How to create a low pass filter on an additive synth?

I'm working on an additive synth and I'm struggling with creating LP/HP filters.


I've got a basic setup where it will amplify the harmonics to a 0 or 1 level creating a cutoff that is all or nothing, essentially a brick wall cutoff, but I'm struggling at trying to get the filter cutoff to fade out harmonics gradually to create a smoother roll off.


Any ideas or suggestion?

Β«1

Comments

  • Michael O'Hagan
    Michael O'Hagan Member Posts: 93 Helper
    edited August 2022

    Delete this post, I figured it out like 10 min after I posted it.πŸ™„

  • ANDREW221231
    ANDREW221231 Member Posts: 295 Advisor

    hot take i think its good that this forum doesn't allow deleting and only allows for edits within a limited time window


    like, i find it annoying for my own posts, but also other people can't go deleting things that were interesting after the fact πŸ€·β€β™‚οΈ

  • Kaiwan_NI
    Kaiwan_NI Administrator Posts: 2,523 admin

    Care to share what the solution is @Michael O'Hagan? πŸ˜ƒ

  • Paule
    Paule Member Posts: 1,328 Expert
    edited August 2022

    Andy, if you reach level saw you can change your posts from now to eternity.

    Kai, Michael is partly in commercial codings active. Mayhaps he's to build one new to sell it?

  • Michael O'Hagan
    Michael O'Hagan Member Posts: 93 Helper

    Does anyone know if there is a set math formula for re creating oscillator sync in additive synthesis?

  • Michael O'Hagan
    Michael O'Hagan Member Posts: 93 Helper

    I'm not really sure how to verbalize this, but it's basically subtracting partials from the spectrum.

  • ANDREW221231
    ANDREW221231 Member Posts: 295 Advisor

    think you might be in uncharted waters here. of course, any given static sync waveform could be constructed additively and the most direct way i can think of to figure out the respective amplitudes would be through spectral or fourier analysis of the waveform


    but that doesn't help you in the case of modulating the 'slave' oscillator to get changing timbres. i think calculating these on the fly would probably look something like the equations for formants in additive, because hard sync's effect on the sound is basically a formant like effect

    what i mean by this is if you take a short clip of sound with arbitrary frequency content and loop it as a periodic waveform now you have a harmonic waveform but the spectrum will have formants in the partials nearest to the frequency content of the original clip.... if that makes sense

  • Michael O'Hagan
    Michael O'Hagan Member Posts: 93 Helper

    I Got the sync issue taken care of, and it sounds great.


    I need to find a way to modulate 256 harmonics individually without using 256 modulators or 256 voices as the cpu wouldn't handle that well.

  • Michael O'Hagan
    Michael O'Hagan Member Posts: 93 Helper

    I'd like to ask for a mod to change the thread title to "Lot's Of Questions About Additive Synthesis".


    I'm going to be asking a lot of questions about additive in the near future and I'd like to keep it all in one thread.


    Thanks

  • ANDREW221231
    ANDREW221231 Member Posts: 295 Advisor
    edited September 2022

    i assumed you were using the sine bank as its the most optimized module for additive in reaktor.


    yeah, using a bank primary sync oscillators is a good easy way to get it done, but i don't believe there's going to be an easy way to do the same with the sine bank


    not to say its not possible, im sure it probably is. just not easy. maybe a solution would be in updating the sine bank parameters at audio rate

  • Michael O'Hagan
    Michael O'Hagan Member Posts: 93 Helper


    Updating the sine bank at audio rate seems like it would be a a cpu scorching overload.

  • ANDREW221231
    ANDREW221231 Member Posts: 295 Advisor

    not really, actually its pretty manageable. generally you only need to worry about updating around once per cycle


    for instance you can output an FFT to sinebank, which means updating anywhere from 80-300 times per second, and its definitely less CPU intensive to do that than to do the same thing with the same amount of primary sine oscillators

  • ANDREW221231
    ANDREW221231 Member Posts: 295 Advisor

    now that i'm thinking about this, it might even be easy... every time you need to reset with a sync pulse you just update the parameters again and it should (in theory) start drawing whatever waveform from the beginning

  • Michael O'Hagan
    Michael O'Hagan Member Posts: 93 Helper
    edited September 2022

    I'm working on this thing right here...


    It's got some pretty solid harmonic modifiers right now, but I was wondering if anyone had any ideas for something more to be added to it.


    Let me know if you have any ideas for new features and I'll see if I can add them in there.

  • ANDREW221231
    ANDREW221231 Member Posts: 295 Advisor

    did you get the sync thing working? if not, i think i can see a way to make it work which i will try out

Back To Top