[sdiy] Digital ADSR - perceivable staircase?

[Stewart Pye]

As Tom has said if you use a 12 bit DAC like the MCP4921 for about $2 
then your problem should disappear, as long as your look up table is 
wide enough or you use interpolation. Micros with enough flash ROM to 
hold a 12 bit wide table aren't that expensive and would be the simplest 
way to do it.

If you use AVR processors you could use an XMEGA with 12 bit DAC and 
enough ROM for your lookup table.

Cheers,
Stew.


Tom Wiltshire wrote:
> I agree with Harry that it depends what you patch it to. The worse-case scenario is VCO pitch or a high-res filter (sine wave VCO effectively). The ear must be more sensitive to pitch changes than volume or timbre. I also agree that fast modulations are (perceived as) smoother than slow ones.
>
> If the envelope is followed by even some fairly basic filtering, the filter has more effect as the steps get shorter, so they actually get *less* noticeable.
> Imagine you've got a filter smoothing the envelope at 1KHz, to give you reasonably fast attack times. If you stretch the attack so it takes a second, those little steps are spikes at a much lower frequency. If you push it up into the audio range, that 1KHz filter is going to be far more effective and will squash them more throughly.
>
> To go back to your question - 8mS between steps is little spikes at a frequency of 125Hz. If you make the filter cutoff low enough to remove those effectively, you're going to lose the ability to produce really snappy envelopes at the other end. But try it and see. I'm always amazed at what you can get away with. There's probably a workable compromise filter freq in there somewhere.
>
> My PIC VCADSR has some minor artifacts at very slow settings. But it is a 10-bit output and uses interpolation to smooth the output as far as possible. This made a significant improvement over the VCLFO, which didn't use any interp. With 8-bit and no interp, I would expect you'll hear the steps unless they're removed by heavy filtering. I've also worked on 12-bit and better envelope designs, and with those no steps are audible whatever you do.
>
> Good luck with it,
> Tom
>
>
> On 15 Feb 2011, at 13:29, Harry Bissell wrote:
>
>   
>> I guess it would depend more on what you had it patched to....
>>
>> If it was driving a VCO and it was quick, you'd never notice but if it was slow
>> you might hear each individual step as a discrete pitch.
>>
>> My experience with the Wiard MiniWave is that fast moving modulations seem
>> more smooth than slow ones, the artifacts are more noticeable
>>
>> Ian has a point, if the steps move into the audio range, they might be ~very~
>> noticeable...
>>
>> H^) harry
>>
>>
>>
>>
>> ----- Original Message -----
>> From: Ian Fritz <ijfritz at comcast.net>
>> To: Matthew Smith <matt at smiffytech.com>, Synth DIY <synth-diy at dropmix.xs4all.nl>
>> Sent: Tue, 15 Feb 2011 07:10:07 -0500 (EST)
>> Subject: Re: [sdiy] Digital ADSR - perceivable staircase?
>>
>> At 11:12 PM 2/14/2011, Matthew Smith wrote:
>>     
>>> If we consider a digitally-generated attack signal as a staircase, at what 
>>> 'width' of step - in other words the time between increments in level - 
>>> would the increase in volume of the system output become noticeable?
>>>
>>> So as to have something to go on, I am considering a maximum A/D/R time of 
>>> 2 seconds so, at maximum time, a 'step' would be about 8ms long, based on 
>>> 256 levels x 256 steps.
>>>
>>> Really wondering what my maximum time can be before the 256 steps become 
>>> obvious.
>>>       
>> Hmmm ... Wouldn't it actually be more noticable at shorter times. (The old 
>> zipper noise problem.)
>>
>> Ian 
>>
>> _______________________________________________
>> Synth-diy mailing list
>> Synth-diy at dropmix.xs4all.nl
>> http://dropmix.xs4all.nl/mailman/listinfo/synth-diy
>>
>> -- 
>> Harry Bissell & Nora Abdullah 4eva
>> _______________________________________________
>> Synth-diy mailing list
>> Synth-diy at dropmix.xs4all.nl
>> http://dropmix.xs4all.nl/mailman/listinfo/synth-diy
>>     
>
> _______________________________________________
> Synth-diy mailing list
> Synth-diy at dropmix.xs4all.nl
> http://dropmix.xs4all.nl/mailman/listinfo/synth-diy
>
>