Could someone tell me if there\'s an advantage in terms of the resources Gigastudio needs in order to structure a big orchestral instrument in the following two ways:
1) I can build a layered instrument in the normal fashion, assigning a controller to crossfade between instruments.
2) I can do the layering by assigning different instruments to different channels, and then send all midi messages out multiple channels, to initiate note-on/note-off info as well as all continuous controller info. In this case I can get much more complex control of the mix because I can fade channels independently by using unique pots. I can also then change patches by sending program numbers independently to the different channels. It\'s also much easier to experiment with different combinations of instruments this way, since I don\'t have to go through the process of merging and combining instruments.
I\'m wondering, however, if there is somehow a greater overhead to using this method. Of course, I\'m sending much more midi info, but I\'m not having trouble with clogging up the midi. I\'m wondering more about what Gigastudio is doing in both of the above cases, and if one is more efficient than the other.
I\'m playing live using GigaStudio and several libraries,including KH Strings, Vota, QL 56 Strat, and some of my own samples. I use a woodwind controller instead of a keyboard
the only downsides I could think of would be more MIDI info in the second option
and possibly less accuracy in the sample start times. I believe in a layered instrument the sample start times may be close to sample accurate. But accross several channels, in a performance with alot of MIDI data, the timing will probably be less accurate.
In most orchestral contexts, this shouldn\'t be an issue. I\'ts when you\'re using the Xfades in conjunction with some techniques that you dont want to introduce phasing into this *could* be a prob. You\'d most likely have to be hitting some high MIDI data throughput and high poly for this to really happen, tho I\'ve never really tested it.
Thanks, King, for the thoughtful reply. I hadn\'t thought of the phasing, but seem not to be having a problem with it. I expect I\'d see more of this if I loaded identical instruments on more than one channel, but that doesn\'t seem to make much sense anyway, so I don\'t do it.