[FFmpeg-devel] OpenEXR Encoder Intentions

Vincent Olivier vincent at up4.com
Sun Apr 28 23:40:16 CEST 2013


On 2013-04-28, at 5:15 PM, Carl Eugen Hoyos <cehoyos at ag.or.at> wrote:

> Vincent Olivier <vincent <at> up4.com> writes:
> 
>>> The change in line 146 of allcodecs.c is missing.
>> 
>> Yes, I didn't commit that,
> 
> Then how do you expect us to review?


I don't. I was asking you a question. I will expect you to review when I will submit a clean pull request for you to review. I will have punctual questions until then.


>> but when I do and change this line:
>> 
>> REGISTER_DECODER(EXR,               	exr);
>> 
>> to this line
>> 
>> REGISTER_ENCDEC(EXR,               	exr);
>> 
>> The config file actually includes the "exr" encoder, 
>> but exrenc.c and exrdec.c still don't get compiled.
> 
> Rerun configure and add rules to libavcodec/Makefile
> (See the current rule for exr.o)


Well now. Works perfectly, thanks to you. Before I get to the bulk of the code, though, I need to study the libswscale upscaling algorithms to for the chroma channels of a YUV image. I asked those questions in the user mailing list (you already replied to my first message and I replied back with details). If I can get to the bottom of that, I will be able to get back to the OpenEXR encoder and the pixel formats it can accepts. I would like to start from 10-bit 444 to go to 32-bit float RGB (and do the 444 to 32-bit float RGB conversion in variable precision space in the encoder directly) so as to minimize the rounding losses.

v


More information about the ffmpeg-devel mailing list