[sdiy] VCF model/passband gain compensation

[Scott Nordlund]

Since I am only aiming for a half-assed model, it doesn't have to be perfect (probably won't include the tanh distortion), but it's certainly welcome, at least for educational purposes.  There's no  better way to understand/appreciate something than to actually make it.  I am familiar enough with DSP, root locus, IIR filters, etc. to follow along and understand what's going on, but the complexities of specific implementations still elude me.  I'm glad to know at least that my understanding was correct about using oversampling.

I still need information though about passband gain compensation, as I don't see any way for the "multi-mode transfer function from lowpass filter" thing to work without it.


----------------------------------------
> Date: Sun, 21 Dec 2008 01:05:17 +0200
> From: ajhuovil at cc.hut.fi
> To: martin at martinfay.com
> CC: gsn10 at hotmail.com; synth-diy at dropmix.xs4all.nl
> Subject: Re: [sdiy] VCF model/passband gain compensation
>
> On Sat, 20 Dec 2008, Martin Fay wrote:
>
>> I suspect Antti may pick this up, in his Moog filter model a simple FIR is
>> used to introduce a ~1/2 sample delay in the feedback loop for phase
>> correction. I've also successfully used an allpass filter in this role.
>
> I changed to the Stilson & Smith "Compromise" version some four years ago
> (see http://ccrma-www.stanford.edu/~stilti/papers/Welcome.html).
>
> The basic idea is to add a two tap FIR filter between each stage. This
> gives more constant resonance and higher max cutoff. For four stages the
> solution works remarkably well. It should work or higher number of stages
> too, but you will need to find the FIR coefficients with some iterative
> search. For four stages they are [1/1.3 0.3/1.3].
>
> I've put some pseudocode here: http://users.tkk.fi/~ajhuovil/moog.txt
>
> Antti
>
> "No boom today. Boom tomorrow. There's always a boom tomorrow"
> -- Lt. Cmdr. Ivanova

_________________________________________________________________
It’s the same Hotmail®. If by “same” you mean up to 70% faster.
http://windowslive.com/online/hotmail?ocid=TXT_TAGLM_WL_hotmail_acq_broad1_122008