How do I flam a trigger to multiple inputs?
  • I am trying to make a clone of Verbos Electronics Random Sampling module. I would include the patch but apparently it has bloated to over 50 mb, so I will merely include a screenshot. But I am trying to make an variable length array of random values where you
    548 x 768 - 208K
    Screen Shot 2016-11-18 at 10.00.18 AM.png
    860 x 796 - 109K
  • It's tough to know how to help without seeing the patch first - you have a computer version, which means you can delete the history. Just go to edit -> clear history and then save. If you made this on your iPad and can't get it off, sync with iCloud or Google Drive (since you can't email it at that size).
  • geez i clicked post in error. to continue...

    where you ping the module and it grabs you one of the 2-64 values one at random. The problem i am having is that when I ping the sample and holds, they seem to be grab the same number from the white noise generator. So I recalled a youtube video where Mark Verbos said he had to flam the triggers for his analog unit to prevent the same thing from happening. I have no idea how I would go about this in Audulus,
    Screen Shot 2016-11-18 at 10.02.53 AM.png
    902 x 752 - 117K
  • @biminiroad LOL you replied before i even finished.
  • Ok, I deleted the history and now the patch is 30mb. I think that means I'm just making inefficient patches, but for now you can see what I am talking about.
  • Sounds like you need to change the seed - just change it from 0 to 1, 2, 3, 4, etc on however many random nodes you have. Tap/click on it and select "set seed."

    Digital randomness is algorithmically generated, which means you have to start with a number. If you start with the same number, the algorithm will spit out the same string of numbers no matter how many instances you run of it. You might not have noticed this when you were first building it because when you create a noise node and then shortly after create another, although they are spitting out the same string of numbers, they are out of sync. Exit and re-enter the patch, however, and they'll reset and be in sync again.

    You don't need to set the seed to a "random" number. 0, 1, 2, 3, 4, ...etc. will be just as random as 1249346, 981765, 3681010, 478257 for seeds.
  • Go inside the RHS Stored Random things and replace them with a simple random node (less CPU usage, smaller file), and then set the seeds differently like this for each one.

    (It's backwards as I uploaded the screen caps, go right to left)
    Screen Shot 2016-11-18 at 1.23.12 PM.png
    755 x 420 - 52K
    Screen Shot 2016-11-18 at 1.23.06 PM.png
    801 x 523 - 62K
    Screen Shot 2016-11-18 at 1.22.58 PM.png
    746 x 473 - 53K
  • Thanks for the tip about the random node vs the white noise.

    If I cut and paste that bank of stored randoms, I will get 8 banks of 8 identical randoms. Since random(x) = the random(x) elsewhere in the patch, right? That's why I was looking slightly delay the one trigger signal sent to the bank of banks of random numbers. Plus I would like to learn how to flam because then I can say flam more, which makes life more fun. In conclusion, I will rewrite the patch from scratch anyways but I would still like to know how I would flam.
  • A flam is just a gate going through a delay that has 1 repeat. All you'd need to do to make a flam is to send a gate through a delay node - you can even adjust gate height with the mix knob so you can emphasize the primary or grace note.

    This doesnt seem like what you're doing though. Are you trying to make a burst of notes that have a semi-random quality to them? Just describe the end result a little more for me.
  • The Quantized Randoms module from Buchla and Verbos allow you to grab a set of random numbers then take one of those numbers at random and use it for modulation. EG I want three random numbers to go to a filter cutoff or whatever, I dial in the n knob to 2 and take the n+1 output and attach it to the cutoff. I can ping the module whenever and it goes to one of those three random number with 33.3% probability. I can ping the other input and get three new random numbers. I would like to do this with 2-64 random values, like the Verbos unit does. Am I making sense?

    The way that Verbos described his unit as working was that he pinged the various s&h circuits in rapid succession, which he referred to as flamming, even though it was going to like 64 places. He had to do this because he uses a digital noise source with pseudo random output. The pinging one after another isn't to create bursts of notes but rapidly create new sets of random numbers.

    I guess if you can think of a simpler way to achieve the same end effect I would love to know.
  • Oh yeah this is SUPER easy I'll make a patch and send it to you in a bit. :)
  • Actually can you clarify this:

    "EG I want three random numbers to go to a filter cutoff or whatever, I dial in the n knob to 2 and take the n+1 output and attach it to the cutoff. I can ping the module whenever and it goes to one of those three random number with 33.3% probability."

    What is the n knob?
    What do you mean dial it into 2 and then add one? Do you want 64 separate outputs? Or just one output that can have 3 values out of 64 stored random values sent to it?

    Better still, give me a list of inputs, outputs, and controls, and that will make things clearer.
  • The module I am looking to emulate is pictured in the first post and my version of it is in the second. The salient portion is labeled "Quantized Randoms."


    1) Gate input to trigger grabbing a random number from the set.
    2) Gate input to generate a new set of random numbers.
    3) Knob which is actually a rotary switch outputting the value of "n," which ranges from 1 to 6


    1) labeled "2^n" outputs the grabbed random from a group of 2,4,8,... (depending on the value the n knob is set to)
    2) labeled "n+" outputs the grabbed random from a group of 2-7 (depending on the value the n knob is set to)
  • Well I eventually got the patch working but it's still 30 Mb. sigh. If the interface requirements were as follows would it be simpler?

    1) Set size (1-64)
    2) Trig grab a random number from the set
    3) Trig create a new set of random numbers


    1) The grabbed number.
  • You can do this in a much, much, much smaller patch. If you want to put it on the backburner for now, I will have the patch to you sometime tonight, possibly with a build tutorial video with it as well. Even the most enormous sequencers and crazy physical modelling filter stuff that @macromachines does only comes out to like 15-20mb. 30 seems crazy large to just grab and spit out some random numbers.
  • It was the patch history! I deleted that and it was down to 3.6 Mb. I would still love to see how you would do it, but to be honest, I am chomping at the bit to watch your next sequencer-building tutorial for building sequencers of arbitrary length. I think I watched the first one like 3 times.
  • I should have said, thanks again for being so helpful. I look forward to seeing whatever you post.
  • Time got away from me I'll have to do this tomorrow! :)