On Sunday 12 October 2003 7:48 pm, Guillaume Laurent wrote: > Looking at updateClocks() it shouldn't be too hard to diminish the > number of DCOP setPointerPosition calls sent to the GUI. Another > possibility would be to make the GUI process those calls faster by > just updating an internal int and update the position pointer from > that only when refreshing the window, thus collapsing several > consecutive setPointerPosition calls when possible.
My reply to Chris still stands - where's the proof? Nothing has changed with updateClocks per se so I'm sure it's a red herring unless of course something has been bolted on to the updateClocks at the gui side (at say setPositionPointer in RosegardenGUI or further downstream). Changing how this call is made would break a lot of gui stuff that already depends on it. updateClocks being the heartbeat of the gui during playback. Saying all that - using cachegrind was inconclusive the other day - I certainly couldn't make sense of the results anyway. Perhaps someone else wants to try and characterise this? Certainly someone needs to come up with something better than this IMHO. R ------------------------------------------------------- This SF.net email is sponsored by: SF.net Giveback Program. SourceForge.net hosts over 70,000 Open Source Projects. See the people who have HELPED US provide better services: Click here: http://sourceforge.net/supporters.php _______________________________________________ Rosegarden-devel mailing list [EMAIL PROTECTED] - use the link below to unsubscribe https://lists.sourceforge.net/lists/listinfo/rosegarden-devel
