Well I guess Carl got that right a while ago, these matrizes are circulating and I think the obligatory RGB conversion matrix is even part of DCI/SMPTE spec now. REC709 may have been a bit special. And maybe the issue was not so much the matrix itself but the interpretation of the source data.


When I asked people about that problem last year, some suggested to me that there is not a single REC709 on consumer media. I found that hard to believe, because on all tv sets I've come around I've never seen gamma so far off.

I can still try a test with the RGB conversion, but I think I tried that last year already and it didn't make a difference. The source MKV in VLC player showed okay.

What I did since then was to have the BluRay on site for a quick comparison BluRay through BD-Player/HDMI, and converted DCP.

- Carsten


Am 24.06.2014 um 12:26 schrieb Gérald Maruccia:

It's for the moment beyond my skills but thismight help :

http://www.brucelindbloom.com/Eqn_RGB_XYZ_Matrix.html

G.
Le 24/06/2014 12:22, Carl Hetherington a écrit :
On Tue, 24 Jun 2014 02:39:21 +0200
Carsten Kurz <audiovisual@t-online.de> wrote:


Ah... as Carsten suggested: are we talking about drastic
completely-wrong colours, or a slight tint/gamma shift?

[snip]


I still have that machine with DVD-o-matic on it, so I recently did
it again with the same setting, and it turned up again.

Today I installed a recent version of DCP-o-matic on the same machine
and did the same scene with it, using same source file, scaler,
REC709 preset,  datarate setting, etc

[snip]

I'm fairly sure that DVD-o-matic had incorrect REC709 conversion
matrices for a large part of its life (maybe all of it...) so I think
that would explain it.

Regards
Carl


_______________________________________________
DCPomatic mailing list
DCPomatic@carlh.net
http://main.carlh.net/cgi-bin/mailman/listinfo/dcpomatic

_______________________________________________
DCPomatic mailing list
DCPomatic@carlh.net
http://main.carlh.net/cgi-bin/mailman/listinfo/dcpomatic