________________________________________________________________
Post by Edgardo Hoszowski
darktable user mailing list
to unsubscribe send a mail to
____________________________________________________________________________
darktable user mailing list
to unsubscribe send a mail to
--
André Felipe
https://www.flickr.com/photos/andrefelipecarvalho/
____________________________________________________________________________
darktable user mailing list to unsubscribe send a mail to
--
regards
Bernhard
http://www.bilddateien.de
____________________________________________________________________________
darktable user mailing list to unsubscribe send a mail to
Bernhard
2016-08-02 21:04:31 UTC
Permalink
either you double-click in Nautilus or Nemo or type
bash$ gcm-import <filepath>
in a terminal and you should see the attached dialog, you click "view
details" and can start investigating ...
--
regards
Bernhard

http://www.bilddateien.de
What's the command to open a .icm file in gnome-color-manager? I'd
like to look at one of my Nikon D810 files.
I opened this in gnome color manager and viewed the profile.
• The white of the output medium is distorted by blobs of color
Nevertheless it's interesting to view the gamut and color curves
of the profile ...
DispCalGUI refuses to open the files.
--
regards
Bernhard
http://www.bilddateien.de
Hello,
I've following this thread with great interest, since I have a
Nikon 7200 and I was even thinking about developing from jpegs
instead of raws.
Could you please instruct me with what to do with those .icm
files? e.g., where to put it in my system. I know it's for the
D7100, but since Saint Germain reccomended it, I would like to
git it a try.
And about the generation of .icm files specifically for the
D7200, if someone could send them to me, I would greatly
appreciate it.
Thank you in advance!
André Felipe
Hello,
I've tried your profiles with my D7200, and I can confirm that the
results are much closer to the camera jpeg.
I can see a slight blue/green cast and the colors are less
saturated, but it's a much better starting point than the base curve.
I'll have to generate my own profiles to be sure (perhaps D7200 is a
bit different from D7100 ?).
Thanks to Edgardo and Vidar !
On Mon, 1 Aug 2016 15:10:22 -0300, Edgardo Hoszowski
Post by Edgardo Hoszowski
Sure, see attached
2016-08-01 5:46 GMT-03:00 Rafa García
Hi Edgardo,
I wrote you related about 7100 icc profiles. I have a
mac updated
Post by Edgardo Hoszowski
to capitan and I couldn't install view NX 2 to generate the
profiles.
Could you send me your 7100 generated profiles?
Thanks!
2016-07-31 19:13 GMT+02:00 Edgardo Hoszowski
Post by Edgardo Hoszowski
Same here with a D5100, with a D7100 all went fine, I
have 2 or 3
Post by Edgardo Hoszowski
Post by Edgardo Hoszowski
different profiles that give different colors cast, but
so far
Post by Edgardo Hoszowski
Post by Edgardo Hoszowski
there's always one closer to the camera jpg.
Using the D7100 profiles on a D5100 raw seems ok too,
much closer
Post by Edgardo Hoszowski
Post by Edgardo Hoszowski
to the camera jpg.
2016-07-31 8:18 GMT-03:00 Colin Adams
Post by Colin Adams
So I tried this (I have a Nikon D810).
When selecting any of the generated icc profiles except
the last
Post by Edgardo Hoszowski
Post by Edgardo Hoszowski
Post by Colin Adams
"unsupported input profile has been replaced by linear
Rec709
Post by Edgardo Hoszowski
Post by Edgardo Hoszowski
Post by Colin Adams
RGB!" So I selected the last one, and set unbreak input
profile
Post by Edgardo Hoszowski
Post by Edgardo Hoszowski
Post by Colin Adams
to 0.0 as you instruct.
The resulting image indeed looks to have slightly
better quality,
Post by Edgardo Hoszowski
Post by Edgardo Hoszowski
Post by Colin Adams
but it has a green cast.
This is with darktable 2.0.5-1 rather than 2.0.4. I'm
doubtful if
Post by Edgardo Hoszowski
Post by Edgardo Hoszowski
Post by Colin Adams
that should be making the difference though.
On Wed, 27 Jul 2016 at 15:39 Jean-Luc CECCOLI <
Post by Jean-Luc CECCOLI
Hello,
Very interesting.
Thanks for sharing.
Rgds,
J.-Luc
Message du 26/07/16 20:41
Objet : Re: [darktable-user] A better starting-point
for Nikon
Post by Edgardo Hoszowski
Post by Edgardo Hoszowski
Post by Colin Adams
Post by Jean-Luc CECCOLI
RAW-files
Post by Jean-Luc CECCOLI
Hello,
Hi!
Post by Jean-Luc CECCOLI
Just one thing that puzzles me : Nikon ViewNX2
generates a
Post by Edgardo Hoszowski
Post by Edgardo Hoszowski
Post by Colin Adams
Post by Jean-Luc CECCOLI
Post by Jean-Luc CECCOLI
profile
for
Post by Jean-Luc CECCOLI
each picture it processes, so each
profile is based upon the parameters of this picture.
Almost. It generates a different profile for each camera
setting (it seems like). So of mine 98 pictures, I
got 10
Post by Edgardo Hoszowski
Post by Edgardo Hoszowski
Post by Colin Adams
Post by Jean-Luc CECCOLI
different profiles.
Post by Jean-Luc CECCOLI
That would mean that, for each picture, one would
have to
Post by Edgardo Hoszowski
Post by Edgardo Hoszowski
Post by Colin Adams
Post by Jean-Luc CECCOLI
Post by Jean-Luc CECCOLI
use the corresponding imported profile, else one
might get even worse results than using the DT
dedicated
Post by Edgardo Hoszowski
Post by Edgardo Hoszowski
Post by Colin Adams
Post by Jean-Luc CECCOLI
Post by Jean-Luc CECCOLI
basecurve, isn't it ?
After applying all 10 different profiles to the same
image, I
Post by Edgardo Hoszowski
Post by Edgardo Hoszowski
Post by Colin Adams
Post by Jean-Luc CECCOLI
discovered
it was only minor differences, and so small I can
neglect
Post by Edgardo Hoszowski
Post by Edgardo Hoszowski
Post by Colin Adams
Post by Jean-Luc CECCOLI
them. I
have
never seen them make a worse result than the
dedicated base
Post by Edgardo Hoszowski
Post by Edgardo Hoszowski
Post by Colin Adams
Post by Jean-Luc CECCOLI
curves.
But
this is a subjective observation!
Regards,
Vidar Hoel
Post by Jean-Luc CECCOLI
That said, I think that that is the way profiles
work...
Post by Edgardo Hoszowski
Post by Edgardo Hoszowski
Post by Colin Adams
Post by Jean-Luc CECCOLI
Post by Jean-Luc CECCOLI
Regards,
J.-Luc
Message du 26/07/16 18:45
Objet : [darktable-user] A better starting-point
for Nikon
Post by Edgardo Hoszowski
Post by Edgardo Hoszowski
Post by Colin Adams
Post by Jean-Luc CECCOLI
RAW-files
Post by Jean-Luc CECCOLI
Hi,
some time ago I noticed that the JPGs from my
Nikon were
Post by Edgardo Hoszowski
Post by Edgardo Hoszowski
Post by Colin Adams
Post by Jean-Luc CECCOLI
Post by Jean-Luc CECCOLI
better
then
Post by Jean-Luc CECCOLI
what I could achieve from processing the
RAW-file (NEF) in
Post by Edgardo Hoszowski
Post by Edgardo Hoszowski
Post by Colin Adams
Post by Jean-Luc CECCOLI
Darktable.
Post by Jean-Luc CECCOLI
After some digging on the Internet, I figured
out I could
Post by Edgardo Hoszowski
Post by Edgardo Hoszowski
Post by Colin Adams
Post by Jean-Luc CECCOLI
Post by Jean-Luc CECCOLI
use an alternative to the default base-curves in
Darktable
Post by Edgardo Hoszowski
Post by Edgardo Hoszowski
Post by Colin Adams
Post by Jean-Luc CECCOLI
Post by Jean-Luc CECCOLI
- and that
made me
get a better starting-point when processing my Nikon
RAW-files
(and I
Post by Jean-Luc CECCOLI
was able to achieve a better image than the JPGs
from my
Post by Edgardo Hoszowski
Post by Edgardo Hoszowski
Post by Colin Adams
Post by Jean-Luc CECCOLI
Post by Jean-Luc CECCOLI
camera!)
Today I finished a step-by-step video of what I
did, so
Post by Edgardo Hoszowski
Post by Edgardo Hoszowski
Post by Colin Adams
Post by Jean-Luc CECCOLI
Post by Jean-Luc CECCOLI
anyone can
see,
learn and perhaps do it themselves.
- I am not really sure behind the technology
behind this.
Post by Edgardo Hoszowski
Post by Edgardo Hoszowski
Post by Colin Adams
Post by Jean-Luc CECCOLI
Post by Jean-Luc CECCOLI
I am not
sure
if a Nikon RAW-file has a profile embedded, but
it seems
Post by Edgardo Hoszowski
Post by Edgardo Hoszowski
Post by Colin Adams
Post by Jean-Luc CECCOLI
Post by Jean-Luc CECCOLI
so.
- I am not sure if there is an other way to
achieve this is
Post by Edgardo Hoszowski
Post by Edgardo Hoszowski
Post by Colin Adams
Post by Jean-Luc CECCOLI
Darktable?
Post by Jean-Luc CECCOLI
- As explained in the video, you will need
Windows, or a
Post by Edgardo Hoszowski
Post by Edgardo Hoszowski
Post by Colin Adams
Post by Jean-Luc CECCOLI
Post by Jean-Luc CECCOLI
friends
laptop
with Windows, once, to make this work.
- My native language is not English :-)
http://youtu.be/O1nS3URVSYA
Enjoy :-)
Best regards,
Vidar Hoel
____________________________________________________________________________
Post by Edgardo Hoszowski
Post by Edgardo Hoszowski
Post by Colin Adams
Post by Jean-Luc CECCOLI
Post by Jean-Luc CECCOLI
darktable user mailing list
to unsubscribe send a mail to
____________________________________________________________________________
Post by Edgardo Hoszowski
Post by Edgardo Hoszowski
Post by Colin Adams
Post by Jean-Luc CECCOLI
darktable user mailing list
to unsubscribe send a mail to
____________________________________________________________________________
Post by Edgardo Hoszowski
Post by Edgardo Hoszowski
Post by Colin Adams
Post by Jean-Luc CECCOLI
darktable user mailing list to unsubscribe send a mail to
____________________________________________________________________________
Post by Edgardo Hoszowski
Post by Edgardo Hoszowski
Post by Colin Adams
darktable user mailing list to unsubscribe send a mail to
____________________________________________________________________________
Post by Edgardo Hoszowski
Post by Edgardo Hoszowski
darktable user mailing list to unsubscribe send a mail to
____________________________________________________________________________
Post by Edgardo Hoszowski
darktable user mailing list
to unsubscribe send a mail to
____________________________________________________________________________
darktable user mailing list
to unsubscribe send a mail to
--
André Felipe
https://www.flickr.com/photos/andrefelipecarvalho/
____________________________________________________________________________
darktable user mailing list to unsubscribe send a mail to
--
regards
Bernhard
http://www.bilddateien.de
____________________________________________________________________________
darktable user mailing list to unsubscribe send a mail to
____________________________________________________________________________
darktable user mailing list to unsubscribe send a mail to
Colin Adams
2016-08-03 06:26:39 UTC
Permalink
OK, thanks.
I did that and then used gcm-view to see if there were any problems. All of
them except one were colorspace YCbCr and no problems were reported. One
was colorspace RGB and has the problem "The profile has the following
problems:
• A scum dot is present for media white".

So "unsupported input profile has been replaced by linear Rec709 RGB!" from
darktable is presumably because it is not supporting the YCbCr colorspace.
Would that be correct? And if so, is there a way to convert to another
colorspace?
Post by Bernhard
either you double-click in Nautilus or Nemo or type
bash$ gcm-import <filepath>
in a terminal and you should see the attached dialog, you click "view
details" and can start investigating ...
--
regards
Bernhard
http://www.bilddateien.de
What's the command to open a .icm file in gnome-color-manager? I'd like to
look at one of my Nikon D810 files.
Post by Bernhard
I opened this in gnome color manager and viewed the profile.
• The white of the output medium is distorted by blobs of color
Nevertheless it's interesting to view the gamut and color curves of the
profile ...
DispCalGUI refuses to open the files.
--
regards
Bernhard
http://www.bilddateien.de
Hello,
I've following this thread with great interest, since I have a Nikon 7200
and I was even thinking about developing from jpegs instead of raws.
Could you please instruct me with what to do with those .icm files? e.g.,
where to put it in my system. I know it's for the D7100, but since Saint
Germain reccomended it, I would like to git it a try.
And about the generation of .icm files specifically for the D7200, if
someone could send them to me, I would greatly appreciate it.
Thank you in advance!
André Felipe
Post by Jean-Luc CECCOLI
Hello,
I've tried your profiles with my D7200, and I can confirm that the
results are much closer to the camera jpeg.
I can see a slight blue/green cast and the colors are less
saturated, but it's a much better starting point than the base curve.
I'll have to generate my own profiles to be sure (perhaps D7200 is a
bit different from D7100 ?).
Thanks to Edgardo and Vidar !
On Mon, 1 Aug 2016 15:10:22 -0300, Edgardo Hoszowski
Post by Edgardo Hoszowski
Sure, see attached
Hi Edgardo,
I wrote you related about 7100 icc profiles. I have a mac updated
to capitan and I couldn't install view NX 2 to generate the
profiles.
Could you send me your 7100 generated profiles?
Thanks!
2016-07-31 19:13 GMT+02:00 Edgardo Hoszowski
Post by Edgardo Hoszowski
Same here with a D5100, with a D7100 all went fine, I have 2 or 3
different profiles that give different colors cast, but so far
there's always one closer to the camera jpg.
Using the D7100 profiles on a D5100 raw seems ok too, much closer
to the camera jpg.
Post by Colin Adams
So I tried this (I have a Nikon D810).
When selecting any of the generated icc profiles except the last
"unsupported input profile has been replaced by linear Rec709
RGB!" So I selected the last one, and set unbreak input profile
to 0.0 as you instruct.
The resulting image indeed looks to have slightly better quality,
but it has a green cast.
This is with darktable 2.0.5-1 rather than 2.0.4. I'm doubtful if
that should be making the difference though.
On Wed, 27 Jul 2016 at 15:39 Jean-Luc CECCOLI <
Post by Jean-Luc CECCOLI
Hello,
Very interesting.
Thanks for sharing.
Rgds,
J.-Luc
Message du 26/07/16 20:41
Objet : Re: [darktable-user] A better starting-point for Nikon
RAW-files
Post by Jean-Luc CECCOLI
Hello,
Hi!
Post by Jean-Luc CECCOLI
Just one thing that puzzles me : Nikon ViewNX2 generates a
profile
for
Post by Jean-Luc CECCOLI
each picture it processes, so each
profile is based upon the parameters of this picture.
Almost. It generates a different profile for each camera
setting (it seems like). So of mine 98 pictures, I got 10
different profiles.
Post by Jean-Luc CECCOLI
That would mean that, for each picture, one would have to
use the corresponding imported profile, else one
might get even worse results than using the DT dedicated
basecurve, isn't it ?
After applying all 10 different profiles to the same image, I
discovered
it was only minor differences, and so small I can neglect them. I
have
never seen them make a worse result than the dedicated base curves.
But
this is a subjective observation!
Regards,
Vidar Hoel
Post by Jean-Luc CECCOLI
That said, I think that that is the way profiles work...
Regards,
J.-Luc
Message du 26/07/16 18:45
Objet : [darktable-user] A better starting-point for Nikon
RAW-files
Post by Jean-Luc CECCOLI
Hi,
some time ago I noticed that the JPGs from my Nikon were
better
then
Post by Jean-Luc CECCOLI
what I could achieve from processing the RAW-file (NEF) in
Darktable.
Post by Jean-Luc CECCOLI
After some digging on the Internet, I figured out I could
use an alternative to the default base-curves in Darktable
- and that
made me
get a better starting-point when processing my Nikon
RAW-files
(and I
Post by Jean-Luc CECCOLI
was able to achieve a better image than the JPGs from my
camera!)
Today I finished a step-by-step video of what I did, so
anyone can
see,
learn and perhaps do it themselves.
- I am not really sure behind the technology behind this.
I am not
sure
if a Nikon RAW-file has a profile embedded, but it seems so.
- I am not sure if there is an other way to achieve this is
Darktable?
Post by Jean-Luc CECCOLI
- As explained in the video, you will need Windows, or a
friends
laptop
with Windows, once, to make this work.
- My native language is not English :-)
http://youtu.be/O1nS3URVSYA
Enjoy :-)
Best regards,
Vidar Hoel
____________________________________________________________________________
Post by Edgardo Hoszowski
Post by Edgardo Hoszowski
Post by Colin Adams
Post by Jean-Luc CECCOLI
Post by Jean-Luc CECCOLI
darktable user mailing list
to unsubscribe send a mail to
____________________________________________________________________________
Post by Edgardo Hoszowski
Post by Edgardo Hoszowski
Post by Colin Adams
Post by Jean-Luc CECCOLI
darktable user mailing list
to unsubscribe send a mail to
____________________________________________________________________________
Post by Edgardo Hoszowski
Post by Edgardo Hoszowski
Post by Colin Adams
Post by Jean-Luc CECCOLI
darktable user mailing list to unsubscribe send a mail to
____________________________________________________________________________
Post by Edgardo Hoszowski
Post by Edgardo Hoszowski
Post by Colin Adams
darktable user mailing list to unsubscribe send a mail to
____________________________________________________________________________
Post by Edgardo Hoszowski
Post by Edgardo Hoszowski
darktable user mailing list to unsubscribe send a mail to
____________________________________________________________________________
Post by Edgardo Hoszowski
darktable user mailing list
to unsubscribe send a mail to
____________________________________________________________________________
darktable user mailing list
to unsubscribe send a mail to
--
André Felipe
https://www.flickr.com/photos/andrefelipecarvalho/
____________________________________________________________________________
darktable user mailing list to unsubscribe send a mail to
--
regards
Bernhard
http://www.bilddateien.de
____________________________________________________________________________
darktable user mailing list to unsubscribe send a mail to
____________________________________________________________________________
darktable user mailing list to unsubscribe send a mail to
____________________________________________________________________________
darktable user mailing list to unsubscribe send a mail to
Saint Germain
2016-08-03 18:41:53 UTC
Permalink
I tried today to generate the icm profile for my D7200.
First ViewNX 2 is not compatible with D7200, so I had to use Capture
NX-D, however the process is otherwise identical.

Generating the jpeg for one single nef file, I got as a result 5 icm
files:
- 4 starting with Nkx_D7200_xxx
- 1 starting with the name Nkx_LinearYcc.icm

After remove the duplicates, I ended up with 4 different icm profiles.
- 3 have the colorspace YCbCr and could not be used by darktable
("unsupported input profile has been replaced by linear Rec709 RGB!").
- 1 has the colorspace RGB and could be used by darktable but I got
completely incorrect color (like it was the negative of the picture,
like in old times).

Usuing gcm-view I have exactly the same result as Colin (error on the
one with colorspace RGB).

So it seems that I can not use this method to generate an input profile
for my D7200 ?
Or have I done something wrong ?

Thanks

On Wed, 03 Aug 2016 06:26:39 +0000, Colin Adams
Post by Colin Adams
OK, thanks.
I did that and then used gcm-view to see if there were any problems.
All of them except one were colorspace YCbCr and no problems were
reported. One was colorspace RGB and has the problem "The profile has
• A scum dot is present for media white".
So "unsupported input profile has been replaced by linear Rec709
RGB!" from darktable is presumably because it is not supporting the
YCbCr colorspace. Would that be correct? And if so, is there a way to
convert to another colorspace?
Post by Bernhard
either you double-click in Nautilus or Nemo or type
bash$ gcm-import <filepath>
in a terminal and you should see the attached dialog, you click
"view details" and can start investigating ...
--
regards
Bernhard
http://www.bilddateien.de
What's the command to open a .icm file in gnome-color-manager? I'd
like to look at one of my Nikon D810 files.
Post by Bernhard
I opened this in gnome color manager and viewed the profile.
• The white of the output medium is distorted by blobs of color
Nevertheless it's interesting to view the gamut and color curves
of the profile ...
DispCalGUI refuses to open the files.
--
regards
Bernhard
http://www.bilddateien.de
Hello,
I've following this thread with great interest, since I have a
Nikon 7200 and I was even thinking about developing from jpegs
instead of raws.
Could you please instruct me with what to do with those .icm
files? e.g., where to put it in my system. I know it's for the
D7100, but since Saint Germain reccomended it, I would like to git
it a try.
And about the generation of .icm files specifically for the D7200,
if someone could send them to me, I would greatly appreciate it.
Thank you in advance!
André Felipe
Post by Jean-Luc CECCOLI
Hello,
I've tried your profiles with my D7200, and I can confirm that the
results are much closer to the camera jpeg.
I can see a slight blue/green cast and the colors are less
saturated, but it's a much better starting point than the base curve.
I'll have to generate my own profiles to be sure (perhaps D7200
is a bit different from D7100 ?).
Thanks to Edgardo and Vidar !
On Mon, 1 Aug 2016 15:10:22 -0300, Edgardo Hoszowski
Post by Edgardo Hoszowski
Sure, see attached
Hi Edgardo,
I wrote you related about 7100 icc profiles. I have a mac
updated to capitan and I couldn't install view NX 2 to
generate the profiles.
Could you send me your 7100 generated profiles?
Thanks!
2016-07-31 19:13 GMT+02:00 Edgardo Hoszowski
Post by Edgardo Hoszowski
Same here with a D5100, with a D7100 all went fine, I have 2
or 3 different profiles that give different colors cast, but
so far there's always one closer to the camera jpg.
Using the D7100 profiles on a D5100 raw seems ok too, much
closer to the camera jpg.
2016-07-31 8:18 GMT-03:00 Colin Adams
Post by Colin Adams
So I tried this (I have a Nikon D810).
When selecting any of the generated icc profiles except the
"unsupported input profile has been replaced by linear
Rec709 RGB!" So I selected the last one, and set unbreak
input profile to 0.0 as you instruct.
The resulting image indeed looks to have slightly better
quality, but it has a green cast.
This is with darktable 2.0.5-1 rather than 2.0.4. I'm
doubtful if that should be making the difference though.
On Wed, 27 Jul 2016 at 15:39 Jean-Luc CECCOLI <
Post by Jean-Luc CECCOLI
Hello,
Very interesting.
Thanks for sharing.
Rgds,
J.-Luc
Message du 26/07/16 20:41
Objet : Re: [darktable-user] A better starting-point for Nikon
RAW-files
Post by Jean-Luc CECCOLI
Hello,
Hi!
Post by Jean-Luc CECCOLI
Just one thing that puzzles me : Nikon ViewNX2
generates a profile
for
Post by Jean-Luc CECCOLI
each picture it processes, so each
profile is based upon the parameters of this picture.
Almost. It generates a different profile for each camera
setting (it seems like). So of mine 98 pictures, I got 10
different profiles.
Post by Jean-Luc CECCOLI
That would mean that, for each picture, one would have
to use the corresponding imported profile, else one
might get even worse results than using the DT
dedicated basecurve, isn't it ?
After applying all 10 different profiles to the same image, I
discovered
it was only minor differences, and so small I can neglect them. I
have
never seen them make a worse result than the dedicated
base curves.
But
this is a subjective observation!
Regards,
Vidar Hoel
Post by Jean-Luc CECCOLI
That said, I think that that is the way profiles work...
Regards,
J.-Luc
Message du 26/07/16 18:45
Objet : [darktable-user] A better starting-point for
Nikon
RAW-files
Post by Jean-Luc CECCOLI
Hi,
some time ago I noticed that the JPGs from my Nikon
were better
then
Post by Jean-Luc CECCOLI
what I could achieve from processing the RAW-file
(NEF) in
Darktable.
Post by Jean-Luc CECCOLI
After some digging on the Internet, I figured out I
could use an alternative to the default base-curves
in Darktable
- and that
made me
get a better starting-point when processing my Nikon
RAW-files
(and I
Post by Jean-Luc CECCOLI
was able to achieve a better image than the JPGs
from my camera!)
Today I finished a step-by-step video of what I did,
so anyone can
see,
learn and perhaps do it themselves.
- I am not really sure behind the technology behind
this. I am not
sure
if a Nikon RAW-file has a profile embedded, but it
seems so.
- I am not sure if there is an other way to achieve
this is
Darktable?
Post by Jean-Luc CECCOLI
- As explained in the video, you will need Windows,
or a friends
laptop
with Windows, once, to make this work.
- My native language is not English :-)
http://youtu.be/O1nS3URVSYA
Enjoy :-)
Best regards,
Vidar Hoel
____________________________________________________________________________
Post by Edgardo Hoszowski
Post by Edgardo Hoszowski
Post by Colin Adams
Post by Jean-Luc CECCOLI
Post by Jean-Luc CECCOLI
darktable user mailing list
to unsubscribe send a mail to
____________________________________________________________________________
Post by Edgardo Hoszowski
Post by Edgardo Hoszowski
Post by Colin Adams
Post by Jean-Luc CECCOLI
darktable user mailing list
to unsubscribe send a mail to
____________________________________________________________________________
Post by Edgardo Hoszowski
Post by Edgardo Hoszowski
Post by Colin Adams
Post by Jean-Luc CECCOLI
darktable user mailing list to unsubscribe send a mail to
____________________________________________________________________________
Post by Edgardo Hoszowski
Post by Edgardo Hoszowski
Post by Colin Adams
darktable user mailing list to unsubscribe send a mail to
____________________________________________________________________________
Post by Edgardo Hoszowski
Post by Edgardo Hoszowski
darktable user mailing list to unsubscribe send a mail to
____________________________________________________________________________
Post by Edgardo Hoszowski
darktable user mailing list
to unsubscribe send a mail to
____________________________________________________________________________
darktable user mailing list
to unsubscribe send a mail to
--
André Felipe
https://www.flickr.com/photos/andrefelipecarvalho/
____________________________________________________________________________
darktable user mailing list to unsubscribe send a mail to
--
regards
Bernhard
http://www.bilddateien.de
____________________________________________________________________________
darktable user mailing list to unsubscribe send a mail to
____________________________________________________________________________
darktable user mailing list to unsubscribe send a mail to
____________________________________________________________________________
darktable user mailing list to unsubscribe send a mail to
____________________________________________________________________________
darktable user mailing list
to unsubscribe send a mail to
Colin Adams
2016-08-03 19:13:36 UTC
Permalink
Yes, those were the names I saw too (except 810 rather than 7200 of course).
Post by Saint Germain
I tried today to generate the icm profile for my D7200.
First ViewNX 2 is not compatible with D7200, so I had to use Capture
NX-D, however the process is otherwise identical.
Generating the jpeg for one single nef file, I got as a result 5 icm
- 4 starting with Nkx_D7200_xxx
- 1 starting with the name Nkx_LinearYcc.icm
After remove the duplicates, I ended up with 4 different icm profiles.
- 3 have the colorspace YCbCr and could not be used by darktable
("unsupported input profile has been replaced by linear Rec709 RGB!").
- 1 has the colorspace RGB and could be used by darktable but I got
completely incorrect color (like it was the negative of the picture,
like in old times).
Usuing gcm-view I have exactly the same result as Colin (error on the
one with colorspace RGB).
So it seems that I can not use this method to generate an input profile
for my D7200 ?
Or have I done something wrong ?
Thanks
On Wed, 03 Aug 2016 06:26:39 +0000, Colin Adams
Post by Colin Adams
OK, thanks.
I did that and then used gcm-view to see if there were any problems.
All of them except one were colorspace YCbCr and no problems were
reported. One was colorspace RGB and has the problem "The profile has
• A scum dot is present for media white".
So "unsupported input profile has been replaced by linear Rec709
RGB!" from darktable is presumably because it is not supporting the
YCbCr colorspace. Would that be correct? And if so, is there a way to
convert to another colorspace?
Post by Bernhard
either you double-click in Nautilus or Nemo or type
bash$ gcm-import <filepath>
in a terminal and you should see the attached dialog, you click
"view details" and can start investigating ...
--
regards
Bernhard
http://www.bilddateien.de
What's the command to open a .icm file in gnome-color-manager? I'd
like to look at one of my Nikon D810 files.
Post by Bernhard
I opened this in gnome color manager and viewed the profile.
• The white of the output medium is distorted by blobs of color
Nevertheless it's interesting to view the gamut and color curves
of the profile ...
DispCalGUI refuses to open the files.
--
regards
Bernhard
http://www.bilddateien.de
Hello,
I've following this thread with great interest, since I have a
Nikon 7200 and I was even thinking about developing from jpegs
instead of raws.
Could you please instruct me with what to do with those .icm
files? e.g., where to put it in my system. I know it's for the
D7100, but since Saint Germain reccomended it, I would like to git
it a try.
And about the generation of .icm files specifically for the D7200,
if someone could send them to me, I would greatly appreciate it.
Thank you in advance!
André Felipe
Post by Jean-Luc CECCOLI
Hello,
I've tried your profiles with my D7200, and I can confirm that the
results are much closer to the camera jpeg.
I can see a slight blue/green cast and the colors are less
saturated, but it's a much better starting point than the base curve.
I'll have to generate my own profiles to be sure (perhaps D7200
is a bit different from D7100 ?).
Thanks to Edgardo and Vidar !
On Mon, 1 Aug 2016 15:10:22 -0300, Edgardo Hoszowski
Post by Edgardo Hoszowski
Sure, see attached
Hi Edgardo,
I wrote you related about 7100 icc profiles. I have a mac
updated to capitan and I couldn't install view NX 2 to
generate the profiles.
Could you send me your 7100 generated profiles?
Thanks!
2016-07-31 19:13 GMT+02:00 Edgardo Hoszowski
Post by Edgardo Hoszowski
Same here with a D5100, with a D7100 all went fine, I have 2
or 3 different profiles that give different colors cast, but
so far there's always one closer to the camera jpg.
Using the D7100 profiles on a D5100 raw seems ok too, much
closer to the camera jpg.
2016-07-31 8:18 GMT-03:00 Colin Adams
Post by Colin Adams
So I tried this (I have a Nikon D810).
When selecting any of the generated icc profiles except the
"unsupported input profile has been replaced by linear
Rec709 RGB!" So I selected the last one, and set unbreak
input profile to 0.0 as you instruct.
The resulting image indeed looks to have slightly better
quality, but it has a green cast.
This is with darktable 2.0.5-1 rather than 2.0.4. I'm
doubtful if that should be making the difference though.
On Wed, 27 Jul 2016 at 15:39 Jean-Luc CECCOLI <
Post by Jean-Luc CECCOLI
Hello,
Very interesting.
Thanks for sharing.
Rgds,
J.-Luc
Message du 26/07/16 20:41
Objet : Re: [darktable-user] A better starting-point for
Nikon
RAW-files
Post by Jean-Luc CECCOLI
Hello,
Hi!
Post by Jean-Luc CECCOLI
Just one thing that puzzles me : Nikon ViewNX2
generates a profile
for
Post by Jean-Luc CECCOLI
each picture it processes, so each
profile is based upon the parameters of this picture.
Almost. It generates a different profile for each camera
setting (it seems like). So of mine 98 pictures, I got 10
different profiles.
Post by Jean-Luc CECCOLI
That would mean that, for each picture, one would have
to use the corresponding imported profile, else one
might get even worse results than using the DT
dedicated basecurve, isn't it ?
After applying all 10 different profiles to the same image, I
discovered
it was only minor differences, and so small I can neglect
them. I
have
never seen them make a worse result than the dedicated
base curves.
But
this is a subjective observation!
Regards,
Vidar Hoel
Post by Jean-Luc CECCOLI
That said, I think that that is the way profiles work...
Regards,
J.-Luc
Message du 26/07/16 18:45
Objet : [darktable-user] A better starting-point for
Nikon
RAW-files
Post by Jean-Luc CECCOLI
Hi,
some time ago I noticed that the JPGs from my Nikon
were better
then
Post by Jean-Luc CECCOLI
what I could achieve from processing the RAW-file
(NEF) in
Darktable.
Post by Jean-Luc CECCOLI
After some digging on the Internet, I figured out I
could use an alternative to the default base-curves
in Darktable
- and that
made me
get a better starting-point when processing my Nikon
RAW-files
(and I
Post by Jean-Luc CECCOLI
was able to achieve a better image than the JPGs
from my camera!)
Today I finished a step-by-step video of what I did,
so anyone can
see,
learn and perhaps do it themselves.
- I am not really sure behind the technology behind
this. I am not
sure
if a Nikon RAW-file has a profile embedded, but it
seems so.
- I am not sure if there is an other way to achieve
this is
Darktable?
Post by Jean-Luc CECCOLI
- As explained in the video, you will need Windows,
or a friends
laptop
with Windows, once, to make this work.
- My native language is not English :-)
http://youtu.be/O1nS3URVSYA
Enjoy :-)
Best regards,
Vidar Hoel
____________________________________________________________________________
Post by Colin Adams
Post by Bernhard
Post by Bernhard
Post by Jean-Luc CECCOLI
Post by Edgardo Hoszowski
Post by Edgardo Hoszowski
Post by Colin Adams
Post by Jean-Luc CECCOLI
Post by Jean-Luc CECCOLI
darktable user mailing list
to unsubscribe send a mail to
____________________________________________________________________________
Post by Colin Adams
Post by Bernhard
Post by Bernhard
Post by Jean-Luc CECCOLI
Post by Edgardo Hoszowski
Post by Edgardo Hoszowski
Post by Colin Adams
Post by Jean-Luc CECCOLI
darktable user mailing list
to unsubscribe send a mail to
____________________________________________________________________________
Post by Colin Adams
Post by Bernhard
Post by Bernhard
Post by Jean-Luc CECCOLI
Post by Edgardo Hoszowski
Post by Edgardo Hoszowski
Post by Colin Adams
Post by Jean-Luc CECCOLI
darktable user mailing list to unsubscribe send a mail to
____________________________________________________________________________
Post by Colin Adams
Post by Bernhard
Post by Bernhard
Post by Jean-Luc CECCOLI
Post by Edgardo Hoszowski
Post by Edgardo Hoszowski
Post by Colin Adams
darktable user mailing list to unsubscribe send a mail to
____________________________________________________________________________
Post by Colin Adams
Post by Bernhard
Post by Bernhard
Post by Jean-Luc CECCOLI
Post by Edgardo Hoszowski
Post by Edgardo Hoszowski
darktable user mailing list to unsubscribe send a mail to
____________________________________________________________________________
Post by Colin Adams
Post by Bernhard
Post by Bernhard
Post by Jean-Luc CECCOLI
Post by Edgardo Hoszowski
darktable user mailing list
to unsubscribe send a mail to
____________________________________________________________________________
Post by Colin Adams
Post by Bernhard
Post by Bernhard
Post by Jean-Luc CECCOLI
darktable user mailing list
to unsubscribe send a mail to
--
André Felipe
https://www.flickr.com/photos/andrefelipecarvalho/
____________________________________________________________________________
Post by Colin Adams
Post by Bernhard
Post by Bernhard
darktable user mailing list to unsubscribe send a mail to
--
regards
Bernhard
http://www.bilddateien.de
____________________________________________________________________________
Post by Colin Adams
Post by Bernhard
Post by Bernhard
darktable user mailing list to unsubscribe send a mail to
____________________________________________________________________________
Post by Colin Adams
Post by Bernhard
darktable user mailing list to unsubscribe send a mail to
____________________________________________________________________________
Post by Colin Adams
Post by Bernhard
darktable user mailing list to unsubscribe send a mail to
____________________________________________________________________________
Post by Colin Adams
darktable user mailing list
to unsubscribe send a mail to
____________________________________________________________________________
darktable user mailing list
to unsubscribe send a mail to
Christian Drechsler
2016-08-05 10:42:07 UTC
Permalink
I tried it now, too, but I must say that (at least for my D610) I had expected a bit more.

I let Nikon's software work on more than 500 files and got 60 differing profiles. Most of them looked more or less the same, three were not usable in Darktable, and one comes out completely green—not just a little greenish, but almost looking like a b/w image looked at through a green filter. I wonder which image this profile was for. :-)

I'll keep one of the profiles as a starting point if I have images with difficult colors, but in general, for me and my camera, using one of the Nikon-like, Nikon-like alternative, or neutral base curves seems to be a better starting point.

For the images I looked at (mostly indoor people photography), the profiles had much too few saturation and contrast (and decreasing the linear value to something above 0 alone didn't yield acceptable results, too), so using them would mean I'd have to fiddle with those settings for every image. I'm much quicker with the base curves, choosing one of the three that fits the image, then some exposure correction, and that's it for the color settings of most images (at least for those that don't need to be perfect).

But I guess it's different for different camera models, and probably for different lighting situations, too.

Best regards,

Christian
Saint Germain
2016-08-07 14:06:50 UTC
Permalink
I tried also with rawtherapee, and I can use the icm profiles without
any problem, so there may be something wrong with darktable.

However with rawtherapee, the results is the same as with darktable and
the D7100 icm profile: the result is quite nice, but there is always a
light green tint.

On Wed, 03 Aug 2016 19:13:36 +0000, Colin Adams
Post by Colin Adams
Yes, those were the names I saw too (except 810 rather than 7200 of course).
Post by Saint Germain
I tried today to generate the icm profile for my D7200.
First ViewNX 2 is not compatible with D7200, so I had to use Capture
NX-D, however the process is otherwise identical.
Generating the jpeg for one single nef file, I got as a result 5 icm
- 4 starting with Nkx_D7200_xxx
- 1 starting with the name Nkx_LinearYcc.icm
After remove the duplicates, I ended up with 4 different icm
profiles.
- 3 have the colorspace YCbCr and could not be used by darktable
("unsupported input profile has been replaced by linear Rec709 RGB!").
- 1 has the colorspace RGB and could be used by darktable but I
got completely incorrect color (like it was the negative of the
picture, like in old times).
Usuing gcm-view I have exactly the same result as Colin (error on
the one with colorspace RGB).
So it seems that I can not use this method to generate an input
profile for my D7200 ?
Or have I done something wrong ?
Thanks
On Wed, 03 Aug 2016 06:26:39 +0000, Colin Adams
Post by Colin Adams
OK, thanks.
I did that and then used gcm-view to see if there were any
problems. All of them except one were colorspace YCbCr and no
problems were reported. One was colorspace RGB and has the
• A scum dot is present for media white".
So "unsupported input profile has been replaced by linear Rec709
RGB!" from darktable is presumably because it is not supporting
the YCbCr colorspace. Would that be correct? And if so, is there
a way to convert to another colorspace?
On Tue, 2 Aug 2016 at 22:04 Bernhard
Post by Bernhard
either you double-click in Nautilus or Nemo or type
bash$ gcm-import <filepath>
in a terminal and you should see the attached dialog, you click
"view details" and can start investigating ...
--
regards
Bernhard
http://www.bilddateien.de
What's the command to open a .icm file in gnome-color-manager?
I'd like to look at one of my Nikon D810 files.
On Tue, 2 Aug 2016 at 20:37 Bernhard
Post by Bernhard
I opened this in gnome color manager and viewed the profile.
• The white of the output medium is distorted by blobs of color
Nevertheless it's interesting to view the gamut and color
curves of the profile ...
DispCalGUI refuses to open the files.
--
regards
Bernhard
http://www.bilddateien.de
Hello,
I've following this thread with great interest, since I have a
Nikon 7200 and I was even thinking about developing from jpegs
instead of raws.
Could you please instruct me with what to do with those .icm
files? e.g., where to put it in my system. I know it's for the
D7100, but since Saint Germain reccomended it, I would like to
git it a try.
And about the generation of .icm files specifically for the
D7200, if someone could send them to me, I would greatly
appreciate it.
Thank you in advance!
André Felipe
Post by Jean-Luc CECCOLI
Hello,
I've tried your profiles with my D7200, and I can confirm
that the results are much closer to the camera jpeg.
I can see a slight blue/green cast and the colors are less
saturated, but it's a much better starting point than the base curve.
I'll have to generate my own profiles to be sure (perhaps
D7200 is a bit different from D7100 ?).
Thanks to Edgardo and Vidar !
On Mon, 1 Aug 2016 15:10:22 -0300, Edgardo Hoszowski
Post by Edgardo Hoszowski
Sure, see attached
2016-08-01 5:46 GMT-03:00 Rafa García
Hi Edgardo,
I wrote you related about 7100 icc profiles. I have a
mac updated to capitan and I couldn't install view NX 2 to
generate the profiles.
Could you send me your 7100 generated profiles?
Thanks!
2016-07-31 19:13 GMT+02:00 Edgardo Hoszowski
Post by Edgardo Hoszowski
Same here with a D5100, with a D7100 all went fine, I
have 2 or 3 different profiles that give different
colors cast, but so far there's always one closer to the
camera jpg. Using the D7100 profiles on a D5100 raw
seems ok too, much closer to the camera jpg.
2016-07-31 8:18 GMT-03:00 Colin Adams
Post by Colin Adams
So I tried this (I have a Nikon D810).
When selecting any of the generated icc profiles except
"unsupported input profile has been replaced by linear
Rec709 RGB!" So I selected the last one, and set unbreak
input profile to 0.0 as you instruct.
The resulting image indeed looks to have slightly better
quality, but it has a green cast.
This is with darktable 2.0.5-1 rather than 2.0.4. I'm
doubtful if that should be making the difference though.
On Wed, 27 Jul 2016 at 15:39 Jean-Luc CECCOLI <
Post by Jean-Luc CECCOLI
Hello,
Very interesting.
Thanks for sharing.
Rgds,
J.-Luc
Message du 26/07/16 20:41
Objet : Re: [darktable-user] A better starting-point
for Nikon
RAW-files
Post by Jean-Luc CECCOLI
Hello,
Hi!
Post by Jean-Luc CECCOLI
Just one thing that puzzles me : Nikon ViewNX2
generates a profile
for
Post by Jean-Luc CECCOLI
each picture it processes, so each
profile is based upon the parameters of this picture.
Almost. It generates a different profile for each
camera setting (it seems like). So of mine 98
pictures, I got 10 different profiles.
Post by Jean-Luc CECCOLI
That would mean that, for each picture, one would
have to use the corresponding imported profile,
else one might get even worse results than using
the DT dedicated basecurve, isn't it ?
After applying all 10 different profiles to the same
image, I
discovered
it was only minor differences, and so small I can
neglect them. I
have
never seen them make a worse result than the
dedicated base curves.
But
this is a subjective observation!
Regards,
Vidar Hoel
Post by Jean-Luc CECCOLI
That said, I think that that is the way profiles
work...
Regards,
J.-Luc
Message du 26/07/16 18:45
Objet : [darktable-user] A better starting-point
for Nikon
RAW-files
Post by Jean-Luc CECCOLI
Hi,
some time ago I noticed that the JPGs from my
Nikon were better
then
Post by Jean-Luc CECCOLI
what I could achieve from processing the RAW-file
(NEF) in
Darktable.
Post by Jean-Luc CECCOLI
After some digging on the Internet, I figured
out I could use an alternative to the default
base-curves in Darktable
- and that
made me
get a better starting-point when processing my
Nikon RAW-files
(and I
Post by Jean-Luc CECCOLI
was able to achieve a better image than the JPGs
from my camera!)
Today I finished a step-by-step video of what I
did, so anyone can
see,
learn and perhaps do it themselves.
- I am not really sure behind the technology
behind this. I am not
sure
if a Nikon RAW-file has a profile embedded, but
it seems so.
- I am not sure if there is an other way to
achieve this is
Darktable?
Post by Jean-Luc CECCOLI
- As explained in the video, you will need
Windows, or a friends
laptop
with Windows, once, to make this work.
- My native language is not English :-)
http://youtu.be/O1nS3URVSYA
Enjoy :-)
Best regards,
Vidar Hoel
Johann Spies
2016-08-10 14:10:18 UTC
Permalink
View NX2 also runs on Apple. As I do not have Windows, I have tried to do
the same on the Macbook, but I could not find the temporary files. Does
somebody know where to find them on the Mac?

Regards
Johann
Post by Saint Germain
I tried also with rawtherapee, and I can use the icm profiles without
any problem, so there may be something wrong with darktable.
However with rawtherapee, the results is the same as with darktable and
the D7100 icm profile: the result is quite nice, but there is always a
light green tint.
On Wed, 03 Aug 2016 19:13:36 +0000, Colin Adams
Post by Colin Adams
Yes, those were the names I saw too (except 810 rather than 7200 of course).
Post by Saint Germain
I tried today to generate the icm profile for my D7200.
First ViewNX 2 is not compatible with D7200, so I had to use Capture
NX-D, however the process is otherwise identical.
Generating the jpeg for one single nef file, I got as a result 5 icm
- 4 starting with Nkx_D7200_xxx
- 1 starting with the name Nkx_LinearYcc.icm
After remove the duplicates, I ended up with 4 different icm profiles.
- 3 have the colorspace YCbCr and could not be used by darktable
("unsupported input profile has been replaced by linear Rec709 RGB!").
- 1 has the colorspace RGB and could be used by darktable but I
got completely incorrect color (like it was the negative of the
picture, like in old times).
Usuing gcm-view I have exactly the same result as Colin (error on
the one with colorspace RGB).
So it seems that I can not use this method to generate an input
profile for my D7200 ?
Or have I done something wrong ?
Thanks
On Wed, 03 Aug 2016 06:26:39 +0000, Colin Adams
Post by Colin Adams
OK, thanks.
I did that and then used gcm-view to see if there were any
problems. All of them except one were colorspace YCbCr and no
problems were reported. One was colorspace RGB and has the
• A scum dot is present for media white".
So "unsupported input profile has been replaced by linear Rec709
RGB!" from darktable is presumably because it is not supporting
the YCbCr colorspace. Would that be correct? And if so, is there
a way to convert to another colorspace?
On Tue, 2 Aug 2016 at 22:04 Bernhard
Post by Bernhard
either you double-click in Nautilus or Nemo or type
bash$ gcm-import <filepath>
in a terminal and you should see the attached dialog, you click
"view details" and can start investigating ...
--
regards
Bernhard
http://www.bilddateien.de
What's the command to open a .icm file in gnome-color-manager?
I'd like to look at one of my Nikon D810 files.
On Tue, 2 Aug 2016 at 20:37 Bernhard
Post by Bernhard
I opened this in gnome color manager and viewed the profile.
• The white of the output medium is distorted by blobs of color
Nevertheless it's interesting to view the gamut and color
curves of the profile ...
DispCalGUI refuses to open the files.
--
regards
Bernhard
http://www.bilddateien.de
Hello,
I've following this thread with great interest, since I have a
Nikon 7200 and I was even thinking about developing from jpegs
instead of raws.
Could you please instruct me with what to do with those .icm
files? e.g., where to put it in my system. I know it's for the
D7100, but since Saint Germain reccomended it, I would like to
git it a try.
And about the generation of .icm files specifically for the
D7200, if someone could send them to me, I would greatly
appreciate it.
Thank you in advance!
André Felipe
Post by Jean-Luc CECCOLI
Hello,
I've tried your profiles with my D7200, and I can confirm
that the results are much closer to the camera jpeg.
I can see a slight blue/green cast and the colors are less
saturated, but it's a much better starting point than the base curve.
I'll have to generate my own profiles to be sure (perhaps
D7200 is a bit different from D7100 ?).
Thanks to Edgardo and Vidar !
On Mon, 1 Aug 2016 15:10:22 -0300, Edgardo Hoszowski
Post by Edgardo Hoszowski
Sure, see attached
2016-08-01 5:46 GMT-03:00 Rafa García
Hi Edgardo,
I wrote you related about 7100 icc profiles. I have a
mac updated to capitan and I couldn't install view NX 2 to
generate the profiles.
Could you send me your 7100 generated profiles?
Thanks!
2016-07-31 19:13 GMT+02:00 Edgardo Hoszowski
Post by Edgardo Hoszowski
Same here with a D5100, with a D7100 all went fine, I
have 2 or 3 different profiles that give different
colors cast, but so far there's always one closer to the
camera jpg. Using the D7100 profiles on a D5100 raw
seems ok too, much closer to the camera jpg.
2016-07-31 8:18 GMT-03:00 Colin Adams
Post by Colin Adams
So I tried this (I have a Nikon D810).
When selecting any of the generated icc profiles except
"unsupported input profile has been replaced by linear
Rec709 RGB!" So I selected the last one, and set unbreak
input profile to 0.0 as you instruct.
The resulting image indeed looks to have slightly better
quality, but it has a green cast.
This is with darktable 2.0.5-1 rather than 2.0.4. I'm
doubtful if that should be making the difference though.
On Wed, 27 Jul 2016 at 15:39 Jean-Luc CECCOLI <
Post by Jean-Luc CECCOLI
Hello,
Very interesting.
Thanks for sharing.
Rgds,
J.-Luc
Message du 26/07/16 20:41
Objet : Re: [darktable-user] A better starting-point
for Nikon
RAW-files
Post by Jean-Luc CECCOLI
Hello,
Hi!
Post by Jean-Luc CECCOLI
Just one thing that puzzles me : Nikon ViewNX2
generates a profile
for
Post by Jean-Luc CECCOLI
each picture it processes, so each
profile is based upon the parameters of this
picture.
Almost. It generates a different profile for each
camera setting (it seems like). So of mine 98
pictures, I got 10 different profiles.
Post by Jean-Luc CECCOLI
That would mean that, for each picture, one would
have to use the corresponding imported profile,
else one might get even worse results than using
the DT dedicated basecurve, isn't it ?
After applying all 10 different profiles to the same
image, I
discovered
it was only minor differences, and so small I can
neglect them. I
have
never seen them make a worse result than the
dedicated base curves.
But
this is a subjective observation!
Regards,
Vidar Hoel
Post by Jean-Luc CECCOLI
That said, I think that that is the way profiles
work...
Regards,
J.-Luc
Message du 26/07/16 18:45
Objet : [darktable-user] A better starting-point
for Nikon
RAW-files
Post by Jean-Luc CECCOLI
Hi,
some time ago I noticed that the JPGs from my
Nikon were better
then
Post by Jean-Luc CECCOLI
what I could achieve from processing the RAW-file
(NEF) in
Darktable.
Post by Jean-Luc CECCOLI
After some digging on the Internet, I figured
out I could use an alternative to the default
base-curves in Darktable
- and that
made me
get a better starting-point when processing my
Nikon RAW-files
(and I
Post by Jean-Luc CECCOLI
was able to achieve a better image than the JPGs
from my camera!)
Today I finished a step-by-step video of what I
did, so anyone can
see,
learn and perhaps do it themselves.
- I am not really sure behind the technology
behind this. I am not
sure
if a Nikon RAW-file has a profile embedded, but
it seems so.
- I am not sure if there is an other way to
achieve this is
Darktable?
Post by Jean-Luc CECCOLI
- As explained in the video, you will need
Windows, or a friends
laptop
with Windows, once, to make this work.
- My native language is not English :-)
http://youtu.be/O1nS3URVSYA
Enjoy :-)
Best regards,
Vidar Hoel
____________________________________________________________
________________
darktable user mailing list
lists.darktable.org
--
Because experiencing your loyal love is better than life itself,
my lips will praise you. (Psalm 63:3)
Vidar Hoel
2016-08-10 14:57:21 UTC
Permalink
Hi,

I do not know MacOS, but the trick is to search for files named *.icm,
and changed/modified within the last minutes. I am sure this is possible
on a Mac, but don't know how! Good luck.

Regards,
Vidar Hoel
Post by Johann Spies
View NX2 also runs on Apple. As I do not have Windows, I have tried to
do the same on the Macbook, but I could not find the temporary files.
Does somebody know where to find them on the Mac?
Regards
Johann
I tried also with rawtherapee, and I can use the icm profiles without
any problem, so there may be something wrong with darktable.
However with rawtherapee, the results is the same as with darktable and
the D7100 icm profile: the result is quite nice, but there is always a
light green tint.
On Wed, 03 Aug 2016 19:13:36 +0000, Colin Adams
Post by Colin Adams
Yes, those were the names I saw too (except 810 rather than 7200 of course).
Post by Saint Germain
I tried today to generate the icm profile for my D7200.
First ViewNX 2 is not compatible with D7200, so I had to use Capture
NX-D, however the process is otherwise identical.
Generating the jpeg for one single nef file, I got as a result 5 icm
- 4 starting with Nkx_D7200_xxx
- 1 starting with the name Nkx_LinearYcc.icm
After remove the duplicates, I ended up with 4 different icm profiles.
- 3 have the colorspace YCbCr and could not be used by darktable
("unsupported input profile has been replaced by linear Rec709 RGB!").
- 1 has the colorspace RGB and could be used by darktable but I
got completely incorrect color (like it was the negative of the
picture, like in old times).
Usuing gcm-view I have exactly the same result as Colin (error on
the one with colorspace RGB).
So it seems that I can not use this method to generate an input
profile for my D7200 ?
Or have I done something wrong ?
Thanks
On Wed, 03 Aug 2016 06:26:39 +0000, Colin Adams
Post by Colin Adams
OK, thanks.
I did that and then used gcm-view to see if there were any
problems. All of them except one were colorspace YCbCr and no
problems were reported. One was colorspace RGB and has the
• A scum dot is present for media white".
So "unsupported input profile has been replaced by linear Rec709
RGB!" from darktable is presumably because it is not supporting
the YCbCr colorspace. Would that be correct? And if so, is there
a way to convert to another colorspace?
On Tue, 2 Aug 2016 at 22:04 Bernhard
Post by Bernhard
either you double-click in Nautilus or Nemo or type
bash$ gcm-import <filepath>
in a terminal and you should see the attached dialog, you click
"view details" and can start investigating ...
--
regards
Bernhard
http://www.bilddateien.de
What's the command to open a .icm file in gnome-color-manager?
I'd like to look at one of my Nikon D810 files.
On Tue, 2 Aug 2016 at 20:37 Bernhard
Post by Bernhard
I opened this in gnome color manager and viewed the profile.
• The white of the output medium is distorted by blobs of color
Nevertheless it's interesting to view the gamut and color
curves of the profile ...
DispCalGUI refuses to open the files.
--
regards
Bernhard
http://www.bilddateien.de
Hello,
I've following this thread with great interest, since I have a
Nikon 7200 and I was even thinking about developing from jpegs
instead of raws.
Could you please instruct me with what to do with those .icm
files? e.g., where to put it in my system. I know it's for the
D7100, but since Saint Germain reccomended it, I would like to
git it a try.
And about the generation of .icm files specifically for the
D7200, if someone could send them to me, I would greatly
appreciate it.
Thank you in advance!
André Felipe
2016-08-01 21:59 GMT-03:00 Saint Germain
Post by Jean-Luc CECCOLI
Hello,
I've tried your profiles with my D7200, and I can confirm
that the results are much closer to the camera jpeg.
I can see a slight blue/green cast and the colors are less
saturated, but it's a much better starting point than the base
curve.
I'll have to generate my own profiles to be sure (perhaps
D7200 is a bit different from D7100 ?).
Thanks to Edgardo and Vidar !
On Mon, 1 Aug 2016 15:10:22 -0300, Edgardo Hoszowski
Post by Edgardo Hoszowski
Sure, see attached
2016-08-01 5:46 GMT-03:00 Rafa García
Hi Edgardo,
I wrote you related about 7100 icc profiles. I have a
mac updated to capitan and I couldn't install view NX 2 to
generate the profiles.
Could you send me your 7100 generated profiles?
Thanks!
2016-07-31 19:13 GMT+02:00 Edgardo Hoszowski
Post by Edgardo Hoszowski
Same here with a D5100, with a D7100 all went fine, I
have 2 or 3 different profiles that give different
colors cast, but so far there's always one closer to the
camera jpg. Using the D7100 profiles on a D5100 raw
seems ok too, much closer to the camera jpg.
2016-07-31 8:18 GMT-03:00 Colin Adams
Post by Colin Adams
So I tried this (I have a Nikon D810).
When selecting any of the generated icc profiles except
"unsupported input profile has been replaced by linear
Rec709 RGB!" So I selected the last one, and set unbreak
input profile to 0.0 as you instruct.
The resulting image indeed looks to have slightly better
quality, but it has a green cast.
This is with darktable 2.0.5-1 rather than 2.0.4. I'm
doubtful if that should be making the difference though.
On Wed, 27 Jul 2016 at 15:39 Jean-Luc CECCOLI <
Post by Jean-Luc CECCOLI
Hello,
Very interesting.
Thanks for sharing.
Rgds,
J.-Luc
Message du 26/07/16 20:41
A : "Jean-Luc CECCOLI"
Objet : Re: [darktable-user] A better starting-point
for Nikon
RAW-files
Post by Jean-Luc CECCOLI
Hello,
Hi!
Post by Jean-Luc CECCOLI
Just one thing that puzzles me : Nikon ViewNX2
generates a profile
for
Post by Jean-Luc CECCOLI
each picture it processes, so each
profile is based upon the parameters of this
picture.
Almost. It generates a different profile for each
camera setting (it seems like). So of mine 98
pictures, I got 10 different profiles.
Post by Jean-Luc CECCOLI
That would mean that, for each picture, one would
have to use the corresponding imported profile,
else one might get even worse results than using
the DT dedicated basecurve, isn't it ?
After applying all 10 different profiles to the same
image, I
discovered
it was only minor differences, and so small I can
neglect them. I
have
never seen them make a worse result than the
dedicated base curves.
But
this is a subjective observation!
Regards,
Vidar Hoel
Post by Jean-Luc CECCOLI
That said, I think that that is the way profiles
work...
Regards,
J.-Luc
Message du 26/07/16 18:45
Objet : [darktable-user] A better starting-point
for Nikon
RAW-files
Post by Jean-Luc CECCOLI
Hi,
some time ago I noticed that the JPGs from my
Nikon were better
then
Post by Jean-Luc CECCOLI
what I could achieve from processing the RAW-file
(NEF) in
Darktable.
Post by Jean-Luc CECCOLI
After some digging on the Internet, I figured
out I could use an alternative to the default
base-curves in Darktable
- and that
made me
get a better starting-point when processing my
Nikon RAW-files
(and I
Post by Jean-Luc CECCOLI
was able to achieve a better image than the JPGs
from my camera!)
Today I finished a step-by-step video of what I
did, so anyone can
see,
learn and perhaps do it themselves.
- I am not really sure behind the technology
behind this. I am not
sure
if a Nikon RAW-file has a profile embedded, but
it seems so.
- I am not sure if there is an other way to
achieve this is
Darktable?
Post by Jean-Luc CECCOLI
- As explained in the video, you will need
Windows, or a friends
laptop
with Windows, once, to make this work.
- My native language is not English :-)
http://youtu.be/O1nS3URVSYA
Enjoy :-)
Best regards,
Vidar Hoel
____________________________________________________________________________
darktable user mailing list
to unsubscribe send a mail to
--
Because experiencing your loyal love is better than life itself,
my lips will praise you. (Psalm 63:3)
____________________________________________________________________________
darktable user mailing list to unsubscribe send a mail to
Jean-Luc CECCOLI
2016-08-10 20:45:18 UTC
Permalink
Hello,

 

Maybe into some temp folder within the user's folder after allowing viewing hiden folders ?

 

Rgrds,

 

J.-Luc

 

 

 

 
Message du 10/08/16 16:57
De : "Vidar Hoel"
Objet : Re: [darktable-user] A better starting-point for Nikon RAW-files
Hi,
I do not know MacOS, but the trick is to search for files named *.icm,
and changed/modified within the last minutes. I am sure this is possible
on a Mac, but don't know how! Good luck.
Regards,
Vidar Hoel
View NX2 also runs on Apple. As I do not have Windows, I have tried to
do the same on the Macbook, but I could not find the temporary files.
Does somebody know where to find them on the Mac?
Regards
Johann
I tried also with rawtherapee, and I can use the icm profiles without
any problem, so there may be something wrong with darktable.
However with rawtherapee, the results is the same as with darktable and
the D7100 icm profile: the result is quite nice, but there is always a
light green tint.
On Wed, 03 Aug 2016 19:13:36 +0000, Colin Adams
Post by Colin Adams
Yes, those were the names I saw too (except 810 rather than 7200 of course).
Post by Saint Germain
I tried today to generate the icm profile for my D7200.
First ViewNX 2 is not compatible with D7200, so I had to use Capture
NX-D, however the process is otherwise identical.
Generating the jpeg for one single nef file, I got as a result 5 icm
- 4 starting with Nkx_D7200_xxx
- 1 starting with the name Nkx_LinearYcc.icm
After remove the duplicates, I ended up with 4 different icm profiles.
- 3 have the colorspace YCbCr and could not be used by darktable
("unsupported input profile has been replaced by linear Rec709 RGB!").
- 1 has the colorspace RGB and could be used by darktable but I
got completely incorrect color (like it was the negative of the
picture, like in old times).
Usuing gcm-view I have exactly the same result as Colin (error on
the one with colorspace RGB).
So it seems that I can not use this method to generate an input
profile for my D7200 ?
Or have I done something wrong ?
Thanks
On Wed, 03 Aug 2016 06:26:39 +0000, Colin Adams
Post by Colin Adams
OK, thanks.
I did that and then used gcm-view to see if there were any
problems. All of them except one were colorspace YCbCr and no
problems were reported. One was colorspace RGB and has the
• A scum dot is present for media white".
So "unsupported input profile has been replaced by linear Rec709
RGB!" from darktable is presumably because it is not supporting
the YCbCr colorspace. Would that be correct? And if so, is there
a way to convert to another colorspace?
On Tue, 2 Aug 2016 at 22:04 Bernhard
Post by Bernhard
either you double-click in Nautilus or Nemo or type
bash$ gcm-import
in a terminal and you should see the attached dialog, you click
"view details" and can start investigating ...
--
regards
Bernhard
http://www.bilddateien.de
What's the command to open a .icm file in gnome-color-manager?
I'd like to look at one of my Nikon D810 files.
On Tue, 2 Aug 2016 at 20:37 Bernhard
Post by Bernhard
I opened this in gnome color manager and viewed the profile.
• The white of the output medium is distorted by blobs of color
Nevertheless it's interesting to view the gamut and color
curves of the profile ...
DispCalGUI refuses to open the files.
--
regards
Bernhard
http://www.bilddateien.de
Hello,
I've following this thread with great interest, since I have a
Nikon 7200 and I was even thinking about developing from jpegs
instead of raws.
Could you please instruct me with what to do with those .icm
files? e.g., where to put it in my system. I know it's for the
D7100, but since Saint Germain reccomended it, I would like to
git it a try.
And about the generation of .icm files specifically for the
D7200, if someone could send them to me, I would greatly
appreciate it.
Thank you in advance!
André Felipe
2016-08-01 21:59 GMT-03:00 Saint Germain
Post by Jean-Luc CECCOLI
Hello,
I've tried your profiles with my D7200, and I can confirm
that the results are much closer to the camera jpeg.
I can see a slight blue/green cast and the colors are less
saturated, but it's a much better starting point than the base
curve.
I'll have to generate my own profiles to be sure (perhaps
D7200 is a bit different from D7100 ?).
Thanks to Edgardo and Vidar !
On Mon, 1 Aug 2016 15:10:22 -0300, Edgardo Hoszowski
Post by Edgardo Hoszowski
Sure, see attached
2016-08-01 5:46 GMT-03:00 Rafa García
Hi Edgardo,
I wrote you related about 7100 icc profiles. I have a
mac updated to capitan and I couldn't install view NX 2 to
generate the profiles.
Could you send me your 7100 generated profiles?
Thanks!
2016-07-31 19:13 GMT+02:00 Edgardo Hoszowski
Post by Edgardo Hoszowski
Same here with a D5100, with a D7100 all went fine, I
have 2 or 3 different profiles that give different
colors cast, but so far there's always one closer to the
camera jpg. Using the D7100 profiles on a D5100 raw
seems ok too, much closer to the camera jpg.
2016-07-31 8:18 GMT-03:00 Colin Adams
Post by Colin Adams
So I tried this (I have a Nikon D810).
When selecting any of the generated icc profiles except
"unsupported input profile has been replaced by linear
Rec709 RGB!" So I selected the last one, and set unbreak
input profile to 0.0 as you instruct.
The resulting image indeed looks to have slightly better
quality, but it has a green cast.
This is with darktable 2.0.5-1 rather than 2.0.4. I'm
doubtful if that should be making the difference though.
On Wed, 27 Jul 2016 at 15:39 Jean-Luc CECCOLI <
Post by Jean-Luc CECCOLI
Hello,
Very interesting.
Thanks for sharing.
Rgds,
J.-Luc
Message du 26/07/16 20:41
De : "Vidar Hoel" > >
A : "Jean-Luc CECCOLI"
,
Post by Saint Germain
Post by Colin Adams
Post by Bernhard
Post by Bernhard
Post by Jean-Luc CECCOLI
Post by Edgardo Hoszowski
Post by Edgardo Hoszowski
Post by Colin Adams
Post by Jean-Luc CECCOLI
Objet : Re: [darktable-user] A better starting-point
for Nikon
RAW-files
Post by Jean-Luc CECCOLI
Hello,
Hi!
Post by Jean-Luc CECCOLI
Just one thing that puzzles me : Nikon ViewNX2
generates a profile
for
Post by Jean-Luc CECCOLI
each picture it processes, so each
profile is based upon the parameters of this
picture.
Almost. It generates a different profile for each
camera setting (it seems like). So of mine 98
pictures, I got 10 different profiles.
Post by Jean-Luc CECCOLI
That would mean that, for each picture, one would
have to use the corresponding imported profile,
else one might get even worse results than using
the DT dedicated basecurve, isn't it ?
After applying all 10 different profiles to the same
image, I
discovered
it was only minor differences, and so small I can
neglect them. I
have
never seen them make a worse result than the
dedicated base curves.
But
this is a subjective observation!
Regards,
Vidar Hoel
Post by Jean-Luc CECCOLI
That said, I think that that is the way profiles
work...
Regards,
J.-Luc
Message du 26/07/16 18:45
De : "Vidar Hoel" > >
Objet : [darktable-user] A better starting-point
for Nikon
RAW-files
Post by Jean-Luc CECCOLI
Hi,
some time ago I noticed that the JPGs from my
Nikon were better
then
Post by Jean-Luc CECCOLI
what I could achieve from processing the RAW-file
(NEF) in
Darktable.
Post by Jean-Luc CECCOLI
After some digging on the Internet, I figured
out I could use an alternative to the default
base-curves in Darktable
- and that
made me
get a better starting-point when processing my
Nikon RAW-files
(and I
Post by Jean-Luc CECCOLI
was able to achieve a better image than the JPGs
from my camera!)
Today I finished a step-by-step video of what I
did, so anyone can
see,
learn and perhaps do it themselves.
- I am not really sure behind the technology
behind this. I am not
sure
if a Nikon RAW-file has a profile embedded, but
it seems so.
- I am not sure if there is an other way to
achieve this is
Darktable?
Post by Jean-Luc CECCOLI
- As explained in the video, you will need
Windows, or a friends
laptop
with Windows, once, to make this work.
- My native language is not English :-)
http://youtu.be/O1nS3URVSYA
Enjoy :-)
Best regards,
Vidar Hoel
____________________________________________________________________________
darktable user mailing list
to unsubscribe send a mail to
--
Because experiencing your loyal love is better than life itself,
my lips will praise you. (Psalm 63:3)
____________________________________________________________________________
darktable user mailing list to unsubscribe send a mail to
____________________________________________________________________________
darktable user mailing list
KOVÁCS István
2016-08-03 14:49:21 UTC
Permalink
https://www.darktable.org/usermanual/ch03s04s03.html.php
"You can also supply your own input ICC profiles and put them into
$DARKTABLE/share/darktable/color/in or $HOME/.config/darktable/color/in.
$DARKTABLE is used here to represent darktable's installation directory and
$HOME your home directory."
KOVÁCS István
2016-08-02 05:34:33 UTC
Permalink
Please note that there are legal concerns regarding these profiles. The
RawTherapee guys, who (AFAIK) were the first ones to cine up with a way to
extract them, never shared them. Please don't share files of questionable
legality on the forum, regardless of size.

Kofa
Colin Adams
2016-08-01 08:26:25 UTC
Permalink
All the *.icm files are 1.6MB in size except one which is 22KB in size -
this one was named quite differently in Windows, so i expect that is the
one that is almost working.

Is it possible to determine why darktable thinks it is not a valid profile?
Post by Colin Adams
So I tried this (I have a Nikon D810).
When selecting any of the generated icc profiles except the last one, I
"unsupported input profile has been replaced by linear Rec709 RGB!"
So I selected the last one, and set unbreak input profile to 0.0 as you
instruct.
The resulting image indeed looks to have slightly better quality, but it
has a green cast.
This is with darktable 2.0.5-1 rather than 2.0.4. I'm doubtful if that
should be making the difference though.
Post by Jean-Luc CECCOLI
Hello,
Very interesting.
Thanks for sharing.
Rgds,
J.-Luc
Message du 26/07/16 20:41
Objet : Re: [darktable-user] A better starting-point for Nikon RAW-files
Post by Jean-Luc CECCOLI
Hello,
Hi!
Post by Jean-Luc CECCOLI
Just one thing that puzzles me : Nikon ViewNX2 generates a profile for
each picture it processes, so each
profile is based upon the parameters of this picture.
Almost. It generates a different profile for each camera setting (it
seems like). So of mine 98 pictures, I got 10 different profiles.
Post by Jean-Luc CECCOLI
That would mean that, for each picture, one would have to use the
corresponding imported profile, else one
might get even worse results than using the DT dedicated basecurve, isn't it ?
After applying all 10 different profiles to the same image, I
discovered
it was only minor differences, and so small I can neglect them. I have
never seen them make a worse result than the dedicated base curves. But
this is a subjective observation!
Regards,
Vidar Hoel
Post by Jean-Luc CECCOLI
That said, I think that that is the way profiles work...
Regards,
J.-Luc
Message du 26/07/16 18:45
Objet : [darktable-user] A better starting-point for Nikon RAW-files
Hi,
some time ago I noticed that the JPGs from my Nikon were better then
what I could achieve from processing the RAW-file (NEF) in
Darktable.
Post by Jean-Luc CECCOLI
After some digging on the Internet, I figured out I could use an
alternative to the default base-curves in Darktable - and that
made me
get a better starting-point when processing my Nikon RAW-files (and
I
Post by Jean-Luc CECCOLI
was able to achieve a better image than the JPGs from my camera!)
Today I finished a step-by-step video of what I did, so anyone can
see,
learn and perhaps do it themselves.
- I am not really sure behind the technology behind this. I am not
sure
if a Nikon RAW-file has a profile embedded, but it seems so.
- I am not sure if there is an other way to achieve this is
Darktable?
Post by Jean-Luc CECCOLI
- As explained in the video, you will need Windows, or a friends
laptop
with Windows, once, to make this work.
- My native language is not English :-)
http://youtu.be/O1nS3URVSYA
Enjoy :-)
Best regards,
Vidar Hoel
____________________________________________________________________________
Post by Jean-Luc CECCOLI
darktable user mailing list
to unsubscribe send a mail to
____________________________________________________________________________
darktable user mailing list
to unsubscribe send a mail to
____________________________________________________________________________
darktable user mailing list to unsubscribe send a mail to
Jean-Luc CECCOLI
2016-07-31 16:34:07 UTC
Permalink
Hello,

 

So, I could try this, and was very impressed by the results.

It seems that decreasing the Linear value down to "0" gives results nearer to the actual subject,
whereas letting it to its default value gives more saturated colors, maybe as an attempt to match
the tastes of most people.
I could describe this as accuracy versus feeling.
Some other thing I could notice : after examining the generated icc files, I was expecting the results
to depend upon the sensitivity, however this seems not to be the case, as you stated in your previous
reply. And this seems very strange to me, as each profile seems to be built taking account of :

- 1st group of characters : maker (Nkx)

- 2nd group : camera model (2 to 5~6 digits)

- 3rd group : camera S/N

- 4th & 5th group : resp. always 01 and 1 so far

- 6th group : 2 digits 03 or 10

- 7th group : 1 digit, values so far are 0, 1, 2, 3 or 9

- 8th to 12th group values : resp. always 00, 10, 00, 00, 00 and 00 so far
- 13th group : always 05 or 06
- 14th group : always 00
- 15th group : 4 digits, sensitivity
- 16th to 19th groups : always resp. 0, 0, 32 and 476 for me.

I really don't know what to think about this all...

Rgrds,

J.-Luc

 

 

 
Message du 27/07/16 16:39
De : "Jean-Luc CECCOLI"
Objet : Re: [darktable-user] A better starting-point for Nikon RAW-files
Hello,
 
Very interesting.
Thanks for sharing.
 
Rgds,
 
J.-Luc
 
 
Message du 26/07/16 20:41
De : "Vidar Hoel"
Objet : Re: [darktable-user] A better starting-point for Nikon RAW-files
Post by Jean-Luc CECCOLI
Hello,
Hi!
Post by Jean-Luc CECCOLI
Just one thing that puzzles me : Nikon ViewNX2 generates a profile for
each picture it processes, so each
profile is based upon the parameters of this picture.
Almost. It generates a different profile for each camera setting (it
seems like). So of mine 98 pictures, I got 10 different profiles.
Post by Jean-Luc CECCOLI
That would mean that, for each picture, one would have to use the
corresponding imported profile, else one
might get even worse results than using the DT dedicated basecurve, isn't it ?
After applying all 10 different profiles to the same image, I discovered
it was only minor differences, and so small I can neglect them. I have
never seen them make a worse result than the dedicated base curves. But
this is a subjective observation!
Regards,
Vidar Hoel
Post by Jean-Luc CECCOLI
That said, I think that that is the way profiles work...
Regards,
J.-Luc
Message du 26/07/16 18:45
De : "Vidar Hoel"
Objet : [darktable-user] A better starting-point for Nikon RAW-files
Hi,
some time ago I noticed that the JPGs from my Nikon were better then
what I could achieve from processing the RAW-file (NEF) in Darktable.
After some digging on the Internet, I figured out I could use an
alternative to the default base-curves in Darktable - and that
made me
get a better starting-point when processing my Nikon RAW-files (and I
was able to achieve a better image than the JPGs from my camera!)
Today I finished a step-by-step video of what I did, so anyone can
see,
learn and perhaps do it themselves.
- I am not really sure behind the technology behind this. I am not
sure
if a Nikon RAW-file has a profile embedded, but it seems so.
- I am not sure if there is an other way to achieve this is Darktable?
- As explained in the video, you will need Windows, or a friends
laptop
with Windows, once, to make this work.
- My native language is not English :-)
http://youtu.be/O1nS3URVSYA
Enjoy :-)
Best regards,
Vidar Hoel
____________________________________________________________________________
darktable user mailing list
to unsubscribe send a mail to
____________________________________________________________________________
darktable user mailing list
Loading...