What am I missing? (16 bit, etc.)

classic Classic list List threaded Threaded
6 messages Options
Reply | Threaded
Open this post in threaded view
|

What am I missing? (16 bit, etc.)

Paul Verizzo
I've decided to get out of my comfort zone and start experimenting with 16 bit large sRGB color spaces now that I have a
new printer with a pretty wide gamut (Canon PRO-100.)

So I take an MRW image, select to work in ProPhoto RGB (Why not go all the way?  Been reading about it, especially on
Luminous Landscapte.)  I make a few adjustments, such as Levels, then select the 8>16 bit option in Depth.  When I try
to save it as either JPG or TIFF, the file is kept at 8 bits.  When I try a PNG version, it fails.  I'm sorry, I don't
have the program up to tell you exactly what the failure message is.

I've spent a lot of time in the online help and the Help PDF's, and I can't find anything that specifically address my
simple workflow or why it doesn't work.

Thanks, Paul
_______________________________________________
Digikam-users mailing list
[hidden email]
https://mail.kde.org/mailman/listinfo/digikam-users
Reply | Threaded
Open this post in threaded view
|

Re: What am I missing? (16 bit, etc.)

carl33914
I am fairly certain that JPG (the standard)  is limited to 8 bit x 3 channels (RGB)= 24 bit depth.
I believe there is a 16 bit TIFF standard, but that may not be universally implemented.

On 01/08/2015 09:40 PM, Paul Verizzo wrote:
I've decided to get out of my comfort zone and start experimenting with 16 bit large sRGB color spaces now that I have a new printer with a pretty wide gamut (Canon PRO-100.)

So I take an MRW image, select to work in ProPhoto RGB (Why not go all the way?  Been reading about it, especially on Luminous Landscapte.)  I make a few adjustments, such as Levels, then select the 8>16 bit option in Depth.  When I try to save it as either JPG or TIFF, the file is kept at 8 bits.  When I try a PNG version, it fails.  I'm sorry, I don't have the program up to tell you exactly what the failure message is.

I've spent a lot of time in the online help and the Help PDF's, and I can't find anything that specifically address my simple workflow or why it doesn't work.

Thanks, Paul
_______________________________________________
Digikam-users mailing list
[hidden email]
https://mail.kde.org/mailman/listinfo/digikam-users



_______________________________________________
Digikam-users mailing list
[hidden email]
https://mail.kde.org/mailman/listinfo/digikam-users
Reply | Threaded
Open this post in threaded view
|

Re: What am I missing? (16 bit, etc.)

Gilles Caulier-4
2015-01-09 20:16 GMT+01:00 Carl McGrath <[hidden email]>:
> I am fairly certain that JPG (the standard)  is limited to 8 bit x 3
> channels (RGB)= 24 bit depth.

JPEG photo => 8 bits
JPEG medical = 12 bits (pattented and not open source)

digiKam use libjpeg and is limited to 8 bits.


> I believe there is a 16 bit TIFF standard, but that may not be universally
> implemented.

it's standard. libtiff support 16 bits and more. TIFF support more
color depth and floating encoding (HDR).

digiKam support 8 and 16 bits in RGBA color space.

PNG support 8 and 16 in RGBA. digiKAm support it too.

JPEG2000 and PGF as similar encoding features than TIFF with wavelets
compression (that TIFF do not have). 8 and 16 bits are supported by
digiKam.

Other image format are exotic and do not have any particular interest
in photography.

For the future we plan to add WebP format support from Google which
use wavelets compression and 8/16 bits color depth. It's opensource.

I take a look to M$ JPEG-XR (normalized by JPEG group to replace JPEG
in camera). Yes you read right, to replace JPEG. Read wiki page for
details : http://en.wikipedia.org/wiki/JPEG_XR

==> JPEG XR is supported by and "open source" libary from M$ but it's
a patented solution and M$ ask royalties.
So, we can forget for a while this format...

Voilà...

Gilles Caulier



>
>
> On 01/08/2015 09:40 PM, Paul Verizzo wrote:
>
> I've decided to get out of my comfort zone and start experimenting with 16
> bit large sRGB color spaces now that I have a new printer with a pretty wide
> gamut (Canon PRO-100.)
>
> So I take an MRW image, select to work in ProPhoto RGB (Why not go all the
> way?  Been reading about it, especially on Luminous Landscapte.)  I make a
> few adjustments, such as Levels, then select the 8>16 bit option in Depth.
> When I try to save it as either JPG or TIFF, the file is kept at 8 bits.
> When I try a PNG version, it fails.  I'm sorry, I don't have the program up
> to tell you exactly what the failure message is.
>
> I've spent a lot of time in the online help and the Help PDF's, and I can't
> find anything that specifically address my simple workflow or why it doesn't
> work.
>
> Thanks, Paul
> _______________________________________________
> Digikam-users mailing list
> [hidden email]
> https://mail.kde.org/mailman/listinfo/digikam-users
>
>
>
> _______________________________________________
> Digikam-users mailing list
> [hidden email]
> https://mail.kde.org/mailman/listinfo/digikam-users
>
_______________________________________________
Digikam-users mailing list
[hidden email]
https://mail.kde.org/mailman/listinfo/digikam-users
Reply | Threaded
Open this post in threaded view
|

Re: What am I missing? (16 bit, etc.)

jdd@dodin.org
Le 09/01/2015 20:36, Gilles Caulier a écrit :

> digiKam use libjpeg and is limited to 8 bits.

and screens do not display more (not to speak about eyes), so more is
only good for editing, then 8 bits for display

jdd

_______________________________________________
Digikam-users mailing list
[hidden email]
https://mail.kde.org/mailman/listinfo/digikam-users
Reply | Threaded
Open this post in threaded view
|

Re: What am I missing? (16 bit, etc.)

Gilles Caulier-4
2015-01-09 20:44 GMT+01:00 jdd <[hidden email]>:

> Le 09/01/2015 20:36, Gilles Caulier a écrit :
>
>> digiKam use libjpeg and is limited to 8 bits.
>
>
> and screens do not display more (not to speak about eyes), so more is only
> good for editing, then 8 bits for display
>
> jdd
>
>
> _______________________________________________
> Digikam-users mailing list
> [hidden email]
> https://mail.kde.org/mailman/listinfo/digikam-users
_______________________________________________
Digikam-users mailing list
[hidden email]
https://mail.kde.org/mailman/listinfo/digikam-users
Reply | Threaded
Open this post in threaded view
|

Re: What am I missing? (16 bit, etc.)

Gilles Caulier-4
In reply to this post by jdd@dodin.org
eyes + bran is able to show around 8 billion of colors. No more.

But to perform colors correction without to degrade image quality,
it's always better to have more color depth to operate. So 16 bits is
ideally better.

To resume : you change/improve/fix image in 16 bits, you export final
image in 8 bits.

Gilles Caulier


2015-01-09 20:44 GMT+01:00 jdd <[hidden email]>:

> Le 09/01/2015 20:36, Gilles Caulier a écrit :
>
>> digiKam use libjpeg and is limited to 8 bits.
>
>
> and screens do not display more (not to speak about eyes), so more is only
> good for editing, then 8 bits for display
>
> jdd
>
>
> _______________________________________________
> Digikam-users mailing list
> [hidden email]
> https://mail.kde.org/mailman/listinfo/digikam-users
_______________________________________________
Digikam-users mailing list
[hidden email]
https://mail.kde.org/mailman/listinfo/digikam-users