Author: michael Date: Tue Feb 5 15:14:01 2008 New Revision: 586 Log: Add match_time_delta to the framecode table. Modified: docs/nut.txt Modified: docs/nut.txt ============================================================================== --- docs/nut.txt (original) +++ docs/nut.txt Tue Feb 5 15:14:01 2008 @@ -247,6 +247,7 @@ main_header: tmp_pts=0 tmp_mul=1 tmp_stream=0 + tmp_match=0 for(i=0; i<256; ){ tmp_flag v tmp_fields v @@ -259,7 +260,8 @@ main_header: else tmp_res=0 if(tmp_fields>5) count v else count= tmp_mul - tmp_size - for(j=6; j<tmp_fields; j++){ + if(tmp_fields>6) tmp_match s + for(j=7; j<tmp_fields; j++){ tmp_reserved[i] v } for(j=0; j<count && i<256; j++, i++){ @@ -274,6 +276,7 @@ main_header: data_size_lsb[i]= tmp_size + j; pts_delta[i]= tmp_pts; reserved_count[i]= tmp_res; + match_time_delta[i]= tmp_match; } } reserved_bytes
On Tue, Feb 05, 2008 at 03:14:02PM +0100, michael wrote:
Author: michael Date: Tue Feb 5 15:14:01 2008 New Revision: 586
Log: Add match_time_delta to the framecode table.
In case someone wonders, why i add this obscure feature to the framecode table ... Its because MDCT based audio codecs strictly have a match_time_delta=1 (which is being ignored by everyone it seems) but if we accurately store it. Then having it in the framecode table would just mean O(1) overhead. If its not in the framecode table it would be O(n) overhead, that is requiring an explixit match_time_delta=1 being stored with every audio frame. Not that i expect many people to set this properly, even less so for mp3 with the bit reservoir where its not constant ... [...] -- Michael GnuPG fingerprint: 9FF2128B147EF6730BADF133611EC787040B0FAB Why not whip the teacher when the pupil misbehaves? -- Diogenes of Sinope
participants (2)
-
michael -
Michael Niedermayer