SirSickSik Contributions

#1013

Yes sorry, I'm working on a digital "mini buchla music easel". My patch is already finished, but I would like to put more possibility that the original music easel don't have. And to me, an hard sync or soft sync or even both make totally beautiful complex harmony. So yes, I'm speaking about two oscillator sync together everytime one of them passing trugh zero from negative to positive. So not just a simple reset.
But saddly the two oscillator I've used didn't get the sync inlet... I'll check for changing them but this is again so much work when you do a hard and complex patch like this.

Thanks a lot to you both for your help :slight_smile:

#1014

which ones do you want to use that don't have the inlet?

#1015

These ones :

#1016

I'm afraid that won't come so easy as those are mutable instrument modules. These are pre-written in header/.c files, so you'ld need to dive into that code, apply the syncing overthere and also add the extra sync input, which is probably pretty hard to "just do"..

But then again, the modules you show are somewhat basic modules, so you could also try to rebuild them using basic oscillators and waveshaping.. Then just use the sync of those basic oscillators.

#1017

tekno with 3 boards
no samples, everything is synthesized with my own modules and own effects.
razorshape

#1018

Thanks for the tips, it can seems easy for you but I have really zero skills into coding so I think I can't do it by myself. I choosen these osc because the timbre sounds pretty close than the buchla one. I'm not sure I'm able to make it by myself..
I've tried all other osc with sync inlet in the patcher, nothing other have this particulary type of timbre. So maybe I will not be able to hardsync my mini buchla...

#1019

just hook up a scope and try figure out what's happening when you move the timbre knob.

as you'll see, the FM module is simply two sine oscillators, one modulating the frequency of the other.
That said, using the factory sine oscillator won't be the same as internally it doesn't have an fm-index-ratio that scales the modulation with based on it's internal frequency.
Though, if you go through my oscillators, you'll be able to find several modules that include both a fm input with fm-index internally calculated as well as sync input. So as long as you use a sine wave as basis for both oscillators, it will sound the same as the fm module of mutable instruments. But lots of my oscillators also allow you to morph the waveform so that you can create more complex sounds then just those two modulating sines.

The overfold of that trianglefold oscillator is done by just using a sineshaper over a triangle oscillator and adding gain to the triangle before the sineshaper (use a "multiplier" and "shift" module from the math folder). If you use the factory triangle oscillator together with a gain module and sine shaper, you'll get the same effect.
Though, the triangle oscillator doesn't have a sync input. Again, just use one of my oscillators, add a gain and sineshaper behind it and you'll have the same thing with more options.

Or why not do both MI oscillators in one? Basically the only modulation of the trianglefold module is that it's adding a sineshaper behind the oscillator, so you could do the same with any other oscillator.

so for a basic setup:
-sync oscillator (that gives the wave that syncs the other oscillators, just use a sine oscillator for this)
-any 2 oscillators that have a sync &fm input (both connected to the sync-oscillator) and some wave-morphing/selecting
-connect the output of one of the 2 oscillators to the fm input of the other
-add a mixer to be able to mix the oscillators
-add a gain and sine shaper behind it

#1020

I'm gonna try it for sure tonight. Huge thanks to you sir for wasting your time to me. Hard knowledges in your backgrounds... I will post my buchla patch on the community when it will be totally finished and I will not forget to mention you :wink:

#1021

I've added an extra sync pin to my classic oscillator.
It's based on a sine oscillator, but thanks to internal self-fm, it can create all kinds of waveshapes. Sine, saw, triangle, pulse and what's ever in between as all controls are continuous amounts of modulation.
the patch has a single sync oscillator (sine) and two classic oscillators.
fm demo.axp (8.9 KB)

if it can't load the classic2 oscillator, just sync libraries

#1022

NEW MODULE

OSC
"complex"
An oscillator based on a thru-zero sinewave oscillator with an internal modulation to turn the sine into a saw-like shape and a no-sync-to-softsync-to-hardsync amount to an external input.
Added to this is a sub-harmonic generator and two pulsewidth generators based on the "inf" distortion module.
The sub-harmonic generator has an internal timer, so the sub can be forced into it's own low-frequency region even if the oscillator root is going higher and higher. Due to how it is made, it does not create fixed sub-octaves like a normal frequency divider (eg always one octave below the root), but, as it generates a squarewave based on whether the root goes through zero áfter the timer has ended, it creates logical sub-harmonic intervals based on the root frequency but in a fixed frequency region. In other words, the root becomes the harmonic overtone of the sub.

anyways, lots of harmonic content for a filter to set it's teeth in :slight_smile:

#1023

Is this new OSC "complex" in the community library?

#1024

I've synced the library but I can't find that beast ! :frowning:

#1025

oh woops, forgot to sync myself XD
waiting for the editor to start up, takes a long time nowadays.. somehow my computer is having more and more problems with the editor.. Especially when I go past the "open recent", it often disconnects the axoloti. Just moving over it with the mouse, when only wanting to save, can cause it to hang for a minute.. :frowning:

ah! synced!

#1026

NEW MODULE

OSC
"complexSH"

another complex oscillator, but this one is based on the S&H module instead of the "inf" distortion.
Uses 2 internal tables for S&H times and amplitudes that can be randomised for new spectral content and mixed between for a special sauce of complex pulse-width modulation.

and to combine with this oscillator for more spectral shizzle

FILTER
"armMOD"
an audiorate modulator, controlled by the phase of the oscillator it reads through a randomisable interpolated table to modulate the pitch/resonance of a filter so each part of the waveform gets another filter treatment. Adds lots more sonic possibilities

"armBPsvf"
audiorate modulatable svf BP filter

#1027

Thanks again. This new Complex OSC is insane !

#1028

I've done exactly what I wanted with your help. You can find it right there :
SineTimbre&sync.axs (4.9 KB)

It's a sine OSC made by "toneburst" modified. I've used your technics for the timbre, except that I've replaced the object "math/*c" with a simple VCA, by that way I'v deleted the annoying digital click wich appears when you move the math knob, and also allows me to apply FM on timbre. And I needed an on/off toggle for the FM in, so it's here.

Now I'm completely done with my Buchla music easel Patch. I will build a box and share this in few weeks.

Thank you so much again for your really usefull help.

#1029

no way 0_o
I managed to pull off realtime fft resynthesis :stuck_out_tongue:
It's still a long way from being good, but I can still recognize it enough when I'm talking through it.

anyone knows a way around that "chThdSleepMilliseconds(20);" part of the code in the fft analyser? It's creating a volume tremelo that I want to get rid off (adding a delay does fix it a bit, but not enough and adds a comb filter sound to it)

#1030

Congratulations, Remco, and your achievement might even make it a more realistic proposition for the creation of a spectrum analyser object for the axoloti patcher environment!

I remember when I asked about that in a thread somewhere on here. I got the impression it would be a resource hog due to the FFT aspect required to pull it off.

But wow, being able to do it in realtime might change things a fair bit, or at least push the possibility of having one, that bit closer :stuck_out_tongue_winking_eye:

#1031

it seems you can just increase the size of the factory fft analyser. The longer you set the buffer (in powers of 2), the lower it seems to be able to get frequencies. You do need to add a lot of displays to show the whole range..

#1032

the only problem with the resynthesis is this sleep&terminate thread:

 while (!chThdShouldTerminate()) {
    chThdSleepMilliseconds(10);
    if (state == LENGTH) {

      arm_rfft_q31(&rfft, &inbuf[0], &fftbuf[0]);	
	arm_mult_q31(hanning_q31, &fftbuf[0],&outbuf[0], n);
      state = LENGTH+1;
    }
 }

it imposes a amplitude (on/off) modulation because of the sleeptime.
it's more or less fixable by using an extra delayline, but the comb filtering is pretty clear and annoying..