It'd be nice if the analyser accepted stereophonic sound input. It could display average or separate graphs fro each channel (for example using different colors overlaid).