Learn how easy it is to sync an existing GitHub or Google Code repo to a SourceForge project! See Demo

Close

#9 Clean up ao driver handling

closed-fixed
Joe Drew
2
2011-10-01
2002-02-18
Idcmp
No

mpg321 should not attempt to validate the argument sent
to --output, nor should attempt to assume anything
about the -a argument.

Check out ogg123 on how to implement libao nicely.
Pass the string to ao_driver_id and let ao figure it out.

Discussion

  • Idcmp
    Idcmp
    2002-02-18

    • priority: 5 --> 7
     
  • Joe Drew
    Joe Drew
    2002-02-18

    • priority: 7 --> 2
     
  • Joe Drew
    Joe Drew
    2002-02-18

    • summary: Bad use of libao. --> Clean up ao driver handling
     
  • Joe Drew
    Joe Drew
    2002-02-18

    Logged In: YES
    user_id=330927

    The -o drivername option no longer tries to validate what's
    passed to it (this was finished over the weekend, and is
    available in CVS).

    However, the -a device option cannot be generalized, because
    libao has no generalized interface to set the output device,
    and every driver sets it differently.

    This bug remains open because the driver code still needs to
    be cleaned up to remove most of the MPG321_USE_* options -
    maybe, anyways.

     
  • Idcmp
    Idcmp
    2002-02-20

    Logged In: YES
    user_id=2329

    I tried the CVS version with alsa09 and it seems to
    consistently hiccup, almost once a frame it sounds like.

    Using alsa 0.9's OSS emulation and -o oss it's fine.

     
  • Joe Drew
    Joe Drew
    2002-02-20

    Logged In: YES
    user_id=330927

    See http://bugs.xiph.org/show_bug.cgi?id=153 . This is a
    known problem; I'll see if I can get Xiphophorous to look at it.

     
  • Idcmp
    Idcmp
    2002-02-20

    Logged In: YES
    user_id=2329

    ogg123 on the same machine, same libao doesn't have this
    problem though - although the description is the same.

    The alsa-0.9 plugin for XMMS also doesn't have this problem.

     
    • status: open --> closed-fixed