I\'ve been stuck for weeks now trying to create a real-time playable malleable orchestral GS performance.
So many choices,so many possible combinations.... muted strings, hard attacks, legato, staccato, detache, pizzicato, small ensembles, full sections, solo instruments,
ditto for winds and brass.
I\'ve already created performances that have several different instruments on different channels and ports, and use a several continuous controllers to crossfade them. I\'m doing this mainly for variations of intensity such as sordini, vs \"gentle\" strings, vs thunderous ones etc. This works well.
Now I\'d like to add in variations of articulation, such as detache, staccato, pizzicato, and effects such as tremelo and trills.
Setting up keyswitched instruments is not that difficult, but it restricts my choices. Once the instrument is built, it will always switch between the instruments that have been programmed in.
On the other hand, my controller can send out (as a batch) any number of program changes on arbitrary channels and ports. This would allow me to send out a batch of program changes which would change the instrument loaded into some or all of the channels, thus giving me a new instrument. I could easily audition and develop new instruments merely by changing the program numbers that are sent out.
Is there a downside to using Giga this way? The upside is that I don\'t have to spend time to creating the keyswitching, and I have the possibility of many more combinations at my fingertips.
Re: Keyswitching vs Program Changes (any advantage)?
In my experience, the downside is that program changes are slower. Keyswitching also uses more memory, but it\'s virtually instant. I believe that many folks here handle this by hocketing between channels in GS, so one instrumental part may span 6 or 8 channels, each with a different articulation. Personally, I like keyswitching, but you may need to edit whole libraries so they all agree (legato C1, stacatto C#1, sFz D1, etc...)