|
From: Mark K. <mar...@co...> - 2004-01-08 03:57:57
|
On Wed, 2004-01-07 at 18:59, Vladimir Senkov wrote: > My dirty hack was not about playing release samples from the gig file. It > was actually about 3 things: > 1) Keeping playing whatever is to be played instead of killing the voice > 2) Gradually reduce the volume of that "to be killed" voice using some curve > 3) Eventually kill the voice when it's "would be effective volume" gets low > enough. > > So it is generating a release envelope, not playing one from the sample. Actually, that sort of thinking can be very valuable I think. Most gig files don't have release samples, so good strategies for handling them is a great thing to build up some understanding around. > It does (as far as i can tell) remove the clicks, but for "realistic" sound > you do need release samples from the gig file too as far as i understand. For the pianos, yes... > > I'm adjusting the effective volume from whatever it would otherwise be if i > just kept playing the note forever. > I'm them removing a % from that effective vol Yes, I can see that now with your explanation. Thanks. <SNIP> > It was late at night though so my hearing might not have been at it's best > :) My ears are old. I know their hearing is not the best. ;-) > > > Anyway, very good to have you here and I look very forward to trying > > out your patches. > > I'm currently not doing the right thing about sustain state of that voice > and something else so i'll probably have to research that before creating > any kind of a meaningful patch. > Even playing wit just your computed release envelopes will be interesting. |