Op zaterdag 5 maart 2016 19:39:01 schreef Carlos E. R.:

Hi,

I simply can't find a suitable video editor that works.

My goal is to extract the interesting parts of a TV recording. Ie, mark the start of the movie, the end, the commercials inside, and dump all that into a single file, perhaps recoding.

A few years ago I used kdenlive with success, but did not need two sound tracks on a machine with 4 cores and 4 GB memory. If I remember right, kdenlive just produced a job which got executed when finished and one could postpone the execution, making it possible to change the job. -- fr.gr. member openSUSE Freek de Kruijf -- To unsubscribe, e-mail: opensuse+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse+owner@opensuse.org

-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA256 On 2016-03-05 21:55, jdd wrote:

...

kdenlive - Well... crashes too often. I can mark the cuts, but it does not display what I'm cutting. And only one sound track.

I use kdenlive without problem for at least two years now (13.2, Leap), but for your language problem, I don't know. Avidemux (given his name) should do the job.

No, avidemux only reads one sound track from the mpeg file. It wants a second sound track to be on another different file.

If you can have video + langages, simply paste them in kdenlive audio tracks

How? The source mpeg file has one video track and two sound tracks. I want all three in the resulting edited file. Kdenlive ignores the second audio track. - -- Cheers / Saludos, Carlos E. R. (from 13.1 x86_64 "Bottle" (Minas Tirith)) -----BEGIN PGP SIGNATURE----- Version: GnuPG v2.0.22 (GNU/Linux) iF4EAREIAAYFAlbbVzEACgkQja8UbcUWM1zRCQEAjE65wNcD/FoGtIdoLwNBqDPd QXw7cNP23wijVin29TYBAI1r3A7BsmmllZfKHPEBOUJLG6DeAHrGvZ1C5wf6y420 =mmX/ -----END PGP SIGNATURE----- -- To unsubscribe, e-mail: opensuse+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse+owner@opensuse.org

Carlos E. R. composed on 2016-03-05 23:16 (UTC+0100):

I was reading a magazine comparison article on video editors for Linux, and the chap said that kdenlive crashed a lot on the distro he was using and had to use another distro where it was stable. I'll post the link later, it is on another machine.

Maybe: http://www.theregister.co.uk/2015/12/26/linux_video_editors/ but I get only timeouts trying. https://web.archive.org/web/20160305013750/http://www.theregister.co.uk/2015... only has page 1 -- "The wise are known for their understanding, and pleasant words are persuasive." Proverbs 16:21 (New Living Translation) Team OS/2 ** Reg. Linux User #211409 ** a11y rocks! Felix Miata *** http://fm.no-ip.com/ -- To unsubscribe, e-mail: opensuse+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse+owner@opensuse.org

On 2016-03-05 19:39, Carlos E. R. wrote:

Hi,

I simply can't find a suitable video editor that works.

I want to try "gopchop" (http://gopchop.sourceforge.net/). I don't find it ready built. So I try to build it. Configure fails here: checking mpeg2dec/mpeg2.h usability... no checking mpeg2dec/mpeg2.h presence... no checking for mpeg2dec/mpeg2.h... no configure: error: Couldn't find header mpeg2dec/mpeg2.h! cer@Telcontar:~/Compilaciones/MultiMedia/gopchop-1.1.8> It probably refers to "mpeg2dec - MPEG-2 Decoder" available in packman, but the -devel package, which is missing. I installed "mpeg2dec-debuginfo", but no go. I will have to install that one from source, too (http://libmpeg2.sourceforge.net/) Ok, I built it. I'll continue tomorrow. -- Cheers / Saludos, Carlos E. R. (from 13.1 x86_64 "Bottle" at Telcontar)

On 2016-03-06 02:06, Carlos E. R. wrote:

I want to try "gopchop" (http://gopchop.sourceforge.net/). I don't find it ready built. So I try to build it. Configure fails here:

checking mpeg2dec/mpeg2.h usability... no checking mpeg2dec/mpeg2.h presence... no checking for mpeg2dec/mpeg2.h... no configure: error: Couldn't find header mpeg2dec/mpeg2.h! cer@Telcontar:~/Compilaciones/MultiMedia/gopchop-1.1.8>

It probably refers to "mpeg2dec - MPEG-2 Decoder" available in packman, but the -devel package, which is missing. I installed "mpeg2dec-debuginfo", but no go. I will have to install that one from source, too (http://libmpeg2.sourceforge.net/)

Ok, I built it. I'll continue tomorrow.

mpeg2dec built and installed successfully (without an rpm, 'make install'). However, "make gopchop" produces C syntax errors with which I'm not familiar: ... Main.cpp:91:23: warning: deprecated conversion from string constant to ‘char*’ [-Wwrite-strings] char *GOPchop_cvsid = "$Id: Main.cpp 346 2009-03-02 23:26:02Z keescook $"; ^ Main.cpp:139:1: error: C99 designator ‘video_driver’ outside aggregate initializer }; ^ Main.cpp:152:1: warning: deprecated conversion from string constant to ‘char*’ [-Wwrite-strings] }; ^ Main.cpp:152:1: warning: deprecated conversion from string constant to ‘char*’ [-Wwrite-strings] Main.cpp:152:1: warning: deprecated conversion from string constant to ‘char*’ [-Wwrite-strings] Main.cpp:152:1: warning: deprecated conversion from string constant to ‘char*’ [-Wwrite-strings] Main.cpp:152:1: warning: deprecated conversion from string constant to ‘char*’ [-Wwrite-strings] Main.cpp:152:1: warning: deprecated conversion from string constant to ‘char*’ [-Wwrite-strings] Main.cpp:152:1: warning: deprecated conversion from string constant to ‘char*’ [-Wwrite-strings] Main.cpp:152:1: warning: deprecated conversion from string constant to ‘char*’ [-Wwrite-strings] Main.cpp:152:1: warning: deprecated conversion from string constant to ‘char*’ [-Wwrite-strings] Main.cpp:152:1: warning: deprecated conversion from string constant to ‘char*’ [-Wwrite-strings] Main.cpp: In function ‘void handle_rc_load()’: Main.cpp:828:39: warning: deprecated conversion from string constant to ‘char*’ [-Wwrite-strings] if (rc_load(PACKAGE,parsable_items)) ^ make[1]: *** [Main.o] Error 1 make[1]: Leaving directory `/data/homedvl/cer/MultiMedia/gopchop-1.1.8/src' make: *** [all-recursive] Error 1 cer@Telcontar:~/Compilaciones/MultiMedia/gopchop-1.1.8> -- Cheers / Saludos, Carlos E. R. (from 13.1 x86_64 "Bottle" at Telcontar)

On 03/05/2016 07:32 PM, Carlos E. R. wrote:

On 2016-03-06 02:20, Basil Chupin wrote:

...

On 06/03/16 05:39, Carlos E. R. wrote:

...

I know that you state below that you tried Cinelerra (https://en.wikipedia.org/wiki/Cinelerra) but believe me this is the one that you want to use. I have used it to edit recorded movies (removing commercials, eg) and it is the ants pants for editing videos. Spend just a LITTLE bit of time to read how to use it and you will not regret the time spent. It does not have a "help" menu entry. What do I read? Just resize the "compositor" window and I see redraw artifacts: bad coding.

I just told it to load a video (12 GB), and after a few minutes it still hasn't finished opening the file, nor do I see any change in the display. Ok, it finished loading, but I do not see any preview anywhere. I don't see where to move forward and backward to set the cut points.

I enter what seems cutpoints, and instead what it does is draw a waveform with the mouse. WHAT?

This is WAY OVER MY HEAD!

I just want a SIMPLE editor to remove regions of the movie, not a zillions windows and options I can't understand at all.

A the list of stating a war, I/we use Pinacle in Windows to edit shows for our TV station. Quick and simple. I've never seen a video editor for Linux that can come close for ease of use. -- Fast is fine, but accuracy is final. You must learn to be slow in a hurry. -Wyatt Earp- _ _... ..._ _ _._ ._ ..... ._.. ... .._ -- To unsubscribe, e-mail: opensuse+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse+owner@opensuse.org

Op zaterdag 5 maart 2016 19:39:01 CET schreef Carlos E. R.:

Hi,

I simply can't find a suitable video editor that works.

My goal is to extract the interesting parts of a TV recording. Ie, mark the start of the movie, the end, the commercials inside, and dump all that into a single file, perhaps recoding.

One limitation is that movies here have two sound tracks, one in Spanish, and the other one in the original language, typically English. I want to keep both, obviously.

I don't need any transition or effect, just a simple editor that allows me to remove unwanted sections.

I have tried:

pitivi - crashes on start

LiVES - works for an hour at 100% CPU and can't load the file.

dvbcut - says my mpeg file is not a video (and it certainly is!)

tragtor - can convert, but not edit.

openshot - almost. My video has two sound tracks, and this tool ignores the second track.

cinelerra-cv - I can't make head or tails of it. I tell it to load a video file, it does, but then I don't see how to do anything. It doesn't even display the video. I may have to read the manual, but I dislike having to read a manual for a GUI. Still...

kdenlive - Well... crashes too often. I can mark the cuts, but it does not display what I'm cutting. And only one sound track.

HandBrake - no edit.

Avidemux - one sound track only.

Kino - abandonware.

Lightworks ??

Shotcut - Segmentation fault on start.

cer@Telcontar:~/Videos/Fusion/trabajo> shotcut QSqlDatabase: QSQLITE driver not loaded QSqlDatabase: available drivers: QMYSQL QMYSQL3 QSqlQuery::exec: database not open QSqlQuery::exec: database not open Database::Database(QObject*) Failed to get version. QSqlQuery::exec: database not open bool Database::upgradeVersion1() Failed to create thumbnails table. Database version is 0 Segmentation fault cer@Telcontar:~/Videos/Fusion/trabajo>

Flowblade ?? I don't understand the tool. May need some reading. At least it runs.

I would like something that would produce a command line to ffmpeg in a GUI. Then I could modify further the options. This would be ideal. tragtor works this way, but has no editor.

-- Cheers

Carlos E. R. (from 13.1 x86_64 "Bottle" at Telcontar)

Carlos, I've done all you describe using kdenlive. More than any other application kdenlive needs its deps and recommends and it's own package to be from the Packman repo. -- Gertjan Lettink, a.k.a. Knurpht openSUSE Board Member openSUSE Forums Team -- To unsubscribe, e-mail: opensuse+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse+owner@opensuse.org

On 2016-03-06 02:21, Carlos E. R. wrote:

mpeg2dec built and installed successfully (without an rpm, 'make install'). However, "make gopchop" produces C syntax errors with which I'm not familiar:

...

Main.cpp:91:23: warning: deprecated conversion from string constant to ‘char*’ [-Wwrite-strings] char *GOPchop_cvsid = "$Id: Main.cpp 346 2009-03-02 23:26:02Z keescook $";

The code section was: char *GOPchop_cvsid = "$Id: Main.cpp 346 2009-03-02 23:26:02Z keescook $"; I change it to: const char *GOPchop_cvsid = "$Id: Main.cpp 346 2009-03-02 23:26:02Z keescook $"; and make is happy on this one after also changing that declaration in the main.h file. The next one, an error, not a warning. Main.cpp:139:1: error: C99 designator ‘video_driver’ outside aggregate initializer }; The code is: global_options options={ run_loop: 0, /* don't wrap arround by default */ run_speed: 1, /* one frame at a time by default */ default_run_speed: 1, /* one frame at a time by default */ auto_refresh: 1, /* show gop position changes by default */ ignore_errors: 0, /* don't ignore errors by default */ drop_orphaned_frames: 0, /* orphaned frame dropping can be nasty */ adjust_timestamps: 1, /* adjust by default */ video_driver_ptr: NULL, /* prefered video driver */ video_driver: "\0", force_system_header: 0, /* force prepended system header pack */ drop_trailing_pack_with_system_header: 0, /* drop final pack if it has a system header */ ignore_endcode: 0, /* don't stop parsing when an End Code is seen */ }; What do I do here? I think it refers to this one: video_driver: "\0", Let me see... the declaration is: typedef struct { gboolean run_loop; /* when in "Run" mode, wrap to beginning at end */ int run_speed; /* how many GOPs to jump forward during "run" */ int default_run_speed; /* the speed to start with upon load */ gboolean auto_refresh; /* redraw the frames after moving the slider */ gboolean ignore_errors; /* don't store errors about file loading */ gboolean drop_orphaned_frames; /* get rid of starting B frames on an open GOP */ gboolean adjust_timestamps; /* adjust the GOP timestamps as we write clips */ gchar* video_driver_ptr; --> gchar video_driver[24]; /* this is the name of the video driver to use */ gboolean force_system_header; /* should a system header be prepended */ gboolean drop_trailing_pack_with_system_header;/* should final pack with a system header be dropped? */ gboolean ignore_endcode; /* should parsing continue past an end code? */ } global_options; I have no idea what to write instead of video_driver: "\0", (that is the remaining hurdle) Next. Main.cpp:152:1: warning: deprecated conversion from string constant to ‘char*’ [-Wwrite-strings] }; ^ code is: rc_parse_item parsable_items[] = { { "run_loop", &options.run_loop, RC_TYPE_BOOLEAN }, { "default_run_speed", &options.default_run_speed, RC_TYPE_INTEGER }, { "auto_refresh", &options.auto_refresh, RC_TYPE_BOOLEAN }, { "ignore_errors", &options.ignore_errors, RC_TYPE_BOOLEAN }, { "drop_orphaned_frames", &options.drop_orphaned_frames, RC_TYPE_BOOLEAN }, { "adjust_timestamps", &options.adjust_timestamps, RC_TYPE_BOOLEAN }, { "video_driver_ptr", &options.video_driver_ptr, RC_TYPE_STRING }, { "force_system_header", &options.force_system_header, RC_TYPE_BOOLEAN }, { "drop_trailing_pack_with_system_header", &options.drop_trailing_pack_with_system_header, RC_TYPE_BOOLEAN }, { "ignore_endcode", &options.ignore_endcode, RC_TYPE_BOOLEAN }, { NULL, NULL, 0 } }; main.h: extern rc_parse_item parsable_items[]; rc.h: typedef struct rc_parse_item_t { char * option; /* name of the rc option */ void * variable; /* variable to load into */ int vartype; /* type of option to load */ } rc_parse_item; So I guess I have to change to: const char * option; I make. One error remains: Main.cpp:139:1: error: C99 designator ‘video_driver’ outside aggregate initializer }; ^ (this one is already above, and I don't know what to do about it) Next one is a warning: Main.cpp: In function ‘void handle_rc_load()’: Main.cpp:828:39: warning: deprecated conversion from string constant to ‘char*’ [-Wwrite-strings] if (rc_load(PACKAGE,parsable_items)) ^ The code is: typedef struct rc_parse_item_t { const char * option; /* name of the rc option */ void * variable; /* variable to load into */ int vartype; /* type of option to load */ } rc_parse_item; ... int rc_save(char * package, rc_parse_item * items); int rc_load(char * package, rc_parse_item * items); ... rc_parse_item parsable_items[] = { { "run_loop", &options.run_loop, RC_TYPE_BOOLEAN }, { "default_run_speed", &options.default_run_speed, RC_TYPE_INTEGER }, { "auto_refresh", &options.auto_refresh, RC_TYPE_BOOLEAN }, { "ignore_errors", &options.ignore_errors, RC_TYPE_BOOLEAN }, { "drop_orphaned_frames", &options.drop_orphaned_frames, RC_TYPE_BOOLEAN }, { "adjust_timestamps", &options.adjust_timestamps, RC_TYPE_BOOLEAN }, { "video_driver_ptr", &options.video_driver_ptr, RC_TYPE_STRING }, { "force_system_header", &options.force_system_header, RC_TYPE_BOOLEAN }, { "drop_trailing_pack_with_system_header", &options.drop_trailing_pack_with_system_header, RC_TYPE_BOOLEAN }, { "ignore_endcode", &options.ignore_endcode, RC_TYPE_BOOLEAN }, { NULL, NULL, 0 } }; if (rc_load(PACKAGE,parsable_items)) fprintf(stderr,"%s",_("Could not load rc file -- using defaults.\n")); I don't understand why it is giving a warning there :-? -- Cheers / Saludos, Carlos E. R. (from 13.1 x86_64 "Bottle" at Telcontar)

On 2016-03-06 10:30, Basil Chupin wrote:

...

It does not have a "help" menu entry. What do I read?

It's all here:

http://heroinewarrior.com/cinelerra/cinelerra.html

Huge manual full text, few images. I read a bit. Ok, I load my video. If I press the "play" icon, should it play? It does not. In the track timeline windows I only see Audio tracks, and a flat audio, no waveform. I can draw my own waverform with the mouse, whatever that is good for... In "viewver", play does nothing. Moving the romboid slider does nothing. I assume it should display images from the video, but it does not. I try another movie, much smaller. Same thing, nothing is displayed. I can't do anything with this software if I simply can't view the video... -- Cheers / Saludos, Carlos E. R. (from 13.1 x86_64 "Bottle" at Telcontar)

On 03/06/2016 08:32 AM, Carlos E. R. wrote:

If I define a new project with one video track and two audio tracks, then load a video, after loading it, resets to 4 empty audio tracks.

I forgot to mention that Pinnacle doesn't display two [ stereo ] audio. It shows a kind of composite audio track during edit. To be fair, the version I'm using is about four versions old so the newer versions have a way different GUI so..................... -- Fast is fine, but accuracy is final. You must learn to be slow in a hurry. -Wyatt Earp- _ _... ..._ _ _._ ._ ..... ._.. ... .._ -- To unsubscribe, e-mail: opensuse+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse+owner@opensuse.org

Hi, i use avidemux3_qt4 from the packman repo on opensuse 13.2. It is the only tool i found, that can cut h264 (HD-TV) videos and is not too complicated. I usually set: - Video Output to "Mpeg4 AVC (x264) - Audio Output to "Copy" -> this preserves all audio tracks! You can also use "Audio -> select track" to select tracks and transcode them individually - Output Format to "Mkv muxer" - personal favourite, but mp4 should be good too Greetings On 03/05/2016 07:39 PM, Carlos E. R. wrote:

Hi,

I simply can't find a suitable video editor that works.

My goal is to extract the interesting parts of a TV recording. Ie, mark the start of the movie, the end, the commercials inside, and dump all that into a single file, perhaps recoding.

One limitation is that movies here have two sound tracks, one in Spanish, and the other one in the original language, typically English. I want to keep both, obviously.

I don't need any transition or effect, just a simple editor that allows me to remove unwanted sections.

I have tried:

pitivi - crashes on start

LiVES - works for an hour at 100% CPU and can't load the file.

dvbcut - says my mpeg file is not a video (and it certainly is!)

tragtor - can convert, but not edit.

openshot - almost. My video has two sound tracks, and this tool ignores the second track.

cinelerra-cv - I can't make head or tails of it. I tell it to load a video file, it does, but then I don't see how to do anything. It doesn't even display the video. I may have to read the manual, but I dislike having to read a manual for a GUI. Still...

kdenlive - Well... crashes too often. I can mark the cuts, but it does not display what I'm cutting. And only one sound track.

HandBrake - no edit.

Avidemux - one sound track only.

Kino - abandonware.

Lightworks ??

Shotcut - Segmentation fault on start.

cer@Telcontar:~/Videos/Fusion/trabajo> shotcut QSqlDatabase: QSQLITE driver not loaded QSqlDatabase: available drivers: QMYSQL QMYSQL3 QSqlQuery::exec: database not open QSqlQuery::exec: database not open Database::Database(QObject*) Failed to get version. QSqlQuery::exec: database not open bool Database::upgradeVersion1() Failed to create thumbnails table. Database version is 0 Segmentation fault cer@Telcontar:~/Videos/Fusion/trabajo>

Flowblade ?? I don't understand the tool. May need some reading. At least it runs.

I would like something that would produce a command line to ffmpeg in a GUI. Then I could modify further the options. This would be ideal. tragtor works this way, but has no editor.

-- Cheers

Carlos E. R. (from 13.1 x86_64 "Bottle" at Telcontar)

On 2016-03-06 20:24, jdd wrote:

...

I load the video. If play automatically is selected in the open file dialogue, the application crashes immediately. I see a clip monitor window at the top right, so I try to use that.

don't try this, use directly the timeline, there is a key to cut. soory I write with W10 here, from Madrid :-(

In the timeline there is no guide whatsoever of what I'm marking. At least, when the video is 12GB, with a much smaller one when I click on the timeline I see the picture in the Project monitor. The timeline only displays one icon and a pink region. However, the point is moot: kdenlive only supports one audio track, and most of my videos have two tracks: Spanish and English. So kdenlive is out completely. -- Cheers / Saludos, Carlos E. R. (from 13.1 x86_64 "Bottle" at Telcontar)

Op zondag 6 maart 2016 22:02:58 schreef Carlos E. R.:

On 2016-03-06 20:24, jdd wrote:

...

...

I load the video. If play automatically is selected in the open file dialogue, the application crashes immediately. I see a clip monitor window at the top right, so I try to use that.

don't try this, use directly the timeline, there is a key to cut. soory I write with W10 here, from Madrid :-(

In the timeline there is no guide whatsoever of what I'm marking. At least, when the video is 12GB, with a much smaller one when I click on the timeline I see the picture in the Project monitor.

The timeline only displays one icon and a pink region.

However, the point is moot: kdenlive only supports one audio track, and most of my videos have two tracks: Spanish and English.

So kdenlive is out completely.

You may try the kdenlive mailing list should be on http://forum.kde.org, but I can't reach it. -- fr.gr. member openSUSE Freek de Kruijf -- To unsubscribe, e-mail: opensuse+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse+owner@opensuse.org

On 2016-03-06 16:44, David Haller wrote:

Hello,

On Sat, 05 Mar 2016, Carlos E. R. wrote:

...

No, avidemux only reads one sound track from the mpeg file. It wants a second sound track to be on another different file.

Try this:

Open the video, Cut as you wish. Then:

- goto Audio -> Main Track and select Audio-Track 1 (e.g. es, dubbed) - goto Audio -> Save (e.g. as out_1.ac3) - goto Audio -> Main Track and select Audio-Track 0 (e.g. en, orig) - goto File -> Save (e.g. as out.avi) - mux Video and Audio with avimerge or better mkvmerge, e.g.:

mkvmerge -o out.mkv out.avi out_1.ac3

This is an horrible hack. It must be done in a single shot by software that understand how to deal with multiple languages. Kdenlive does not support multiple language tracks, there is a bugzilla about it. And avidemux3 aborts, it is not usable. Maybe I have to use Windows to get a decent video editor :-/ -- Cheers / Saludos, Carlos E. R. (from 13.1 x86_64 "Bottle" at Telcontar)

On Sun, Mar 6, 2016 at 2:06 AM, Carlos E. R. wrote:

I want to try "gopchop" (http://gopchop.sourceforge.net/). I don't find it ready built. So I try to build it. Configure fails here:

I use gopchop for the edits I do. It depends on your source format. The nice thing about gopchop is that it does not decode/encode the file. It just cuts out whatever is between GOP markers. So it is fast, there is no degredation of the results, and whatever was in the source is in the destination. I just worry that toue media stream is more complicated than gopchop supports. I never had any problems building it. -- Roger Oberholtzer -- To unsubscribe, e-mail: opensuse+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse+owner@opensuse.org

"Carlos E. R." writes: Hi Carlos:

Avidemux - one sound track only.

Not true. If you go to "Audio"-> "Select Track", you should be able to select the tracks you want. Regards, Charles -- "By golly, I'm beginning to think Linux really *is* the best thing since sliced bread." (By Vance Petree, Virginia Power)

On 2016-03-07 07:46, Roger Oberholtzer wrote:

On Sun, Mar 6, 2016 at 2:06 AM, Carlos E. R. <> wrote:

...

I want to try "gopchop" (http://gopchop.sourceforge.net/). I don't find it ready built. So I try to build it. Configure fails here:

I use gopchop for the edits I do. It depends on your source format. The nice thing about gopchop is that it does not decode/encode the file. It just cuts out whatever is between GOP markers. So it is fast, there is no degredation of the results, and whatever was in the source is in the destination.

Exactly. That's why I want it.

I never had any problems building it.

What version? I have 1.1.8 from sourceforge, and the compiler throws an error: Main.cpp:139:1: error: C99 designator ‘video_driver’ outside aggregate initializer }; The code is: global_options options={ run_loop: 0, /* don't wrap arround by default */ run_speed: 1, /* one frame at a time by default */ default_run_speed: 1, /* one frame at a time by default */ auto_refresh: 1, /* show gop position changes by default */ ignore_errors: 0, /* don't ignore errors by default */ drop_orphaned_frames: 0, /* orphaned frame dropping can be nasty */ adjust_timestamps: 1, /* adjust by default */ video_driver_ptr: NULL, /* prefered video driver */ video_driver: "\0", force_system_header: 0, /* force prepended system header pack */ drop_trailing_pack_with_system_header: 0, /* drop final pack if it has a system header */ ignore_endcode: 0, /* don't stop parsing when an End Code is seen */ }; I think it refers to this one: video_driver: "\0", Let me see... the declaration is: typedef struct { gboolean run_loop; /* when in "Run" mode, wrap to beginning at end */ int run_speed; /* how many GOPs to jump forward during "run" */ int default_run_speed; /* the speed to start with upon load */ gboolean auto_refresh; /* redraw the frames after moving the slider */ gboolean ignore_errors; /* don't store errors about file loading */ gboolean drop_orphaned_frames; /* get rid of starting B frames on an open GOP */ gboolean adjust_timestamps; /* adjust the GOP timestamps as we write clips */ gchar* video_driver_ptr; --> gchar video_driver[24]; /* this is the name of the video driver to use */ gboolean force_system_header; /* should a system header be prepended */ gboolean drop_trailing_pack_with_system_header;/* should final pack with a system header be dropped? */ gboolean ignore_endcode; /* should parsing continue past an end code? */ } global_options; I have no idea what to write instead of video_driver: "\0", -- Cheers / Saludos, Carlos E. R. (from 13.1 x86_64 "Bottle" at Telcontar)

On 2016-03-07 13:21, Charles Philip Chan wrote:

"Carlos E. R." <> writes:

Hi Carlos:

...

Avidemux - one sound track only.

Not true. If you go to "Audio"-> "Select Track", you should be able to select the tracks you want.

avidemux2 only allows one track from the current file. Another track must come from an external file. avidemux3 can pick several tracks, apparently, but I have to specify which file for each track, instead of the thing picking automatically all tracks in the current video file. And anyway, avidemux3 aborts in the middle of the film, never finishes the job. -- Cheers / Saludos, Carlos E. R. (from 13.1 x86_64 "Bottle" at Telcontar)

Op maandag 7 maart 2016 15:41:11 CET schreef Carlos E. R.:

On 2016-03-07 13:21, Charles Philip Chan wrote:

...

"Carlos E. R." <> writes:

Hi Carlos:

...

Avidemux - one sound track only.

Not true. If you go to "Audio"-> "Select Track", you should be able to select the tracks you want.

avidemux2 only allows one track from the current file. Another track must come from an external file.

avidemux3 can pick several tracks, apparently, but I have to specify which file for each track, instead of the thing picking automatically all tracks in the current video file.

And anyway, avidemux3 aborts in the middle of the film, never finishes the job.

Knowing you're familiar with CLI, here's http://zoid.cc/12/12/ffmpeg-audio-video/ I tested this with a 2 minute video and 2 extra sound tracks. -- Gertjan Lettink, a.k.a. Knurpht openSUSE Board Member openSUSE Forums Team -- To unsubscribe, e-mail: opensuse+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse+owner@opensuse.org

I made the needed changes and gopchop compiles. At least on Leap. I can send you the modified files. When I get a chance I can put it on OBS. But you are welcome to the mods if you want them. On Mon, Mar 7, 2016 at 4:27 PM, Roger Oberholtzer wrote:

On Mon, Mar 7, 2016 at 3:01 PM, Carlos E. R. wrote:

...

On 2016-03-07 07:46, Roger Oberholtzer wrote:

...

...

I never had any problems building it.

What version? I have 1.1.8 from sourceforge, and the compiler throws an error:

I will check what I have at home. IIRC, the laptop I was using for this was recently replaced with Leap. So I guess I have a newer c++ compiler. I will see what I can come up with. I know that string constants do throw lots of compiler warnings if they are not specified as C++ now likes thenm. Not sure about the other errors.

Maybe I should add gopchop to OBS. If I get it to compile.

-- Roger Oberholtzer

-- Roger Oberholtzer -- To unsubscribe, e-mail: opensuse+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse+owner@opensuse.org

Le 07/03/2016 10:58, Florian Gleixner a écrit :

I can only guess that the recording has some problems like missing frames. For the audio: avidemux3 can have problems encoding audio when the audio tracks change during recording.

For broken files you may have a chance to fix it by cutting it with avidemux and select copy for both Video and Audio output. Then try to postprocess it with HandBrake. Or try to preprocess it with HandBrake and then cut it with avidemux3.

MTS (aka numerical TV files) often have faulty frames, and only a MTS reader can cope with this. This gives with most readers out of synv audio... I know of no solution. Was one in SD time, never ported to HD jdd -- To unsubscribe, e-mail: opensuse+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse+owner@opensuse.org

On 2016-03-07 20:20, jdd wrote:

MTS (aka numerical TV files) often have faulty frames, and only a MTS reader can cope with this. This gives with most readers out of synv audio... I know of no solution.

Most of my videos come from a TV recorder, and I do them using two programs in Java: cridmanager and projectX. It does no conversion, just allows cutting at the GOP point, allowing removal of any section. It is quite fast, I can "adapt" a dozen of videos at the same time. And it runs in Java, so works both in Linux and Windows. And it handles both language tracks, only that the ordering is unpredicatable. After having a file with the material I want, I can use ffmpeg to recode it at a smaller size. However, these new videos come from a different device. Instead of over the air, they are transmitted on fibre, and the result is already an mpeg file, of a huge size, like 12 or more gigabytes. Several resolutions. -- Cheers / Saludos, Carlos E. R. (from 13.1 x86_64 "Bottle" at Telcontar)

On 2016-03-07 16:24, Knurpht - Gertjan Lettink wrote:

Knowing you're familiar with CLI, here's http://zoid.cc/12/12/ffmpeg-audio-video/

I tested this with a 2 minute video and 2 extra sound tracks.

I know how to recode and crop a video with two audio tracks with ffmpeg, I do it often. The problem is to remove the minutes before the start of the movie, those beyond the end, and the commercials in the middle. With ffmpeg I can extract the section in the middle with ffmpeg. Tell it a start point in seconds and the duration, I think. I just need to find those points. However, some of these streams have a bad timeline, the seconds correspond to several seconds in the real world. I have seen a suggestion from someone here (sorry, I forget whom, busy day), about expanding (uncompressing) the video stream first, because some applications are happier. I'm unfamiliar with this technique. Anyone? -- Cheers / Saludos, Carlos E. R. (from 13.1 x86_64 "Bottle" at Telcontar)

On 2016-03-07 20:12, jdd wrote:

Maybe I have to use Windows to get a decent video editor :-/ you will have the same problems.

some notes:

* your file is very large. Can you record some example of 5 or 10 s and post them somewhere for we to try?

Well, I would have to do it with dd for reliability. No tool I know so far can copy video and audio (two tracks) cutting a region without crashing or manipulating the tracks. I can not tell the recorder to start and stop at points of my choosing. I have to tell it to record certain program, and it chooses the points from the official start and end times of the program. If I stop the recording, it deletes the file.

* I can't verify right now, but I'm reasonably sure ffmpeg can extract all video and audio from ant file

Yes, it can, but it does manipulation of the original so that may you would not see the problems I see.

* I think (only?) mkv can have several audio tracks

No, avi can as well. What is not typical is to include subtitles on avi, but it is on mkv.

When demuxing, all the tracks have the very same lengh (about time, not size, of course), so they can be worked on easily

it may be easier, then, to move them to mp4, for example, before editing to have the smaller file possible, because, of course, you can't edit the tracks separately :-(

No tool I know can do it. If I tell ffmpg to convert the entire stream to something, there are losses. If then use another editor to remove sections, there are more losses. -- Cheers / Saludos, Carlos E. R. (from 13.1 x86_64 "Bottle" at Telcontar)

Le 07/03/2016 20:43, Carlos E. R. a écrit :

It seems that if I want to keep a video, it is more hassle free to just use an emule client :-/ don't works anymore, use thepiratebay :-)

jdd -- To unsubscribe, e-mail: opensuse+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse+owner@opensuse.org

"Carlos E. R." writes: Hi Carlos:

Ok, got it, thankyou. Unfortunately, gopchop doesn't like my video files :-(

What device produce such terrible mpeg2 files? Maybe if you tell us, a Google search will yield more info. I uses to do a lot of commercial cutting with the mpeg2 files produced by my WinTV HVR-2250 with Avidemux with no problems (one audio track). Charles -- "I once witnessed a long-winded, month-long flamewar over the use of mice vs. trackballs...It was very silly." (By Matt Welsh)

Le 07/03/2016 21:17, Carlos E. R. a écrit :

Well, I would have to do it with dd for reliability.

dd can't cut any video file. key word is "GOP" (https://en.wikipedia.org/wiki/Group_of_pictures) jdd -- To unsubscribe, e-mail: opensuse+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse+owner@opensuse.org

On 2016-03-05 19:39, Carlos E. R. wrote: cer@Telcontar:~/Videos/Fusion/trabajo> time ffmpeg -i Las\ 13\ claves\ del\ Apolo\ 13\ \ Episodio\ 1.mpeg -target dv dv ffmpeg version 3.0 Copyright (c) 2000-2016 the FFmpeg developers ... Input #0, mpegts, from 'Las 13 claves del Apolo 13 Episodio 1.mpeg': Duration: 01:05:02.62, start: 1.049322, bitrate: 2299 kb/s Program 1 Stream #0:0[0x1e1]: Video: h264 (High) ([27][0][0][0] / 0x001B), yuv420p(tv, bt470bg), 720x576 [SAR 16:11 DAR 20:11], 25 fps, 50 tbr, 90k tbn, 50 tbc Stream #0:1[0x1e2](esp): Audio: mp2 ([3][0][0][0] / 0x0003), 48000 Hz, stereo, s16p, 192 kb/s (clean effects) Stream #0:2[0x1e3](vo): Audio: mp2 ([3][0][0][0] / 0x0003), 48000 Hz, stereo, s16p, 128 kb/s (clean effects) Could not determine norm (PAL/NTSC/NTSC-Film) for target. Please prefix target with "pal-", "ntsc-" or "film-", or set a framerate with "-r xxx". real 0m0.126s user 0m0.113s sys 0m0.012s cer@Telcontar:~/Videos/Fusion/trabajo> Huh? Framerate? Can't it read the framrate from input? Ok... cer@Telcontar:~/Videos/Fusion/trabajo> time ffmpeg -r 25 -i Las\ 13\ claves\ del\ Apolo\ 13\ \ Episodio\ 1.mpeg -target dv dv ... Could not determine norm (PAL/NTSC/NTSC-Film) for target. Please prefix target with "pal-", "ntsc-" or "film-", or set a framerate with "-r xxx". But I'm already using "-r xxx"! Ok, I'll use "-target film-dv" or "pal-dv". Somewhere mid video it finds sync problems. However, I can play the converted file. One audio track, and I'm unsure how to specify more when using "-target dv". Sigh. Reading manual again. By default, ffmpeg includes only one stream of each type (video, audio, subtitle) present in the input files and adds them to each output file. It picks the "best" of each based upon the following criteria: for video, it is the stream with the highest resolution, for audio, it is the stream with the most channels, for subtitles, it is the first subtitle stream. In the case where several streams of the same type rate equally, the stream with the lowest index is chosen. You can disable some of those defaults by using the "-vn/-an/-sn" options. For full manual control, use the "-map" option, which disables the defaults just described. It doesn't say an option to "include all streams"! :-( -map [-]input_file_id[:stream_specifier][,sync_file_id[:stream_specifier]] | [linklabel] (output) The first "-map" option on the command line specifies the source for output stream 0, the second "-map" option specifies the source for output stream 1, etc. So it should be -map 0:0 -map 0:1 -map 0:3 to include the three streams in the same order. But: cer@Telcontar:~/Videos/Fusion/trabajo> time ffmpeg -i Las\ 13\ claves\ del\ Apolo\ 13\ \ Episodio\ 1.mpeg -map 0:0 -map 0:1 -map 0:2 -target pal-dv dv-pal ffmpeg version 3.0 Copyright (c) 2000-2016 the FFmpeg developers built with gcc 4.8 (SUSE Linux) ... Input #0, mpegts, from 'Las 13 claves del Apolo 13 Episodio 1.mpeg': Duration: 01:05:02.62, start: 1.049322, bitrate: 2299 kb/s Program 1 Stream #0:0[0x1e1]: Video: h264 (High) ([27][0][0][0] / 0x001B), yuv420p(tv, bt470bg), 720x576 [SAR 16:11 DAR 20:11], 25 fps, 50 tbr, 90k tbn, 50 tbc Stream #0:1[0x1e2](esp): Audio: mp2 ([3][0][0][0] / 0x0003), 48000 Hz, stereo, s16p, 192 kb/s (clean effects) Stream #0:2[0x1e3](vo): Audio: mp2 ([3][0][0][0] / 0x0003), 48000 Hz, stereo, s16p, 128 kb/s (clean effects) File 'dv-pal' already exists. Overwrite ? [y/N] y [dv @ 0x1bc92c0] Can't initialize DV format! Make sure that you supply exactly two streams: <======================================= video: 25fps or 29.97fps, audio: 2ch/48|44|32kHz/PCM (50Mbps allows an optional second audio stream) Output #0, dv, to 'dv-pal': Metadata: encoder : Lavf57.25.100 Stream #0:0: Video: dvvideo, yuv420p, 720x576 [SAR 16:11 DAR 20:11], q=2-31, 200 kb/s, 25 fps, 25 tbn, 25 tbc Metadata: encoder : Lavc57.24.102 dvvideo Stream #0:1(esp): Audio: pcm_s16le, 48000 Hz, stereo, s16, 1536 kb/s (clean effects) Metadata: encoder : Lavc57.24.102 pcm_s16le Stream #0:2(vo): Audio: pcm_s16le, 48000 Hz, stereo, s16, 1536 kb/s (clean effects) Metadata: encoder : Lavc57.24.102 pcm_s16le Stream mapping: Stream #0:0 -> #0:0 (h264 (native) -> dvvideo (native)) Stream #0:1 -> #0:1 (mp2 (native) -> pcm_s16le (native)) Stream #0:2 -> #0:2 (mp2 (native) -> pcm_s16le (native)) Could not write header for output file #0 (incorrect codec parameters ?): Operation not permitted real 0m2.878s user 0m0.115s sys 0m0.020s cer@Telcontar:~/Videos/Fusion/trabajo> It seems that "DV" only allows two tracks, so I can't do the second language. There is a comment above which I do not understand: "50Mbps allows an optional second audio stream". Where? -- Cheers / Saludos, Carlos E. R. (from 13.1 x86_64 "Bottle" at Telcontar)

On 2016-03-07 22:08, Charles Philip Chan wrote:

"Carlos E. R." writes:

Hi Carlos:

...

Ok, got it, thankyou. Unfortunately, gopchop doesn't like my video files :-(

What device produce such terrible mpeg2 files? Maybe if you tell us, a Google search will yield more info. I uses to do a lot of commercial cutting with the mpeg2 files produced by my WinTV HVR-2250 with Avidemux with no problems (one audio track).

It is not hardware, it is my ISP. You get a client, they do the recording in their servers, not my home. Later I can play it on the sitting room TV, or in my computer, using kodi and a third party plugin. This plugin is able to record locally a recording in the cloud. I may search the forum that recommended this plugin and see what they use for edit, but most of them are Windows people. -- Cheers / Saludos, Carlos E. R. (from 13.1 x86_64 "Bottle" at Telcontar)

Le 07/03/2016 22:20, Carlos E. R. a écrit :

On 2016-03-05 19:39, Carlos E. R. wrote:

cer@Telcontar:~/Videos/Fusion/trabajo> time ffmpeg -i Las\ 13\ claves\ del\ Apolo\ 13\ \ Episodio\ 1.mpeg -target dv dv ffmpeg version 3.0 Copyright (c) 2000-2016 the FFmpeg developers

... Input #0, mpegts, from 'Las 13 claves del Apolo 13 Episodio 1.mpeg': note: mpegts: TS. so MTS file, not real mpg as we usually call them.

is it SD or HD? SD is about to be stopped in France (April 4, from memory). i see 720x576, so SD. use projectX to fix them. http://dodin.info/wiki/pmwiki.php?n=Doc.RecupererUneVideoTNT (sorry, in french) jdd -- To unsubscribe, e-mail: opensuse+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse+owner@opensuse.org

Hello, On Mon, 07 Mar 2016, Roger Oberholtzer wrote:

I made the needed changes and gopchop compiles.

A few ;)

At least on Leap. I can send you the modified files. When I get a chance I can put it on OBS. But you are welcome to the mods if you want them.

On Mon, Mar 7, 2016 at 4:27 PM, Roger Oberholtzer [..]

...

I will check what I have at home. IIRC, the laptop I was using for this was recently replaced with Leap. So I guess I have a newer c++ compiler. I will see what I can come up with. I know that string constants do throw lots of compiler warnings if they are not specified as C++ now likes thenm. Not sure about the other errors.

Maybe I should add gopchop to OBS. If I get it to compile.

I have a clean compile (from 12.1 to Leap) at http://download.opensuse.org/repositories/home:/dnh/ heavily based on your diffs and .spec, even though I changed/reworked quite a bit too ;) Have fun, -dnh -- "HPUX" and "sane" have never been uttered in the same sentence without accompanying negatives. -- Richard Henderson on gcc ml -- To unsubscribe, e-mail: opensuse+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse+owner@opensuse.org

Hello, On Mon, 07 Mar 2016, Carlos E. R. wrote:

On 2016-03-07 13:21, Charles Philip Chan wrote:

...

"Carlos E. R." <> writes:

...

Avidemux - one sound track only.

Not true. If you go to "Audio"-> "Select Track", you should be able to select the tracks you want.

avidemux2 only allows one track from the current file. Another track must come from an external file.

Nope. You can cut the file as descriped in the parallel post, and then save first the second and more audio-tracks to an external file by selecting each as Main audio track (Audio -> Main track), then lastly select the "main" audio track as "Main Audio" and save that together with the video (or also as external file). Then mux it all together with mkvmerge.

avidemux3 can pick several tracks, apparently, but I have to specify which file for each track, instead of the thing picking automatically all tracks in the current video file.

Works for me.

And anyway, avidemux3 aborts in the middle of the film, never finishes the job.

Broken file ... One could try massaging it with mencoder/handbrake or so, depending what's broken (whole stream, just video or audio). -dnh -- Don't tell my momma I'm a sysadmin, she thinks I play piano in a whorehouse." -- Alan J Rosenthal -- To unsubscribe, e-mail: opensuse+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse+owner@opensuse.org

On Tue, Mar 8, 2016 at 8:06 AM, David Haller wrote:

Hello,

On Mon, 07 Mar 2016, Roger Oberholtzer wrote:

...

I made the needed changes and gopchop compiles.

A few ;)

I just did the few to remove compiler errors and warnings. But the printf() corrections you did also seem needed. gopchop has some unneccessarily complicated code, IMHO. But then again, I am a C type of guy.

I have a clean compile (from 12.1 to Leap) at http://download.opensuse.org/repositories/home:/dnh/

heavily based on your diffs and .spec, even though I changed/reworked quite a bit too ;)

I was fighting with the libxml2 stuff. It refused to complete the configure step. I see your different BuildRequires for that and a few other libs. Why are these needing to be specified differently? -- Roger Oberholtzer -- To unsubscribe, e-mail: opensuse+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse+owner@opensuse.org

Hello, On Tue, 08 Mar 2016, Roger Oberholtzer wrote:

On Tue, Mar 8, 2016 at 8:06 AM, David Haller wrote:

...

Hello,

On Mon, 07 Mar 2016, Roger Oberholtzer wrote:

...

I made the needed changes and gopchop compiles.

A few ;)

I just did the few to remove compiler errors and warnings. But the printf() corrections you did also seem needed. gopchop has some unneccessarily complicated code, IMHO. But then again, I am a C type of guy.

I've seen much worse, actually, I find it rather readable...

...

I have a clean compile (from 12.1 to Leap) at http://download.opensuse.org/repositories/home:/dnh/

heavily based on your diffs and .spec, even though I changed/reworked quite a bit too ;)

I was fighting with the libxml2 stuff. It refused to complete the configure step.

Huh? That worked here out of the box after I allowed 'build' to actually install libxml2 (I first commented out all BuildRequires just for the heck of it) and pkgconfig(libxml-2.0) vs. libxml2-devel should not make a difference here. Can you check / mail me / paste.opensuse.org a config.log of one of those failed builds?

I see your different BuildRequires for that and a few other libs. Why are these needing to be specified differently?

Generally, using 'pkgconfig(foo)' for packages that provide it is much cleaner and will work on any distro that supports 'pkgconfig()'-deps (which should be most by now). So, it won't matter if it's foo4-dev, foo-dev, foo-devel, or whatnot that contains the foo.pc, all provide pkgconfig(foo) and all is well. It's sort of being able to use 'BuildRequires: /usr/bin/foo-config' for a package that has a -config script. And the Requires get sorted out by rpm itself (depending on e.g. 'libmpeg2.so.0()(64bit)'), so generally you should not use those unless you know you need them. Which is why the trick with the dummy libmpeg2 (just headers and dummy libs) works, so even as libmpeg2 is "forbidden" in OBS, we can build gopchop "legally" in OBS. And BTW: the libmpeg2 stuff is in libmpeg2-0 and libmpeg2-devel and mpeg2dec, that latter contains (just) the binaries in /usr/bin/ (plus manpages). We should better take this to the -packaging list or PM. -dnh -- / "I dunno, it strikes me that whatever direction you read the Bible, \ \ it ends with the destruction of the world." -- Dave Brown / -- To unsubscribe, e-mail: opensuse+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse+owner@opensuse.org

On 2016-03-05 19:39, Carlos E. R. wrote:

Hi,

I simply can't find a suitable video editor that works.

My goal is to extract the interesting parts of a TV recording. Ie, mark the start of the movie, the end, the commercials inside, and dump all that into a single file, perhaps recoding.

Well, taking info from several posts here, I'm trying with ffmpeg in the CLI. I use avidemux to find the cut points and the crop values. In the sample I'm testing, I have to remove 138 lines top and bottom. The filter says that it is cropping from "1920x1080 ==> 1920*804" Thus the CLI option for ffmpeg has then to be "-vf crop=1920:804:0:138" For start and end times, also use avidemux. It starts on 00:05:44.600 and ends on 01:58:30.520. The parameter for ffmpeg is start and duration, so calculate the duration in LO calc. It becomes "-ss 00:05:44.600 -t 01:52:45.92" I include "-maps" for all streams, and I add "-async" to protect against sync errors. The full line becomes: ffmpeg -i "Star Trek.mpeg" \ -ss 00:05:44.600 -t 01:52:45.92 \ -map 0:0 -map 0:1 -map 0:2 \ -q:v 4 -q:a 1 -async 1 -vcodec libxvid \ -vf crop=1920:804:0:138 \ -acodec libmp3lame StarTrek.avi However, the movie ends short: the timestamps in the resulting video does not match those in the input. I had to convert the entire input file, after failing on 3 earlier attempts that also ended short. The stop point in the output is 3:14:25, which is absurd. The problem may be different. (It would be nice to be able to do a fast and dirty conversion for these tests, instead of waiting an hour or two for each test, but I don't know how) The problem is that sound is choppy, but not at the start, but since the point 15:30. There is a chirp, then the sound plays and mutes every second. Maybe those silent periods every second are adding to the total play time. As it is, of course, the movie is unwatchable. The full output on the CLI is: cer@Telcontar:~/Videos/Fusion/trabajo> ffmpeg -i "La pelicula de la semana Star Trek.mpeg" -ss 00:05:44.600 -map 0:0 -map 0:1 -map 0:2 -q:v 4 -q:a 1 -async 1 -vcodec libxvid -vf crop=1920:804:0:138 -acodec libmp3lame StarTrek.avi ffmpeg version 3.0 Copyright (c) 2000-2016 the FFmpeg developers built with gcc 4.8 (SUSE Linux) configuration: --prefix=/usr --libdir=/usr/lib64 --shlibdir=/usr/lib64 --incdir=/usr/include/ffmpeg --extra-cflags='-fmessage-length=0 -grecord-gcc-switches -fstack-protector -O2 -Wall -D_FORTIFY_SOURCE=2 -funwind-tables -fasynchronous-unwind-tables -g' --optflags='-fmessage-length=0 -grecord-gcc-switches -fstack-protector -O2 -Wall -D_FORTIFY_SOURCE=2 -funwind-tables -fasynchronous-unwind-tables -g' --disable-htmlpages --enable-pic --disable-stripping --enable-shared --disable-static --enable-runtime-cpudetect --enable-gpl --disable-openssl --enable-avresample --enable-libcdio --enable-gnutls --enable-ladspa --enable-libass --enable-libbluray --enable-libcelt --enable-libcdio --enable-libdc1394 --enable-libfreetype --enable-libgsm --enable-libopenjpeg --enable-libopus --enable-libpulse --enable-libschroedinger --enable-libspeex --enable-libtheora --enable-libvorbis --enable-libvpx --enable-libwebp --enable-pic --enable-pthreads --enable-vaapi --enable-vdpau --disable-decoder=dca -- enable-libdcadec --enable-libmp3lame --enable-libopencore-amrnb --enable-libopencore-amrwb --enable-version3 --enable-libtwolame --enable-libx264 --enable-libx265 --enable-libxvid --enable-x11grab libavutil 55. 17.103 / 55. 17.103 libavcodec 57. 24.102 / 57. 24.102 libavformat 57. 25.100 / 57. 25.100 libavdevice 57. 0.101 / 57. 0.101 libavfilter 6. 31.100 / 6. 31.100 libavresample 3. 0. 0 / 3. 0. 0 libswscale 4. 0.100 / 4. 0.100 libswresample 2. 0.101 / 2. 0.101 libpostproc 54. 0.100 / 54. 0.100 [h264 @ 0x703280] mmco: unref short failure [h264 @ 0x703280] Increasing reorder buffer to 2 [h264 @ 0x703280] Increasing reorder buffer to 3 [h264 @ 0x703280] mmco: unref short failure Input #0, mpegts, from 'La pelicula de la semana Star Trek.mpeg': Duration: 03:01:02.68, start: 1.017833, bitrate: 9902 kb/s Program 1 Stream #0:0[0x1e1]: Video: h264 (High) ([27][0][0][0] / 0x001B), yuv420p(tv, bt709), 1920x1080 [SAR 1:1 DAR 16:9], 25 fps, 25 tbr, 90k tbn, 50 tbc Stream #0:1[0x1e2](esp): Audio: mp2 ([3][0][0][0] / 0x0003), 48000 Hz, stereo, s16p, 192 kb/s (clean effects) Stream #0:2[0x1e3](AU2): Audio: mp2 ([3][0][0][0] / 0x0003), 48000 Hz, stereo, s16p, 128 kb/s (clean effects) File 'StarTrek.avi' already exists. Overwrite ? [y/N] y -async is forwarded to lavfi similarly to -af aresample=async=1:min_hard_comp=0.100000:first_pts=0. Last message repeated 1 times Output #0, avi, to 'StarTrek.avi': Metadata: ISFT : Lavf57.25.100 Stream #0:0: Video: mpeg4 (libxvid) (xvid / 0x64697678), yuv420p, 1920x804 [SAR 1:1 DAR 160:67], q=2-31, 200 kb/s, 25 fps, 25 tbn, 25 tbc Metadata: encoder : Lavc57.24.102 libxvid Stream #0:1(esp): Audio: mp3 (libmp3lame) (U[0][0][0] / 0x0055), 48000 Hz, stereo, s16p (clean effects) Metadata: encoder : Lavc57.24.102 libmp3lame Stream #0:2(AU2): Audio: mp3 (libmp3lame) (U[0][0][0] / 0x0055), 48000 Hz, stereo, s16p (clean effects) Metadata: encoder : Lavc57.24.102 libmp3lame Stream mapping: Stream #0:0 -> #0:0 (h264 (native) -> mpeg4 (libxvid)) Stream #0:1 -> #0:1 (mp2 (native) -> mp3 (libmp3lame)) Stream #0:2 -> #0:2 (mp2 (native) -> mp3 (libmp3lame)) Press [q] to stop, [?] for help [h264 @ 0x25fa5e0] mmco: unref short failure [h264 @ 0x26cf940] mmco: unref short failure [h264 @ 0x25a1300] mmco: unref short failureme=00:15:29.44 bitrate=5716.4kbits/s speed=0.858x [mpegts @ 0x6feee0] PES packet size mismatch Last message repeated 1 times [mp2 @ 0x705e80] Header missing Error while decoding stream #0:1: Invalid data found when processing input [mp2 @ 0x706340] Header missing Error while decoding stream #0:2: Invalid data found when processing input [h264 @ 0x2688800] mmco: unref short failure [h264 @ 0x25fa5e0] left block unavailable for requested intra mode at 0 12bits/s dup=0 drop=10 speed=1.08x [h264 @ 0x25fa5e0] error while decoding MB 0 12, bytestream 2428 [h264 @ 0x25fa5e0] concealing 6769 DC, 6769 AC, 6769 MV errors in I frame [h264 @ 0x26416c0] reference picture missing during reorder Last message repeated 1 times [h264 @ 0x26416c0] Missing reference picture, default is 65608 Last message repeated 1 times [h264 @ 0x2688800] mmco: unref short failure [h264 @ 0x26cf940] co located POCs unavailable=00:25:51.08 bitrate=4344.6kbits/s dup=0 drop=10 speed=1.08x [h264 @ 0x25a1300] co located POCs unavailable [h264 @ 0x25fa5e0] mmco: unref short failure [h264 @ 0x2688800] Reference 2 >= 22864kB time=00:25:51.84 bitrate=4343.8kbits/s dup=0 drop=10 speed=1.08x [h264 @ 0x2688800] error while decoding MB 73 18, bytestream 13349 [h264 @ 0x2688800] concealing 5976 DC, 5976 AC, 5976 MV errors in B frame [h264 @ 0x26cf940] co located POCs unavailable [h264 @ 0x25fa5e0] mmco: unref short failureme=00:25:52.89 bitrate=4344.7kbits/s dup=0 drop=10 speed=1.09x [h264 @ 0x26cf940] mmco: unref short failure [h264 @ 0x25a1300] left block unavailable for requested intra mode at 0 22bits/s dup=0 drop=10 speed= 1.3x [h264 @ 0x25a1300] error while decoding MB 0 22, bytestream 46017 [h264 @ 0x25a1300] concealing 5569 DC, 5569 AC, 5569 MV errors in P frame [h264 @ 0x25fa5e0] co located POCs unavailable [h264 @ 0x26416c0] co located POCs unavailable [h264 @ 0x25a1300] co located POCs unavailable=00:57:47.72 bitrate=3501.0kbits/s dup=0 drop=10 speed= 1.3x [h264 @ 0x2688800] reference picture missing during reorder Last message repeated 1 times [h264 @ 0x2688800] Missing reference picture, default is 65550 Last message repeated 1 times [h264 @ 0x26416c0] mmco: unref short failure [mpegts @ 0x6feee0] PES packet size mismatchme=00:57:49.20 bitrate=3499.9kbits/s dup=0 drop=10 speed= 1.3x [h264 @ 0x25fa5e0] concealing 960 DC, 960 AC, 960 MV errors in B frame [h264 @ 0x26416c0] co located POCs unavailable [h264 @ 0x2688800] number of reference frames (0+5) exceeds max (4; probably corrupt input), discarding one [mp2 @ 0x705e80] Header missing Error while decoding stream #0:1: Invalid data found when processing input [h264 @ 0x2688800] co located POCs unavailable [h264 @ 0x25fa5e0] co located POCs unavailable [h264 @ 0x25a1300] mmco: unref short failure [h264 @ 0x2688800] mmco: unref short failureme=00:57:51.57 bitrate=3497.5kbits/s dup=0 drop=10 speed= 1.3x Past duration 0.999992 too large 3016122kB time=02:05:30.28 bitrate=3281.2kbits/s dup=0 drop=10 speed=1.19x Last message repeated 3 times Past duration 0.999992 too large 3016455kB time=02:05:30.56 bitrate=3281.4kbits/s dup=0 drop=10 speed=1.19x Last message repeated 6 times Past duration 0.999992 too large 3016510kB time=02:05:30.84 bitrate=3281.3kbits/s dup=0 drop=10 speed=1.19x Last message repeated 1 times [mpegts @ 0x6feee0] PES packet size mismatchime=02:08:40.64 bitrate=3253.7kbits/s dup=0 drop=10 speed= 1.2x [mp2 @ 0x705e80] Header missing Error while decoding stream #0:1: Invalid data found when processing input [h264 @ 0x25fa5e0] concealing 1880 DC, 1880 AC, 1880 MV errors in I frame [h264 @ 0x26416c0] co located POCs unavailable [h264 @ 0x2688800] co located POCs unavailable [h264 @ 0x25fa5e0] co located POCs unavailable [h264 @ 0x26cf940] reference picture missing during reorder [h264 @ 0x26cf940] Missing reference picture, default is 65646 [h264 @ 0x25a1300] mmco: unref short failure [h264 @ 0x26416c0] reference picture missing during reorder [h264 @ 0x26416c0] Missing reference picture, default is 65662 [h264 @ 0x2688800] mmco: unref short failure [h264 @ 0x25a1300] left block unavailable for requested intra mode at 0 32kbits/s dup=0 drop=10 speed= 1.2x [h264 @ 0x25a1300] error while decoding MB 0 32, bytestream 32731 [h264 @ 0x25a1300] concealing 4369 DC, 4369 AC, 4369 MV errors in I frame [h264 @ 0x25fa5e0] co located POCs unavailable [h264 @ 0x26416c0] co located POCs unavailable [h264 @ 0x2688800] co located POCs unavailable [mpegts @ 0x6feee0] PES packet size mismatch [h264 @ 0x26cf940] co located POCs unavailable [mp2 @ 0x706340] Header missing Error while decoding stream #0:2: Invalid data found when processing input [h264 @ 0x25fa5e0] reference picture missing during reorder Last message repeated 1 times [h264 @ 0x25fa5e0] Missing reference picture, default is 65566 Last message repeated 1 times [h264 @ 0x25a1300] mmco: unref short failure [h264 @ 0x26cf940] left block unavailable for requested intra4x4 mode -1 at 0 18s dup=0 drop=10 speed=1.23x [h264 @ 0x26cf940] error while decoding MB 0 18, bytestream 10738 [h264 @ 0x26cf940] concealing 6049 DC, 6049 AC, 6049 MV errors in I frame [h264 @ 0x25a1300] co located POCs unavailable [h264 @ 0x26416c0] reference picture missing during reorder bitrate=3216.8kbits/s dup=0 drop=10 speed=1.23x Last message repeated 1 times [h264 @ 0x26416c0] Missing reference picture, default is 65646 Last message repeated 1 times [h264 @ 0x2688800] mmco: unref short failure [h264 @ 0x25fa5e0] mmco: unref short failureime=02:24:43.32 bitrate=3215.8kbits/s dup=0 drop=10 speed=1.23x [h264 @ 0x26cf940] concealing 4464 DC, 4464 AC, 4464 MV errors in B frame7kbits/s dup=0 drop=10 speed=1.23x [h264 @ 0x25a1300] co located POCs unavailable [h264 @ 0x25fa5e0] co located POCs unavailable [h264 @ 0x26cf940] co located POCs unavailablee=02:24:44.60 bitrate=3216.2kbits/s dup=0 drop=10 speed=1.23x [mpegts @ 0x6feee0] PES packet size mismatch [mp2 @ 0x705e80] Header missing Error while decoding stream #0:1: Invalid data found when processing input [h264 @ 0x26416c0] reference picture missing during reorder Last message repeated 1 times [h264 @ 0x26416c0] Missing reference picture, default is 65550 Last message repeated 1 times [h264 @ 0x25a1300] reference picture missing during reorder [h264 @ 0x25a1300] Missing reference picture, default is 65566 [h264 @ 0x25fa5e0] mmco: unref short failure [h264 @ 0x26cf940] concealing 2131 DC, 2131 AC, 2131 MV errors in B frame7kbits/s dup=0 drop=10 speed=1.25x [mpegts @ 0x6feee0] DTS 633356805 < 633587205 out of order [h264 @ 0x25a1300] co located POCs unavailable [h264 @ 0x25fa5e0] co located POCs unavailable [mpegts @ 0x6feee0] PES packet size mismatch [mp2 @ 0x706340] Header missing= 4510495kB time=03:12:06.40 bitrate=3205.7kbits/s dup=0 drop=10 speed=1.25x Error while decoding stream #0:2: Invalid data found when processing input [h264 @ 0x26416c0] co located POCs unavailable [h264 @ 0x2688800] co located POCs unavailable [h264 @ 0x25a1300] reference picture missing during reorder Last message repeated 1 times [h264 @ 0x25a1300] Missing reference picture, default is 65710 Last message repeated 1 times [h264 @ 0x26cf940] mmco: unref short failure [h264 @ 0x26cf940] Reference 2 >= 210806kB time=03:12:09.02 bitrate=3205.2kbits/s dup=0 drop=10 speed=1.25x [h264 @ 0x26cf940] error while decoding MB 11 46, bytestream 93852 [h264 @ 0x26cf940] concealing 2678 DC, 2678 AC, 2678 MV errors in B frame [h264 @ 0x25a1300] co located POCs unavailable [h264 @ 0x25fa5e0] number of reference frames (0+5) exceeds max (4; probably corrupt input), discarding one [h264 @ 0x25a1300] Reference 2 >= 272955kB time=03:43:53.40 bitrate=3093.6kbits/s dup=0 drop=10 speed=1.27x [h264 @ 0x25a1300] error while decoding MB 99 38, bytestream 189042 [h264 @ 0x25a1300] concealing 3550 DC, 3550 AC, 3550 MV errors in P frame [h264 @ 0x25fa5e0] co located POCs unavailable [h264 @ 0x2688800] reference picture missing during reorder bitrate=3093.5kbits/s dup=0 drop=10 speed=1.27x Last message repeated 1 times [h264 @ 0x2688800] Missing reference picture, default is 65694 Last message repeated 1 times [h264 @ 0x26cf940] mmco: unref short failure [h264 @ 0x26416c0] mmco: unref short failureime=03:43:54.68 bitrate=3093.5kbits/s dup=0 drop=10 speed=1.27x [mpegts @ 0x6feee0] PES packet size mismatchime=03:43:55.28 bitrate=3093.7kbits/s dup=0 drop=10 speed=1.27x [h264 @ 0x2688800] Reference 3 >= 2 [h264 @ 0x2688800] error while decoding MB 58 28, bytestream 67150 [h264 @ 0x2688800] concealing 4791 DC, 4791 AC, 4791 MV errors in B frame [h264 @ 0x26cf940] co located POCs unavailable [h264 @ 0x25a1300] number of reference frames (0+5) exceeds max (4; probably corrupt input), discarding one [mp2 @ 0x706340] Header missing Error while decoding stream #0:2: Invalid data found when processing input [h264 @ 0x25fa5e0] co located POCs unavailable [mpegts @ 0x6feee0] PES packet size mismatchime=03:59:51.86 bitrate=3030.2kbits/s dup=0 drop=10 speed=1.29x Last message repeated 1 times [mp2 @ 0x706340] Header missing Error while decoding stream #0:2: Invalid data found when processing input [h264 @ 0x25fa5e0] Reference 2 >= 2 [h264 @ 0x25fa5e0] error while decoding MB 77 62, bytestream 34200 [h264 @ 0x25fa5e0] concealing 692 DC, 692 AC, 692 MV errors in B frame [h264 @ 0x26416c0] co located POCs unavailable [h264 @ 0x2688800] co located POCs unavailable [mp2 @ 0x705e80] Header missing= 5323692kB time=03:59:52.96 bitrate=3030.1kbits/s dup=0 drop=10 speed=1.29x Error while decoding stream #0:1: Invalid data found when processing input [h264 @ 0x26cf940] co located POCs unavailable [h264 @ 0x25a1300] co located POCs unavailable [h264 @ 0x25fa5e0] co located POCs unavailable [h264 @ 0x2688800] reference picture missing during reorder Last message repeated 1 times [h264 @ 0x2688800] Missing reference picture, default is 65614 Last message repeated 1 times [h264 @ 0x26416c0] mmco: unref short failure [mpegts @ 0x6feee0] PES packet size mismatch [h264 @ 0x25a1300] concealing 2151 DC, 2151 AC, 2151 MV errors in B frame [h264 @ 0x26416c0] co located POCs unavailable [mp2 @ 0x705e80] Header missing Error while decoding stream #0:1: Invalid data found when processing input [h264 @ 0x25a1300] mmco: unref short failure [mpegts @ 0x6feee0] PES packet size mismatchime=04:47:32.48 bitrate=3106.2kbits/s dup=0 drop=10 speed=1.31x [h264 @ 0x25a1300] Reference 2 >= 2 [h264 @ 0x25a1300] error while decoding MB 18 6, bytestream 1959 [h264 @ 0x25a1300] concealing 7471 DC, 7471 AC, 7471 MV errors in B frame [h264 @ 0x25fa5e0] co located POCs unavailablee=04:47:33.20 bitrate=3106.3kbits/s dup=0 drop=10 speed=1.31x [h264 @ 0x26416c0] co located POCs unavailable [h264 @ 0x2688800] co located POCs unavailable [h264 @ 0x26cf940] co located POCs unavailable [h264 @ 0x25a1300] co located POCs unavailable [mp2 @ 0x706340] Header missing Error while decoding stream #0:2: Invalid data found when processing input [h264 @ 0x2688800] mmco: unref short failure [h264 @ 0x25fa5e0] mmco: unref short failure [mpegts @ 0x6feee0] PES packet size mismatch [h264 @ 0x25a1300] cabac decode of qscale diff failed at 2 25 [h264 @ 0x25a1300] error while decoding MB 2 24, bytestream 20871 [h264 @ 0x25a1300] concealing 5327 DC, 5327 AC, 5327 MV errors in B frame [mp2 @ 0x705e80] Header missing Error while decoding stream #0:1: Invalid data found when processing input [h264 @ 0x25fa5e0] co located POCs unavailable [h264 @ 0x2688800] mmco: unref short failure [h264 @ 0x25fa5e0] mmco: unref short failureime=04:47:41.60 bitrate=3104.8kbits/s dup=0 drop=10 speed=1.31x frame=262498 fps= 19 q=4.0 Lsize= 7090588kB time=05:09:59.08 bitrate=3123.1kbits/s dup=0 drop=10 speed=1.33x video:6468443kB audio:586667kB subtitle:0kB other streams:0kB global headers:0kB muxing overhead: 0.502870% cer@Telcontar:~ I converted 3 other movies, without stop point, for further verification: a) Same choppy sound after 00:17:46 (different TV channel) b) Same choppy sound after 00:40:02 (same channel as 'a') c) Same choppy sound after 00:48:15 (different TV channel). With a commercial in the middle. So, unless I find an option to ffmpeg that corrects the choppiness, conversion is impossible. -- Cheers / Saludos, Carlos E. R. (from 13.1 x86_64 "Bottle" at Telcontar)

On 2016-03-07 22:12, jdd wrote:

Le 07/03/2016 21:17, Carlos E. R. a écrit :

...

Well, I would have to do it with dd for reliability.

dd can't cut any video file. key word is "GOP" (https://en.wikipedia.org/wiki/Group_of_pictures)

I have done it once, I think, and it partially works. The viewer traverses the file till it finds a point at which it can "start". Doesn't work with all formats. -- Cheers / Saludos, Carlos E. R. (from 13.1 x86_64 "Bottle" at Telcontar)

On 2016-03-07 22:36, jdd wrote:

...

... Input #0, mpegts, from 'Las 13 claves del Apolo 13 Episodio 1.mpeg': note: mpegts: TS. so MTS file, not real mpg as we usually call them.

is it SD or HD? SD is about to be stopped in France (April 4, from memory). i see 720x576, so SD. use projectX to fix them.

Depends on the channel. That one above is 720x576 pixels at 16:9 nominally, but in fact it is 4:3, it would need vertical crop. I will have to see about using projectX on it... -- Cheers / Saludos, Carlos E. R. (from 13.1 x86_64 "Bottle" at Telcontar)

On 2016-03-10 11:43, jdd wrote:

upload your smaller file :-)

A one gig file? Yes, sure, tell me where to upload. The only place I have is google drive, but I don't think it would take it. -- Cheers / Saludos, Carlos E. R. (from 13.1 x86_64 "Bottle" at Telcontar)

On 2016-03-10 11:47, jdd wrote:

Le 10/03/2016 11:24, Carlos E. R. a écrit :

...

I will have to see about using projectX on it...

projectX is specially designed to deal with such problem, and works very well, once you get how to use it

http://dodin.info/wiki/pmwiki.php?n=Doc.RecupererUneVideoTNT

it's in french, but google translate is probably enough to understand it. I always find easier to correct first, then edit with any application. ProjectX gives separately video and audio. Not tested with several tracks

I can almost read French, but not completely, so I used google translate. I have ProjectX 0.90.4 installed from source, combined with cridmanager-1.4.3, which is a tool to work with the Gigaset M740AV. This cridmanager calls ProjectX with the appropriate list of files for movies recorded in the gigaset. This combo works perfectly. I tried in the past ProjectX from packman, but it failed somehow. I tried the ProjectX that I have with two of the movies I have now, and it is unable to load them: Error while decoding frame not enough data in buffer -- Cheers / Saludos, Carlos E. R. (from 13.1 x86_64 "Bottle" at Telcontar)

On 2016-03-05 19:39, Carlos E. R. wrote:

Hi,

I simply can't find a suitable video editor that works.

My goal is to extract the interesting parts of a TV recording. Ie, mark the start of the movie, the end, the commercials inside, and dump all that into a single file, perhaps recoding.

With help from jdd, I got ProjectX to half work. I had already ProjectX 0.90.4 installed from source, which works fine with movies made by my Gigaset M740AV, but was unable to even load the videos made from my sitting room setup that is the origin of this thread. jdd was able to load a sample I sent him (one gigabyte) So I looked again and found ProjectX_0.91.0.zip at https://sourceforge.net/projects/project-x/, dated 2011-03-27. I installed this one and run it, and it can load these video streams. Good! I thought. However, it is unable to decode the video stream, only the sound. It can demux the file and produce two audio tracks, apparently correct, but no video at all. The error it reports is this, by the ton: !> dropping video data, GOP larger than 6MB !> dropping video data, GOP larger than 6MB !> dropping video data, GOP larger than 6MB which is very unfortunate :-( -- Cheers / Saludos, Carlos E. R. (from 13.1 x86_64 "Bottle" at Telcontar)

On 2016-03-11 02:36, Carlos E. R. wrote:

On 2016-03-10 09:22, Carlos E. R. wrote:

...

On 2016-03-05 19:39, Carlos E. R. wrote:

...

Well, taking info from several posts here, I'm trying with ffmpeg in the CLI.

I found that the output is choppy or jerky somewhere after 15 minutes of play, on the four movies I tried. There is a chirp, then it plays a second, stops a second, plays a second... the play time increases about double. It does not depend on the player, it seems, but it does depend a bit on the video codec.

I got a suggestion offlist to try to code each stream separately. I did: ffmpeg -i "input-file.mpeg" -ss 00:20:00.00 -t 00:5:00.00 \ -map 0:0 -c:v libx264 -preset ultrafast -vf crop=1920:804:0:138 \ -c:a copy output-file_0.mkv then -map 0:1 y luego map 0:2 And yes indeed, the video stream is choppy, but not the two audio streams. This matches the results with ProjectX, that finds the video unusable. -- Cheers / Saludos, Carlos E. R. (from 13.1 x86_64 "Bottle" at Telcontar)

On 2016-03-11 03:04, Carlos E. R. wrote:

On 2016-03-11 02:36, Carlos E. R. wrote:

...

On 2016-03-10 09:22, Carlos E. R. wrote:

...

On 2016-03-05 19:39, Carlos E. R. wrote:

I got a suggestion offlist to try to code each stream separately. I did:

ffmpeg -i "input-file.mpeg" -ss 00:20:00.00 -t 00:5:00.00 \ -map 0:0 -c:v libx264 -preset ultrafast -vf crop=1920:804:0:138 \ -c:a copy output-file_0.mkv

then -map 0:1 y luego map 0:2

And yes indeed, the video stream is choppy, but not the two audio streams. This matches the results with ProjectX, that finds the video unusable.

I got another suggestion from the same person: -copyts Do not process input timestamps, but keep their values without trying to sanitize them. In particular, do not remove the initial start time offset value. Note that, depending on the vsync option or on specific muxer processing (e.g. in case the format option avoid_negative_ts is enabled) the output timestamps may mismatch with the input timestamps even when this option is selected. The idea is that it seems that the timestamps would simply be passed on from the transport stream to the output, as players are able to correctly play the input stream, instead of trying to write the correct timestamp, and failing. It would be like this: ffmpeg -i "inputstream.mpeg" -copyts-ss 00:20:00.00 -t 00:5:00.00 \ -map 0:0 -map 0:1 -map 0:2 -c:v libx264 \ -preset ultrafast -vf crop=1920:804:0:138 -c:a copy output.mkv And indeed, this test section now plays well :-) With this method, I think I can easily process videos on which I only have to remove a region from the start and end, not the middle. For those I would have to design a modification. -- Cheers / Saludos, Carlos E. R. (from 13.1 x86_64 "Bottle" at Telcontar)

Hello, On Sat, 12 Mar 2016, Carlos E. R. wrote:

On 2016-03-11 18:19, David Haller wrote: [..]

...

mkvmerge --split parts:+00:20:00.00-00:25:00.00 -o output.mkv input.mpeg

Try this too!

...

Just leave out the cropping for now.

Or try mencoder, I'm not quite sure about the '-map' parameter of ffmpeg, but this should be a first try (including cropping):

I had a look at mencoder and several web pages, and found the methodology and syntax very complex. That's why I haven't tried yet. ffmpeg I more or less understand, but this one, no.

They're quite similar actually ;)

...

mencoder -ss 00:20:00.00 -endpos 5:00.00 -oac copy \ -ovc x264 -x264encopts crf=23:trellis=1 \ -vf crop=1920:804:0:138 -o output.avi input.mpeg

(-x264encopts crf=23 is rather high quality, can be adjusted)

This copies just the first audio-track, but more can be done seperately, e.g.

mencoder -ss 00:20:00.00 -endpos 5:00.00 -oac copy -aid 129 \ -ovc frameno -o output_aid129.avi input.mpeg

(mux with mkvmerge -o output.mkv output.avi -D output_aid129.avi)

This is more meant as a test.

I'm running it now. But it appears it is producing a different section at 00:20:00 than ffmpeg. :-? Yes, the above command "-ss 00:20:00.00" appears to start coding at timestamp 11:00 in the input stream, so the 5 minute section does not contain the bad part.

That's weird. That it's off a bit due to GOPs is normal, but ISTR, that I had this too once that it seemed to take ~half of the timestamp ...

I'll try with -ss 00:40:00.00

Ok, done. The audio is out of sync to the video by several minutes. :-(

Damn. mencoder was broken in that respect for quite a long time which is why I still use an old version MPlayer-1.0rc4_r31930, the brokenness came a few commits later, and was supposedly fixed just recently, but I haven't tested that yet. Do you have the current packman version?

...

PS: no comments on my previous mails?

Huh, because I'm running tests, and each one may take hours :-)

The mkvmerge one should take just a minute ;) And using avidemux with "copy" (+mkvmerge) should also take just minutes. -dnh -- printk(KERN_WARNING "Multi-volume CD somehow got mounted.\n"); linux-2.2.16/fs/isofs/inode.c -- To unsubscribe, e-mail: opensuse+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse+owner@opensuse.org

Le 13/03/2016 19:47, Carlos E. R. a écrit :

On 2016-03-05 19:39, Carlos E. R. wrote:

I've got a 21 MB sample copy that can be used for study. Small enough to email to anybody that wants to check the problem, after splitting a bit. Or perhaps upload to google drive, dunno.

I used this to generate it:

ffmpeg -i "Input file.mpeg" -ss 00:21:12.00 -t 00:00:30.00 -map 0 -c:v copy -c:a copy FailureTest.ts

it works with your apollo film, but this film is very bad to check, because there is very few lip sync, so it's impossible to hear is the sound is in sync :-) by the way, you can copy the apollo (or any other file) to a public_html folder in your account on my server and link to it like this :-) http://dodin.org/~crobin/ I copied to your ~/bin a very useful (for me) script that build a html index page that link to any file you give as argument. I use it in my temp folders when I want to have a link to give ~/bin/lstohtml.sh test.html to create the index page jdd -- To unsubscribe, e-mail: opensuse+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse+owner@opensuse.org

Basil Chupin composed on 2016-03-06 12:20 (UTC+1100):

I know that you state below that you tried Cinelerra (https://en.wikipedia.org/wiki/Cinelerra) but believe me this is the one that you want to use. I have used it to edit recorded movies (removing commercials, eg) and it is the ants pants for editing videos. Spend just a LITTLE bit of time to read how to use it and you will not regret the time spent.

Where did you get yours? It's not in Packman, at least not for 42.2. -- "The wise are known for their understanding, and pleasant words are persuasive." Proverbs 16:21 (New Living Translation) Team OS/2 ** Reg. Linux User #211409 ** a11y rocks! Felix Miata *** http://fm.no-ip.com/ -- To unsubscribe, e-mail: opensuse+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse+owner@opensuse.org

On 2017-06-06 22:28, Felix Miata wrote:

Basil Chupin composed on 2016-03-06 12:20 (UTC+1100):

...

I know that you state below that you tried Cinelerra (https://en.wikipedia.org/wiki/Cinelerra) but believe me this is the one that you want to use. I have used it to edit recorded movies (removing commercials, eg) and it is the ants pants for editing videos. Spend just a LITTLE bit of time to read how to use it and you will not regret the time spent.

Where did you get yours? It's not in Packman, at least not for 42.2.

Can cinelerra work with dual sound track videos? Several good video apps can only work with one language, they discard the other. In Spain, many films are "dual": say, English and Spanish, and of course, I want to keep both tracks. Oh, I see this is an old thread of mine. At the time, cinelerra-cv was available. I posted:

cinelerra-cv - I can't make head or tails of it. I tell it to load a video file, it does, but then I don't see how to do anything. It doesn't even display the video. I may have to read the manual, but I dislike having to read a manual for a GUI. Still...

My notes on it say that I could not make it work at all. -- Cheers / Saludos, Carlos E. R. (from 42.2 x86_64 "Malachite" at Telcontar)

On Tue, 6 Jun 2017 16:28:31 -0400 Felix Miata wrote:

Basil Chupin composed on 2016-03-06 12:20 (UTC+1100):

...

I know that you state below that you tried Cinelerra (https://en.wikipedia.org/wiki/Cinelerra) but believe me this is the one that you want to use. I have used it to edit recorded movies (removing commercials, eg) and it is the ants pants for editing videos. Spend just a LITTLE bit of time to read how to use it and you will not regret the time spent.

Where did you get yours? It's not in Packman, at least not for 42.2.

I believe Packman may have removed Cinelerra because it has been forked and at least one branch is building it's own packages for the major distributions. The two websites I know of are these, with the latter one containing articles and links about the fork: http://cinelerra.org/ and https://cinelerra-cv.org/ My system is running 64-bit openSUSE Leap 42.2 and I was able to install Cinelerra after adding the following repository and using zypper to install package 'cinelerra' the current version (recently updated) is 5.1-20170531 Repository Name: cincv URL: https://cinelerra-cv.org/five/pkgs/leap/ Category: YUM hth & regards, Carl -- To unsubscribe, e-mail: opensuse+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse+owner@opensuse.org

On 2017-06-13 13:51, gumb wrote:

On 06/06/17 22:28, Felix Miata wrote:

...

Basil Chupin composed on 2016-03-06 12:20 (UTC+1100):

...

I know that you state below that you tried Cinelerra (https://en.wikipedia.org/wiki/Cinelerra) but believe me this is the one that you want to use. I have used it to edit recorded movies (removing commercials, eg) and it is the ants pants for editing videos. Spend just a LITTLE bit of time to read how to use it and you will not regret the time spent.

Where did you get yours? It's not in Packman, at least not for 42.2.

Peculiar, I can no longer find the beginning of this thread either in my mail program (which discards everything more than 30 days old) or on the ML archives. But I seem to recall it being a question about a simple way of cutting/joining videos.

Because it goes back a good way, more than a year ;-) Date: Sat, 5 Mar 2016 19:39:01 +0100 (CET) From: Carlos E. R. To: OS-en Subject: [opensuse] Again, trying to find a suitable video editor.

And I think everybody piled in with the usual suggestions of Kdenlive, etc. But you might be better off with a more simple, dedicated tool such as Avidemux, or Vidcutter: https://www.linux-apps.com/p/1180755

The problem with avidemux is that it does only one sound track. I need two, because movies in Spain come dubbed to Spanish, but usually also contain the original sound track. Naturally I want to keep both. Vidcutter I haven't tried, wasn't mentioned in the original thread. Description: The simplest & sexiest tool for cutting and joining your videos without the need for re-encoding or a diploma in multimedia. VidCutter focuses on getting the job done using tried and true tech in its arsenal via mpv and FFmpeg. I like that idea, I do not want to recode, just remove commercials. -- Cheers / Saludos, Carlos E. R. (from 42.2 x86_64 "Malachite" at Telcontar)

gumb composed on 2017-06-13 13:51 (UTC+0200): .

And I think everybody piled in with the usual suggestions of Kdenlive, etc. But you might be better off with a more simple, dedicated tool such as Avidemux, or Vidcutter: https://www.linux-apps.com/p/1180755 . As I told Carlos, I couldn't get Vidcutter to work at all, while so far I've been unable to get Avidemux to feign the same quality as the source in recode mode required to get frame accurate cuts, or write out in a reasonable time, less than 2-3 times the length of the original. Avidemux took several hours to reduce a 29 minute original to 20 minutes sans commercials.

I saw a suggestion somewhere that for best results I should decompress the original before trying to edit it, but so far I've been unable to find an appropriate example how to do that. Man pages to me are little better than useless for lack of sufficient examples. -- "The wise are known for their understanding, and pleasant words are persuasive." Proverbs 16:21 (New Living Translation) Team OS/2 ** Reg. Linux User #211409 ** a11y rocks! Felix Miata *** http://fm.no-ip.com/ -- To unsubscribe, e-mail: opensuse+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse+owner@opensuse.org

On 2017-06-13 18:00, Felix Miata wrote:

...

Vidcutter I haven't tried, wasn't mentioned in the original thread. . Tried. No good. http://lists.links2linux.de/pipermail/packman/2017-June/015248.html . I do not want to recode, just remove commercials. . I have to recode in most cases. The commercials are typically shoehorned in without enough black between commercials to keep the entirety of the program starts and ends unless doing frame accurate cutting. Avidemux in copy mode won't. In frame accurate mode it takes hours to decode, the two times I tried, it produced a quarter-sized file that plays worse than an LP VHS tape, and the other time a gigantic file 9X the size of the original that I didn't even try to

Carlos E. R. composed on 2017-06-13 14:05 (UTC+0200): . play.

From some reads it appears that it is able to cut sections without recoding most of the video, only about the region of the cut if it doesn't happen at th exact place that gets a full image (the rest of the frames contain the differences to the key frame). That region can be recoded using the same settings and codecs as the original video, that's what I want. Such an editor run fast. Otherwise, it is hours. -- Cheers / Saludos, Carlos E. R. (from 42.2 x86_64 "Malachite" at Telcontar)

On Tuesday, 13 June 2017 9:13:57 PM AEST Carlos E. R. wrote:

From some reads it appears that it is able to cut sections without recoding most of the video, only about the region of the cut if it doesn't happen at th exact place that gets a full image (the rest of the frames contain the differences to the key frame).

That region can be recoded using the same settings and codecs as the original video, that's what I want.

Such an editor run fast. Otherwise, it is hours.

I found Flowblade to be excellent and simple to use. There are some Youtube tutorials that helped me a lot in getting started. -- Regards, Graham Smith -- To unsubscribe, e-mail: opensuse+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse+owner@opensuse.org

"David C. Rankin" writes:

On 06/13/2017 07:05 AM, Carlos E. R. wrote:

...

The problem with avidemux is that it does only one sound track. I need two, because movies in Spain come dubbed to Spanish, but usually also contain the original sound track. Naturally I want to keep both.

Agreed,

Although I have no need for it. You should be able to enable up to 4 sound tracks in Avidemux Audio-> "Select Track" Charles -- "All language designers are arrogant. Goes with the territory..." (By Larry Wall)

On 06/14/2017 03:23 AM, jdd@dodin.org wrote:

long time ago, wow... may be 10 years ago? when I was still making video on windows, I got an utility with some software or hardware that just did hat: cut mpg file without recoding, and almost instantly. Dunno if it was only on master frames but it worked very well

never found the same again.

That's one of the main reasons I still keep and love my old Sony 3-ccd video camera that stores uncompressed video on digital tape (mini-dv) (and one that still uses digital-8 same content format, just different tape size). The .dv format provides separate stereo audio tracks along with the video track. Never had to work about demuxing the contents. Just read the clips in, and they automatically separate into separate video and audio tracks. (it takes 31 MB/s for storage, but not a problem on Terabyte drives) When I started video editing 80 GB drives were huge (and 5400 rpm slow...), but they worked. I'll have to look at kdenlive next time I put the arch drive in my box. That may be one area where FW5 outdoes KDE3, it will be the first check-mark in the FW5 column :) -- David C. Rankin, J.D.,P.E. -- To unsubscribe, e-mail: opensuse+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse+owner@opensuse.org

On 06/14/2017 03:39 AM, Charles Philip Chan wrote:

...

On 06/13/2017 07:05 AM, Carlos E. R. wrote:

...

The problem with avidemux is that it does only one sound track. I need two, because movies in Spain come dubbed to Spanish, but usually also contain the original sound track. Naturally I want to keep both. Agreed, Although I have no need for it. You should be able to enable up to 4 sound tracks in Avidemux

Audio-> "Select Track"

Where cinelerra shines is it provides a great interface for both video and audio that is almost identical to Audacity for sound (as far as the multiple track layout, etc..) Avidemux has no timeline or multitrack layout. The cli interface can do it, it just doesn't look like the gui can. From their Quickstart: Avidemux is a simple tool for simple video processing tasks. The keyword here is simple: it does not offer tools like a timeline, multitrack editing, you cannot freely move or splice audio and video clips from various sources. So it may be the desktop interface that's missing, the cli looks very capable. -- David C. Rankin, J.D.,P.E.

On 2017-06-14 10:12, Graham Smith wrote:

On Tuesday, 13 June 2017 9:13:57 PM AEST Carlos E. R. wrote:

...

From some reads it appears that it is able to cut sections without recoding most of the video, only about the region of the cut if it doesn't happen at th exact place that gets a full image (the rest of the frames contain the differences to the key frame).

That region can be recoded using the same settings and codecs as the original video, that's what I want.

Such an editor run fast. Otherwise, it is hours.

I found Flowblade to be excellent and simple to use. There are some Youtube tutorials that helped me a lot in getting started.

Good grief. I started it. I try to open a video file. The dialogue is capable of opening directories, it does not see the videos inside. It wants to open "Flowblade projects", not videos. If I create a new project, it wants me to type settings: Like DV/DVD PAL. This is nonsense: I want to use the exact same settings as the original video has, not recode to a new setting. This program is not good to me. -- Cheers / Saludos, Carlos E. R. (from 42.2 x86_64 "Malachite" at Telcontar)

On 2017-06-14 10:23, jdd@dodin.org wrote:

Le 14/06/2017 à 10:05, David C. Rankin a écrit :

...

You simply cannot go wrong with cinerella. Y

if you can get it to work... it's fitted for film makers https://en.wikipedia.org/wiki/Cinelerra

kdenlive do the same with much more in/out options.

but by the way the Carlos problem is not this.

it's keeping *the choice* on the final file between two language tracks. Kdenlive is said to be able to do it, but I'm not sure this is really possible right now, even on last appimage one.

Yes. Many developers, being from the USA, simply do not understand the problem. I think kdenlive failed on this a year ago. And besides the two language audio tracks, there are the subtitles tracks, which can also be two. I have a video with about a dozen language tracks, both audio and subtitles.

Most are not even able to load the two audio tracks of the original file. One have to first demux image/audio1/audio2, load them in the editor, export them separately and build after that a mkv file

clearly too much work only for removing unwanted advertisement contents

Indeed. I'm having to store the videos whole, commercials et all, and just hit the fast-forward on playback.

long time ago, wow... may be 10 years ago? when I was still making video on windows, I got an utility with some software or hardware that just did hat: cut mpg file without recoding, and almost instantly. Dunno if it was only on master frames but it worked very well

never found the same again.

Do you remember the name of that thing? Maybe it works with Wine.

I would like one because may compact camera videos do have wrong start or end, having to edit and recompile them waste time

Yes. -- Cheers / Saludos, Carlos E. R. (from 42.2 x86_64 "Malachite" at Telcontar)

On 2017-06-14 11:09, David C. Rankin wrote:

On 06/14/2017 03:39 AM, Charles Philip Chan wrote:

...

...

On 06/13/2017 07:05 AM, Carlos E. R. wrote:

...

The problem with avidemux is that it does only one sound track. I need two, because movies in Spain come dubbed to Spanish, but usually also contain the original sound track. Naturally I want to keep both. Agreed, Although I have no need for it. You should be able to enable up to 4 sound tracks in Avidemux

Audio-> "Select Track"

It only processes one.

...

Where cinelerra shines is it provides a great interface for both video and audio that is almost identical to Audacity for sound (as far as the multiple track layout, etc..) Avidemux has no timeline or multitrack layout. The cli interface can do it, it just doesn't look like the gui can.

From their Quickstart:

Avidemux is a simple tool for simple video processing tasks. The keyword here is simple: it does not offer tools like a timeline, multitrack editing, you cannot freely move or splice audio and video clips from various sources.

I do not need any of that. -- Cheers / Saludos, Carlos E. R. (from 42.2 x86_64 "Malachite" at Telcontar)

On 05/03/2016 20:39, Carlos E. R. wrote:

Shotcut - Segmentation fault on start.

cer@Telcontar:~/Videos/Fusion/trabajo> shotcut QSqlDatabase: QSQLITE driver not loaded QSqlDatabase: available drivers: QMYSQL QMYSQL3 QSqlQuery::exec: database not open QSqlQuery::exec: database not open Database::Database(QObject*) Failed to get version. QSqlQuery::exec: database not open bool Database::upgradeVersion1() Failed to create thumbnails table. Database version is 0 Segmentation fault cer@Telcontar:~/Videos/Fusion/trabajo> Which shotcut are you using, I just updated it to 17.06, it works fine for me on 42.2. Regards Dave P

-- To unsubscribe, e-mail: opensuse+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse+owner@opensuse.org

On 05/03/2016 20:39, Carlos E. R. wrote:

Shotcut - Segmentation fault on start.

cer@Telcontar:~/Videos/Fusion/trabajo> shotcut QSqlDatabase: QSQLITE driver not loaded QSqlDatabase: available drivers: QMYSQL QMYSQL3 QSqlQuery::exec: database not open QSqlQuery::exec: database not open Database::Database(QObject*) Failed to get version. QSqlQuery::exec: database not open bool Database::upgradeVersion1() Failed to create thumbnails table. Database version is 0 Segmentation fault cer@Telcontar:~/Videos/Fusion/trabajo>

If you install libQt5Sql5-sqlite, it may work. I've required it in the package. Upstream gives no idea of run time dependencies and expect you to install a massive snap package. I have plasma5 so I can't uninstall libQt5Sql5-sqlite to test. You should file a bug against multimedia:apps/shotcut via web ui anyway. https://build.opensuse.org/package/show/multimedia:apps/shotcut it will automatically get assigned to me. Regards Dave P -- To unsubscribe, e-mail: opensuse+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse+owner@opensuse.org

On 2017-06-14 14:44, Charles Philip Chan wrote:

"Carlos E. R." <> writes:

...

...

Although I have no need for it. You should be able to enable up to 4 sound tracks in Avidemux

Audio-> "Select Track"

No.

What do you mean. With Avidemux 2.62.0 from packman, it clearly allows me to enable up to 4 tracks. Example image of dialog:

http://lh4.ggpht.com/-5uzi8PEr_74/UHOqpP3cVEI/AAAAAAAAVew/Wpc0UMnhhK0/avidem...

Unless it has been improved, last time I tried it did not. Let's try again. Loading... takes minutes to index the video. Says: H.264 detected. If the file is using B-frames as reference it can lead to a crash of stuttering. Avidemux can use another mode which is safe but YOU WILL LOSE FRAME ACCURACY Do you want to use that mode? What can I do but use safe mode? Now it says: Multiple audio tracks. The file you just loaded contains several audio tracks. Go to Audio->Main Track to select the active one. Ok. Nothing new there. I do, that, then "select second audio track". I see nothing like your photo. Instead, I get one dialogue: audio source: choice of none, External AC3, External MP3, External WAW. So, no, it is impossible. Ah, the "frame" just displays a green box, the video does not display AT ALL. There is not timeline graph, either. I don't know how I may select cut points anywhere. This is Avidemux 2.5.6. Ah, I see the slider. When I move it the video appears. So, single audio track, difficult to set start end points, and only two, I can not remove commercials in the middle. I try to move the point with the keyboard. CPU goes to 100% on four cores and doesn't move the slider anymore. Killing app. Signal 15. -- Cheers / Saludos, Carlos E. R. (from 42.2 x86_64 "Malachite" at Telcontar)

On 10/03/2016 15:00, Carlos E. R. wrote:

On 2016-03-10 11:11, Carlos E. R. wrote:

...

On 2016-03-10 09:22, Carlos E. R. wrote:

...

On 2016-03-05 19:39, Carlos E. R. wrote:

...

In this stream, things are correct till 00:15:30, then choppiness starts.

Does that tell something? :-?

Using the codec "copy", I get more interesting info:

time ffmpeg -i "La pelicula de la semana Star Trek.mpeg" -ss 00:05:44.600 -map 0:0 -map 0:1 -map 0:2 -c:v copy -c:a copy StarTrek-Copy.mkv ... Input #0, mpegts, from 'La pelicula de la semana Star Trek.mpeg': Duration: 03:01:02.68, start: 1.017833, bitrate: 9902 kb/s Program 1 Stream #0:0[0x1e1]: Video: h264 (High) ([27][0][0][0] / 0x001B), yuv420p(tv, bt709), 1920x1080 [SAR 1:1 DAR 16:9], 25 fps, 25 tbr, 90k tbn, 50 tbc Stream #0:1[0x1e2](esp): Audio: mp2 ([3][0][0][0] / 0x0003), 48000 Hz, stereo, s16p, 192 kb/s (clean effects) Stream #0:2[0x1e3](AU2): Audio: mp2 ([3][0][0][0] / 0x0003), 48000 Hz, stereo, s16p, 128 kb/s (clean effects) Output #0, matroska, to 'StarTrek-Copy.mkv': Metadata: encoder : Lavf57.25.100 Stream #0:0: Video: h264 (H264 / 0x34363248), yuv420p, 1920x1080 [SAR 1:1 DAR 16:9], q=2-31, 25 fps, 25 tbr, 1k tbn, 90k tbc Stream #0:1(esp): Audio: mp2 (P[0][0][0] / 0x0050), 48000 Hz, stereo, 192 kb/s (clean effects) Stream #0:2(AU2): Audio: mp2 (P[0][0][0] / 0x0050), 48000 Hz, stereo, 128 kb/s (clean effects) Stream mapping: Stream #0:0 -> #0:0 (copy) Stream #0:1 -> #0:1 (copy) Stream #0:2 -> #0:2 (copy) Press [q] to stop, [?] for help [matroska @ 0x25a47c0] Timestamps are unset in a packet for stream 0. This is deprecated and will stop working in the future. Fix your code to set the timestamps properly [matroska @ 0x25a47c0] Non-monotonous DTS in output stream 0:0; previous: 929720, current: 929680; changing to 929720. This may result in incorrect timestamps in the output file. [matroska @ 0x25a47c0] Non-monotonous DTS in output stream 0:0; previous: 929720, current: 929680; changing to 929720. This may result in incorrect timestamps in the output file. [matroska @ 0x25a47c0] Non-monotonous DTS in output stream 0:0; previous: 929720, current: 929680; changing to 929720. This may result in incorrect timestamps in the output file. [matroska @ 0x25a47c0] Non-monotonous DTS in output stream 0:0; previous: 929720, current: 929680; changing to 929720. This may result in incorrect timestamps in the output file. [mpegts @ 0x259cea0] PES packet size mismatch Last message repeated 1 times [matroska @ 0x25a47c0] Can't write packet with unknown timestamp av_interleaved_write_frame(): Invalid argument [matroska @ 0x25a47c0] Can't write packet with unknown timestamp Error writing trailer of StarTrek-Copy.mkv: Invalid argumentframe=23239 fps=2508 q=-1.0 Lsize= 1060888kB time=00:15:29.72 bitrate=9347.8kbits/s speed= 100x video:1024132kB audio:36313kB subtitle:0kB other streams:0kB global headers:0kB muxing overhead: 0.041773% Conversion failed!

real 0m9.804s user 0m4.904s sys 0m2.079s cer@Telcontar:~/Videos/Fusion/trabajo>

It crashes that instant that the video and sound starts to be choppy.

Try starting your ffmpeg with -fflags genpts before anything else. Dave P -- To unsubscribe, e-mail: opensuse+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse+owner@opensuse.org

On 2017-06-15 09:28, Dave Plater wrote:

On 11/03/2016 04:04, Carlos E. R. wrote:

...

ffmpeg -i "input-file.mpeg" -ss 00:20:00.00 -t 00:5:00.00 \ -map 0:0 -c:v libx264 -preset ultrafast -vf crop=1920:804:0:138 \ -c:a copy output-file_0.mkv

then -map 0:1 y luego map 0:2

And yes indeed, the video stream is choppy, but not the two audio streams. This matches the results with ProjectX, that finds the video unusable. To merge two identical videos use ffmpeg -f concat -i textfile -c copy where textfile has lines "file video1" next line is "file video2" and so on. See: https://trac.ffmpeg.org/wiki/Concatenate

You mean that's the trick to copy sections from the same stream and concatenate again into a single file? Yes, that's something I wanted to know, thanks :-) -- Cheers / Saludos, Carlos E. R. (from 42.2 x86_64 "Malachite" at Telcontar)

On 15/06/2017 10:52, Carlos E. R. wrote:

On 2017-06-15 09:28, Dave Plater wrote:

...

On 11/03/2016 04:04, Carlos E. R. wrote:

...

ffmpeg -i "input-file.mpeg" -ss 00:20:00.00 -t 00:5:00.00 \ -map 0:0 -c:v libx264 -preset ultrafast -vf crop=1920:804:0:138 \ -c:a copy output-file_0.mkv

then -map 0:1 y luego map 0:2

And yes indeed, the video stream is choppy, but not the two audio streams. This matches the results with ProjectX, that finds the video unusable. To merge two identical videos use ffmpeg -f concat -i textfile -c copy where textfile has lines "file video1" next line is "file video2" and so on. See: https://trac.ffmpeg.org/wiki/Concatenate

You mean that's the trick to copy sections from the same stream and concatenate again into a single file? Yes, that's something I wanted to know, thanks :-)

Don't forget to put -fflags genpts first it fixes a host of timing problems. I use it all the time. Dave P -- To unsubscribe, e-mail: opensuse+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse+owner@opensuse.org

"Carlos E. R." writes:

Ok. Nothing new there.

I do, that, then "select second audio track". I see nothing like your photo. Instead, I get one dialogue:

audio source: choice of none, External AC3, External MP3, External WAW.

So, no, it is impossible.

It looks like you are using the old GTK interface which, IIRC, is not being developed any more. Move on to the QT5 one. Charles -- "On a normal ascii line, the only safe condition to detect is a 'BREAK' - everything else having been assigned functions by Gnu EMACS." (By Tarl Neustaedter)

On 2017-06-15 11:44, Charles Philip Chan wrote:

"Carlos E. R." <> writes:

...

Ok. Nothing new there.

I do, that, then "select second audio track". I see nothing like your photo. Instead, I get one dialogue:

audio source: choice of none, External AC3, External MP3, External WAW.

So, no, it is impossible.

It looks like you are using the old GTK interface which, IIRC, is not being developed any more. Move on to the QT5 one.

Maybe you are using avidemux3? When using version 3, on select audio tracks, the fields are empty, it does not see any audio track in the already loaded video file. Apparently it wants me to load the track from an external file. So I select track 1 and 2 to be the same mpeg file. There is no way to select the audio tracks that exist in that stream. It should say that there are two tracks and allow me to select each one. This is useless. -- Cheers / Saludos, Carlos E. R. (from 42.2 x86_64 "Malachite" at Telcontar)

On 2017-06-15 12:31, Charles Philip Chan wrote:

...

This is Avidemux 2.5.6.

Just noticed, this is old. Try the Avidemux3 (2.6.20-15) from Packman.

I did. It is impossible to select two audio tracks from the already loaded video file. It does not see the tracks. -- Cheers / Saludos, Carlos E. R. (from 42.2 x86_64 "Malachite" at Telcontar)

Le 15/06/2017 à 12:31, Charles Philip Chan a écrit :

...

This is Avidemux 2.5.6.

Just noticed, this is old. Try the Avidemux3 (2.6.20-15) from Packman.

Charles

only avidemux3 - do not forget to select gui... and yes, have better audio selection. but only select keyframe as start and stop one may need to recompile with only keyframes to get the cut right jdd -- http://dodin.org -- To unsubscribe, e-mail: opensuse+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse+owner@opensuse.org

On 2017-06-15 12:40, jdd@dodin.org wrote:

Le 15/06/2017 à 12:31, Charles Philip Chan a écrit :

...

...

This is Avidemux 2.5.6.

Just noticed, this is old. Try the Avidemux3 (2.6.20-15) from Packman.

Charles

only avidemux3 - do not forget to select gui...

and yes, have better audio selection.

No, it doesn't. I had to imput the video file in the selection box, instead of it automatically using the already loaded video file. Then, it presented me a selection with a hundred languages, instead of just the two that were in the file. When I did, it said that some tracks are used multiple times. It wants different external audio files as input! This is impossible to use.

but only select keyframe as start and stop

And impossible to remove commercials in the middle.

one may need to recompile with only keyframes to get the cut right

I don't know about that. -- Cheers / Saludos, Carlos E. R. (from 42.2 x86_64 "Malachite" at Telcontar)

On 2017-06-15 13:05, gumb wrote:

On 15/06/17 12:51, Carlos E. R. wrote:

...

This is impossible to use.

...

but only select keyframe as start and stop

And impossible to remove commercials in the middle.

...

one may need to recompile with only keyframes to get the cut right

I don't know about that.

I think in this case you'd be best off contacting directly one of the developers and presenting your specific workflow. (Probably better to try first with a smaller dedicated tool like Avidemux than a bigger program with several contributors). It may well be that nobody's come up against this exact issue, and like you say, a lot of State-side developers don't always think so much about elements that might only be relevant to a non-English userbase.

There are some simple tools that would do the task, but they are broken or abandoned. For instance, ProjectX. kdenlive people were adding dual language support last time I looked, perhaps at the start of this thread.

I've had a few successes in the past when I ask directly lone developers of smaller applications for feature requests or bug fixes. It's only when something becomes abandonware and the developers have moved on or lost motivation that I get no response.

I would need another round of testing video software. One year I do it, I find nothing suitable, perhaps contact some project, then forget for another year. A GUI thing that would simply generate a script for ffmpeg could be adjusted, adding the proper switches to make ffmpeg work perfectly. I have not found such an editor. I have to try gopchop again. Oopps... version is March 2016, so the start of this thread. Pointless to test it again. Another abandonware? Yes. https://sourceforge.net/projects/gopchop/files/ Last dev version is 2009. I see three other recommended projects: dvbcut --> 2013 Linux Video Editor (lve) --> 2013 Mpg2Cut2 --> 2015 Not very promising, is it? -- Cheers / Saludos, Carlos E. R. (from 42.2 x86_64 "Malachite" at Telcontar)

On 2017-06-15 09:20, Dave Plater wrote:

On 10/03/2016 15:00, Carlos E. R. wrote:

...

On 2016-03-10 11:11, Carlos E. R. wrote:

...

On 2016-03-10 09:22, Carlos E. R. wrote:

...

On 2016-03-05 19:39, Carlos E. R. wrote:

........**** Thanks for retaking this old problem, but... see below.

...

...

In this stream, things are correct till 00:15:30, then choppiness starts.

Does that tell something? :-?

Using the codec "copy", I get more interesting info:

time ffmpeg -i "La pelicula de la semana Star Trek.mpeg" -ss 00:05:44.600 -map 0:0 -map 0:1 -map 0:2 -c:v copy -c:a copy StarTrek-Copy.mkv ... Input #0, mpegts, from 'La pelicula de la semana Star Trek.mpeg': Duration: 03:01:02.68, start: 1.017833, bitrate: 9902 kb/s Program 1 Stream #0:0[0x1e1]: Video: h264 (High) ([27][0][0][0] / 0x001B), yuv420p(tv, bt709), 1920x1080 [SAR 1:1 DAR 16:9], 25 fps, 25 tbr, 90k tbn, 50 tbc Stream #0:1[0x1e2](esp): Audio: mp2 ([3][0][0][0] / 0x0003), 48000 Hz, stereo, s16p, 192 kb/s (clean effects) Stream #0:2[0x1e3](AU2): Audio: mp2 ([3][0][0][0] / 0x0003), 48000 Hz, stereo, s16p, 128 kb/s (clean effects) Output #0, matroska, to 'StarTrek-Copy.mkv': Metadata: encoder : Lavf57.25.100 Stream #0:0: Video: h264 (H264 / 0x34363248), yuv420p, 1920x1080 [SAR 1:1 DAR 16:9], q=2-31, 25 fps, 25 tbr, 1k tbn, 90k tbc Stream #0:1(esp): Audio: mp2 (P[0][0][0] / 0x0050), 48000 Hz, stereo, 192 kb/s (clean effects) Stream #0:2(AU2): Audio: mp2 (P[0][0][0] / 0x0050), 48000 Hz, stereo, 128 kb/s (clean effects) Stream mapping: Stream #0:0 -> #0:0 (copy) Stream #0:1 -> #0:1 (copy) Stream #0:2 -> #0:2 (copy) Press [q] to stop, [?] for help [matroska @ 0x25a47c0] Timestamps are unset in a packet for stream 0. This is deprecated and will stop working in the future. Fix your code to set the timestamps properly [matroska @ 0x25a47c0] Non-monotonous DTS in output stream 0:0; previous: 929720, current: 929680; changing to 929720. This may result in incorrect timestamps in the output file. [matroska @ 0x25a47c0] Non-monotonous DTS in output stream 0:0; previous: 929720, current: 929680; changing to 929720. This may result in incorrect timestamps in the output file. [matroska @ 0x25a47c0] Non-monotonous DTS in output stream 0:0; previous: 929720, current: 929680; changing to 929720. This may result in incorrect timestamps in the output file. [matroska @ 0x25a47c0] Non-monotonous DTS in output stream 0:0; previous: 929720, current: 929680; changing to 929720. This may result in incorrect timestamps in the output file. [mpegts @ 0x259cea0] PES packet size mismatch Last message repeated 1 times [matroska @ 0x25a47c0] Can't write packet with unknown timestamp av_interleaved_write_frame(): Invalid argument [matroska @ 0x25a47c0] Can't write packet with unknown timestamp Error writing trailer of StarTrek-Copy.mkv: Invalid argumentframe=23239 fps=2508 q=-1.0 Lsize= 1060888kB time=00:15:29.72 bitrate=9347.8kbits/s speed= 100x video:1024132kB audio:36313kB subtitle:0kB other streams:0kB global headers:0kB muxing overhead: 0.041773% Conversion failed!

real 0m9.804s user 0m4.904s sys 0m2.079s cer@Telcontar:~/Videos/Fusion/trabajo>

It crashes that instant that the video and sound starts to be choppy.

Try starting your ffmpeg with -fflags genpts before anything else. Dave P

As this was a year ago, I'm unsure I still have that video file, the symlink is broken. I have to try see if I renamed it. cer@Telcontar:~/Videos/Fusion/trabajo> locate Star\ Trek.mpeg /home_aux/cer/Fusion/Cine Star Trek.mpeg /home_aux/cer/Fusion/Hacer/Trek/Cine Star Trek.mpeg /home_aux/cer/Fusion/trabajo/La pelicula de la semana Star Trek.mpeg /home_aux/cer/Fusion/trabajo/La pelicula de la semana Star Trek.mpeg.idx2 cer@Telcontar:~/Videos/Fusion/trabajo> cer@Telcontar:~/Videos/Fusion/trabajo> l La\ pelicula\ de\ la\ semana\ \ Star\ Trek.mpeg lrwxrwxrwx 1 cer users 79 Mar 13 2016 La pelicula de la semana Star Trek.mpeg -> /home_aux/cer/Fusion/Hechos.originales/La pelicula de la semana Star Trek.mpeg The symlink is broken, ie, it seems I no longer have that video sample. Maybe I already converted it and deleted the original. It might be this one: cer@Telcontar:~/Videos/Fusion/trabajo> time ffmpeg -i ../Cine\ \ Star\ Trek.mpeg -ss 00:05:44.600 -map 0:0 -map 0:1 -map 0:2 -c:v copy -c:a copy StarTrek-Copy.mkv ... Input #0, mpegts, from '../Cine Star Trek.mpeg': Duration: 02:12:01.94, start: 1.118100, bitrate: 7901 kb/s Program 1 Stream #0:0[0x1e1]: Video: h264 (High) ([27][0][0][0] / 0x001B), yuv420p(tv, bt709, top first), 1920x1080 [SAR 1:1 DAR 16:9], 25 fps, 25 tbr, 90k tbn, 50 tbc Stream #0:1[0x1e2](esp): Audio: mp2 ([3][0][0][0] / 0x0003), 48000 Hz, stereo, s16p, 192 kb/s (clean effects) Stream #0:2[0x1e3](vo): Audio: mp2 ([3][0][0][0] / 0x0003), 48000 Hz, stereo, s16p, 128 kb/s (clean effects) Stream #0:3[0x246](esp): Subtitle: dvb_subtitle ([6][0][0][0] / 0x0006) It is not the same video, the old one was "Duration: 03:01:02.68, start: 1.017833, bitrate: 9902 kb/s " The current one also crashes - but I haven't checked if the cut point and sizes are correct for this video: Output #0, matroska, to 'StarTrek-Copy.mkv': Metadata: encoder : Lavf57.71.100 Stream #0:0: Video: h264 (High) (H264 / 0x34363248), yuv420p(tv, bt709, top first), 1920x1080 [SAR 1:1 DAR 16:9], q=2-31, 25 fps, 25 tbr, 1k tbn, 90k tbc Stream #0:1(esp): Audio: mp2 (P[0][0][0] / 0x0050), 48000 Hz, stereo, s16p, 192 kb/s (clean effects) Stream #0:2(vo): Audio: mp2 (P[0][0][0] / 0x0050), 48000 Hz, stereo, s16p, 128 kb/s (clean effects) Stream mapping: Stream #0:0 -> #0:0 (copy) Stream #0:1 -> #0:1 (copy) Stream #0:2 -> #0:2 (copy) Press [q] to stop, [?] for help [matroska @ 0x21c2b40] Timestamps are unset in a packet for stream 0. This is deprecated and will stop working in the future. Fix your code to set the timestamps properly [matroska @ 0x21c2b40] Non-monotonous DTS in output stream 0:0; previous: 180480, current: 180440; changing to 180480. This may result in incorrect timestamps in the output file. [matroska @ 0x21c2b40] Non-monotonous DTS in output stream 0:0; previous: 180480, current: 180440; changing to 180480. This may result in incorrect timestamps in the output file. [matroska @ 0x21c2b40] Non-monotonous DTS in output stream 0:0; previous: 180480, current: 180440; changing to 180480. This may result in incorrect timestamps in the output file. [matroska @ 0x21c2b40] Non-monotonous DTS in output stream 0:0; previous: 180480, current: 180440; changing to 180480. This may result in incorrect timestamps in the output file. [matroska @ 0x21c2b40] Non-monotonous DTS in output stream 0:0; previous: 180480, current: 180440; changing to 180480. This may result in incorrect timestamps in the output file. [matroska @ 0x21c2b40] Non-monotonous DTS in output stream 0:0; previous: 180480, current: 180440; changing to 180480. This may result in incorrect timestamps in the output file. [matroska @ 0x21c2b40] Can't write packet with unknown timestamp av_interleaved_write_frame(): Invalid argument [matroska @ 0x21c2b40] Can't write packet with unknown timestamp Error writing trailer of StarTrek-Copy.mkv: Invalid argument frame= 4518 fps=2759 q=-1.0 Lsize= 161519kB time=00:03:00.92 bitrate=7313.5kbits/s speed= 110x video:154874kB audio:7046kB subtitle:0kB other streams:0kB global headers:0kB muxing overhead: unknown Conversion failed! This message worries me: «Timestamps are unset in a packet for stream 0. This is deprecated and will stop working in the future. Fix your code to set the timestamps properly» It is impossible to "fix the code", as the video is a stream directly copied from digital TV. Ie, it is what the broadcasters use. From a conversation off-list about the original problem the solution then was to use "-copyts". Like this: time nice ffmpeg -i "La pelicula de la semana Star Trek.mpeg" -copyts -ss 00:05:45 -t 01:52:45.92 -map 0:0 -map 0:1 -map 0:2 -c:v libx264 -preset medium -vf crop=19 20:804:0:138 -c:a copy StarTrek.mkv -- Cheers / Saludos, Carlos E. R. (from 42.2 x86_64 "Malachite" at Telcontar)

On Sun, 18 Jun 2017 13:03:52 +0200 (CEST) Carlos E. R. wrote:

Well, I asked on the ffmpeg mail list. The idea seems to be to use a video viewer, such as xine, and write down the cutpoints in paper, then compose your command line to ffmpeg by hand.

But it seems an almost dead mail list, anyhow.

I keep thinking about the relative ease with which one could accomplish these kinds of edits 'back in the day' with the right (albeit expensive) analog equipment. :) I think a good start would be to write a script which reads a simple text input file, e.g. - - - - - 8< - - - - - simple text input file - - - - - >8 - - - - - ### Source ### # Source path and filename, e.g. /home/user/docs/videos/myvideo.mp4 ### Target ### # Target path and filename or directory for 'slices', e.g. # /external-usb/slicedvideo.mp4 or /external-usb/slices ### Slices ### # Create one new line with a value pair for each desired 'slice' using # hh:mm:ss format, e.g. # 01:59:00 02:00:00 'slices' one minute from the top of the 2nd hour: # cut-out-time cut-in-time ... # Inverse operation? 'yes' or 'no' ('yes' = keep the slices, # concatenate them into a single file and toss the rest) # Save 'slices' as clips? 'yes' or 'no' ('yes' disables concatenation) - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - The script reads these inputs and generates the appropriate 'one liner.' Of course, the options supported by this system can be expanded to accommodate different needs. Geez! I'll bet you could publish this to github and it could become famous like youtube-dl :) -- To unsubscribe, e-mail: opensuse+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse+owner@opensuse.org

On 19/06/17 03:16, Carlos E. R. wrote:

On 2017-06-18 14:42, Carl Hartung wrote:

...

On Sun, 18 Jun 2017 13:03:52 +0200 (CEST) Carlos E. R. wrote:

...

Well, I asked on the ffmpeg mail list. The idea seems to be to use a video viewer, such as xine, and write down the cutpoints in paper, then compose your command line to ffmpeg by hand.

But it seems an almost dead mail list, anyhow. I keep thinking about the relative ease with which one could accomplish these kinds of edits 'back in the day' with the right (albeit expensive) analog equipment. :) I think a good start would be to write a script which reads a simple text input file, e.g.

- - - - - 8< - - - - - simple text input file - - - - - >8 - - - - - ### Source ### # Source path and filename, e.g. /home/user/docs/videos/myvideo.mp4

### Target ### # Target path and filename or directory for 'slices', e.g. # /external-usb/slicedvideo.mp4 or /external-usb/slices

### Slices ### # Create one new line with a value pair for each desired 'slice' using # hh:mm:ss format, e.g. # 01:59:00 02:00:00 'slices' one minute from the top of the 2nd hour: # cut-out-time cut-in-time ... # Inverse operation? 'yes' or 'no' ('yes' = keep the slices, # concatenate them into a single file and toss the rest)

# Save 'slices' as clips? 'yes' or 'no' ('yes' disables concatenation) - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -

The script reads these inputs and generates the appropriate 'one liner.' I know how to do that. That is not the problem. The problem is generating the text input file with the timestamps to cut at. At the moment, it is paper and pen.

As I already told you, with all the time you have spent on posting here and reading there and everywhere, you could have looked at one or two videos on YouTube and had your videos edited by now using Cinelerra. Cinelerra does what you want with ease. I did several 3GB+ videos to rid them of commercials using Cinlelerra. But "you do what a man has to do" :-). BC -- Government has become a committee for managing the affairs of the rich. -- To unsubscribe, e-mail: opensuse+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse+owner@opensuse.org

On 20/06/17 12:49, Carlos E. R. wrote:

I just found on the ffmepg mail list that avidemux has a cute hidden feature: you can remove sections in the middle, not just the start and end cuts.

Just define a region from A to B that you want to remove. Then go to Menu/Edit/cut. The section goes instantly! You can repeat for as many sections as needed, apparently. Finally, do "Save" to apply it all.

Works in versions 2 and 3.

However. With my source video in version 2, I got a video with the audio taken from one of the removed sections, so totally out of sync. With version 3 I get no audio whatsoever.

Haven't tried it myself, but just saw this pop up on OMG Ubuntu. I'm hedging my bets that it won't do what you need, because ... http://www.omgubuntu.co.uk/2017/06/curlew-gtk-media-converter-gnome-desktop -- To unsubscribe, e-mail: opensuse+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse+owner@opensuse.org

On 13/06/17 13:51, gumb wrote:

And I think everybody piled in with the usual suggestions of Kdenlive, etc. But you might be better off with a more simple, dedicated tool such as Avidemux, or Vidcutter: https://www.linux-apps.com/p/1180755

One for Carlos - I hope this reply goes into the right place in this ancient thread. I believe you were looking for video tools that would treat/export all the original subtitle tracks. I just saw a piece on OMG!Ubuntu! about an update to VidCutter, which *may* - if the notes hold true - provide this functionality, but you'll have to try it out yourself. http://www.omgubuntu.co.uk/2017/11/vidcutter-5-0-released-improved-ui-frame-... gumb -- To unsubscribe, e-mail: opensuse+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse+owner@opensuse.org

Hi, DaVinci Resolve is freeware, and is much more powerful then any other non-paid alternative. With free version you can't output 4k and 8k videos, but I don't think its necessary for a hobby. There are a lot of tutorials on youtube. Installation on Linux requires nVidia card and CUDA support. On 11/20/2017 09:09 PM, gumb wrote:

On 13/06/17 13:51, gumb wrote:

...

And I think everybody piled in with the usual suggestions of Kdenlive, etc. But you might be better off with a more simple, dedicated tool such as Avidemux, or Vidcutter: https://www.linux-apps.com/p/1180755

One for Carlos - I hope this reply goes into the right place in this ancient thread.

I believe you were looking for video tools that would treat/export all the original subtitle tracks. I just saw a piece on OMG!Ubuntu! about an update to VidCutter, which *may* - if the notes hold true - provide this functionality, but you'll have to try it out yourself.

http://www.omgubuntu.co.uk/2017/11/vidcutter-5-0-released-improved-ui-frame-...

gumb

-- To unsubscribe, e-mail: opensuse+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse+owner@opensuse.org

On 20/11/17 21:13, Carlos E. R. wrote:

On Monday, 2017-11-20 at 20:09 +0100, gumb wrote:

...

One for Carlos - I hope this reply goes into the right place in this ancient thread.

I'll tell you later ;-)

...

I believe you were looking for video tools that would treat/export all the original subtitle tracks. I just saw a piece on OMG!Ubuntu! about an update to VidCutter, which *may* - if the notes hold true - provide this functionality, but you'll have to try it out yourself.

http://www.omgubuntu.co.uk/2017/11/vidcutter-5-0-released-improved-ui-frame-...

Thanks :-)

I'll try it out when it comes to packman :-)

Version 5 is in the Packman repos now. Added yesterday.

It is not only subtitles that matter for me, but different language audio tracks. Most movies here (on TV) comes in the original language (typically English) and the local one, Spanish. The subtitles they most often come in Spanish, and sometimes also in English (typically the ones for deaf people), which I prefer for me.

I like to watch in English and if I have difficulties following the speech, I turn English subtitles on. If those are not available, I use the Spanish subtitles, but listening in English and reading in Spanish demands more effort for me. It is different from people that don't speak English, they are content with Spanish subtitles only.

But then, most people here prefer Spanish audio; thus my recordings should have all options - and few video editors in Linux allow this. I don't know in Windows, but I guess they have them, seeing that I get recordings done precisely the way I want them.

From the article: 'Stream mapping helps to ensure all of source media streams (e.g, video, audio, subtitles) are included in the finished export.' In theory, sounds good. -- To unsubscribe, e-mail: opensuse+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse+owner@opensuse.org

Carlos E. R. wrote:

I only need a simple editor for streams recorded from TV. I want to remove the start lead, the tail end, and the commercials in the middle.

I was reading some thread about cutting on the mythtv list - I can't find it again, but in mythtv, you mark the lead and the tail, and you can have mythtv mark the commercials. I am pretty certain I saw someone mention that he used ffmpeg to do the cutting based on the marks from mythtv. Somebody also mentioned https://www.mythtv.org/wiki/MythDVBcut I don't do any video cutting myself - mythtv has the marks, that is sufficient. -- Per Jessen, Zürich (10.9°C) http://www.cloudsuisse.com/ - your owncloud, hosted in Switzerland. -- To unsubscribe, e-mail: opensuse+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse+owner@opensuse.org

-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 On Thursday, 2017-11-23 at 13:37 -0000, Paul Groves wrote:

I just use ffmpeg to remove adverts from my recordings.

ffmpeg -i mov.mp4 -ss 01:00 -t 60 -c copy cut.mp4

-ss (which means start stream) will start at 1 min into the video (so removing the beginning). -t (time duration) will stream for 60 seconds -c copy will keep the encoding and not re-encode the video.

So this command will output 60 seconds of video starting at 1 minute in. I use the slider in VLC player to find the exact times. Then just run the command again until you have all the segments of video between the adverts.

Then join the videos together with concatenate command in ffmpeg. (Remember to specify -c copy to not re encode the video).

You could write a bash script and input the times. Then it can cut out the sepcified sections and join them at the end.

Yes, this involved procedure works, but I have to add another step: a quick pass over the video stamping the time on the display, because otherwise there are errors. I just hopped that on the XXI century there would be simple software to do it with click click the mouse, in Linux. Even if the GUI is just a frontend that writes a script for ffmpeg. That would be very wonderful, I could then use any needed extra switches to my heart happiness. - -- Cheers, Carlos E. R. (from openSUSE 42.2 x86_64 "Malachite" at Telcontar) -----BEGIN PGP SIGNATURE----- Version: GnuPG v2 iEYEARECAAYFAloXIKcACgkQtTMYHG2NR9WdVACcD6IL7xp11NDx881eDv2qV1B7 w/UAnRQ33XmTJNynYDHGVszw6SIeiMNZ =g1P2 -----END PGP SIGNATURE----- -- To unsubscribe, e-mail: opensuse+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse+owner@opensuse.org

If you're simply re-formatting/converting, ffmpeg for sure. If you're doing much else, well...  Not so much. I had this second case come up with a relative on Windows 10 who'd previously used a now removed-by-MS utility.  I pointed him to avidemux and this tutorial: https://www.avidemux.org/admWiki/doku.php?id=tutorial:create_video_from_stil... On 11/23/2017 11:25 AM, Carlos E. R. wrote:

-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1

On Thursday, 2017-11-23 at 13:37 -0000, Paul Groves wrote:

...

I just use ffmpeg to remove adverts from my recordings.

ffmpeg -i mov.mp4 -ss 01:00 -t 60 -c copy cut.mp4

-ss (which means start stream) will start at 1 min into the video (so removing the beginning). -t (time duration) will stream for 60 seconds -c copy will keep the encoding and not re-encode the video.

So this command will output 60 seconds of video starting at 1 minute in. I use the slider in VLC player to find the exact times. Then just run the command again until you have all the segments of video between the adverts.

Then join the videos together with concatenate command in ffmpeg. (Remember to specify -c copy to not re encode the video).

You could write a bash script and input the times. Then it can cut out the sepcified sections and join them at the end.

Yes, this involved procedure works, but I have to add another step: a quick pass over the video stamping the time on the display, because otherwise there are errors.

I just hopped that on the XXI century there would be simple software to do it with click click the mouse, in Linux.

Even if the GUI is just a frontend that writes a script for ffmpeg. That would be very wonderful, I could then use any needed extra switches to my heart happiness.

- -- Cheers,        Carlos E. R.        (from openSUSE 42.2 x86_64 "Malachite" at Telcontar)

-----BEGIN PGP SIGNATURE----- Version: GnuPG v2

iEYEARECAAYFAloXIKcACgkQtTMYHG2NR9WdVACcD6IL7xp11NDx881eDv2qV1B7 w/UAnRQ33XmTJNynYDHGVszw6SIeiMNZ =g1P2 -----END PGP SIGNATURE-----

-- To unsubscribe, e-mail: opensuse+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse+owner@opensuse.org

On Thu, 23 Nov 2017 20:32:01 +0100 (CET) "Carlos E. R." wrote:

-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1

On Tuesday, 2017-11-21 at 07:49 +0100, gumb wrote:

...

On 20/11/17 21:13, Carlos E. R. wrote:

...

On Monday, 2017-11-20 at 20:09 +0100, gumb wrote:

...

One for Carlos - I hope this reply goes into the right place in this ancient thread.

I'll tell you later ;-)

...

I believe you were looking for video tools that would treat/export all the original subtitle tracks. I just saw a piece on OMG!Ubuntu! about an update to VidCutter, which *may* - if the notes hold true - provide this functionality, but you'll have to try it out yourself.

http://www.omgubuntu.co.uk/2017/11/vidcutter-5-0-released-improved-ui-frame-...

...

...

>

Thanks :-)

I'll try it out when it comes to packman :-)

Version 5 is in the Packman repos now. Added yesterday.

Installing :-))

Crashes on loading:

cer@Telcontar:~/Fusion/Hacer/Trek> vidcutter & [1] 16260 cer@Telcontar:~/Fusion/Hacer/Trek> Traceback (most recent call last): File "/usr/bin/vidcutter", line 9, in <module> load_entry_point('vidcutter==5.0.0', 'gui_scripts', 'vidcutter')() File "/usr/lib/python3.4/site-packages/pkg_resources/__init__.py", line 558, in load_entry_point return get_distribution(dist).load_entry_point(group, name) File "/usr/lib/python3.4/site-packages/pkg_resources/__init__.py", line 2682, in load_entry_point return ep.load() File "/usr/lib/python3.4/site-packages/pkg_resources/__init__.py", line 2355, in load return self.resolve() File "/usr/lib/python3.4/site-packages/pkg_resources/__init__.py", line 2361, in resolve module = __import__(self.module_name, fromlist=['__name__'], level=0) File "/usr/lib64/python3.4/site-packages/vidcutter/__main__.py", line 40, in <module> from vidcutter.videocutter import VideoCutter File "/usr/lib64/python3.4/site-packages/vidcutter/videocutter.py", line 30, in <module> from typing import Union ImportError: No module named 'typing'

[1]+ Exit 1 vidcutter cer@Telcontar:~/Fusion/Hacer/Trek>

FWIW, I just installed vidcutter (on 42.2) and got the same error. 'typing' is apparently a standard Python module that provides type hints. But ... it's apparently only standard after version 3.6.5. So I tried: $ python3 '/usr/bin/vidcutter' but that gives the same error. Apparently I've only got 3.4.6 There is https://pypi.python.org/pypi/typing to download a backport but that's further than I'm going to try. HTH, Dave

:-((

- -- Cheers, Carlos E. R. (from openSUSE 42.2 x86_64 "Malachite" at Telcontar) -----BEGIN PGP SIGNATURE----- Version: GnuPG v2

iEYEARECAAYFAloXIjEACgkQtTMYHG2NR9VvzACeJbLa6to3dUAd4knTjFm/scLl p3kAoIjVnoR4Nrs37V3qwRTQnuteGJEK =lLdj -----END PGP SIGNATURE-----

-- To unsubscribe, e-mail: opensuse+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse+owner@opensuse.org

-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 On Tuesday, 2017-11-21 at 01:28 +0100, Carlos E. R. wrote:

I only need a simple editor for streams recorded from TV. I want to remove the start lead, the tail end, and the commercials in the middle. It also has to keep the two audio tracks (Spanish and English) and the subtitles. Best if it doesn't recode the video stream, for speed - just remove sections, copy the rest as is.

I got, offlist, the idea of testing a Windows app, Smart Cutter for DV and DVB (http://www.fame-ring.com/). It is payware, but has a free trial version (I don't know what are the limitations of the trial version). It does not recode. You mark the points for removal, then generate a resulting file. I'm testing it inside a vmware virtual machine. Marking the edit points is a bit slow, but once done the recording is very fast, being basically a file copy operation. What I have not being sucessful yet is in removing sections, the result has the entire stream. I don't know if I'm not doing it right or that's a limitation of the "trial" version. Ah, apparently it is a water mark, says the web page - but I see no water mark in the result. I'll test a bit more. If I can make it work, the next step would be to run it in Wine instead. According to the help, I mark a start point, an end point, then hit delete to remove that section. It doesn't, instead it deletes the mark points, and the conversion applies to the entire stream. This software is the exact thing I want, but it fails to work. - -- Cheers, Carlos E. R. (from openSUSE 42.2 x86_64 "Malachite" at Telcontar) -----BEGIN PGP SIGNATURE----- Version: GnuPG v2 iEYEARECAAYFAloXRS0ACgkQtTMYHG2NR9X/oACfS9V0kNUD9FOfnHeBGvCJshnU S9gAoISgoeNm4pB0bSoSyGLlMadsE7nv =aLVd -----END PGP SIGNATURE----- -- To unsubscribe, e-mail: opensuse+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse+owner@opensuse.org