[sdiy] Wavetable Design Update

[ASSI]

On Saturday 05 February 2011, ASSI wrote:
> Matt needs 256x that frequency to feed the synchronous counters for the
> waveform generator.  Just doing squarewaves at the note frequency is a
> bit boring, isn't it?

I guess I should explain that in more detail because it is somewhat 
counterintuitive: for a design like Matt is doing (or a Walsh Osc, or any 
organ with top octaves), you need to provide the highest harmonic and the 
divider chain will produce perfect octaves to follow suit.  That way, tuning 
error translates into a relative frequency error on the highest frequency 
that will then stay constant throughout all the divider chain outputs.  A 
phase accumulator design on the other hand sets the accumulator value for 
the fundamental.  Any tuning error (say, in cents) at the fundamental also 
translates into the same tuning error for all octaves higher up.

In case it isn't obvious by now: it is much easier to make that tuning error 
small at the fundamental than it is for the highest harmonic.  Actually the 
peak tuning error for harmonics is the same for both designs.  However, it 
shows up as jitter (so the tuning error at each harmonic _averages_ to the 
tuning error of the fundamental) in the phase accumulator, while it is a 
constant tuning error in the other case.  In both cases the limit is imposed 
by the timing resolution of the clock.


Achim.
-- 
+<[Q+ Matrix-12 WAVE#46+305 Neuron microQkb Andromeda XTk Blofeld]>+

SD adaptations for Waldorf Q V3.00R3 and Q+ V3.54R2:
http://Synth.Stromeko.net/Downloads.html#WaldorfSDada