Re: mencoder HEAD: tv audio recording broken?

On Fri, Jan 30, 2004 at 01:23:29PM +0100, Oswald Buddenhagen wrote:
> surprise surprise, it still does not work. :[
> 
> On Fri, Jan 16, 2004 at 02:52:34PM +0100, Oswald Buddenhagen wrote:
> > this must have been introduced somewhen between my last cvs update (no
> > more than two weeks) and today.
> > 
> more concretely, it was between jan 8 and jan 16.
> 
i tracked it down to rev 1.7 of libmpdemux/ai_oss.c

> things do work fine with -oac pcm, so the problem is possibly in the
> liblame interface code. liblame is current cvs HEAD (works fine when i
> downgrade mencoder).
> 
the problem is the combination of my soundcard with liblame: my card (or
the driver, whatever) says the actual sample rate is 44101, instead of
44100. as a consequence, liblame cannot handle the data:

> > my command line is (as ever):
> > 
> >  mencoder tv:// -tv \
> >   driver=v4l:width=768:height=576:volume=65535:bass=32768:treble=32768 \
> >   -vf crop=720:540:24:18,pp=lb,denoise3d -ovc lavc -lavcopts \
> >   vcodec=mpeg4:vqscale=4:keyint=300 -oac mp3lame -lameopts fast:vbr=4:q=4
> > 
> > output is: [...]
> > 
> > Video stream:  665.036 kbit/s  (83129 bps)  size: 3358431 bytes  40.400 secs  1009 frames
> > 

v4l source: don't mess up volume

heya,

the attached patch removes several superflous volume settings - setting
the mute flag should be sufficient. *)
this fixes the major annoyance that every time i start xawtv after
mencoder had run, the volume is zero and i have to manually increase it
again.

i'm not sure that having tv_param_volume -1 is a good default, but one
can't set it to -1 explicitly ... maybe a big fat warning should be
emited if the volume is detected to be (close to) zero.

*) if it's not on some (broken) setups, one could go a different route:
save the current setting and restore it at shutdown.

greetings

--

-- 
Chaos, panic, and disorder - my work here is done.

Index: tvi_v4l.c
===================================================================
RCS file: /cvsroot/mplayer/main/libmpdemux/tvi_v4l.c,v
retrieving revision 1.70
diff -U2 -r1.70 tvi_v4l.c
--- tvi_v4l.c	20 Jan 2004 13:32:05 -0000	1.70
+++ tvi_v4l.c	1 Feb 2004 02:58:13 -0000
 <at>  <at>  -333,5 +333,4  <at>  <at> 

add support for audio format 0x6771

Hi

I have some videos with no audio in mplayer:
Playing toma.caragiu.dvdrip.cd1.avi.
AVI file format detected.
VIDEO:  [XVID]  720x576  24bpp  25.000 fps  918.9 kbps (112.2 kbyte/s)
Clip info:
 Software: VirtualDubMod 1.5.4.1 (build 2066/release)
==========================================================================
Cannot find codec for audio format 0x6771.
Read DOCS/HTML/en/codecs.html!
==========================================================================

I've found the explanation here:
http://www.xiph.org/archives/vorbis/200305/0045.html

Downloaded codec from 
http://www.wma-decoder.com/free/divx_ac3/AC3_ogg_vorbis_audio_codec/
and some another OGG dlls, but it still doesn't work. 
Could support for this be added or it's deprecated?

I'm using mplayer 1.0pre2

Thanks,
--

-- 
Marius Andreiana
Galuna - Solutii Linux in Romania
http://www.galuna.ro

Re: Release schedule

Heya,,

> Tomorrow :)

Well then - I've committed the fix. Happy releasin' :)

Mosu

--

-- 
If Darl McBride was in charge, he'd probably make marriage
unconstitutional too, since clearly it de-emphasizes the commercial
nature of normal human interaction, and probably is a major impediment
to the commercial growth of prostitution. - Linus Torvalds

_______________________________________________
MPlayer-dev-eng mailing list
MPlayer-dev-eng <at> mplayerhq.hu
http://mplayerhq.hu/mailman/listinfo/mplayer-dev-eng

Re: [PATCH] Donald Graft's kerndeint adaptive deinterlacing filter

Diego Biurrun wrote:
> Tobias Diedrich writes:
>  > D Richard Felker III wrote:
>  > > IMO apply it _before_ release, not after. But the -ao stuff in the man
>  > > page doesn't belong with this patch...
>  > 
>  > I was fiddling around with ao_pcm some time ago and it seems the manpage
>  > part slipped through.  Applied without it.
> 
> Feel free to commit the rest separately, but I did not get the man

It's not really worth it.

> page update, did you really commit it?

Doh. I somehow managed to forget the commit for the manpage.

--

-- 
Tobias						PGP: http://9ac7e0bc.2ya.com
This mail is made of 100% recycled bits.

Re: Re: [PATCH] Donald Graft's kerndeint adaptive deinterlacing filter

Tobias Diedrich writes:
 > Doh. I somehow managed to forget the commit for the manpage.

OK, I just replied to this commit on mplayer-docs, but then noticed
that Tobias Diedrich is not subscribed to it, so I am repeating this
here.  The comments about filter parameter design are of general
interest, too, so please forgive me..

Tobias Diedrich CVS writes:
 > Modified Files:
 > 	mplayer.1 
 > Log Message:
 > kerndeint documentation

Thanks.  Some comments below.

 > +.PD 1

This one needs to come below the .PD 0, not above.

 > +.TP
 > +.B kerndeint[=map[:order[:thresh[:sharp[:twoway]]]]]
 > +Donald Graft's adaptive kernel deinterlacer.
 > +Deinterlaces parts of video if a configurable threshold is exceeded.
 > +.PD 0
 > +.RSs
 > +.IPs map (0 or 1)

You need to put the argument of .IPs in "" quotes unless it is only
one word.

[PATCH] libavcodec compilation fix

Hi,

Here are just 2 small compilation fix for libavcodec. The first one
is just a classic #ifdef CONFIG_ENCODERS missing.
The second one allow to build lcl.c even when zlib is not installed
in the standard path. But it is specific to mplayer and would probably
need to be a bit different for building into ffmpeg.

Aurelien

diff -Naur MPlayer-20040129.orig/libavcodec/i386/dsputil_mmx.c MPlayer-20040129/libavcodec/i386/dsputil_mmx.c
--- MPlayer-20040129.orig/libavcodec/i386/dsputil_mmx.c	2004-02-01 13:56:42.000000000 +0100
+++ MPlayer-20040129/libavcodec/i386/dsputil_mmx.c	2004-02-01 15:37:07.000000000 +0100
 <at>  <at>  -2158,7 +2158,9  <at>  <at> 
                 c->put_no_rnd_pixels_tab[1][2] = put_no_rnd_pixels8_y2_mmx2;
                 c->avg_pixels_tab[0][3] = avg_pixels16_xy2_mmx2;
                 c->avg_pixels_tab[1][3] = avg_pixels8_xy2_mmx2;
+#ifdef CONFIG_ENCODERS
                 c->vsad[0] = vsad16_mmx2;
+#endif //CONFIG_ENCODERS
             }

 #if 1

diff -Naur MPlayer-20040129.orig/libavcodec/Makefile MPlayer-20040129/libavcodec/Makefile
--- MPlayer-20040129.orig/libavcodec/Makefile	2004-02-01 13:56:42.000000000 +0100
+++ MPlayer-20040129/libavcodec/Makefile	2004-02-01 15:38:43.000000000 +0100
 <at>  <at>  -7,7 +7,7  <at>  <at> 

Re: Re: [PATCH] Donald Graft's kerndeint adaptive deinterlacing filter

On Sun, Feb 01, 2004 at 05:17:49PM -0300, Diego Biurrun wrote:
>  > +Paint pixels which exceed the threshold white (default 0).
>  > +.IPs order (0 or 1)
>  > +Swap fields if 1 (default 0).
> 
> Hmm, why does a flag get a value?  The other filters do it by name,
> i.e.
> 
>   -vf kerndeint=order
> 
> instead of
> 
>   -vf kerndeint=1
> 
> IMHO this is easier to use and it is also more consistent with the
> rest of the filters.

Huh? There is no filter that works like this!

>  > +.IPs thresh (0 - 255)
>  > +Threshold (default 10).
> 
> This appears to be the most important parameter and thus the one that
> will be set the most.  IMHO it should be the first parameter then, so
> that you can do things like
> 
>   -vf kerndeint=100
> 
> instead of
> 

Re: Re: [PATCH] Donald Graft's kerndeint adaptive deinterlacing filter

D Richard Felker III writes:
 > On Sun, Feb 01, 2004 at 05:17:49PM -0300, Diego Biurrun wrote:
 > >  > +Paint pixels which exceed the threshold white (default 0).
 > >  > +.IPs order (0 or 1)
 > >  > +Swap fields if 1 (default 0).
 > > 
 > > Hmm, why does a flag get a value?  The other filters do it by name,
 > > i.e.
 > > 
 > >   -vf kerndeint=order
 > > 
 > > instead of
 > > 
 > >   -vf kerndeint=1
 > > 
 > > IMHO this is easier to use and it is also more consistent with the
 > > rest of the filters.
 > 
 > Huh? There is no filter that works like this!

Well, hmm, yes, you are right, of course.  Dunno what I was thinking,
just ignore this comment.

 > > Since your filter does not support things like -vf kerndeint=help or
 > 
 > WTF? This is not allowed! Filters are not supposed to print help &
 > exit. G2 will have a proper way for filters to export help text.

It's called the new config layer (or so I believe):

Re: Re: [PATCH] Donald Graft's kerndeint adaptive deinterlacing filter

On Sun, Feb 01, 2004 at 07:32:53PM -0300, Diego Biurrun wrote:
>  > > Since your filter does not support things like -vf kerndeint=help or
>  > 
>  > WTF? This is not allowed! Filters are not supposed to print help &
>  > exit. G2 will have a proper way for filters to export help text.
> 
> It's called the new config layer (or so I believe):
> 
> 
> akira:/usr/src/mplayer/dev$ ./mplayer -vf scale=help
> MPlayer dev-CVS-031215-21:24-3.3.2 (C) 2000-2004 MPlayer Team
> 
> CPU: Intel Pentium MMX P55C (new) 233.8 MHz (Family: 5, Stepping: 1)
> Detected cache-line size is 32 bytes
> CPUflags:  MMX: 1 MMX2: 0 3DNow: 0 3DNow2: 0 SSE: 0 SSE2: 0
> Compiled for x86 CPU with extensions: MMX
> 
> Reading config file /usr/local/etc/mplayer/mplayer.conf: No such file or directory
> Reading config file /home/diego/.mplayer/config
> 
>  Name                 Type            Min        Max
> 
>  w                    Integer         -3         No        
>  h                    Integer         -3         No        
>  interlaced           Integer         0          1         
>  chr-drop             Integer         0          3         
>  param                Integer         0          100       
>  presize              Object presets  No         No        
> 
>