A little dream of a continously morphable multipurpose oscilator...

[Gone soft]

I dream of a continuously morphable multipurpose oscillator. I may be wrong, but it seems the classic basic waveforms can be broken down to placement of a maximum of five points along time and level axises with choices for «shape» similar to in the multistage envelope. These points would as a possible starting point (lots of points here...) be spaced evenly along the (horizontal) time axis with the starting point at zero, point 3 in the middle and the last one at the very end. Thus a sine wave would be made up of a level/value setting (at the vertical axis) of zero for the first point, some shaping of the line that moves to the next point, maximum positive value for the second and appropriate shaping as it approaches zero again for the middle. Then a shaped curve would head down for the maximum (or lowest if you will) negative value at point four and shaped it makes its way up to zero again at the last point. Off course each of these parameters should have their own modulation input. With appropriate mapping of modulation sources the shapers can move to their middle positions (thus not shaping at all) as level one approaches full level and if I'm not way off here the second point should have a level halfway between that and zero while the middle point remains at zero. The fourth point gets modulated to halfway between zero and bottom while the fifth and last hits the bottom. Thus a sine wave should have been morphed to a sawtooth, if I haven't gotten anything all wrong along the way. The points should be able to move along the time axis also, to get PWM amongst other things. I'm not sure if the first, the middle and the end points should always be fixed while 2 and 4 could be moved along the time axis through faders/knobs and dedicated modulation inputs. If the points would simply move within the space between the points before and the point following them it should work well, and the middle point can just be left at the default position whenever that seems convenient. I'm not sure if the total length should be modulatable so reducing it to half would provide what I think is called a double sawtooth, or if that would have undesirable side effects and complicate matters more than necessary. I may also off course have misunderstood something as to how oscillators actually works as I'm just looking at the waveforms and don't know too much about the inner workings of an oscillator.

Then I keep on dreaming - it should have one input for frequency that works in the usual pitch range, one that works in the usual FM range and.. one for sweeping the frequency at greater range with two knobs for limiting the range down and up. Each of these with modulation inputs. If neither the pitch nor the FM range input range is used setting the downward range limiter to it's lowest value will make it possible to sweep the oscillator downwards all the way to actually stopping when the lowest modulation value is received. Limit the upper range appropriately and it would remain in LFO range, expand the upper range to allow modulation into audio range etc. The upper range should be able to go relatively far down, and if it is set to a value lower than the bottom range it would stop affecting it at the point the ranges meet and the oscillator would remain at this frequency until the values take their separate ways again. Thus you can have a relatively random input and make crazy modulations of filter or other parameters, make it settle at an appropriate LFO frequency for a chosen time and go all over the place or just straight up into audio frequency range in a manner as stable or random as desirable.

It would be very helpful if the range limiting knobs would show values in frequency somehow, either with small windows or just by being able to see the value as you twist them such as for instance the decimator does. With no input an appropriate frequency can be chosen by either turning the upper range knob downwards below the default middle value or the lower range knob upwards and that would affect frequency as soon as the value has a higher value than the default.

It would be a very flexible little beast, and if you steer away from sinewaves and just disregard the shape parameters it wouldn't be all that hard to operate and achieve relatively predictable results. It would have a sort of simplicity that hopefully wouldn't make the learning curve to steep. It shouldn't have a single switch as the concept is for it to be continuously morphable.

Any opinions on this? Does it even make sense to the rest of you people? I've tried to present my idea clearly, I apologize if I've failed. What do you think, Martin? A horrible idea or something you would like to implement?

[Gone soft]

Behold! My amazing Paint skills will help those who may have found that first part somewhat fuzzy but not so logic. I meant something simple like this:



[/img]

edit: not sure if that made it any better...

[bmot]

sorry, that was way too long for me to read, but from the bit I did read, perhaps you should try cableguys curve where you have loads of LFOs you can draw yourself assignable to loads of paramters which could pretty well do what you want I guess:

http://www.cableguys.de/curve.html

i.e. draw in the morphing yourself

[EvilDragon]

Check out QuikQuak Glass Viper. Its oscillators are working EXACTLY how you propose here. You get 4 LFOs to modulate segments of the oscillator waveform. It is very cool. There's a LOT of flexibility in this under-the-radar synth...

[Gone soft]

Thank you for the tips, but none of these seem to offer quite what I'm looking for. They don't provide flexible oscillators that could also modulate anything within a modular environment (at least not as far as I can see), and for the QuickQuack it doesn't seem I would be able to lock the waveform as for instance a triangle wave for five seconds or so (or longer or shorter) before altering it into something else. You know, letting things be as molten or as solid as desired at any given moment through use of combinations of control converters and assorted modulation sources that can affect various aspects to various degrees through amplifier b modules and mixers etc. Also, this is a product forum for the Vaz software synths so recommending other products here might seem a bit out of place..

[bmot]

this is a product forum for the Vaz software synths so recommending other products here might seem a bit out of place..

yeah fair point! i actually didn't notice that until i'd just sent the post