A little feature request for Zebra2

[trance_lucent]

Just found interesting thread "Alternative to Sylenth1"http://www.kvraudio.com/forum/viewtopic ... highlight=, there was some good ideas, which I proposed too (such as not to do dedicated EDM synth but improve Zebra a little). Also soundpalace wrote there:

I personally feel that the main thing Zebra is missing in comparison to Sylenth1 is the more edgy distortion and some other effects in Sylenth1's effects section. I would love to hear some awesome Foldback and Clip distortion units in Zebra, as opposed to the more polite and subtle Shaper. VCF Vintage 2 drive is the closest I could find in Zebra2 for such distortion.

While I think Vintage 2 is identical to XMF's XMF besides of knobs behaviour, I agree that it has very good character of distortion, and the even harder distortion is the thing I want too.

And although other requested features are better to implement in Z3, it is too long to wait it, and may be U-he can implement just one feature in one of upcoming Z2 updates? I mean harder distortion or may be even clipping (and may be triangle foldback would be great too, but it can wait for Z3). Just simple algorithm, without heavy antialiasing routines, uh? In XMF or may be in Shaper? I think it is just couple of hours of worktime.

[Urs]

We'll see... I've recently implemented a first idea of distortion in my weekend project. Maybe Sascha can take it on once Satin 1.1 is out and about.

[Sendy]

Isn't the Wedge mode basically a triangle shaper/foldback distortion?

[trance_lucent]

Isn't the Wedge mode basically a triangle shaper/foldback distortion?

It seems for me that Wedge is similar to XMF's Fold. And I like this mode, it can add some FMish character to the sound, but I'd prefer if the output(input) function would not foldbacking to zero just one time but lasting forever like bipolar or unipolar (with DC offset multiplied by input's sign) sine. Right now it is like one-cycle sine with fades at the sides. Also I'd like to have triangle version of it with harder edge. All Zebra distortion modes are too "rounded", and I often need more "edged" sound.

[trance_lucent]

I have measured the transfer functions of Zebra foldback modes. As I wrote, they are not cyclic (this is desired sometimes and sometimes not) and smooth, not triangled (which is also good, but I want more hard edge sometimes).

Edit: I measured the functions by feeding sawtooth into units, but as I set the frequency too low it seems the result was distorted by builtin HPF. At higher freqs the negative overshootings are much lesser.

Edit2 (higher freq):

[Sendy]

There is the wrap osc FX if you want to wrap multiple cycles, it can fold like a triangle or jump to zero and change sign producing squarish sounds, though of course it's limited in that it can't work on an arbitrary signal. It covers some very interesting territory between FM, sync, and PWM if you use small modulation values and set it up carefully.

[hollo]

We'll see... I've recently implemented a first idea of distortion in my weekend project. Maybe Sascha can take it on once Satin 1.1 is out and about.

This - and that old supresaw Osc you'v talked in the other thread,
and the wait for Z3 will be much more relaxed.

[trance_lucent]

Some other Zebra's drive curves:

Also I've tried Sylenth1. Good synth indeed, but so limited even for EDM!
But I like its distortion.
Here is some curves:

Overdrive is good sounding, somewhat similar to XMF's Analogue.
Clipping is what I want! So edged!
But Foldback is strange. While it is sounding good, it has not so much "foldbacking" fmish character I like. It is somewhere between foldback and usual overdrive. But I like the fact it's limit isn't zero if the signal approaches infinity. And it is smoothcurved too without pronounced edge.

There is the wrap osc FX if you want to wrap multiple cycles, it can fold like a triangle or jump to zero and change sign producing squarish sounds, though of course it's limited in that it can't work on an arbitrary signal. It covers some very interesting territory between FM, sync, and PWM if you use small modulation values and set it up carefully.

Yes, I recently discovered the might of Wrap. Feeding Hypercomb at high resolution into small values of wrap creates distorted supersaw with just one osc. And you can stack it further and distort with XMF. Cool for some plucked sounds.
But I like the dynamic character of shaping (that is why I feed animating hypercomb into Wrap in the previous example). I like to place it onto beating stacked oscillators and then also some at the master section (so if there are several sounds play together the shaper/overdrive makes it more alive).

[Sendy]

Yeah, when you start getting into feeding Osc FX into eachother you really start to wish for three FX slots, or at least, I do

[3ee]

Yeah, when you start getting into feeding Osc FX into eachother you really start to wish for three FX slots, or at least, I do

Me too!
+ it would be great if we could set the order (+modulation if any) as well.

[Howard]

@Sendy: OK, from which synth is the waveform in your avatar? Guessing Rayblaster.

[Sendy]

@Sendy: OK, from which synth is the waveform in your avatar? Guessing Rayblaster.

It's the combined Triangle/Pulse waveform from the SID chip with a low pulsewidth. One of the most interesting raw waveforms you can get off a chip, IMO.

[klinik]

Trance_lucent, how do you sample the transfer functions?? Thanks!!

[trance_lucent]

Trance_lucent, how do you sample the transfer functions?? Thanks!!

Well, as the shaper is basically function = output(input), to plot the output you need to feed it with the linear function. So I fed it with plain sawtooth wave from usual osc. The plot of one cycle of this distorted sawtooth is the transfer function of shaper.
But if the shaper has highpass filter on its output, it is better use higher frequency of oscillator. Also this way isn't very good if the distortion unit isn't plain shaper but has dynamic behaviour.