@mdvries
Thanks for the attempt
. What you need to do is connect to the machine from another PC via SSH and run xbmc from there so you can issue commands to the debugger, like so
Code:
ssh user@host
startx &
export DISPLAY=:0
export AE_ENGINE=SOFT
gdb /path/to/xbmc.bin
As for the AE_ENGINE question, no, there will not be for quite some time due to the nature of how the audio code works, it needs to be selected during start-up. You can however uninstall pulse, or add "export AE_ENGINE=SOFT" to /home/user/.profile so it is always set.
@DeViLRuNNeR
Down-mixing reduces volume, this is why...
Say we have two channels we wish to mix into one... that means, two audio samples. All audio resolutions have a minimum and maximum range, for example, float's range is -1.0 to 1.0, where 0.0 is silent, anything beyond this range is distortion, the audio card cant render it without possibly damaging your speakers by clipping the audio at its peaks.
When we mix two samples together the math is very simple, it does include RMS power calculations, but for this example we will keep it simple.
output = (sample1 + sample2) / 2
So lets say sample1 = 1.0, and sample2 = 1.0, this will give us a final result of 1.0. The division by 2 is required to keep the result inside the range of the data format, but it also has the effect of attenuating the output. There is no way to overcome this, it is a hardware limitation. The boost option in XBMC is a hack, it just removes the divisor, so the final result in this example will be 2.0, which depending on your sound card, may be clipped, cause distortion, etc.
The only safe way to increase the volume is to perform DRC (Dynamic Range Compression) which muddies the quality of the signal and many people don't like it. But even so, it is a planned feature for later.