[sdiy] Imperfect VCO

[Theo]

The paper mentioned is not exactly the latest in the art.
Just google for "band limited waveform generation" to find some other
approaches.

There is one method worth to mention that is little published about,
probably because it is not interesting  enough from a math side.

Phase distortion of a sine wave is lean on CPU load and gives rather good
results.
Indeed the same thing the pp from Casio did 15 years ago.
One of the nice things is how flexible phase distortion is,
classic Casio modulated from saw to sine but it does not end there.
PD of a sinewave to get a variable saw <--> tri or saw <--> square is also
possible.
On the down side, you never get a "pure" saw wave,
in the end every ramp fast or slow is a segment of a sine.
The higher harmonics carry a bit less energy compared to a analogue saw.

Just as with any other method, computing on a higher sample rate and then
down sample would help to reduce aliasing.
But even if you don't PD can out perform most other methods on the aliasing
side.

Theo



From: Bert Schiettecatte <bert.schiettecatte at esat.kuleuven.ac.be>


> >Dig up "Alias-Free Digital Synthesis of Classic Analog Waveforms" by
> >Tim Stilson and Julius Smith to get an impression of how hard it really
>
> >is to produce a sawtooth or square wave in the digital domain with
> >proper bandlimiting.
>
> Exactly, quite hard I have to say. But I don't think the approach taken
> in this paper is the best computational load/quality wise. I sort of
> have a feeling that oversampling the whole synth loop and then
> downsampling in the end might be a better approach ...
>
> bert
>
> ---
> Outgoing mail is certified Virus Free.
> Checked by AVG anti-virus system (http://www.grisoft.com).
> Version: 6.0.504 / Virus Database: 302 - Release Date: 24/07/2003
>
>