[Ffmpeg-devel-irc] ffmpeg-devel.log.20180806

[burek]

[02:46:47 CEST] <JEEB> michaelni_: did you note anything else about that ismv patch?
[06:50:39 CEST] <cone-958> ffmpeg 03Gyan Doshi 07master:076b19660e7c: avcodec/libaomenc: correct range for cpu-used
[11:13:57 CEST] <loosechainsaw> i am attempting to use gpu acceleration in my app to decode video. How do I copy the entire frame->data buffer into cuda to i can convert it to bgra?
[11:37:10 CEST] <michaelni_> JEEB, no, only what i wrote
[12:08:36 CEST] <JEEB> michaelni: fate-lavf-ismv? that I fixed
[12:08:48 CEST] Action: JEEB has been traveling around so hasn't noticed anything else
[12:32:22 CEST] <jdarnley> How is a decoder supposed to signal an interlaced stream and field order to avcodec and the user?
[12:33:07 CEST] <nevcairiel> Avframe has fields for that
[12:33:16 CEST] <jdarnley> There is the AVFieldOrder enum in AVCodecContext but the doxy says that is set by the user on decoding (which doesn't make sense to me).
[12:35:54 CEST] <jdarnley> nevcairiel: you mean the "int interlaced_frame" and "int top_field_first" members?
[12:36:07 CEST] <nevcairiel> Yeah
[12:36:15 CEST] <jdarnley> I guess you do and it is what I need.
[12:36:58 CEST] <nevcairiel> Those things can change midstream in most codecs, which is why it's not mirrored as a stream level property
[19:53:39 CEST] <pi-> (REPOST) Seeking ffmpeg consultant https://paste.pound-python.org/show/SIeD9VyutK8MeIB6Immd/
[20:54:10 CEST] <cone-341> ffmpeg 03Vittorio Giovara 07master:ce7ca726b2ca: vf_tonemap: Fix logic for detecting the maximum peak of untagged sources
[20:54:10 CEST] <cone-341> ffmpeg 03Vittorio Giovara 07master:e467179cfbe2: vf_tonemap[_opencl]: Move determine_signal_peak() to a shared file
[20:54:10 CEST] <cone-341> ffmpeg 03Vittorio Giovara 07master:572ef567a528: vf_tonemap_opencl: Move update_metadata() to a shared file
[20:54:10 CEST] <cone-341> ffmpeg 03Vittorio Giovara 07master:aa8e5eecac97: vf_tonemap: Update hdr metadata with the new peak value
[00:00:00 CEST] --- Tue Aug  7 2018