channel isolated wav output can be out-of-sync

[pclalv]

game: Robopon: Sun Version track: 10

I'm using gbsplay to generate channel isolated wav output (one wav file containing only channel 1, one wav file containing only channel 2, etc.) of this track. the pulse 0 and wav channels of this track both begin with silence, so I've been passing -T 10 to gbsplay, like so:

gbs="/path/to/Robopon - Sun Version (1998)(Red, Hudson Soft).gbs"

./gbsplay -T10 -owav "$gbs" 10 10
mv ./gbsplay-10.wav robopon-10.wav

./gbsplay -T10 -234 -owav "$gbs" 10 10
mv ./gbsplay-10.wav robopon-10-1.wav

./gbsplay -T10 -134 -owav "$gbs" 10 10
mv ./gbsplay-10.wav robopon-10-2.wav

./gbsplay -T10 -124 -owav "$gbs" 10 10
mv ./gbsplay-10.wav robopon-10-3.wav

./gbsplay -T10 -123 -owav "$gbs" 10 10
mv ./gbsplay-10.wav robopon-10-4.wav

my ears say that channels 1, 2, and 4 are all in sync with one another (and with the full mix wav), but that channel 3 is out of sync.

one way that you might be able to hear this discrepancy is by:

1. loading all five wav files into a program like Audicity;
2. panning the full mix hard left and the separated channels hard right; and
3. muting and unmuting channels to hear what sounds in sync and what doesn't.

it might be challenging for a listener unfamiliar with this song to hear the discrepancy, as the unfamiliar listener may hear the discrepancy as an instance of the track employing a delayed note effect.

[pclalv]

uh, nevermind. Audacity seems to correctly interpret the channel 3 wav file, whereas "smarter" software screws it up.