Wavetable scanning: is this possible?

[justin3am]

Some folks are describing wavetable oscillators which have two or more banks of wavetables, spread across X, Y and sometimes Z axes. This is just a way of navigating a large number of waves, by splitting them into more than one table.

I could be mistaken but sounds like the OP is asking if there are oscillators which will allow you to scan through the start point of wave (how I interpret the left-righ axis in their graphic). In that case I would suppose the phase parameter would do the trick.

I could also interpret the OPs description as some kind of windowing of the wave. I've seen that Serum and some others can apply operations to a waveform, to mimic things like PWM, sync and other modifications.

[Sound_Bear]

(Redacted)

[foosnark]

Wondering if scanning of a wavetable has to be always from bottom-to-top or vice versa when we look at a 3D-representation of a table like shown in numerous WT-synths.

Why not - alternatively - from left-to-right or vice versa? *scratches head*

See the well known scanning we're used to:


Now ... would it be impossible to scan from left to right or right to left here - meaning implement such a readout alternative?

The results are going to range from not very good, to not actually better than scanning it "properly" in most cases.

Let's say you have a very simple but useful wavetable that morphs from a sine to a nice fat saturated triangle. At phase=0 for the sine, the amplitude is 0. At phase=0 for the triangle, the amplitude is still 0. So if you turn it "sideways" so your oscillator phase is sweeping the other way, you'll just get silence. As you start moving to the right, you'll have small waves (the difference between the sine and triangle) but with a DC offset. As you approach 90 degrees of what was the original phase, you'll move back toward silence again... and again at 180 and 270 degrees.

Taking the example in your graphic, it looks like there won't be any dead space but there'll be some DC offsets, and shapes that aren't really any more interesting than reading the table as it was meant to be.

Also, a lot of wavetables have more samples than table indices -- since I know that one is from Serum, it has 2048 samples left-to-right and only 4 different waveshapes top-to-bottom, so it's going to rely a lot more on interpolation. So they're not particularly well suited to scanning in the other direction.

Of course you could design wavetables that work either way, but there's really not that much benefit in it.

Oscillators that are designed for 2D or 3D maps rather than linear tables are another story. There are a lot of those in Eurorack -- Mutable Instruments Braids/Sheep/Plaits/Beads, SynthTech E350/E352/E370, IME Piston Honda, 4ms SWN and some others -- as well as Anyma Phi/V (which I'm 100% sure must have evolved from Mutable Instruments open-source code).

(But honestly, when I had the SynthTech VCOs I mostly preferred the other modes which let you do wavefolding, swarms, phase mod etc. on a linear wavetable rather than the multi-dimensional stuff. Likewise with Mutable Instruments wavetables, I tend to modulate just one axis with big movements anyway, and keep any modulation of the other axis really subtle or nonexistent.)

[swilow11]

I'm wondering if the additional functions in Hive 2 allow something like what the OP is describing.