[delay] what are the rules / basic principles?


    Mar 31 2013 | 2:44 pm
    Hi all,
    Using a [1ms] delay; how reasonable is that? What can i use it for?
    Example, When using [sfrecord~] you have to;
    1. set/create a filename to save to
    2. set the record toggle ON
    3. feed the signal you want to record
    Now assuming i am automating all of that what are the reasonable delay values i should setup?
    -----------| > createFile/name >
    -BANG-| > [delay 100?] > set toggle ON (record)
    -----------| > [delay 101?] > startloop (play)
    I guess it matters on the machine processor? system load? what should be reasonable minimum values?
    Also in this SFrecord example i am Normalizing a file and applying a negative gain right after.(using mxj buf.Op)
    Is [1ms] delay between normalization and gain reduction sufficient?
    Is it a matter of, aslong as the "normalize" message is received, and the "gain" message follows, both commands will be executed regardless of the processor hangup/system load?
    Thanks!
    Happy Easter!

    • Mar 31 2013 | 3:25 pm
      Using deferlow is more appropriate in most of the cases you mention, I believe.
    • Mar 31 2013 | 3:55 pm
      Hey I didn't know about deferlow, sounds very handy!
      I guess you can also check the trigger [t ] object and [pipe] for more flexibility.
    • Apr 01 2013 | 5:21 am
      errm ... in most cases the right order of messages is all you need.
      sending "write sound.aif, start" as two consecutive messages to sfplay~will it make write the file, and queue the start message internally until the file creation is finished.
      -110
    • Apr 01 2013 | 1:38 pm
      Yes, the order of messages is so important, i had all forgotten about it. (still noobing)
      Thanks for the [deferlow], cool object.