random and change - help please


    Nov 30 2006 | 10:45 am
    Hi, Tis is probably a total newbie question but i really need your help as i have tried all the ways i can think of.
    Here is my problem: I need to generate a random number betveen 0 and 3 on each bang to random and filter the numbers so that there are no immediate repetitions ever (so just using [deal] does not work). I'm trying to do that by connecting [random] and [change]. Change filters out the repetitions nicely but i need a way to send bang to random when [change] has filtered out the int random sent so the [random] would trie automatically again until a different int is generated and change lets it pass.
    Here is what i have now:
    thanks Taaniel

    • Nov 30 2006 | 11:03 am
      On 30 nov. 06, at 11:45, Taaniel wrote:
      > Hi, > Tis is probably a total newbie question but i really need your help > as i have tried all the ways i can think of. > > Here is my problem: > I need to generate a random number betveen 0 and 3 on each bang to > random and filter the numbers so that there are no immediate > repetitions ever (so just using [deal] does not work). I'm trying > to do that by connecting [random] and [change]. Change filters out > the repetitions nicely but i need a way to send bang to random when > [change] has filtered out the int random sent so the [random] would > trie automatically again until a different int is generated and > change lets it pass.
      Hi,
      There's an abstraction called "urn-jb" in the examples folder of Max. It uses the urn object which output random numbers without duplicates, and some "intelligence" to filter the possible repetition at the start of a new serie.
      HTH, ej
    • Nov 30 2006 | 11:20 am
      Like this ?
      cheers Roger
      --- Taaniel wrote:
      > > > Here is my problem: > I need to generate a random number betveen 0 and 3 > on each bang to random and filter the numbers so > that there are no immediate repetitions ever (so > just using [deal] does not work). I'm trying to do > that by connecting [random] and [change]. Change > filters out the repetitions nicely but i need a way > to send bang to random when [change] has filtered > out the int random sent so the [random] would trie > automatically again until a different int is > generated and change lets it pass. >
    • Nov 30 2006 | 1:11 pm
      an annotated solution without urn
      Roby
      On 30 nov. 06, at 11:45, Taaniel wrote: > Here is my problem: > I need to generate a random number betveen 0 and 3 on each bang to > random and filter the numbers so that there are no immediate > repetitions ever (so just using [deal] does not work). I'm trying > to do that by connecting [random] and [change]. Change filters out > the repetitions nicely but i need a way to send bang to random when > [change] has filtered out the int random sent so the [random] would > trie automatically again until a different int is generated and > change lets it pass. > > Here is what i have now: > >
    • Nov 30 2006 | 1:38 pm
      so much for your help !
      Taaniel
    • Nov 30 2006 | 6:34 pm
      On Nov 30, 2006, at 5:11 AM, Roby Steinmetzer wrote:
      >> I need to generate a random number betveen 0 and 3 on each bang to >> random and filter the numbers so that there are no immediate >> repetitions ever (so just using [deal] does not work).
      I think this works...if someone finds a bug please let me know...
    • Nov 30 2006 | 9:14 pm
      You can also check out "anti-bis" from RTC-lib by Karlheinz Essl.
      Best, Trond
      >>> /I need to generate a random number betveen 0 and 3 on each bang to >>> random and filter the numbers so that there are no immediate >>> repetitions ever (so just using [deal] does not work). / >>>
    • Dec 01 2006 | 7:31 am
      a simple arithmetic solution (uncommented) //yac
    • Dec 01 2006 | 8:30 am
      try this
      Lorenzo
    • Dec 01 2006 | 2:25 pm
      Taaniel wrote: > Here is my problem: I need to generate a random number betveen 0 and > 3 on each bang to random and filter the numbers so that there are no > immediate repetitions ever (so just using [deal] does not work).
      Yet another one:
      -- Stefan Tiedje------------x------- --_____-----------|-------------- --(_|_ ----|-----|-----()------- -- _|_)----|-----()-------------- ----------()--------www.ccmix.com
    • Dec 04 2006 | 5:51 pm
      Yacine Sebti wrote: > a simple arithmetic solution (uncommented)
      nice one... took a while to understand why it works... it's probably faster than any comparison for duplicates...
      You win the price for the most elegant solution. Could be a wonderful task for Max classes... It shows that there are a lot of very different kind of obvious solutions, but there might be still an even more elegant solution out there...
      Stefan
      -- Stefan Tiedje------------x------- --_____-----------|-------------- --(_|_ ----|-----|-----()------- -- _|_)----|-----()-------------- ----------()--------www.ccmix.com