[sdiy] DCO Question

[Olivier Gillet]

> i.e. spectral purity

a digital system will never be able to reproduce a perfect sawtooth
anyway. only a band-limited one.

you're just choosing the lesser of two evils: eliminate all harmonics
above nyquist, instead of folding them back.

> I don't want to calculate polynomials in floating point on the fly so I think I'll pre-calculate a table. Flash memory is cheaper than cycles.

flash memory access has a cost too - table offset calculation, the
actual vldr, and some nasty processor-specific side effects (messing
with flash prefetching/caching/DMA access). A couple of vmul.f32 and
vfma.f32 are not that bad for low order polynomials.

don't assume, benchmark!

On Thu, Nov 2, 2017 at 5:15 PM, Tim Ressel <timr at circuitabbey.com> wrote:
> Ah, so if I am understanding polyBlep, one sacrifices waveform linearity
> i.e. spectral purity for jitter reduction. The last few samples get morphed
> so that the accumulator maxes out every cycle. In this case the saw wave
> will nose over a bit at the top. Another way of saying it is the fraction
> left at the top is amortized over a few samples. Since I am not trying to
> hit exact frequencies this should work nice and easy.
>
> I don't want to calculate polynomials in floating point on the fly so I
> think I'll pre-calculate a table. Flash memory is cheaper than cycles.
>
> Thanks!
>
> -t-i-m-b-o-
>
>
>
> On 11/2/2017 8:39 AM, Eric Brombaugh wrote:
>>
>> Jitter / Aliasing - it's both!
>>
>> Eliminating that "feature" of digital oscillators is the primary
>> motivation behind the proliferation of antialiased oscillator algorithms and
>> the proliferation thereof. Check out all the papers on BLEP, MinBLEP,
>> PolyBLEP, DPW, etc.
>>
>> Eric
>>
>> On 11/02/2017 08:26 AM, Tim Ressel wrote:
>>>
>>> Hi,
>>>
>>> I am making a digital oscillator (DCO) and was wondering if there is a
>>> way to reduce the jitter one gets at higher frequencies. I understand it
>>> comes from round off error (for lack of a better term): as the accumulator
>>> reaches the top of the range the remaining amount is less than the value
>>> being added, and that fraction changes every cycle. This causes jitter, or
>>> is it aliasing? Anyway, is there a clever way to deal with it?
>>>
>>> Thanks!
>>>
>>
>> _______________________________________________
>> Synth-diy mailing list
>> Synth-diy at synth-diy.org
>> http://synth-diy.org/mailman/listinfo/synth-diy
>>
>
> --
> --Tim Ressel
> Circuit Abbey
> timr at circuitabbey.com
>
> _______________________________________________
> Synth-diy mailing list
> Synth-diy at synth-diy.org
> http://synth-diy.org/mailman/listinfo/synth-diy