Thanks folks -- I removed the LPF which did seem to have an effect on the tuning but didn't entirely correct it. I think I'll take a step back from this code and learn a bit more from Romain's example instead. I was initially scared by the length of it but I now see it consists mostly of helpful comments!
All the best, Rich On Wed, Oct 14, 2020 at 8:30 AM Oliver Larkin <olilar...@googlemail.com> wrote: > I think it’s the type of low pass filter that you have inserted that is > the problem. Si.smooth Is an iir, and it’s phase response will cause issues > with the tuning I think. There are other variations of low pass filter that > you can use that will be better in this respect > > On Wed, 14 Oct 2020 at 07:40, Romain Michon <rmnmic...@gmail.com> wrote: > >> Ah, that makes more sense. First thing first, nMax should be a fixed >> number and you're lucky that your code actually works ;). As for the >> tuning, that's one of the things with waveguide models: their tuning often >> has to be adjusted "by hand." This is due to the fact that extra delay is >> added by the various filters in the algorithm. So your best bet here is to >> adjust the tuning using an external oscillator and comparing the produced >> pitch to it. >> >> Cheers, >> >> Romain >> >> On Tue, Oct 13, 2020 at 11:44 PM Rich Cochrane <rich.cochr...@bigi.org.uk> >> wrote: >> >>> Thanks to both of you for replying -- it was probably unhelpful of me >>> not to post the whole code, which I'm attaching here. The whole thing does >>> produce a note, but the pitch is off (octaves are very flat). >>> However, asking you to pick my altered version of someone else's code isn't >>> very reasonable. >>> >>> Romain -- I'll look at the example you suggested, I imagine that's a >>> better starting-point anyway. >>> >>> Cheers, >>> >>> Rich >>> >>> On Tue, Oct 13, 2020 at 3:10 PM Romain Michon <rmnmic...@gmail.com> >>> wrote: >>> >>>> Hi Rich, >>>> >>>> This indeed won't make any sound. The waveguide object of the physical >>>> modeling library in Faust is just 2 delay lines in parallel : it's just a >>>> simple waveguide without terminations. For a simple string example, have a >>>> look at the openString function: >>>> https://github.com/grame-cncm/faustlibraries/blob/master/physmodels.lib#L471 >>>> >>>> Cheers, >>>> >>>> Romain >>>> >>>> On Sat, Oct 10, 2020 at 4:05 AM Julius Smith <julius.sm...@gmail.com> >>>> wrote: >>>> >>>>> Hi Rich, >>>>> >>>>> I am not familiar with this code, but thanks for the pointers. >>>>> >>>>> You have the right idea. ma.SR/freq needs to equal the round-trip >>>>> delay up and down the waveguide, adding up all delay line lengths, all >>>>> filter delays, and the feedback delay from the Faust compiler (there's >>>>> always at least one sample of that in any closed loop). >>>>> >>>>> Cheers, >>>>> Julius >>>>> >>>>> On Fri, Oct 9, 2020 at 6:55 AM Rich Cochrane < >>>>> rich.cochr...@bigi.org.uk> wrote: >>>>> >>>>>> Hello, >>>>>> >>>>>> I'm trying to learn Faust's physical modeling features "by >>>>>> experimentation" so am blundering around a bit. >>>>>> >>>>>> I found the last example on this page comprehensible: >>>>>> "Waveguide-Strings in Faust": >>>>>> https://hvc.berlin/Sound_Synthesis/Physical_Modeling/physical-modeling-faust-examples/ >>>>>> >>>>>> I wanted to be able to play it with a keyboard. I thought this might >>>>>> work: >>>>>> >>>>>> segment(pos) = waveguide(nMax,n) >>>>>> with{ >>>>>> nMax = ma.SR/freq; >>>>>> n = pos * nMax; >>>>>> }; >>>>>> >>>>>> Where freq is the usual frequency value triggered by a MIDI keyboard. >>>>>> But the result is a long way out of tune. >>>>>> >>>>>> I tried "nMax = freq : pm.f2l : pm.l2s;" instead but this appears to >>>>>> have the same net effect -- at least, it's also out of tune. >>>>>> >>>>>> What's the proper way to do this? Or is this base code not the right >>>>>> starting point for this? >>>>>> >>>>>> Cheers, >>>>>> >>>>>> Rich >>>>>> >>>>>> >>>>>> _______________________________________________ >>>>>> Faudiostream-users mailing list >>>>>> Faudiostream-users@lists.sourceforge.net >>>>>> https://lists.sourceforge.net/lists/listinfo/faudiostream-users >>>>>> >>>>> >>>>> >>>>> -- >>>>> "Anybody who knows all about nothing knows everything" -- Leonard >>>>> Susskind >>>>> _______________________________________________ >>>>> Faudiostream-users mailing list >>>>> Faudiostream-users@lists.sourceforge.net >>>>> https://lists.sourceforge.net/lists/listinfo/faudiostream-users >>>>> >>>> >>>> >>>> -- >>>> >>>> Romain Michon >>>> +33 (0)7 67 39 72 40http://grame.fr/~michon >>>> >>>> >> >> -- >> >> Romain Michon >> +33 (0)7 67 39 72 40http://grame.fr/~michon >> >> _______________________________________________ >> Faudiostream-users mailing list >> Faudiostream-users@lists.sourceforge.net >> https://lists.sourceforge.net/lists/listinfo/faudiostream-users >> >
_______________________________________________ Faudiostream-users mailing list Faudiostream-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/faudiostream-users
