do you know some oscillators i can retrig the phase ?
i used one from bog, xco but strangely when i adjust the phase and retriger it the signal get delayed by the amount of phase shift, never seen that behavior, i m looking for alternatives.
if you know some osc that can retrig the phase i would appreciate it very much
I guess your main issue is that generally oscillators in modular are free-running. And we generally don’t trigger the oscillator (to “start” it), but instead open some gate to clear the way for the already running oscillator signal to the next link in the chain. Whereever it is in its phase cycle.
I guess you would just need an oscillator that can reset its phase (restart its cycle) on demand. Which is effectively what oscillator sync is. Just syncing an oscillator with another will just have both cycles synced, but still free running. And keep syncing every cycle. You would need to reset it with a single pulse/gate.
Not sure at what point (at what voltage) the actual sync/restart would take place. Or more specific, if this is the same for all osc sync implementations.
For the Fundamental oscillator the documentation states: “The SYNC input applies hard-syncing to the waveforms by resetting the phase of the waveform when the sync signal passes upward through 0V”. So, I guess a pulse or gate signal would work just fine for that.
Maybe also read this thread about many things on Phase and Phase Shift
Maybe the Waveguide Delay (see the Phase shift thread) can give you access to any other then the initial/start phase angle (for basically any oscillator). Since with that you can introduce arbitrary delays with great precision.
thanks steve ,awesome ! i make psytrance also and i use already shapemaster for my kicks recently, but i also done bassline with it with great result.
thanks for the patch i m gonna try it, mine is very similar , just the bog and shape master, i didn t know this osc could output saw as well , thanks a lot for the tip
thanks for the tips i will try ! with using the sync input i can t retrig where i want in the cycle but depending the osc if the waveform is suited, it might be fine ! i will do some testing thanks a lot.
Since FM is mostly implemented as PM, you can imagine that the phase modulation results are highly dependent on the relative phases. To get predictable results, the phase needs to be reset (when notes are triggered). To get variation (if so desired), you can have the added option to NOT reset and have freerunning oscillators/operators.
The most obvious example is: (0 Hz carrier) phase modulation “waveshaping”, where the relative phase determines where in the carriers cycle “compression” or “expansion” will take place. Therefore determining the resulting waveshape and its spectral content. Generally: if the waveform changes (check oscilloscope), the spectrum changes (check frequency analyzer), and vice versa.
You can hear it happening in “normal” FM/PM (though cyclic/periodically) when detuning the modulator/carrier frequencies a tiny bit. But…back to basics…
Compare to Wavetable “scanning”. But in PM, instead of just linear scanning from start to end of the carrier’s waveform, you move the scanner’s start point (phase), thus modulation somewhere and thus something else.
In the case of FM/PM you also have the option to modulate the “scan” speed during the scan cycle. In case of the original Yamaha “DX FM” (=PM), the scan speed was modulated by a sine (since it only had sines), and the wave to be scanned was also a sine. But why stop there? You can use any scanner “shape” (think modulator) to modulate the scanning “speed” of scanning any other “shape” (think carrier).
Some more on this “effect” (including various scanner/phasor variations) from the Valley Terrorform documentation:
“Zero Frequency? What is the purpose of that?!” I hear you say
The eagle eyed amongst you may have noticed one of the benefits of including the DX style phase mod flavour of FM. Remember, in phase modulation, the phase of the of the read phasor is shifted, regardless of frequency. Therefore, if the frequency is zero and you shift the phase of the read phasor, you will still read the contents of the wavetable but with an external source. When this feature is enabled, the read phasor is “parked” at the beginning of the wavetable so that no hidden phase offset is introduced. The following two sub-sections discuss example uses for this feature.
Appendix A: How it works
Terrorform generates a tone using “wavetable lookup synthesis”. This is the exact same method used in each operator in Dexter. As the name suggests, the tone is generated by looking up and reading back the contents of a wavetable. The lookup process is done by a rising ramp wave, known as a read phasor, that reads the table from beginning to end at a set frequency. Different waves in a bank can be scanned through by cross fading between adjacent tables within the same bank. The read phasor can be shaped in several ways that can alter the output waveform. Terrorform offers 27 shaping modes. After being read, the resulting signal is passed through the Enhancer. This further shapes the signal in different ways such as bit-crushing, Chebyshev waveshaping, folding, and more.
So, something similar happens when phase is not reset at note trigger. The resulting waveshape would be different each time the carrier output goes tot the end mix / amp (where you can then hear it). This is why many “FM” (PM) implementations (like the DX7) offer a choice between synced and freerunning operators (oscillators). Sometimes you want it synced (e.g. tight bass) and somtimes you don’t (e.g. “acoustic” instruments like guitar/piano)
A completely different phenomena, but somewhat comparable, recognizable and explanatory nonetheless: relative phase when just summing/mixing. Free running oscillators will generally cause phase differences and resulting phase (combfilter) effects. Especially when detuning/drifting is involved. To get predictable results, osc sync was invented.
A simple way to experience phase “difference” effects is to take two oscillators, setup mix, FM, PM or sync or whatever interaction between them, and then modulate the frequency (and thus phase) of one of them a bit.
E,g. setup a modulator (ENV, LFO or button or whatever) that modulates the frequency of one of the oscillators a tiny bit for a short while (which will also change the relative phase). Then see (FFT/Spectrum Analyzer) or hear (ears) if the spectrum changes.
Remember that lowering the frequency of just one, just means it is running (cycling) slower the the other, thus lagging behind in phase. If the difference is a fixed frequency, this phase in/out will cycle at the difference in frequency.
If it is a temperary difference/detune (after which they are running at the same frequency again, in tune), then afterwards they will most likely be out of sync (producing a static spectrum again, unlike detuning where it would cycle forever).
Since this whole phase difference thing effectively results in spectral effects, it is also a crude way to create new (static) spectra from two (or more) fixed spectrum oscillators (assuming they are running at the exact same speed/ratio).
In “our” digital world you could also just use a precision delay (sampledelay or milliseconds) on one of the signals to effectively cause phase differences. E.g. the Sckitam Waveguide Delay.
Anyway, way too many words, again, but consider it just my “two cents” on the subject…
I hope all this is usefull/inspirational to someone out there…
Not sure what you mean. Maybe I’m wrong somewhere. Maybe my words were ambiguous somewhere. The whole story was about relative phase. Phase difference between 2 oscillators. I don’t wanna spread confusion…or misinformation.
Sure, but all I’m saying, and @carbon14 is also saying (I think), is that we have a single “VCO”. It’s made up of a carrier and a modulator. When there is a change in the relative phase of the carrier and the modulator, the result is just a change in the absolute phase of the output, which is difficult to detect.
It is true that for bass sound you can get popping from any VCO with a very fast envelope, and when you are mixing multiple VCOs you can get time dependent cancelation, both of which can be solved with a phase reset.
But this isn’t particular to “Phase Distortion Synthesis”, it’s true of any VCO.
So, I don’t see why this case you are taking about is special. In my mind (which may be incorrect) it’s all about resetting the absolute phase of the output.
I haven’t done too much with phase modulation, but FM and PM are closely related, yes?
I don’t think those differences due to phase differences are necessarily difficult to detect.
When I was experimenting with Neoni FM and others, I heard distinctly different timbres, and saw different waveforms on the scope, with the same settings, the only difference being the phase of the modulator relative to the carrier at the onset. The timbre would be static, but if I varied the frequency of the carrier or modulator and then returned to the same original settings, the timbre would be static again, but different.
I was trying to make different oscillators sound the same with linear FM, and sometimes they would be nearly identical, and others definite differences. The only thing I think that was varying was the relative phase of the modulator to the carrier at the onset of whatever ratio I was establishing. When I added sync, the differences disappeared.
So if I could hear differences with FM, it seems intuitive that I could also hear differences with PM.
@Fred66 From skimming the replies already posted, I’m not sure I understand the question, but what about Submarine’s PO-101 (VCO version) and PO-102 (LFO version)?
PO-102 looks identical.
Of particular interest is the bottom section. Four CV inputs for phase control, four manual phase control knobs, and output jacks for the same waveform at up to four different phases. Any stepped change to the CV gives an immediate phase shift in the output. Two different modules with the same phase control CV will give you more choices of waveforms. (As in, say, a reference Sine plus a phase-shifted Sawtooth.)
I have not checked it out far enough to see if it’s possible to synchronize two module from the very first cycle.
Is this at all like what you had in mind?
Ooh, now that I think about it, PO-102 is giving me Delusions of Being Steve Reich.