[MPlayer-users] Bugs, questions

Bogdán Zaválnij bogdan at renyi.hu
Sat Mar 23 22:05:02 CET 2002

> I rather like to read too much than trying to guess what system/error
> a user has. Beside, if it's really not needed i can skipp it :)
OK, you know it.. ;-)

> > the make error again:
> > 
> > libmpdemux/libmpdemux.a(demux_ogg.o): In function `demux_ogg_add_packet':
> > demux_ogg.o(.text+0x9b): undefined reference to `vorbis_packet_blocksize'
> > collect2: ld returned 1 exit status
> > make: *** [mplayer] Error 1
> ok, check your installation of vorbis (looks like something is missing)
> man ldconfig
> man ld.so.conf
> tell ./configure where libvorbis and libogg is
well, I doubt the problem is there, since the CVS 20020315 compileing
perfectly.. I allready got a hint, to update the vorbis, but even if it
would help, I think it is a problem:

because there is no hint in the documentation, that I _should_ use the
newest one, as it is written about gcc or binutils!!!

(I just compiled libogg and libvobris 1.0.rev3, and after overwriting the
old ones the mplayer compiled OK.) (by the way: my other problem was, that
simply compile-ing and installing with default setups the ogg and vorbis
the mplayer still did not compiled well, because the old lib files was in
/usr/lib, and the new got in /usr/local/lib.. many people may have
problems with all this.. :-( I sholud've reinstall obb and vorbis with
--prefix=/usr ... rpm -e did not worked because of the dependencies..
and vorbis rev3 is not available in rpm..)

I don't know if other systems have these problems. once again I had vorbis
files 1.0beta4.


> You are using sdl with X11 drivers, that's afaik the slowest possible
> driver you can use. Even -vo x11 is faster. Try to get Xvideo support
> for your card (dunno whether it's possible) or get a better graphic
> card. Beside you have a system that is really at the lower end (so you need
> to have a good card)
OK, thats right. (althought I heard that the mplayer is the best.. ;-))

> > My additional question was, why is the sound decoding using 50-60% of cpu
> > time, while mpg123 uses only 4-6% while playing mp3 files.. (it is
> > perhaps some sort of lame question, but there is still no word about it in
> > the DOCS.)
> Hmm.. should not...
playing without framedrop:
A: 172.8 V: 171.8 A-V:  1.071 ct:  0.200  4295/4295  43% 94%  7.3% 42 0 0%

and with framedrop allowed:
A: 175.6 V: 175.6 A-V: -0.008 ct:  0.258  4391/4391  19% 42%  4.5% 79 0 0%

quite strange, that the sound processing time is decreased by the
"-framedrop" option.. and ther is still the problem about the mp3
decoding.. (or I misunderstood the numbers above, althought I read the
faq.html for a couple of times to be sure..)

> Dunno what's wrong here... never used nas
> (btw: it's not a good idea to put to much between mplayer
> and the sound card as the delay get's unpredictable)
You're right for a moment, but on the other hand I sholud disagree with

if I have two computers (the second is a 486..), and the "main" one is not
able to handle the sound perfectly (for example because of some
sound card driver problems), than it seems to be a perfect solution, to
deal with the actual sound processing on an other computer, and give the
mplayer a perfectly clean and specified api layer to play the sound
throu.. am I wrong?


More information about the MPlayer-users mailing list