[FFmpeg-devel] [PATCH]Support decoding and encoding of 64bit tiff, v2

Jean First jeanfirst at gmail.com
Thu Jan 12 23:27:19 CET 2012


On Thu Jan 12 2012 05:55:02 GMT+0100 (CET), Ehmann Jochen HSLU D&K wrote:
> Carl Eugen Hoyos<cehoyos<at>  ag.or.at>  writes:
>
>> @@ -498,7 +506,7 @@ AVCodec ff_tiff_encoder = {
>>                                 PIX_FMT_YUV420P, PIX_FMT_YUV422P,
>>                                 PIX_FMT_YUV444P, PIX_FMT_YUV410P,
>>                                 PIX_FMT_YUV411P, PIX_FMT_RGB48LE,
>> -                              PIX_FMT_RGBA, PIX_FMT_NONE},
>> +                              PIX_FMT_RGBA, PIX_FMT_RGBA64, PIX_FMT_NONE},
> This patch and the changed version of the "64bit png decoding" patch (with
> PIX_FMT_RGBA64) allow to reencode the sample from ticket 639 (64bit png) to tiff
> on LE and BE, it also allows to reencode the sample from ticket 503 (64 bit le
> tiff) to tiff on LE.
> I tested encoding to RGB48 tiff with both lena and the 48bit dpx sample, and
> both worked on LE and BE (tested with ImageMagick and OSX preview).
>
> Any objections or better tests or may I apply?

AFAIR the tiff encoder supports only LE. Should this value not be 
PIX_FMT_RGBA64LE ?
Jean


More information about the ffmpeg-devel mailing list