Physical Modeling: Guitar, Piano, Bass


    Apr 06 2014 | 9:36 pm
    Hey everyone!
    I am in a beginner max/msp class. Our final project is to do physical modeling of guitar, piano, and bass. My partner and I have been working on this constantly for the last week and a half, and have gotten nowhere. We were looking at using biquads~ to make filters, but our class never covered it, and it seems like there is an easier way to go. My main questions are:
    Why is it when we remove the biquad, no signal is passed through at all. Biquad is just a filter, right?
    The biquads appear to be controlling most of how the timbre of the guitar is formed, so is there anywhere that talks about piano and bass parameters for a biquad, or what is another way of forming these timbres? I feel that the filtergraph~ and amplitude envelope don't have as much of an impact on the sound.
    Once in a while, our sound stops working even when we won't change anything about the patch. Anyone see a solution?
    Check out our patch, and any help would be greatly appreciated! Thanks!

    • Apr 07 2014 | 6:35 am
      edit
      server crashed
    • Apr 07 2014 | 6:40 am
      Hi
      your feedback algorithm was incorrect, causing infinite feedback - and the [dac~] was shutting itself down.
      I also think [biquad~] is a little heavy-handed for string modelling. You're tutor may disagree but I would favour [onepole~] for the string and several [reson~]'s for the soundboard.
      J Acous Soc Am
      106/5 1999
      T.D.Rossing G.Eban
      Normal Modes of a radially braced guitar determined by electronic TV holography
      And the science:
      donald e hall
      musical acoustics (2nd ed.)
      California:Brooks/Cole 1991
      Sorted.
      Brendan
    • Apr 07 2014 | 6:46 am
      ps
      Just type "Karplus Strong" or "physical modelling" into the search box up there. Endless resources.
    • Apr 07 2014 | 1:02 pm
      Thank you so much for your help!