Re: [FFmpeg-user] More Liberal Licensing

2023-11-28 Thread Reindl Harald




Am 28.11.23 um 07:28 schrieb Suminda Sirinath Salpitikorala Dharmasena:

- You cannot just borrow some code than using the library


thanks god

___
ffmpeg-user mailing list
ffmpeg-user@ffmpeg.org
https://ffmpeg.org/mailman/listinfo/ffmpeg-user

To unsubscribe, visit link above, or email
ffmpeg-user-requ...@ffmpeg.org with subject "unsubscribe".


Re: [FFmpeg-user] More Liberal Licensing

2023-11-28 Thread Suminda Sirinath Salpitikorala Dharmasena
For a project like this, forking into a private or proprietary code base
with substantial modifications is not feasible as maintaining it will be
difficult and keeping up and modifying upstream changes will be very
difficult. Also maintaining a private hard fork will be very costly.

 > copyleft so everyone actually improves it

Being copyleft does not mean everyone improves it though many believe so.
Increase in the number of users is what will drive improvements. Usable
improvements are generally what gets actually upstreamed. Beyond a certain
threshold, keeping private improvements has a larger cost than upstreaming
it for large projects and when the modifications become extensive. *GPL
means that all forks need to be public, not that the modifications need to
be upstreamed. LGPL encourages to use a library as is without modification
when used in proprietary code. Encouraging modifications and
adaptation even in proriotery context will increase potential contributors
who are familiar with the code base. Even if a project they work on is
proprietary they can very well become contributors when those working on it
in their own time. *GPL reduces local improvements either local, private or
public from commential users and also subsequently upstreaming of the
changes once the changes become substatial. Once a library project grows
beyond a critical mass LGPL hurts the project more than helps.
___
ffmpeg-user mailing list
ffmpeg-user@ffmpeg.org
https://ffmpeg.org/mailman/listinfo/ffmpeg-user

To unsubscribe, visit link above, or email
ffmpeg-user-requ...@ffmpeg.org with subject "unsubscribe".


Re: [FFmpeg-user] More Liberal Licensing

2023-11-28 Thread Jean-Baptiste Kempf
On Tue, 28 Nov 2023, at 11:23, Suminda Sirinath Salpitikorala Dharmasena wrote:
> changes once the changes become substatial. Once a library project grows
> beyond a critical mass LGPL hurts the project more than helps.

Let me introduce you to a small project:
http://kernel.org/


-- 
Jean-Baptiste Kempf -  President
+33 672 704 734
___
ffmpeg-user mailing list
ffmpeg-user@ffmpeg.org
https://ffmpeg.org/mailman/listinfo/ffmpeg-user

To unsubscribe, visit link above, or email
ffmpeg-user-requ...@ffmpeg.org with subject "unsubscribe".


Re: [FFmpeg-user] More Liberal Licensing

2023-11-28 Thread Suminda Sirinath Salpitikorala Dharmasena
This is not meant to be used as a library. This is an application.
___
ffmpeg-user mailing list
ffmpeg-user@ffmpeg.org
https://ffmpeg.org/mailman/listinfo/ffmpeg-user

To unsubscribe, visit link above, or email
ffmpeg-user-requ...@ffmpeg.org with subject "unsubscribe".


Re: [FFmpeg-user] More Liberal Licensing

2023-11-28 Thread Suminda Sirinath Salpitikorala Dharmasena
In application software fragmentation can hurt. *GPL reduces the risk of
fragmentation. This is how the Linux Kernel succeeded.
___
ffmpeg-user mailing list
ffmpeg-user@ffmpeg.org
https://ffmpeg.org/mailman/listinfo/ffmpeg-user

To unsubscribe, visit link above, or email
ffmpeg-user-requ...@ffmpeg.org with subject "unsubscribe".


Re: [FFmpeg-user] More Liberal Licensing

2023-11-28 Thread Suminda Sirinath Salpitikorala Dharmasena
This is the reason OpenJDK is successful though at this point of time they
are very well at a point they can change the license.

GPL nature of the license prevented fragmentation by de-incentivising
competing incompatible hard forks.

>
___
ffmpeg-user mailing list
ffmpeg-user@ffmpeg.org
https://ffmpeg.org/mailman/listinfo/ffmpeg-user

To unsubscribe, visit link above, or email
ffmpeg-user-requ...@ffmpeg.org with subject "unsubscribe".


Re: [FFmpeg-user] More Liberal Licensing

2023-11-28 Thread Reindl Harald




Am 28.11.23 um 11:32 schrieb Suminda Sirinath Salpitikorala Dharmasena:

This is the reason OpenJDK is successful though at this point of time they
are very well at a point they can change the license.

GPL nature of the license prevented fragmentation by de-incentivising
competing incompatible hard forks.


leave us in peace - a license change won't happen even if it would be 
possible but it isn't


case closed
___
ffmpeg-user mailing list
ffmpeg-user@ffmpeg.org
https://ffmpeg.org/mailman/listinfo/ffmpeg-user

To unsubscribe, visit link above, or email
ffmpeg-user-requ...@ffmpeg.org with subject "unsubscribe".


Re: [FFmpeg-user] More Liberal Licensing

2023-11-28 Thread David Bernat
Please unsubscribe. I have sent numerous requests.

On Tue, Nov 28, 2023 at 5:34 AM Reindl Harald 
wrote:

>
>
> Am 28.11.23 um 11:32 schrieb Suminda Sirinath Salpitikorala Dharmasena:
> > This is the reason OpenJDK is successful though at this point of time
> they
> > are very well at a point they can change the license.
> >
> > GPL nature of the license prevented fragmentation by de-incentivising
> > competing incompatible hard forks.
>
> leave us in peace - a license change won't happen even if it would be
> possible but it isn't
>
> case closed
> ___
> ffmpeg-user mailing list
> ffmpeg-user@ffmpeg.org
> https://ffmpeg.org/mailman/listinfo/ffmpeg-user
>
> To unsubscribe, visit link above, or email
> ffmpeg-user-requ...@ffmpeg.org with subject "unsubscribe".
>
___
ffmpeg-user mailing list
ffmpeg-user@ffmpeg.org
https://ffmpeg.org/mailman/listinfo/ffmpeg-user

To unsubscribe, visit link above, or email
ffmpeg-user-requ...@ffmpeg.org with subject "unsubscribe".


Re: [FFmpeg-user] More Liberal Licensing

2023-11-28 Thread Suminda Sirinath Salpitikorala Dharmasena
Noted.

No worries. Just asked to see if it is a possibility.


On Tue, 28 Nov 2023 at 16:04, Reindl Harald  wrote:

>
>
> Am 28.11.23 um 11:32 schrieb Suminda Sirinath Salpitikorala Dharmasena:
> > This is the reason OpenJDK is successful though at this point of time
> they
> > are very well at a point they can change the license.
> >
> > GPL nature of the license prevented fragmentation by de-incentivising
> > competing incompatible hard forks.
>
> leave us in peace - a license change won't happen even if it would be
> possible but it isn't
>
> case closed
> ___
> ffmpeg-user mailing list
> ffmpeg-user@ffmpeg.org
> https://ffmpeg.org/mailman/listinfo/ffmpeg-user
>
> To unsubscribe, visit link above, or email
> ffmpeg-user-requ...@ffmpeg.org with subject "unsubscribe".
>
___
ffmpeg-user mailing list
ffmpeg-user@ffmpeg.org
https://ffmpeg.org/mailman/listinfo/ffmpeg-user

To unsubscribe, visit link above, or email
ffmpeg-user-requ...@ffmpeg.org with subject "unsubscribe".


Re: [FFmpeg-user] More Liberal Licensing

2023-11-28 Thread Reindl Harald



Am 28.11.23 um 11:35 schrieb David Bernat:

Please unsubscribe. I have sent numerous requests.


what is your problem?

nobody needs another 10 mails within minutes in the style of "This is 
not meant to be used as a library. This is an application"


it's not possible to change the license, it won't happen and both where 
statet clearly - so no reason to ride the dead horse in another 10 mails



On Tue, Nov 28, 2023 at 5:34 AM Reindl Harald 
wrote:



Am 28.11.23 um 11:32 schrieb Suminda Sirinath Salpitikorala Dharmasena:

This is the reason OpenJDK is successful though at this point of time

they

are very well at a point they can change the license.

GPL nature of the license prevented fragmentation by de-incentivising
competing incompatible hard forks.


leave us in peace - a license change won't happen even if it would be
possible but it isn't

case closed


___
ffmpeg-user mailing list
ffmpeg-user@ffmpeg.org
https://ffmpeg.org/mailman/listinfo/ffmpeg-user

To unsubscribe, visit link above, or email
ffmpeg-user-requ...@ffmpeg.org with subject "unsubscribe".


Re: [FFmpeg-user] More Liberal Licensing

2023-11-28 Thread David Bernat
Licensing is not the issue. Show of force required.

On Tue, Nov 28, 2023 at 5:41 AM Reindl Harald 
wrote:

>
>
> Am 28.11.23 um 11:35 schrieb David Bernat:
> > Please unsubscribe. I have sent numerous requests.
>
> what is your problem?
>
> nobody needs another 10 mails within minutes in the style of "This is
> not meant to be used as a library. This is an application"
>
> it's not possible to change the license, it won't happen and both where
> statet clearly - so no reason to ride the dead horse in another 10 mails
>
> > On Tue, Nov 28, 2023 at 5:34 AM Reindl Harald 
> > wrote:
> >
> >>
> >> Am 28.11.23 um 11:32 schrieb Suminda Sirinath Salpitikorala Dharmasena:
> >>> This is the reason OpenJDK is successful though at this point of time
> >> they
> >>> are very well at a point they can change the license.
> >>>
> >>> GPL nature of the license prevented fragmentation by de-incentivising
> >>> competing incompatible hard forks.
> >>
> >> leave us in peace - a license change won't happen even if it would be
> >> possible but it isn't
> >>
> >> case closed
>
> ___
> ffmpeg-user mailing list
> ffmpeg-user@ffmpeg.org
> https://ffmpeg.org/mailman/listinfo/ffmpeg-user
>
> To unsubscribe, visit link above, or email
> ffmpeg-user-requ...@ffmpeg.org with subject "unsubscribe".
>
___
ffmpeg-user mailing list
ffmpeg-user@ffmpeg.org
https://ffmpeg.org/mailman/listinfo/ffmpeg-user

To unsubscribe, visit link above, or email
ffmpeg-user-requ...@ffmpeg.org with subject "unsubscribe".


Re: [FFmpeg-user] More Liberal Licensing

2023-11-28 Thread Rob Hallam
As has been stated, it seems vanishingly unlikely there will be a relicense.

However, it seems there is a misconception about the GPL here:

On Tue, 28 Nov 2023 at 10:24, Suminda Sirinath Salpitikorala
Dharmasena  wrote:
>*GPL means that all forks need to be public, not that the modifications need to
> be upstreamed.

The GPL does not require "all forks need to be public". A private
entity (company, organisation) can use a GPL'd project internally and
they never have to release their code. [0]

This comes up so frequently it has a FAQ:
https://www.gnu.org/licenses/gpl-faq.en.html#GPLRequireSourcePostedPublic

Of course, if the product is released, users have the right to request
a copy of the source to the product. That is of course the point of
copyleft- if you build on the freely-available work of other people,
others should be able to use your work freely too.

Cheers,
Rob

--

[0]: Note that another option would be calling an ffmpeg binary from a
(publically-released) proprietary program, but naturally the source of
ffmpeg (including any modifications) would need to be available on
request; and if the two programs had sufficient shared state, they may
be considered as a single program and so again covered by the GPL.
___
ffmpeg-user mailing list
ffmpeg-user@ffmpeg.org
https://ffmpeg.org/mailman/listinfo/ffmpeg-user

To unsubscribe, visit link above, or email
ffmpeg-user-requ...@ffmpeg.org with subject "unsubscribe".


Re: [FFmpeg-user] More Liberal Licensing

2023-11-28 Thread Phil Rhodes via ffmpeg-user
 As has been said, it would be essentially impossible to track down everyone 
who's contributed to a project the size of ffmpeg and gain permission of each 
to alter the licence, so the discussion is effectively moot. Another reason 
it's moot is that open source is effectively a religion to a lot of people, 
Stallman is one of its prophets, and the licence he originated is one of its 
sacred texts. The question being asked here is perhaps a little naïve but that 
explains the vehemence of the response. Asking people to give up GPL is like 
pissing on their sainted aunt.

The thing that always makes me laugh about this is a piece written years ago 
for O'Reilly in which it was argued that the GPL is unnecessarily confusing, 
and the person making that argument had been senior in the Perl community. When 
Perl people are criticising something for being impenetrable, it's probably 
time to take note.

The thing which makes all this a bit difficult is whether it's possible (or 
easy) to prove that a given distributed binary is actually compiled from the 
source code one is offering. I'm not sure that this has ever been tested in 
court - most of the (L)GPL stuff hasn't, much - but it's not obvious to how 
someone could prove that if there were ever a dispute. It would be difficult to 
create a build system which would create a bit-identical binary twice from the 
same source tree, and other approaches (use of the strings command is often 
suggested) are obviously far from conclusive and tremendously easy to fake.
In the end, most open source licences work because people mostly obey the 
spirit of them and, I suspect, for no other reason. I suspect many of them are 
broken constantly and it's okay because nobody's reading the program flash out 
of most of the world's shipped SoCs and trying to figure out what's in there.
P
On Tuesday, 28 November 2023 at 11:40:39 GMT, Rob Hallam 
 wrote:  
 
 As has been stated, it seems vanishingly unlikely there will be a relicense.

However, it seems there is a misconception about the GPL here:

On Tue, 28 Nov 2023 at 10:24, Suminda Sirinath Salpitikorala
Dharmasena  wrote:
>*GPL means that all forks need to be public, not that the modifications need to
> be upstreamed.

The GPL does not require "all forks need to be public". A private
entity (company, organisation) can use a GPL'd project internally and
they never have to release their code. [0]

This comes up so frequently it has a FAQ:
https://www.gnu.org/licenses/gpl-faq.en.html#GPLRequireSourcePostedPublic

Of course, if the product is released, users have the right to request
a copy of the source to the product. That is of course the point of
copyleft- if you build on the freely-available work of other people,
others should be able to use your work freely too.

Cheers,
Rob

--

[0]: Note that another option would be calling an ffmpeg binary from a
(publically-released) proprietary program, but naturally the source of
ffmpeg (including any modifications) would need to be available on
request; and if the two programs had sufficient shared state, they may
be considered as a single program and so again covered by the GPL.
___
ffmpeg-user mailing list
ffmpeg-user@ffmpeg.org
https://ffmpeg.org/mailman/listinfo/ffmpeg-user

To unsubscribe, visit link above, or email
ffmpeg-user-requ...@ffmpeg.org with subject "unsubscribe".
  
___
ffmpeg-user mailing list
ffmpeg-user@ffmpeg.org
https://ffmpeg.org/mailman/listinfo/ffmpeg-user

To unsubscribe, visit link above, or email
ffmpeg-user-requ...@ffmpeg.org with subject "unsubscribe".


Re: [FFmpeg-user] does ffmpeg support MVC (3D BluRays)?

2023-11-28 Thread Paul B Mahol
On Mon, Nov 27, 2023 at 11:31 PM Philippe Cerfon  wrote:

> Hey there.
>
> I couldn't find a definitive answer on this.
> Some Wikipedia articles and stackoverflow questions claim ffmpeg would
> not support playback of MVC (and thus 3D BluRays).
>
>
IIRC MVC in h264 is not supported.


> I don't have a 3D projector or anything the likes yet, but still
> wanted to check whether some Videos would actually contain 3D
> information.
>
> So basically I though via ffmpeg's stereo3d filter
> (https://trac.ffmpeg.org/wiki/Stereoscopic) I might be able to get the
> 3D BluRay into a side-by-side picture, which should then be similar
> but of course a bit different.
>
> However, when doing so the results are not really as I'd expect them:
> Each side is only roughly half of the full picture. And the same
> happens also when I do it on a BluRay that for sure has no 3D.
>

stereo3d filter is for anaglyph presentation of decoded 3d frames and some
other minor features.
supporting alternating, top/bottom, left/right etc.

But as usual:
- missing ffmpeg version
- missing full uncut ffmpeg command line and its output
- missing input files


>
>
> I further saw that there are probably some OpenSource MVC decoders:
> https://github.com/carrardt/h264-tools
>
> https://github.com/Britz/FFmpeg (was a master thesis with H.264 MVC
> support for ffmpeg)
>
>
>
> So should that work already, or could I open a feature request somewhere?
> :-)
>

IIRC H.264 MVC is alternating frames 3d format.


>
>
> Thanks,
> Philippe
> ___
> ffmpeg-user mailing list
> ffmpeg-user@ffmpeg.org
> https://ffmpeg.org/mailman/listinfo/ffmpeg-user
>
> To unsubscribe, visit link above, or email
> ffmpeg-user-requ...@ffmpeg.org with subject "unsubscribe".
>
___
ffmpeg-user mailing list
ffmpeg-user@ffmpeg.org
https://ffmpeg.org/mailman/listinfo/ffmpeg-user

To unsubscribe, visit link above, or email
ffmpeg-user-requ...@ffmpeg.org with subject "unsubscribe".


Re: [FFmpeg-user] More Liberal Licensing

2023-11-28 Thread Carl Zwanzig

On 11/28/2023 4:09 AM, Phil Rhodes via ffmpeg-user wrote:

The thing which makes all this a bit difficult is whether it's possible
(or easy) to prove that a given distributed binary is actually compiled
from the source code one is offering. I'm not sure that this has ever
been tested in court - most of the (L)GPL stuff hasn't, much - but it's
not obvious to how someone could prove that if there were ever a
dispute.


FWIW, BusyBox has taken on some commercial organizations to court and 
prevailed, search "busybox copyright lawsuit" and you'll get things like


https://softwarefreedom.org/news/2009/dec/14/busybox-gpl-lawsuit/
https://arstechnica.com/information-technology/2010/08/court-rules-gpl-part-of-a-well-pleaded-case/ 
(summary judgement)

https://www.computerworld.com/article/2537947/open-source-legal-group-strikes-again-on-busybox--suing-verizon.html

So whether or not you agree with the current licensing structure, if you 
want to use ffmpeg components you have to play by those rules.


Later,

z!
___
ffmpeg-user mailing list
ffmpeg-user@ffmpeg.org
https://ffmpeg.org/mailman/listinfo/ffmpeg-user

To unsubscribe, visit link above, or email
ffmpeg-user-requ...@ffmpeg.org with subject "unsubscribe".


Re: [FFmpeg-user] More Liberal Licensing

2023-11-28 Thread Phil Rhodes via ffmpeg-user
 > So whether or not you agree with the current licensing structure, if you> 
 >want to use ffmpeg components you have to play by those rules.
Well, of course, but that's not really my concern: it's more that even if you 
were playing by the rules, it might actually be quite difficult to prove that. 
Someone could claim that the source code being distributed was not the source 
code used to compile the binary in question, and it would be next to impossible 
to prove otherwise - especially as you'd be trying to persuade a court full of 
people who are not software engineers and have only the haziest understanding 
of what code and compiled objects even are.
This is not a problem specific to ffmpeg, it's just a piece of software that 
seems especially likely to be targeted for this sort of thing, given what it 
does.
P
___
ffmpeg-user mailing list
ffmpeg-user@ffmpeg.org
https://ffmpeg.org/mailman/listinfo/ffmpeg-user

To unsubscribe, visit link above, or email
ffmpeg-user-requ...@ffmpeg.org with subject "unsubscribe".


[FFmpeg-user] drawtext filter missing

2023-11-28 Thread Leo Butler via ffmpeg-user
Hello,

I ran into this problem with the current Debian testing build, so I
thought I would try the static build. Both are compiled with
--enable-libfreetype but both barf on a drawtext filter. I am attaching
the log file from the static build.

According to:

https://ffmpeg.org/ffmpeg-all.html#drawtext-1

the libfreetype library should be sufficient.

See also:
https://bugs.debian.org/cgi-bin/bugreport.cgi?bug=1056597
https://trac.ffmpeg.org/ticket/10705#ticket

Leo

ffmpeg started on 2023-11-28 at 14:38:33
Report written to "ffmpeg-20231128-143833.log"
Log level: 48
Command line:
./ffmpeg -y -report -f lavfi -i "testsrc=duration=5.1:size=hd720:rate=10" -f mpegts out.ts
ffmpeg version 6.1-static https://johnvansickle.com/ffmpeg/  Copyright (c) 2000-2023 the FFmpeg developers
  built with gcc 8 (Debian 8.3.0-6)
  configuration: --enable-gpl --enable-version3 --enable-static --disable-debug --disable-ffplay --disable-indev=sndio --disable-outdev=sndio --cc=gcc --enable-fontconfig --enable-frei0r --enable-gnutls --enable-gmp --enable-libgme --enable-gray --enable-libfribidi --enable-libass --enable-libfreetype --enable-libmp3lame --enable-libopencore-amrnb --enable-libopencore-amrwb --enable-libopenjpeg --enable-librubberband --enable-libsoxr --enable-libspeex --enable-libsrt --enable-libvorbis --enable-libopus --enable-libtheora --enable-libvidstab --enable-libvo-amrwbenc --enable-libvpx --enable-libwebp --enable-libx264 --enable-libx265 --enable-libxml2 --enable-libxvid --enable-libzimg
  libavutil  58. 29.100 / 58. 29.100
  libavcodec 60. 31.102 / 60. 31.102
  libavformat60. 16.100 / 60. 16.100
  libavdevice60.  3.100 / 60.  3.100
  libavfilter 9. 12.100 /  9. 12.100
  libswscale  7.  5.100 /  7.  5.100
  libswresample   4. 12.100 /  4. 12.100
  libpostproc57.  3.100 / 57.  3.100
Splitting the commandline.
Reading option '-y' ... matched as option 'y' (overwrite output files) with argument '1'.
Reading option '-report' ... matched as option 'report' (generate a report) with argument '1'.
Reading option '-f' ... matched as option 'f' (force format) with argument 'lavfi'.
Reading option '-i' ... matched as output url with argument 'testsrc=duration=5.1:size=hd720:rate=10'.
Reading option '-f' ... matched as option 'f' (force format) with argument 'mpegts'.
Reading option 'out.ts' ... matched as output url.
Finished splitting the commandline.
Parsing a group of options: global .
Applying option y (overwrite output files) with argument 1.
Applying option report (generate a report) with argument 1.
Successfully parsed a group of options.
Parsing a group of options: input url testsrc=duration=5.1:size=hd720:rate=10.
Applying option f (force format) with argument lavfi.
Successfully parsed a group of options.
Opening an input file: testsrc=duration=5.1:size=hd720:rate=10.
[AVFilterGraph @ 0xdad98c0] Setting 'duration' to value '5.1'
[AVFilterGraph @ 0xdad98c0] Setting 'size' to value 'hd720'
[AVFilterGraph @ 0xdad98c0] Setting 'rate' to value '10'
detected 8 logical cores
[Parsed_testsrc_0 @ 0xdada8c0] size:1280x720 rate:10/1 duration:5.10 sar:1/1
[AVFilterGraph @ 0xdad98c0] query_formats: 2 queried, 1 merged, 0 already done, 0 delayed
[lavfi @ 0xdad92c0] All info found
Input #0, lavfi, from 'testsrc=duration=5.1:size=hd720:rate=10':
  Duration: N/A, start: 0.00, bitrate: N/A
  Stream #0:0, 1, 1/10: Video: wrapped_avframe, rgb24, 1280x720 [SAR 1:1 DAR 16:9], 10 fps, 10 tbr, 10 tbn
Successfully opened the file.
Parsing a group of options: output url out.ts.
Applying option f (force format) with argument mpegts.
Successfully parsed a group of options.
Opening an output file: out.ts.
[out#0/mpegts @ 0xdadb700] No explicit maps, mapping streams automatically...
[vost#0:0/mpeg2video @ 0xdade600] Created video stream from input stream 0:0
[file @ 0xdae2140] Setting default whitelist 'file,crypto,data'
Successfully opened the file.
Stream mapping:
  Stream #0:0 -> #0:0 (wrapped_avframe (native) -> mpeg2video (native))
Press [q] to stop, [?] for help
[graph 0 input from stream 0:0 @ 0xdae3400] Setting 'video_size' to value '1280x720'
[graph 0 input from stream 0:0 @ 0xdae3400] Setting 'pix_fmt' to value '2'
[graph 0 input from stream 0:0 @ 0xdae3400] Setting 'time_base' to value '1/10'
[graph 0 input from stream 0:0 @ 0xdae3400] Setting 'pixel_aspect' to value '1/1'
[graph 0 input from stream 0:0 @ 0xdae3400] Setting 'frame_rate' to value '10/1'
[graph 0 input from stream 0:0 @ 0xdae3400] w:1280 h:720 pixfmt:rgb24 tb:1/10 fr:10/1 sar:1/1
[format @ 0xdae39c0] Setting 'pix_fmts' to value 'yuv420p|yuv422p'
[auto_scale

Re: [FFmpeg-user] drawtext filter missing

2023-11-28 Thread Paul B Mahol
On Tue, Nov 28, 2023 at 9:40 PM Leo Butler via ffmpeg-user <
ffmpeg-user@ffmpeg.org> wrote:

> Hello,
>
> I ran into this problem with the current Debian testing build, so I
> thought I would try the static build. Both are compiled with
> --enable-libfreetype but both barf on a drawtext filter. I am attaching
> the log file from the static build.
>
> According to:
>
> https://ffmpeg.org/ffmpeg-all.html#drawtext-1
>
> the libfreetype library should be sufficient.
>
>
Citation from that link:

To enable compilation of this filter, you need to configure FFmpeg with
--enable-libfreetype and --enable-libharfbuzz. To enable default font
fallback and the font option you need to configure FFmpeg with
--enable-libfontconfig. To enable the text_shaping option, you need to
configure FFmpeg with --enable-libfribidi.

So your statements are invalid.


> See also:
> https://bugs.debian.org/cgi-bin/bugreport.cgi?bug=1056597
> https://trac.ffmpeg.org/ticket/10705#ticket
>
> Leo
>
> ___
> ffmpeg-user mailing list
> ffmpeg-user@ffmpeg.org
> https://ffmpeg.org/mailman/listinfo/ffmpeg-user
>
> To unsubscribe, visit link above, or email
> ffmpeg-user-requ...@ffmpeg.org with subject "unsubscribe".
>
___
ffmpeg-user mailing list
ffmpeg-user@ffmpeg.org
https://ffmpeg.org/mailman/listinfo/ffmpeg-user

To unsubscribe, visit link above, or email
ffmpeg-user-requ...@ffmpeg.org with subject "unsubscribe".


Re: [FFmpeg-user] does ffmpeg support MVC (3D BluRays)?

2023-11-28 Thread Philippe Cerfon
Hey Paul.

On Tue, Nov 28, 2023 at 4:51 PM Paul B Mahol  wrote:
> IIRC MVC in h264 is not supported.

What a pity. I've filed https://trac.ffmpeg.org/ticket/10706 in case
someone is ever going to work on this.

> stereo3d filter is for anaglyph presentation of decoded 3d frames and some
> other minor features.
> supporting alternating, top/bottom, left/right etc.

I'm really new to 3D, so may be completely wrong: but as far as I
understood, quite some end devices (e.g. 3D capable projectors or so)
expect e.g. side-by-side input, too (i.e. than cannot cope with MVC
either).

Also, it would have been nice to convert the MVC to a side-by-side
version and see whether it actually works. Seems so far, there are
only few ways to decode this. Most people apparently use BD3D2MK3D[0],
but that's windows only, and as far as I can see not open.

> But as usual:

Well I guess the following are not really useful, if it anyway doesn't
support MVC, but here they are:

> - missing ffmpeg version

$ ffmpeg -version
ffmpeg version 6.1-4 Copyright (c) 2000-2023 the FFmpeg developers
built with gcc 13 (Debian 13.2.0-6)
configuration: --prefix=/usr --extra-version=4 --toolchain=hardened
--libdir=/usr/lib/x86_64-linux-gnu
--incdir=/usr/include/x86_64-linux-gnu --arch=amd64 --enable-gpl
--disable-stripping --enable-gnutls --enable-ladspa --enable-libaom
--enable-libass --enable-libbluray --enable-libbs2b --enable-libcaca
--enable-libcdio --enable-libcodec2 --enable-libdav1d
--enable-libflite --enable-libfontconfig --enable-libfreetype
--enable-libfribidi --enable-libglslang --enable-libgme
--enable-libgsm --enable-libjack --enable-libmp3lame
--enable-libmysofa --enable-libopenjpeg --enable-libopenmpt
--enable-libopus --enable-libpulse --enable-librabbitmq
--enable-librist --enable-librubberband --enable-libshine
--enable-libsnappy --enable-libsoxr --enable-libspeex --enable-libsrt
--enable-libssh --enable-libtheora --enable-libtwolame
--enable-libvidstab --enable-libvorbis --enable-libvpx
--enable-libwebp --enable-libx265 --enable-libxml2 --enable-libxvid
--enable-libzimg --enable-libzmq --enable-libzvbi --enable-lv2
--enable-omx --enable-openal --enable-opencl --enable-opengl
--enable-sdl2 --disable-sndio --enable-libjxl --enable-pocketsphinx
--enable-librsvg --enable-libvpl --disable-libmfx --enable-libdc1394
--enable-libdrm --enable-libiec61883 --enable-chromaprint
--enable-frei0r --enable-libsvtav1 --enable-libx264
--enable-libplacebo --enable-librav1e --enable-shared
libavutil  58. 29.100 / 58. 29.100
libavcodec 60. 31.102 / 60. 31.102
libavformat60. 16.100 / 60. 16.100
libavdevice60.  3.100 / 60.  3.100
libavfilter 9. 12.100 /  9. 12.100
libswscale  7.  5.100 /  7.  5.100
libswresample   4. 12.100 /  4. 12.100
libpostproc57.  3.100 / 57.  3.100


> - missing full uncut ffmpeg command line and its output

I merely tried on some MKV (which was generated from a BluRay):
$ ffplay -i a2.mkv -vf stereo3d=sbsr:sbsr

and various oher outbuts. But obviously, the MVC stream is not any of
the input formats that the filter supports.

> - missing input files

That might be difficult, I mean these are from BluRays, so the
material is a) super big, b) copyrighted.


> IIRC H.264 MVC is alternating frames 3d format.

As far as I understood it, it's rather that the H264 contains two
streams: a "normal" (AVC) one, which I think it s the left eye, and
the MVC stream, which is like a diff from the left to the right eye.

Thanks,
Philippe.

[0] https://download.videohelp.com/r0lZ/BD3D2AVS/index.html
___
ffmpeg-user mailing list
ffmpeg-user@ffmpeg.org
https://ffmpeg.org/mailman/listinfo/ffmpeg-user

To unsubscribe, visit link above, or email
ffmpeg-user-requ...@ffmpeg.org with subject "unsubscribe".