2 (probably pedestrian) questions about random samples which will be granularly synthesized


    Mar 23 2006 | 4:18 pm
    I am working on a patch which randomly (using a metro and a urn-jb) chooses
    a different sample out of 10 every so often which will be granularly
    synthesized. I am using the play~ object. I am trying to automate
    crossfades between the currently playing sample and the just randomly
    selected sample (as soon as it is chosen), but I am getting really
    frustrutrated with this right now. If anyone could give me some
    suggestions, that'd be great. Also, I'd like to give the user 10 toggle
    boxes, which would opt certain of the sample in and out of the random
    process, is there any easy way to do this? (so for example, only 1, 4, 8,
    would be randomally chosen). I can send my patch if need be.
    Thanks in advance.

    • Mar 23 2006 | 9:51 pm
      On 23-Mar-2006, at 17:18, baby.squirrel wrote:
      > Also, I'd like to give the user 10 toggle boxes, which would opt
      > certain of the sample in and out of the random process, is there
      > any easy way to do this? (so for example, only 1, 4, 8, would be
      > randomally chosen).
      Like this?
      Requires lp.ernie from Litter Pro. Ernie has lots of other
      enhancements over the standard urn object.
      Sorry about the alpha channel in the image, I know it's a drag for
      some mail clients, but I can't readily change it on this machine.
      Forum readers are also attachment-disadvantaged.-(
      Best -- Peter
      -------------- http://www.bek.no/~pcastine/Litter/ -------------
      Peter Castine | +--> Litter Power & Litter Bundle for Jitter
      |....................................................
      p@castine.de | iCE: Sequencing, Recording, and Interface Building
      pcastine@gmx.net | for Max/MSP
      pcastine@bek.no | http://www.dspaudio.com/ Extremely cool
      4-15@kagi.com |....................................................
      | home|chez nous|wir|i nostri http://www.castine.de/
    • Mar 23 2006 | 11:59 pm
      This is what I want to do, but is there any way to do this without buying
      litter pro?
    • Mar 24 2006 | 2:58 pm
      this is a less elegant solution but i think it does what you need. it uses the radio group instead of toggles
      max v2;
    • Mar 24 2006 | 3:06 pm
      additionally, for your crossfades i think you could use info~ and use the file length to create a time function for your random selection. ie, the crossfades could be a percentage of your file length,. this could also be used to calculate the the time for the next random sample choice, if you get my drift...
    • Mar 24 2006 | 3:29 pm
      I understand the patch example you posted, but I don't have the listfunnel object! is there anything substitutable, i have 4.3 on windows, bear in mind (plus a lot of free externals). Also I'm not really sure what you mean about using info~ as I'm somewhat of a beginner at Max, could you post an example?
    • Mar 24 2006 | 3:32 pm
      Also, I don't think the percentage of the file length would not adequate for crossfading, since considering I'm going to use these files for granular synthesis, only a small bit of each sample will be played anyways (using the line~ object, etc.).
    • Mar 24 2006 | 3:46 pm
      ah, ok. listfunnel is a 4.5 external, but you can use this workaround without much trouble:
      max v2;
      also, the efficiency of that info~ method may be curtailed by the the size of file chunk you are using. how small are these chunks typically?
    • Mar 24 2006 | 4:01 pm
      probably something like 100 ms, maybe a little more, maybe a little less...
    • Mar 24 2006 | 4:57 pm
      perhaps you could post/send me this patch. it would be interesting to see exactly what you are trying to implement regarding the crossfade issues.
      k
    • Mar 24 2006 | 5:37 pm
      I don't have my patch offhand, but here's an implementation of the idea you suggested re: the radio buttons, etc. i did have an error when changing the randomness from the sampleness to randomness of buffers as you'll see in the patch. thanks a lot, you're very much help.
      max v2;
    • Mar 24 2006 | 6:24 pm
      ok. a few little fixes to this. it works for me. haven't tackled the xfades as you will see:
      max v2;
    • Mar 24 2006 | 8:34 pm
      xfade version. not sure if it's quite what yr after:
      max v2;
    • Mar 25 2006 | 2:53 am
      thank you very much for the thorough suggestions, i'm sure this will help
      out.