FYI, there is a new version of pythm available on opkg.org 2009-02-26 - Version 0.5.4-dmr: ------------------------------- * Fixed many issues with skip and previous. They should behave as expected now. * Removed dependence on pyalsaaudio. Volume changing is handled via the back-ends once again. * Reduced CPU overhead of gstreamer by using a custom pipeline instead of playbin.
There is not to much more performance I can squeeze out of the program based on its architecture, I think. The python code itself does very little and hands off all the heavy lifting to gstreamer and gtk. (Speaking of GTK, it takes 5% CPU, as reported from top, to simply redraw the track time in the GUI). Gstreamer and alsa are not the most efficient ways to get sound to come out of a device, I believe. However, my goal is to build upon "sane" components thus leaving only doing the work that needs to be done. -- Dylan Maxwell Reilly _______________________________________________ Openmoko community mailing list [email protected] http://lists.openmoko.org/mailman/listinfo/community
