1 |
Hello, |
2 |
|
3 |
On Fri, Oct 20, 2017 at 11:17 AM, Helmut Jarausch <jarausch@××××××.be> wrote: |
4 |
> Hi, |
5 |
> I'm considering buying a new monitor (and graphics card) which supports |
6 |
> 10 bits per color channel. |
7 |
> Will Gimp on a Linux machine (X11) support this now or in the near future. |
8 |
> Or is it just waste of money to buy a monitor with more than 8 bits/color channel? |
9 |
> Many thanks for some hints, |
10 |
> Helmut |
11 |
> |
12 |
|
13 |
It is my understanding the software itself needs to support this. It |
14 |
is rare enough that I do not know how it is done in practice, but, |
15 |
generally, it would be necessary to specify relative colors or colors |
16 |
by name and leave implementation up to a layer that translates the |
17 |
color spaces. |
18 |
|
19 |
This would be similar to what has to happen for high DPI displays. If |
20 |
a program is not aware that pixels can be different sizes, it will not |
21 |
make use of the extra resolution properly. |
22 |
|
23 |
High end graphics cards have implemented features related to what you |
24 |
are asking about. Some support transformations that happen directly on |
25 |
the tuples that represent the color model of the human eye, or the |
26 |
model of the reflectivity of pigments used in printing. I have yet to |
27 |
see any programs which actually use these features. |
28 |
|
29 |
Checking the Wiki, "deep color" capable graphics cards have existed |
30 |
since the 1990s. Also at |
31 |
https://en.wikipedia.org/wiki/Color_depth#Deep_color_.2830.2F36.2F48-bit.29 |
32 |
are compatibility notes that seem to indicate that color depth is |
33 |
mediated by the graphics card and screen. It is up to the program to |
34 |
ask for colors of that depth to be displayed. |
35 |
|
36 |
|
37 |
Short answer: Open bug reports on the trackers for the programs you |
38 |
are interested in. |
39 |
|
40 |
Cheers, |
41 |
R0b0t1. |