mpg123-users — Mailing list for users of mpg123

[Bob A. <bo...@di...>]

I have been trying to find a solution for the audio defect that I have 
experiencing with mpg123 0.59r. I am trying to implement the open source 
asterisk pbx on a Solaris system and I am confronted with this problem 
with no apparent alternative but to use mpg123. Yes, I am aware the 
mpg321 is a potential alternative but it relies on libao and it will not 
build on a Solaris box.

The problem with mpg123 based on all of my testing appears to be a fault 
in the decoding logic that is causing bit errors to varying degrees. I 
suspect that the problem is somehow related to running on a sparc 
processor since I would otherwise expect that if this problem occured 
when running on intel processors it would have been fixed a long time ago.

Description of the problem

While playing any mp3 file with mpg123 I hear the music with a fairly 
regular tick, tick, tick sound occuring about 2-3 times per second. The 
severity of the tick sounds vary from minor to severe where they almost 
mask the music for a split second. I have verified that my audio 
hardware is ok. I regularly use mxaudio (xaudio) to play the same mp3 
files and they sound perfect.

I don't believe the issue is related to the audio output hardware in my 
computer. The reason I am now confronted with this is that I have been 
testing asterisk (the open source voip pbx) which relies on mpg123 to 
play music on hold and other sounds. The mp3 audio is converted to an 
rtp bit stream and sent directly to the phone bypassing any local sound 
hardware. The same ticking noises are quite noticable as well in this 
instance even though the audio stream is being reduced to an 8KHz sample 
rate.

Build information is as follows:

make solaris-gcc

SunOS arrow 5.8 Generic_117350-05 sun4u sparc SUNW,UltraSPARC-IIi-Engine
Using built-in specs.
Target: sparc-sun-solaris2.8
Configured with: ../configure --prefix=/opt
Thread model: posix
gcc version 4.0.2

FYI, I had previously compiled with gcc 2.9.5 and had the same problem 
so it does not appear to be compiler related.

I sure wish someone could take a look at this. I will be glad to compile 
and test any potential fixes.

Thanks,
Bob