![]() My AVC skills have improved (+how AVC is embedded in theses containers), this seems no more logic, so I change this. I remember I did this because some files were different (not showing this "unknown profile") and a user said there is some compatibility issues between some MKV files and players, but I can not find them anymore. OK, this is very old code, and maybe bad code. But this is code from x264's flv and mkv writer. Sorry for replying to this one-year-old post. I noticed that adding a raw TrueHD (not a hybrid TrueHD+AC3) file into Mediainfo fails to show any metadata other than size.Īnd I find no reason for mkv and flv to use extra data for AVC different from MP4. I have this MOV file with LPCM 5.1 audio track, mediainfo seems to incorrectly report it as 3ch, sampling rate 1Hz I see the number of channels "NumberChannelDescriptions", but difficult to know the channel map (and map them to L R, Rear L or R)Įdit: Oups, I had a bug, code edited, now I understand better the values! This if from the file provided by Vincent Vegaīut if someone can help to decrypt the values (I don't understand what I decode compared to specs.). I know, and this is the data I will use to know the real number of audio channels:Ġ0000233 Bytes per packet: 40E77000 (1088909312)Ġ0000267 NumberChannelDescriptions: 6 (6) Who knows where all this "transformation" information is stored? Here 1 is the version number, 64 is profile, and there is no "reserved" byte between compatible profile and level.Īll my AVC in flv shows Unknown as well as AVC in mkv except those in VFW compatible mode. And I find no reason for mkv and flv to use extra data for AVC different from MP4. X264_put_byte( c, 0xe1 ) // 3 bits reserved (111) + 5 bits number of sps (00001)ĪvcC = 0xff // nalu size length is four bytes I am interested if you have any explaination about this. if they are different, I display "Container profile" (this value is not used by any player I know), and 1 is un unknown profile from specs. Theses files are reported to be NOT playable on some HW players, so users want to see if the "Profile" field is the same as in the seq_parameter_set. This is something I didn't yet understand.Ġ0001161 Size of NALU length minus 1: 3 (3)Ġ000064F Size of NALU length minus 1: 3 (3) Yes, some people can live from this kind of development! Not a big company, but enough for 1-2 people. I don't live from donations, but from paid feature requests (most of them are not visible in the GUI, you must look inside the code to find them -), GUI will be implemented for theses features when I have free time, people who pay use mainly the library). ![]() ![]() For GUI, I win less, so this is low priority for (I don't forget, but low priority). You win a better software, I win good support. I work more on feedback I have here about wrongly analyzed files because I know that people paying me can have the same issue later. This is not "just" a small information utility, because I support lot of formats (same API for all formats, rare), I provide a lot of information not easy to find (B-Frames, GOP, interleave method.) and I provide a library (and important: cross platform), used by companies for automation on hundred of thousands of files. I can imagine several people sending some donations, but this? Whoa.Īctually, donations are not enough, far from it. (.) It's "just" a small information utility like tens of others.
0 Comments
Leave a Reply. |
AuthorWrite something about yourself. No need to be fancy, just an overview. ArchivesCategories |