1. Please do not install macOS 10.15 Catalina yet, as Native Instruments software and hardware products are not supported under macOS 10.15 yet. For more info, please go HERE!
    Dismiss Notice

Storing Audio In A Buffer?

Discussion in 'REAKTOR' started by zoobooboozoo, May 15, 2014.

  1. zoobooboozoo

    zoobooboozoo Member

    Messages:
    56
    Hi all, the journey of glitching audio and harressing reaktor forum memebrs is still on.

    I wish to be able to store a short clip from the inocming audio(using Reaktor as FX) in somekind of buffer for futher manipulations.

    I worked based on salamndra's glitch tutorial but if I manipulate the glitch further, it gets fed into the delay, manipulated again, then fed into the delay again etc. and as a result the sound is gradually ruined.
    So for example: If I'm adding a phaser to the glitch, it gets "rephased" with every trigger of the glitch effect(when the glitch triggers are not spaced: "legato") and then as I said, the sound gets screwed up or at least manipulated in an undesired way since glitch is a sound that got screwed up in the 1st place ;-)

    I'm thinking that If I store the to-be-glitched audio in a buffer I can trigger it and then feed the clean signal through the rest of the FX with each gate trigger and solve my problem.

    Any suggestions? (that hopefully don't involve core...)
     
  2. salamanderanagram

    salamanderanagram NI Product Owner

    Messages:
    3,438
    a few thoughts - one, it's not necessary to put your other audio modifiers in the feedback path! you can simply have a glitch from the video and manipulate the output.

    second, if you need another option i would take a look at the audio table, which can store audio buffers that aren't constantly being written to.
     
  3. zoobooboozoo

    zoobooboozoo Member

    Messages:
    56
    Can I do a similar thing with sample & hold? seems simpler...

    How do I determine the length of the held sample in Sample &Hold or the Audi Table?
    For example if I want a 100ms clip: I need to have a "1" value arrive at Trig / W for 100ms and then have an 0 arrive at Trig / W ? I'm thinking accumulator, system info-CR, and value modules which will trig 0 and 1 depending if the accumulator equals 100ms...

    A general question: When triggers are triggered by "a rising zero" or "positive value", what exactly does that mean? does a regular trigger triggers whenever an event comes in regardless of the value? "rising zero" is when a zero is changed to anything higher than zero? "positive value" includes 0?(I guess not)
     
    Last edited: May 16, 2014
  4. salamanderanagram

    salamanderanagram NI Product Owner

    Messages:
    3,438
    no, a sample and hold is like an audio rate value module, it doesn't store sequences of values.

    and again, i don't see what you're going to gain out of this - you're not going to get anything that functions any differently than the glitch you've already got. as i said above, just don't place the extra effects in the feedback path and you'll be fine.
     
  5. zoobooboozoo

    zoobooboozoo Member

    Messages:
    56
    Hey salamandra, I'm not trying to be ungrateful or stubborn, I appreciate your videos and your forum help very much.

    My main problem is: and I'm not at my DAW right now to retest it but I remember that I played around with the glitch FX and If I had 2 consqeuent notes triggering different glitch rates than the recursive feeding into the delay would alter the sound. this is undesirable for me, and I will upload an example later on today.

    Besides, As I said in previous posts, I'm trying to educate my self about reaktor by building this FX and not just "to get it over with". It's a good opprtunity for me to learn about Audio Table. On top of it, The audio table solution seems like a more flexible one for future features.

    I watched your audio table tutorials and found the phrase looper one to be very helpful. I think I can easily base a similar glitch FX on it. My question is(and I couldn't find info about it in the manual or elsewhere) is why do you use different interpulations in different builds? basing the loop on sample rates like you did in the phrase looper video seems like the best choice IMHO. since, as I understand it, the audio is sampled at the sample rate anyhow so the best audio "resolution" will always be the sample rate and not an arbitrary amound of ms. Am I missing something here? The goal for me at least, is to have the sound as close to it's original quality as possible.


    I'm trying to test the glitch functionallity fully(or at least as close as I can get to "fully") so I don't run into problems in the future. I want to build it as bug-free as I can so I can trust it would do everything I need without problems in the future and I can have an eased mind when dropping it into a track :)
     
  6. salamanderanagram

    salamanderanagram NI Product Owner

    Messages:
    3,438
    hey, not sure exactly what you mean...

    interpolation is for when you are reading back at a different speed than you read it in at - i believe the tutorials you're mentioning have a speed knob. if you want the glitch to always be at the same speed, there is no reason to interpolate. that said, clicking the interpolation option in the audio table properties is a pretty easy task that takes next to no CPU so you might as well have it on.

    don't worry i didn't think you were ;)
     
    Last edited: May 17, 2014
  7. zoobooboozoo

    zoobooboozoo Member

    Messages:
    56
    For testing it I made a brand new ensemble with no extra features. just the simple structure required for the glitch, a knob, and a switch.
    My ableton set and my .ens: Link

    So here are a few problems I've encountered that made me look for an alternative glitching mecahnism(i.e. audio table),

    1) Usually I do but Sometimes I don't want the audio to be resampled. for example if I want to glitch a kick and keep it glitching while I change the rate of the glitch - I don't want the snare or HH or w/e to be resampled instead of the kick while I'm playing with it and with the rate. In this case I need to have a long note while "resample" is on or long note/ several legato notes while "no resample" is on. Then and I turn the knob while the audio is glitching and the sound gets messed up.

    2) When "resample" is on, legato notes are "playing" and each note correlates to the same bit of audio (for example a repeating kick) a wierd thing happens: the 1st glitch sounds fine but the 2ns one and every glitched kick after the 2nd one had gone through a change of timber and it also have click(s) at the beginning of each glitch kick.

    3) Sometimes with spaced notes(no legato), whether its "resample" or "no resample", there are clicks appearaing at the end of the glitch.
     
  8. zoobooboozoo

    zoobooboozoo Member

    Messages:
    56
    Gentle Bump ? :)