This seems like a great opportunity to calibrate memory usage between Giga users.
If you have loaded only the instruments that the Vota Utility requires (8 vowels instruments and 2 consonant instruments for each of the male and female choirs), I\'d like to know if you get a reading of 68% memory usage as I do.
It looks as if I\'m going to have to add another workstation now. The choir is very beautiful, and the Vota Utility makes it all the more desirable. With 1 Gig, I can just about load the strings, brass, and winds I use as well, but I believe when I start playing in earnest (and in real time), with layered choirs, strings, winds and brass, I\'m going to find the polyphony (at 160) not adequate. I\'m just finishing my midi setup which involves using several instances of MidiOx to help in routing issues. When that\'s done, I\'ll see what the story is with polyphony.
This makes me wonder, again, if Tascam is considering dropping the streaming from hard disk as the Itanium and Hammer chips make it into the market. Memory will no longer be an issue (other than a financial one.... and even the money to run an orchestra from RAM should be within reach in the forseeable future. Wouldn\'t this make dropouts, clicks, pops, etc a thing of the past, even with huge instruments? You\'d think it might make GS more stable as well. Anyone have any thoughts about this?
I\'d always render the choir to a wave file and work with that
If I need to adjust timing I can cut the words into different phrases and trigger them with a newly built GigaInstrument with only those.
Another thing I thought about was taht you could build instruments with the words sung at low expression only and then at high expression only, then build layered Gigainstruments to get EXP control after you build the words [img]images/icons/smile.gif[/img]
Quite a bit of a memory saver [img]images/icons/smile.gif[/img] ....since there\'s quite a bit less samples.
Maybe I\'ll build \"lite\" versions of the instruments stretched at 3rds or somethign, for realtime use with other instruments
Please expand a little on your thoughts. I think I grasp some of what you\'re saying, and it makes a lot of sense.
What I think you might be driving at seems to relate to the somewhat non-understandability of the words a choir sings. You can tell they are words, but often you can\'t really tell what they are saying. Given that, i suppose you could create a limited number of words in the utility... then somehow figure out which samples are not being used and do a limited save of just the ones that _are_ being used.
On the other hand, maybe you could say more about the idea of extracting the wavs. How does that aid in saving memory.. or are you saying you\'d record words and make them into wavs to be used as individual samples... maybe with regions wider than chromatic regions. I sense something here I don\'t fully understand, but want to learn more about.