I've discovered this while debugging: if you're blending between two sequences that are the exact same pose, you will not attain that pose unless at least one of the two blendfactors is at one ( feel free to try that in hlmv ).
Can anyone tell me how the blending works, and if it's possible to have it working the way I'd expect ( e.g. 0.5 * sequence a + 0.5 * a sequence identical to a = a. ). Any insight in the animation blending process is appreciated. Maarten > The bottom one is probably at 0.6, but to eliminate all doubt: > > http://i336.photobucket.com/albums/n330/jfkeats/ol_testmap0041.jpg > > idle_, walk_ and lower_ are identical, the sum of their weights is 1, and > still the blend is not 'complete', there's a part of jesus in there which > causes the gap. > >> It doesn't look like the two sliders are both on 0.5: the top one is >> further to the left than the bottom one. >> >> Maarten De Meyer wrote: >>> Hi list! >>> >>> I'm trying to understand a blending problem. Simple question: >>> >>> http://i336.photobucket.com/albums/n330/jfkeats/problem.jpg >>> >>> In this setup, run_lower only affects the legs ( so the upper pose is >>> Jesus, weightlists are at 0 for upper body ). run_upper_minigun and >>> walk_upper_minigun are both a simple, correct pose ( correct being with >>> the hand on the front handle ). >>> >>> Why is that, if they're both blended with weight 0.50, and the >>> run_lower >>> should have a weight of 0 for the upper body, I still have a >>> gap/incorrect >>> pose ( caused by some of the 'Jesus' pose still being blended in )? If >>> the >>> sum of their weights is 1.0, shouldn't the pose be exact? Ingame this >>> translates in glitches when we start/stop running, other than that it's >>> fine. >>> >>> Thanks for any insight, >>> >>> Maarten >>> >>> >>> _______________________________________________ >>> To unsubscribe, edit your list preferences, or view the list archives, >>> please visit: >>> http://list.valvesoftware.com/mailman/listinfo/hlcoders >>> >>> >>> >> >> >> _______________________________________________ >> To unsubscribe, edit your list preferences, or view the list archives, >> please visit: >> http://list.valvesoftware.com/mailman/listinfo/hlcoders >> >> >> > > > > _______________________________________________ > To unsubscribe, edit your list preferences, or view the list archives, > please visit: > http://list.valvesoftware.com/mailman/listinfo/hlcoders > > > _______________________________________________ To unsubscribe, edit your list preferences, or view the list archives, please visit: http://list.valvesoftware.com/mailman/listinfo/hlcoders
