Accum


    Feb 22 2007 | 11:31 pm
    Does anyone use accum? One of my students found it and used in his first patch. It doesn't seem to do what I expected and the help and manual don't help or man...Is it me or is funky?
    Cheers Gary Lee Nelson Oberlin College www.timara.oberlin.edu/GaryLeeNelson

    • Feb 22 2007 | 11:49 pm
      it's a pretty basic object. it's cleaner than using a bunch of [int] and [+] and [*] objects.
      try opening the help patch on it. send a value into the left inlet, and that will set (and output) the "initial value" stored in the accum object. then, send a value into the middle inlet. that value will be added to the "initial value", and then that sum will be stored as the "current value". however, nothing is output until it receives a bang in its left inlet. similarly, values sent into the right inlet will multiply the current value.
      if your confusion comes from the fact that nothing is output when the object's value changes, then try using a [t b i] object to trigger a bang to the left inlet whenever a value is sent to the middle or right inlets.
      -scott
    • Feb 22 2007 | 11:59 pm
      I've used it a few times. It's like a very odd counter. The trick is you have to always send a value to the center inlet (the amount you want to add to the "accumulation", then bang the left inlet to get the result. e.g.
      David
      On Feb 22, 2007, at 3:31 PM, Gary Lee Nelson wrote:
      > Does anyone use accum? One of my students found it and used in his > first > patch. It doesn't seem to do what I expected and the help and > manual don't > help or man...Is it me or is funky? > > Cheers > Gary Lee Nelson > Oberlin College > www.timara.oberlin.edu/GaryLeeNelson > >
    • Feb 23 2007 | 12:02 am
      Ah, then it IS funky! The manual should say in big red letters that, unlike all other max objects inputs to the right must happen before action at the left-most inlet causes output. I learn once again that blind faith is dangerous. :-)
      On 2/22/07 6:49 PM, "swieser1" wrote:
      > > it's a pretty basic object. it's cleaner than using a bunch of [int] and [+] > and [*] objects. > > try opening the help patch on it. send a value into the left inlet, and that > will set (and output) the "initial value" stored in the accum object. then, > send a value into the middle inlet. that value will be added to the "initial > value", and then that sum will be stored as the "current value". however, > nothing is output until it receives a bang in its left inlet. similarly, > values sent into the right inlet will multiply the current value. > > if your confusion comes from the fact that nothing is output when the object's > value changes, then try using a [t b i] object to trigger a bang to the left > inlet whenever a value is sent to the middle or right inlets. > > -scott
      Cheers Gary Lee Nelson Oberlin College www.timara.oberlin.edu/GaryLeeNelson
    • Apr 01 2007 | 6:21 pm
      gimmee back my patchcords you thiefs !!!
      OK OK there must have been some copypaste mistake from me ;-)
      Thanks for your answers anyway.
      Here is the real thing (makes more sense) :
    • Apr 01 2007 | 7:59 pm
      At 8:21 PM +0200 4/1/07, raphael seguin wrote: >gimmee back my patchcords you thiefs !!!
      OK, that _is_ strange. It seems like accum, after a certain point, refuses to acknowledge any number less than 16. (in your example it comes after processing step 31)
      Here's your patch set up for single stepping with a bunch of debugging stuff scattered around. This version runs until the accum number and the + number are different. You notice that once it breaks (at step 31) accum is a little wacky. You can bang in numbers < 17 with no output, but numbers >= 17 cause output, as normal.
      -- Chris Muir | "There are many futures and only one status quo. cbm@well.com | This is why conservatives mostly agree, http://www.xfade.com | and radicals always argue." - Brian Eno
    • Apr 01 2007 | 8:58 pm
    • Apr 01 2007 | 9:50 pm
      I use accum all the time...