PWMing sound examples?

[Zamise]

Anyone have some good demos of how PWMing affects a saw form? Or any waveform that doesn't just sound like two OSCs with one being detuned for the beefyness?

I've been trying to do it with samples by inverting one waveform and playing them together thinking they'd effectivly cancel themselfs out and get silence, but that don't seem to be happening. I do get some audible changes in using an inverted waveform over non-inverted and changing the cycle start times, but they mostly seem to more or less add distortion. So I tried doing some PWMing on my CS-15 and I just don't seem to hear much if any change in the timbre, so I try it on my AN1X not much going on there either with PWMing. FMing on the other hand seems to kick PWM's a*s.

What am I doing wrong? I'd like to hear some examples and maybe some explanation on how so I can do it too and know what I am doing if anyone has the time.

Thanks!

[smoothcriminal]

No idea if its helpful or not, but this is what an ax60 oscillator looks like with 100% pulse width on a saw wave. If the width were being modulated, that dip in the center of the saw ramp would be expanding and contracting. i was under the impression most synths only had it available for square (aka pulse) wave, so maybe thats your problem with the CS-15.

[Zamise]

Oh craplatude! That must be what I was doing wrong, saws don't do poo when PWMing. Its the squares. Switch the CS-15 over to square and got quite a noticeable change when sweeping. Same deal on my Oddy too, PWM only seems to work only on the square VCOs or just way more noticeable. Learned me something new today.

Funny thing is, I was looking at an SY77 manual on its PMing and it shows pictures of saws inverted for PWMing that equal squares, so I think that might of got the saws stuck in my head for trying to do PWMing on them. I'm going to give squares a shot on sampling now instead of saws, see if any thing better comes of it. Unfortunately I didn't sample the square on my CS-15 cause it looked more like a freaky sine, so I may have to find a better square/pulse.

Thanks!

[Roby31]

Some synths do PWM on every wave, if you're just looking for an example, demo Novation's V-Station and you'll hear what goes on. Of course the more obvious thing (and probably the most useful) is pwm on a pulse.

[Stab Frenzy]

You only get PWM on pulse waveforms. The clue was hidden in the name.

You can't modulate the pulse width of a waveform that doesn't have a pulse width.

[smoothcriminal]

I have found that the "modulation" aspect of pulse width isn't very useful on anything other than squares, since a narrow pulse distortion in a different waves sounds really harsh. But with wide pulse values that aren't modulated, you can get interesting waveforms like the sawsquare I posted. At least this is how it works on an AX60.

[Zamise]

Right Oh! My old cooledit 2000 had a tone generator, with square

[smoothcriminal]

Since I'm a miserable procrastinator, the fact that I have other things to do means I couldn't resist making a PWM patch. This is a single track with a PWM+filter envelope square wave with no FX or EQ.

[Alex E]

The MS2000 did have width modulation on the saw wave and a few other waves. If I still had mine I'd make a demo.

[madtheory]

...PWMing...FMing

Why did you add "ing" to them? That's appalling abuse of the language. I will have to tell AG.

[Zamise]

Modulating.. its sort of like turning lactation to lactating... PWLing, good yeah/no good? Its what AG does when he hears someone else's 2600 screaming

Anyhow, thanks for the song listening to the Pulse Width Modulationing on it right now. Also, I did sort of a flow chart of how it is, but isn't working so hot using samples like I heard is supposedly possible to do, but I don't think it is really doing the trick like a real PWM. Or, maybe it is, just not really sure what exactly is going on but here:


I'm U/Lnating a vid to youtube on it right now, I'll post if it came out ok, but maybe it'll show some of my total confusionism and un-understadulation on being a Pulse Width Motard.

[smoothcriminal]

I have no idea what you're trying to do here. The only "PWM but not really PWM" trick I've heard of had something to do with OSC sync where you use the rate of one OSC to retrigger the second OSC's square wave at different widths. I can't imagine any reason why two inverted square waves would produce PWM.

[Zamise]

The idea was that a perfectly inverted waveform should cancel out the other to 0, then if the width or percentage was changed it would start PWM[ing]. However, they never seemed to cancel each other out, so unsure if worth bothering to use variations of offset inverted waveform samples to do any more kinds of PWM using this method. We/I got this idea from the SY77 users manual, it shows a figure example using saws on page 30 to program PWM using sampled waveforms. Check it:



So We/I thought it could be done on another sampler, being our/my beloved RS7000

[projectwoofer]

I can't imagine any reason why two inverted square waves would produce PWM.

They don't, but two inverted saws will do it!

[Zamise]

Yeah I tried saws first, did stuff on the samples, but wasn't doing anything on CS-15, got massively confused, brain started melting, came here to get answers, got some answers but brain still melting.