May 2005 - MPlayer-dev-eng - lists.mplayerhq.hu

Synchronization of players in LAN
by J.A. Gutierrez 18 Jun '08

18 Jun '08

Hello I'm trying to synchronise several players running on several computers connected to a dedicated 10/100 Ethernet switch. Each one reads a different movie (video only) from its internal hard disk, and resolution will be 768x576. I'm thinking of a simple solution: - Add some code at mplayer startup so it will wait for an UDP broadcast packet (sent by a central master in the LAN). This way all the players would start at the same time. - Use NTP to keep in sync all the computers' clocks, and trust mplayer's rtc synchronisation feature. With this simple setup, could we be sure all the players would keep the synchronisation? Or should the master computer track them all and take control on their speeds? How difficult would be to add code so all the player synchronise using the LAN? Or should I use some external clock connected to all serial ports? Or use a totally different approach? Thanks. -- finger spd(a)shiva.cps.unizar.es for PGP / .mailcap tip of the day: / La vida es una carcel application/ms-tnef; cat '%s' > /dev/null / con las puertas abiertas text/x-vcard; cat '%s' > /dev/null / (A. Calamaro)

14 40

NOVIRUS Bug 188 - [Patch] add vob subtitle to mencoder.
by ha shao 30 Apr '07

30 Apr '07

Vobsub is quite popular but mencoder don't support it. This patch add the -vobsub -vobsubid to mencode. The codes were mostly copy & paste from mplayer's vobsub sessions (FIXME: tags included). I don't really know what I was doing but it simply works for me. __________________________________________________ Do You Yahoo!? Tired of spam? Yahoo! Mail has the best spam protection around http://mail.yahoo.com Index: cfg-common.h =================================================================== RCS file: /cvsroot/mplayer/main/cfg-common.h,v retrieving revision 1.137 diff -u -r1.137 cfg-common.h --- cfg-common.h 18 Oct 2004 20:41:05 -0000 1.137 +++ cfg-common.h 2 Jan 2005 10:24:51 -0000 @@ -251,6 +251,9 @@ {"sub-bg-alpha", &sub_bg_alpha, CONF_TYPE_INT, CONF_RANGE, 0, 255, NULL}, {"sub-no-text-pp", &sub_no_text_pp, CONF_TYPE_FLAG, 0, 0, 1, NULL}, {"sub-fuzziness", &sub_match_fuzziness, CONF_TYPE_INT, CONF_RANGE, 0, 2, NULL}, + {"vobsub", &vobsub_name, CONF_TYPE_STRING, 0, 0, 0, NULL}, + {"vobsubid", &vobsub_id, CONF_TYPE_INT, CONF_RANGE, 0, 31, NULL}, + #endif #ifdef USE_OSD {"font", &font_name, CONF_TYPE_STRING, 0, 0, 0, NULL}, Index: cfg-mplayer.h =================================================================== RCS file: /cvsroot/mplayer/main/cfg-mplayer.h,v retrieving revision 1.236 diff -u -r1.236 cfg-mplayer.h --- cfg-mplayer.h 27 Dec 2004 19:34:42 -0000 1.236 +++ cfg-mplayer.h 2 Jan 2005 10:24:51 -0000 @@ -327,10 +327,6 @@ {"menu", "OSD menu support was not compiled in.\n", CONF_TYPE_PRINT,0, 0, 0, NULL}, #endif - // these should be moved to -common, and supported in MEncoder - {"vobsub", &vobsub_name, CONF_TYPE_STRING, 0, 0, 0, NULL}, - {"vobsubid", &vobsub_id, CONF_TYPE_INT, CONF_RANGE, 0, 31, NULL}, - {"sstep", &step_sec, CONF_TYPE_INT, CONF_MIN, 0, 0, NULL}, // set a-v distance, should be moved to -common and supported in MEncoder Index: mencoder.c =================================================================== RCS file: /cvsroot/mplayer/main/mencoder.c,v retrieving revision 1.259 diff -u -r1.259 mencoder.c --- mencoder.c 27 Dec 2004 17:30:12 -0000 1.259 +++ mencoder.c 2 Jan 2005 10:24:53 -0000 @@ -196,6 +196,7 @@ float sub_delay=0; float sub_fps=0; int sub_auto = 0; +char *vobsub_name=NULL; int subcc_enabled=0; int suboverlap_enabled = 1; @@ -489,6 +490,28 @@ vo_init_osd(); + //==================== Open VOB-Sub ============================ + if (vobsub_name){ + vo_vobsub=vobsub_open(vobsub_name,spudec_ifo,1,&vo_spudec); + if(vo_vobsub==NULL) + mp_msg(MSGT_CPLAYER,MSGL_ERR,MSGTR_CantLoadSub,vobsub_name); + }else if(sub_auto && filename && (strlen(filename)>=5)){ + /* try to autodetect vobsub from movie filename ::atmos */ + char *buf = malloc((strlen(filename)-3) * sizeof(char)); + memset(buf,0,strlen(filename)-3); // make sure string is terminated + strncpy(buf, filename, strlen(filename)-4); + vo_vobsub=vobsub_open(buf,spudec_ifo,0,&vo_spudec); + free(buf); + } + + if(vo_vobsub){ + sub_auto=0; // don't do autosub for textsubs if vobsub found + // inited_flags|=INITED_VOBSUB; + vobsub_set_from_lang(vo_vobsub, dvdsub_lang); + // check if vobsub requested only to display forced subtitles + forced_subs_only=vobsub_get_forced_subs_flag(vo_vobsub); + } + stream=open_stream(filename,0,&file_format); if(!stream){ @@ -1131,7 +1154,11 @@ else sscanf(seek_to_sec, "%f", &d); - demux_seek(demuxer, d, 1); + if (demux_seek(demuxer, d, 1)){ + /* FIXME there should be real seeking for vobsub */ + if (vo_vobsub) + vobsub_seek(vo_vobsub, d_video->pts); + } // there is 2 way to handle the -ss option in 3-pass mode: // > 1. do the first pass for the whole file, and use -ss for 2nd/3rd pases only // > 2. do all the 3 passes with the same -ss value @@ -1548,14 +1575,41 @@ // DVD sub: if(vo_spudec||vobsub_writer){ unsigned char* packet=NULL; - int len; - while((len=ds_get_packet_sub(d_dvdsub,&packet))>0){ - mp_msg(MSGT_MENCODER,MSGL_V,"\rDVD sub: len=%d v_pts=%5.3f s_pts=%5.3f \n",len,sh_video->pts,d_dvdsub->pts); - if (vo_spudec) - spudec_assemble(vo_spudec,packet,len,90000*d_dvdsub->pts); - if (vobsub_writer) - vobsub_out_output(vobsub_writer,packet,len,mux_v->timer + d_dvdsub->pts - sh_video->pts); - } + int len, timestamp; + // Get a sub packet from the dvd or a vobsub and make a timestamp relative to sh_video->timer + while(1) { + // FIXME: messed up timer and pts????????? + // Vobsub + if(vo_vobsub) { + if(sh_video->pts+sub_delay>=0) { + len = vobsub_get_packet(vo_vobsub,sh_video->pts+sub_delay,(void**)&packet,&timestamp); + if(len > 0) { + timestamp -= (sh_video->pts + sub_delay - sh_video->timer)*90000; + mp_dbg(MSGT_CPLAYER,MSGL_V,"\rVOB sub: len=%d v_pts=%5.3f v_timer=%5.3f sub=%5.3f ts=%d \n",len,sh_video->pts,sh_video->timer,timestamp / 90000.0,timestamp); + } + } + } else { + // DVD sub + len=ds_get_packet_sub(d_dvdsub,&packet); + if(len > 0) { + float x = d_dvdsub->pts - sh_video->pts; + if (x < -10 || x > 10) // prevent missing subs on pts reset + timestamp = 90000*(sh_video->timer + d_dvdsub->pts + sub_delay - sh_video->pts); + else + timestamp = 90000*(sh_video->timer + sub_delay); + + mp_msg(MSGT_MENCODER,MSGL_V,"\rDVD sub: len=%d v_pts=%5.3f s_pts=%5.3f ts=%d \n",len,sh_video->pts,d_dvdsub->pts,timestamp); + } + } + if (len <= 0 || !packet) break; + if (timestamp < 0) timestamp = 0; + else{ + if (vo_spudec) + spudec_assemble(vo_spudec,packet,len,timestamp); + if (vobsub_writer) + vobsub_out_output(vobsub_writer,packet,len,mux_v->timer + d_dvdsub->pts - sh_video->pts); + } + } if (vo_spudec) { spudec_heartbeat(vo_spudec,90000*sh_video->pts); vo_osd_changed(OSDTYPE_SPU); @@ -1603,6 +1657,11 @@ mp_msg(MSGT_MENCODER,MSGL_FATAL,MSGTR_ErrorWritingFile, out_filename); mencoder_exit(1, NULL); } + +if(vo_vobsub) + vobsub_close(vo_vobsub); + vo_vobsub = NULL; + if(vobsub_writer) vobsub_out_close(vobsub_writer);

5 4

[PATCH] Better ARM support for mplayer/ffmpeg, ported from atty fork
by Bernhard Rosenkraenzer 27 May '06

27 May '06

Hi, while playing with some new hardware, I found it's running a forked mplayer -- and it looks like they're following the GPL. The maintainer's page is here: http://atty.jp/?Zaurus/mplayer Unfortunately it's mostly in Japanese, so it's hard to figure out any details. Their code looks quite interesting (at least to those of us w/ ARM CPUs). The patches I've attached are the patches from atty.jp with a couple of modifications by myself: - ported to current CVS - reverted their change of removing SNOW support from ffmpeg - cleaned up their bswap mess - removed DOS-style linebreaks from various files - reverted their moving "audiocodec mad" in front of "audiocodec ffmp3" in codecs.conf The mplayer part contains some hunks that aren't just for arm optimizations (e.g. they put some stuff into demux_audio.c to detect a Xing mp3 header), so this shouldn't be applied blindly. LLaP bero

5 11

[PATCH] supress zero frames in index regeneration
by Yue Shi Lai 06 Apr '06

06 Apr '06

Hi, This patch supresses zero-length frames being mistaken for keyframes during regeneration of the AVI index ("-forceidx"). Yue Shi Lai -- ______________________________________________ Check out the latest SMS services @ http://www.linuxmail.org This allows you to send and receive SMS through your mailbox. Powered by Outblaze

4 3

[PATCH] missing marker bit
by Christoph Egger 31 Dec '05

31 Dec '05

Hello! Synopsis: When mplayer should player a mpeg video, some of them result in this: MPEG-PS file format detected. missing marker bit! MPEG: bad sequence header Video: Cannot read properties. ========================================================================== Opening audio decoder: [mp3lib] MPEG layer-2, layer-3 AUDIO: 48000 Hz, 2 ch, 16 bit (0x20), ratio: 24000->192000 (192.0 kbit) Selected audio codec: [mp3] afm:mp3lib (mp3lib MPEG layer-2, layer-3) ========================================================================== ... and mplayer only plays the audio, but not the video. VLC and Xine also claim about the missing marker bit, but nonetheless they play the video. With the attached patch, mplayer can do this, too. This is the output with the patch: MPEG-PS file format detected. missing marker bit! VIDEO: MPEG1 480x576 (aspect 2) 25.000 fps 4280.4 kbps (535.0 kbyte/s) ========================================================================== Opening audio decoder: [mp3lib] MPEG layer-2, layer-3 AUDIO: 48000 Hz, 2 ch, 16 bit (0x20), ratio: 24000->192000 (192.0 kbit) Selected audio codec: [mp3] afm:mp3lib (mp3lib MPEG layer-2, layer-3) ========================================================================== Opening video decoder: [mpegpes] MPEG 1/2 Video passthrough ... -- NEU +++ DSL Komplett von GMX +++ http://www.gmx.net/de/go/dsl GMX DSL-Netzanschluss + Tarif zum superg�nstigen Komplett-Preis!

3 6

[PATCH] remove-logo filter
by Robert Edele 22 Sep '05

22 Sep '05

Hi, This is a resubmission (with a few very minor changes) or my remove-logo filter that I submitted a few months ago. Would someone mind having a look at it. The filter works by loading an image which tells it which pixels are in the logo or not, and blur-filling them using nearby non-logo pixels. A stronger blur is used for pixels further from the edge of the logo. While it isn't as effective as a reverse-alpha-blending filter for translucent logos, it will work on all logos and makes a substantially smaller and less visible blur effect than the existing de-logo. Attached are the patch files. Sincerely, Robert Edele

4 5

[PATCH] vobsub: keep aspect ratio
by Csillag Kristof 25 Aug '05

25 Aug '05

Hi there! One more little thing: When we rip a movie, black bars are often removed from the top and the bottom of the original picture. This is OK, but it means that the aspect ratio of the movie is changed. ON the other side, when we dump a VOBsub subtitle, no cropping takes place, so it's aspect rate will match the original (un-cropped) movie's. So, when we display a subtitle with the cropped movie, and rescale the vobsub's frame to match the movie's sizes, we are distorting the subtitle's aspect ratio. (Typically, we get shorter letters.) This makes reading the text harder. * * * So, I added the option "-spu-keep-aspect". If it is switched on, the VOBsub's original aspect ratio is used to display it. If it's off (default), the aspect ratio of the movie is applied to the subtitles. See the attached patch. (I updated the man page, too.) (Note: this patch depends on the following two pending patches of mine: a0_positioning-fix-and-cleanup.patch, a1_scaling.patch) I uploaded a sample file here: http://rht.bme.hu/~lunkwill/computer/mplayer/aspect-demo.mkv The default "mplayer -slang eng aspect-demo.mkv" gives this: http://rht.bme.hu/~lunkwill/computer/mplayer/without-keep-aspect.png "mplayer -slang eng aspect-demo.mkv -spu-keep-aspect" gives this: http://rht.bme.hu/~lunkwill/computer/mplayer/with-keep-aspect.png * * * What do you think? IMHO the subtitles look much better this way. Shouldn't this be the default behavior? -- Csillag Kristof <fenwick(a)freemail.hu>

6 22

[PATCH] mp_msg cleanup
by Oded Shimon 25 Jul '05

25 Jul '05

I just cleaned up mp_msg a little, just about only modified it so it uses 2 parameters instead of 1, and still use a tiny little less ram than the previous implementation (umm, was that the reason for the original very bizarre implementation? its the only reason i could think of.). - ods15

5 5

[PATCH] leak fixes for loop mode
by Timothy Lee 21 Jul '05

21 Jul '05

Dear all, This patch fixes leaks which are noticeable when MPlayer runs in loop mode. The failure to clean up the data buffer used by Tremor after file detection loses 6.4 kbyte for every loop. The leaks were detected using valgrind. Regards, Timothy Lee diff -Naur -wbBE main-cvs/libmpdemux/demuxer.c main-leakfix/libmpdemux/demuxer.c --- main-cvs/libmpdemux/demuxer.c 2005-01-20 08:50:12.000000000 +0800 +++ main-leakfix/libmpdemux/demuxer.c 2005-02-04 17:31:26.000000000 +0800 @@ -219,13 +219,13 @@ demux_close_rtp(demuxer); break; #endif case DEMUXER_TYPE_SMJPEG: - demux_close_smjpeg(demuxer); return; + demux_close_smjpeg(demuxer); break; case DEMUXER_TYPE_DEMUXERS: - demux_close_demuxers(demuxer); return; + demux_close_demuxers(demuxer); break; case DEMUXER_TYPE_AVI: case DEMUXER_TYPE_AVI_NI: case DEMUXER_TYPE_AVI_NINI: - demux_close_avi(demuxer); return; + demux_close_avi(demuxer); break; #ifdef HAVE_XMMS case DEMUXER_TYPE_XMMS: demux_close_xmms(demuxer); break; @@ -250,10 +250,10 @@ #endif } // free streams: - for(i=0;i<256;i++){ + for(i=0;i<MAX_A_STREAMS;i++) if(demuxer->a_streams[i]) free_sh_audio(demuxer->a_streams[i]); + for(i=0;i<MAX_V_STREAMS;i++) if(demuxer->v_streams[i]) free_sh_video(demuxer->v_streams[i]); - } //if(sh_audio) free_sh_audio(sh_audio); //if(sh_video) free_sh_video(sh_video); // free demuxers: @@ -1059,6 +1059,7 @@ } //=============== Try to open as MPEG-ES file: ================= if(file_format==DEMUXER_TYPE_MPEG_ES || file_format==DEMUXER_TYPE_MPEG4_ES || file_format==DEMUXER_TYPE_H264_ES){ // little hack, see above! + if (demuxer) free_demuxer(demuxer); demuxer=new_demuxer(stream,file_format,audio_id,video_id,dvdsub_id); if(!ds_fill_buffer(demuxer->video)){ mp_msg(MSGT_DEMUXER,MSGL_ERR,MSGTR_InvalidMPEGES); diff -Naur -wbBE main-cvs/libmpdemux/demux_ogg.c main-leakfix/libmpdemux/demux_ogg.c --- main-cvs/libmpdemux/demux_ogg.c 2005-01-24 11:30:58.000000000 +0800 +++ main-leakfix/libmpdemux/demux_ogg.c 2005-02-04 17:31:26.000000000 +0800 @@ -764,6 +764,7 @@ /// Error if(np < 0) { mp_msg(MSGT_DEMUX,MSGL_DBG2,"Ogg demuxer : Bad page sync\n"); + ogg_sync_clear(sync); free(ogg_d); return 0; } @@ -773,6 +774,7 @@ buf = ogg_sync_buffer(sync,BLOCK_SIZE); len = stream_read(s,buf,BLOCK_SIZE); if(len == 0 && s->eof) { + ogg_sync_clear(sync); free(ogg_d); return 0; } @@ -1057,6 +1059,7 @@ } if(!n_video && !n_audio) { + ogg_sync_clear(sync); free(ogg_d); return 0; } diff -Naur -wbBE main-cvs/mplayer.c main-leakfix/mplayer.c --- main-cvs/mplayer.c 2005-01-24 11:30:56.000000000 +0800 +++ main-leakfix/mplayer.c 2005-02-04 17:31:26.000000000 +0800 @@ -501,7 +501,13 @@ current_module="exit_player"; // free mplayer config - free(mconfig); + m_config_free(mconfig); + + // free playtree + play_tree_free(playtree, 1); + + // free OSD + free_osd_list(); #ifdef USE_EDL if(edl_records != NULL) free(edl_records); // free mem allocated for EDL diff -Naur -wbBE main-cvs/osdep/getch2.c main-leakfix/osdep/getch2.c --- main-cvs/osdep/getch2.c 2004-04-06 05:33:08.000000000 +0800 +++ main-leakfix/osdep/getch2.c 2005-02-04 17:31:26.000000000 +0800 @@ -73,6 +73,7 @@ getch2_keys[getch2_key_db].code=code; ++getch2_key_db; /* printf("%s=%s\n",id,p); */ + if (p != term_p) free(p); } static int success=0;

3 3

[PATCH] Preliminary musepack support
by Amiaux Benoit 10 Jul '05

10 Jul '05

Hello, I'm one of the guys who ported libmusepack to *nix. Here's my opinion on the topic. Current decoder is able to playback musepack-only audio files fine. Yes, sv7 bitstream is bad. Yes, seeking is an issue. Current decoder has a slow but bit-perfect implementation of seeking. Another issue with seeking is that there are no limits to block dependencies. We had once a file with something like 1024 blocks dependencies, which obviously isn't nice for packetizing the bitstream. I wish you all good luck if you want to make a proper implementation around that in mpc-in-avi. So what about musepack support in mplayer ? A library of the encoder is currently in the making (which will be open-source too btw). Hopefully it'll make it much more easier to make a real bitstream for musepack, and why not, even choose nut or matroska as a 'native' container. So if any of you wants to work on musepack, I can only suggest to either do it completely, come help us, and put up a new bitstream, we'll modify decoder lib accordingly, mplayer will be able to mux/demux/play it properly. Or mplayer could be restricted to play audio-only musepack files, and seeking will be relatively slow. We already have vorbis in AVI, please no broken mpc in AVI. Grettings, Benoît Amiaux

5 6