February 2008 - NUT-devel - lists.mplayerhq.hu

[nut]: r630 - docs/nut.txt
by diego 13 Feb '08

13 Feb '08

Author: diego Date: Wed Feb 13 19:49:27 2008 New Revision: 630 Log: Fix some typos. Modified: docs/nut.txt Modified: docs/nut.txt ============================================================================== --- docs/nut.txt (original) +++ docs/nut.txt Wed Feb 13 19:49:27 2008 @@ -701,24 +701,24 @@ match_time_delta (s) match_time_delta[frame_code] (s) If FLAG_MATCH_TIME is not set then this value shall be used for match_time_delta, otherwise this value is ignored. - MUST be <32768 and >-32768 or =1-(1<<62) + MUST be <32768 and >-32768 or =1-(1<<62). header_idx[frame_code] (v) The index into the elision_header table. - MUST by <128 + MUST be <128. header_count_minus1 (v) The number of distinct non empty elision headers. - MUST be <128 + MUST be <128. elision_header[header_idx] (vb) For frames with a final size <= 4096 this header is prepended to the frame data. That is if the stored frame is 4000 bytes and the elision_header is 96 bytes then it is prepended, if it is 97 byte then it is not. - elision_header[0] is fixed to a length 0 header - The length of each elision_header except header 0 MUST be < 256 and >0 - The sum of the lengthes of all elision_headers MUST be <=1024 + elision_header[0] is fixed to a length 0 header. + The length of each elision_header except header 0 MUST be < 256 and >0. + The sum of the lengthes of all elision_headers MUST be <=1024. lsb_pts Least significant bits of the pts in time_base precision. @@ -1068,11 +1068,11 @@ duplicate the role of a container. Any s For example Vorbis in Ogg in NUT is invalid, as is mpegvideo + mpegaudio in MPEG-PS/TS in NUT or dvvideo + dvaudio in DV in NUT. -A demuxer MUST NOT demux a nut frame containing anything else than exactly a +A demuxer MUST NOT demux a NUT frame containing anything else than exactly a single codec frame. A muxer MUST NOT store anything except exactly 1 single -codec frame per nut frame. +codec frame per NUT frame. For RAW audio up to 4096 samples shall be considered 1 codec frame, the actual -number of samples MAY vary between nut frames. +number of samples MAY vary between NUT frames. Sample code (Public Domain, & untested):

1 0

[nut]: r604 - docs/nutissues.txt
by michael 13 Feb '08

13 Feb '08

Author: michael Date: Tue Feb 12 00:33:21 2008 New Revision: 604 Log: 3 more issues which have come up in the past but have IIRC never been resolved. Modified: docs/nutissues.txt Modified: docs/nutissues.txt ============================================================================== --- docs/nutissues.txt (original) +++ docs/nutissues.txt Tue Feb 12 00:33:21 2008 @@ -110,3 +110,42 @@ Solutions: A. Store such alternative playlists of scenes in info packets somehow. B. Design a separate layer for it. C. Do not support this. + + +Issue header-compression +------------------------ +Headers of codec frames often contain low entropy information or things +we already know like the frame size. + +A. Store header bytes and length in the framecode table. +B. Leave things as they are. + + +Issue small-frames +------------------ +The original intent of nut frames was that 1 container frame == 1 codec +frame. Though this does not seem to be explicitly written in nut.txt. +Also it is inefficient for very small frames, AMR-NB for example has 6-32 +bytes per frame. + +Solutions: +A. Enforce 1 container frame == 1 codec frame even if it causes 10% overhead. +B. Allow multiple frames as long as the whole packet is less than some + fixed minimum in bytes (like 256byte) +C. Allow multiple frames as long as the whole packet is less than some + fixed minimum in bytes (like 256byte) and the codec uses a constant + framesize in samples. +D. Use header compression, that is allow to store the first (few) bytes + of a codec frame together with its size in the framecode table. This + would allow us to store the size of a frame without any redundancy. + Thus effectivly avoiding the overhead small frames cause. + + +Issue pcm-frames +---------------- +No word is said about how many or few PCM samples should be in a frame. + +Solutions: +A. Define an maximum number of samples (like 512) +B. Define an maximum timespam (like 0.1 sec) +C. Define an maximum number of bytes (like 1024)

6 19

[nut]: r624 - docs/nutissues.txt
by michael 13 Feb '08

13 Feb '08

Author: michael Date: Wed Feb 13 15:17:09 2008 New Revision: 624 Log: header elision update Modified: docs/nutissues.txt Modified: docs/nutissues.txt ============================================================================== --- docs/nutissues.txt (original) +++ docs/nutissues.txt Wed Feb 13 15:17:09 2008 @@ -115,6 +115,8 @@ we already know like the frame size. A. Store header bytes and length in the framecode table. B. Leave things as they are. +Implemented + Issue small-frames ------------------ @@ -135,6 +137,8 @@ D. Use header compression, that is allow would allow us to store the size of a frame without any redundancy. Thus effectivly avoiding the overhead small frames cause. +It seems the consensus is: A+D + Issue pcm-frames ----------------

2 1

[nut]: r622 - src/trunk/nututils/nutparse.c
by cladisch 13 Feb '08

13 Feb '08

Author: cladisch Date: Wed Feb 13 13:40:44 2008 New Revision: 622 Log: check frame code table delta match time restriction Modified: src/trunk/nututils/nutparse.c Modified: src/trunk/nututils/nutparse.c ============================================================================== --- src/trunk/nututils/nutparse.c (original) +++ src/trunk/nututils/nutparse.c Wed Feb 13 13:40:44 2008 @@ -324,6 +324,9 @@ static void parse_main_header(void) if (tmp_fields > 6) { tmp_match = read_svar(); printf(" tmp_match: %"PRId64"\n", tmp_match); + if ((tmp_match <= -32768 || tmp_match >= 32768) && + tmp_match != 1 - (INT64_C(1) << 62)) + error("absolute delta match time must be less than 32768"); } for (j64 = 7; j64 < tmp_fields; ++j64) printf(" tmp_reserved[%"PRIu64"]: %"PRIu64"\n", j64, read_var());

2 1

[nut]: r614 - docs/nut4cc.txt
by michael 13 Feb '08

13 Feb '08

Author: michael Date: Tue Feb 12 16:30:55 2008 New Revision: 614 Log: Choose a 4cc for mpeg4. Bikeshed flames welcome Modified: docs/nut4cc.txt Modified: docs/nut4cc.txt ============================================================================== --- docs/nut4cc.txt (original) +++ docs/nut4cc.txt Tue Feb 12 16:30:55 2008 @@ -32,6 +32,7 @@ MJPG ITU JPEG MPG4 MS MPEG4v1 (!NOT ISO MPEG4!) MP42 MS MPEG4v2 MP43 MS MPEG4v3 +MP4V ISO MPEG4 Video mpg1 ISO MPEG1 Video mpg2 ISO MPEG2 Video MRLE MS RLE @@ -93,4 +94,4 @@ P<type><interleaving><bits> little endia PFD[32] would for example be signed 32 bit little endian IEEE float -FIXME mpeg4/dv/gray16/audio codecs/multi chan interleaving +FIXME dv/gray16/audio codecs/multi chan interleaving

3 3

[nut]: r629 - docs/nut.txt
by cladisch 13 Feb '08

13 Feb '08

Author: cladisch Date: Wed Feb 13 16:17:52 2008 New Revision: 629 Log: fix typo Modified: docs/nut.txt Modified: docs/nut.txt ============================================================================== --- docs/nut.txt (original) +++ docs/nut.txt Wed Feb 13 16:17:52 2008 @@ -1072,7 +1072,7 @@ A demuxer MUST NOT demux a nut frame con single codec frame. A muxer MUST NOT store anything except exactly 1 single codec frame per nut frame. For RAW audio up to 4096 samples shall be considered 1 codec frame, the actual -number of sampes MAY vary between nut frames. +number of samples MAY vary between nut frames. Sample code (Public Domain, & untested):

1 0

[nut]: r628 - docs/nutissues.txt
by michael 13 Feb '08

13 Feb '08

Author: michael Date: Wed Feb 13 16:09:10 2008 New Revision: 628 Log: Issue pcm-frames solution Modified: docs/nutissues.txt Modified: docs/nutissues.txt ============================================================================== --- docs/nutissues.txt (original) +++ docs/nutissues.txt Wed Feb 13 16:09:10 2008 @@ -151,6 +151,8 @@ A. Define an maximum number of samples ( B. Define an maximum timespam (like 0.1 sec) C. Define an maximum number of bytes (like 1024) +Implemented: A + Issue pcm-sample-format -----------------------

1 0

[nut]: r627 - docs/nut.txt
by michael 13 Feb '08

13 Feb '08

Author: michael Date: Wed Feb 13 16:05:46 2008 New Revision: 627 Log: Strict 1 codec frame rule and RAW audio size limit Modified: docs/nut.txt Modified: docs/nut.txt ============================================================================== --- docs/nut.txt (original) +++ docs/nut.txt Wed Feb 13 16:05:46 2008 @@ -1068,6 +1068,11 @@ duplicate the role of a container. Any s For example Vorbis in Ogg in NUT is invalid, as is mpegvideo + mpegaudio in MPEG-PS/TS in NUT or dvvideo + dvaudio in DV in NUT. +A demuxer MUST NOT demux a nut frame containing anything else than exactly a +single codec frame. A muxer MUST NOT store anything except exactly 1 single +codec frame per nut frame. +For RAW audio up to 4096 samples shall be considered 1 codec frame, the actual +number of sampes MAY vary between nut frames. Sample code (Public Domain, & untested):

1 0

[nut]: r626 - docs/nutissues.txt
by michael 13 Feb '08

13 Feb '08

Author: michael Date: Wed Feb 13 15:20:55 2008 New Revision: 626 Log: Issue stream-chapter-overlap consensus (again scream if you disagree!) Modified: docs/nutissues.txt Modified: docs/nutissues.txt ============================================================================== --- docs/nutissues.txt (original) +++ docs/nutissues.txt Wed Feb 13 15:20:55 2008 @@ -51,6 +51,8 @@ issue, that chapters might not end exact all streams. There could be a few samples difference. Do we care about this? +It seems the consensus is: no, we dont care + Issue info-stream-subsets -------------------------

1 0

[nut]: r625 - docs/nutissues.txt
by michael 13 Feb '08

13 Feb '08

Author: michael Date: Wed Feb 13 15:19:05 2008 New Revision: 625 Log: Issue pcm-sample-format consensus If i misinterpreted anyones position please say so! Modified: docs/nutissues.txt Modified: docs/nutissues.txt ============================================================================== --- docs/nutissues.txt (original) +++ docs/nutissues.txt Wed Feb 13 15:19:05 2008 @@ -158,6 +158,8 @@ A. fourcc B. extradata C. New field in the stream header +It seems the consensus is: A + Issue pcm-interleaving ----------------------

1 0