|
 Kyma Forum
 Kyma Support
 Random based sounds & MIDIVoice
|
| next newest topic | next oldest topic |
| Author | Topic: Random based sounds & MIDIVoice |
|
Sylvain KEPLER Member |
posted 29 July 2005 10:04
 
Hi, I was creating a random values based sound today. I started by creating the single sound I imagined by putting together the different prototypes as usual, among which a lot of them are random generated values by the mean of the randExp Capytalk function and constant value prototypes... Once I was satisfied in a first stage (the "mono" sound was playing about exactly how I imagined) I then wished to make it polyphonic by placing the MIDIVoice at the end of the arborescence. My wish is to have independant random behaviors for each voice played. However, by the time I 'm writing the post, any voice I play has the exact same 'randomised' behavior (an identical result but playing at a different frequency) than another played at the same time. Is there any trick to make each voice playing independantly in terms of random generated values from the other ones ? Otherwise I'm afraid I should duplicate many times the same sound ( and make enter different settings for each) and mix the overall which would be then too much DSP consumming... Thanks for you input in this matter. [This message has been edited by Sylvain KEPLER (edited 29 July 2005).] IP: Logged |
|
keph Member |
posted 29 July 2005 10:42
The idea that comes to mind, but I haven't tested is: !KeyDown nextRandExpWithSeed: (?VoiceNumber) That way it provides a unique seed to the random generator each voice. However, I am not sure if the seed value can be scripted that way. I know you can with the Chaotic generator. IP: Logged |
|
Sylvain KEPLER Member |
posted 29 July 2005 11:52
BTW, it happens that even encapsulating a class out of a single sound, I end up with exact same random behaviors when I play many instances of it. I mean , for instance,that random events occure at the same time, whereas I wish I could hear different events at different times. So It seems that the way of reaching the goal is about having a good understanding about random numbers generated with Kyma. I would even end up by asking a simpler question : How could I succeed making True random numbers generators in Kyma... A couple of minutes after having wrote that post : I managed to divide the randExp Capytalk function by an interger controlled with a Fader which I turned as a Hotvalue. Encapsulated the Class and tried the result playing two instances of that Class. Hence I succeeded to control individually with two faders each of the random generated numbers...so each instance 'behaves' separately. Feel free to submit your opinion about this. Maybe SSC could give their opinion too ? (What I'm trying to make is something that sounds like a sample cloud) [This message has been edited by Sylvain KEPLER (edited 29 July 2005).] IP: Logged |
|
SSC Administrator |
posted 29 July 2005 13:17
Sylvain-- Keph's seeding idea does work (using ?VoiceNumber as the seed). Also, if the random expression depends in some way on a key event, it should create an independent random number generator for each voice. Please let us know the expression you are using and we can suggest a way to modify it to depend on a key event. If the random expression depends on a MIDI continuous controller, those are shared among all voices, so the expression expression would also be shared. IP: Logged |
|
Sylvain KEPLER Member |
posted 30 July 2005 08:40
Thank you keph and SSC. I 'll be trying this next week and will let you know how I get on. Indeed it should be successful. IP: Logged |
All times are CT (US) | next newest topic | next oldest topic |
  |
|
Contact Us | Symbolic Sound Home
This forum is provided solely for the support and edification of the customers of Symbolic Sound Corporation.
Ultimate Bulletin Board 5.45c