Mercurial > code > home > repos > light9
diff show/dance2010/readme @ 521:fd4f06a93c95
start 2010 show
Ignore-this: e50fb61985226922b124ce1f50c73fee
author | drewp@bigasterisk.com |
---|---|
date | Tue, 08 Jun 2010 06:48:25 +0000 |
parents | |
children | ab2ab3a54c10 |
line wrap: on
line diff
--- /dev/null Thu Jan 01 00:00:00 1970 +0000 +++ b/show/dance2010/readme Tue Jun 08 06:48:25 2010 +0000 @@ -0,0 +1,30 @@ +for local testing on plus: + +sudo -u mpd -s +export LD_LIBRARY_PATH=/home/drewp/score +./mpd --no-daemon + +fix networking.py to look for player on plus + +PYTHONPATH=../pympd bin/ascoltami --show http://light9.bigasterisk.com/show/dance2010 + +plus(pts/3):~/projects/light9% bin/dmxserver -n + +vidref +grab time from asco +always record picture against {songuri}/vid/{take}/{songtime}, so we never miss a recording (but at night when they're all gone, we don't need any recordings?) +play prev videos +let me tag a good rehearsal or toss junk rehearsals. if we juggle the playback time too much, you can be sure it's not a good pass +qt window with one live pane and any number of synced playback panes. +Do we need to detach from current song+time to view something else? +curvecalc should be able to fetch a sample of a lit frame to stick in its timeline +need to move curvecalc to qt? +can i dynamically change the output filename of a filesink? that might be the way to steer the ouptut correctly. But, i might want to append one take's frames into one file. Maybe use a standard compressor like mjpeg, and separately map the in-movie timestamp to our playback timestamp in case they drift. +new take for every single restart? i guess so, since they could pass over the same song time. +get twisted qt wrapper for our networking + +show is 70min of music: at 8fps and 50k/pic, 1.6MB of image data per pass. + + +todo: +- check if mpd has a working precise-time system yet, so we could get off the patched one