[sdiy] Filter cut off DAC

[rsdio at audiobanshee.com]

Totally agree with Richie on this one. Any decent VCF should always be ready to double as a VCO.

Brian


On May 21, 2018, at 9:36 AM, rburnett at richieburnett.co.uk wrote:
> I'd say that the answer is the same as the number of bits required to control a VCO :-)  The justification is this....
> 
> The human ear isn't particularly sensitive to "brightness" of a sound.  So you could get away with quite a coarse control over a simple *non resonant* low-pass filter.  However, once you increase the resonance, you then start to boost a narrow band of frequencies.  Then your ear gets much more sensitive to changes in the cutoff frequency.  If you take this to the extreme, with enough resonance the filter self-oscillates and becomes an oscillator.  The human ear is very sensitive to changes in *pitch* though!  So the question now becomes how fine a resolution of pitch do I need?
> 
> That is how I would think about this problem.  The worst case scenario is when the filter is self-oscillating and then "oscillator rules" apply in terms of what pitch step size is perceptible.  Maybe something around 10 cents? ...or as low as 1 cent even?
> 
> -Richie,
> 
> 
> On 2018-05-21 15:53, paula at synth.net wrote:
>> Random question,
>> How many bits is "enough" for controlling the filter cut off on a VCF?
>> I would start with say 12, but would 16 be better? or 24?
>> At some point there's a cost trade off.
>> Thoughts?