Hi Loïc, I’ll reply here since there might still be some Faust interest but yes we should go off-list for further discussion as necessary.
Your comments on crosstalk are helpful but I don’t quite understand why delay effects are problematic. My impression from your comments is that crosstalk could be beneficially improved. I had not considered that the crosstalk characteristics might change depending on player technique such as string muting. Hmmm…. Yes, I listened to the hexfuzz demo on soundcloud. Pretty cool! I co-developed crosstalk cancellation technology a number of years ago in connection with loudspeakers and listeners. This is a linear system which is why I asked about the linearity of the hex pickup. I generalized the technique from 2x2 (two loudspeakers and two ears) to MxN. The hex pickup would be of course 6x6 but might be simplified if for example only adjacent strings had significant crosstalk. In any case I would expect the hex pickup problem to be much easier than acoustic propagation involving heads. I will reference a paper I wrote for the Journal of the Audio Engineering Society. If you are interested I can send you a copy. Here is the link to the abstract: https://www.aes.org/e-lib/browse.cfm?elib=7888 Jerry > On Dec 7, 2020, at 4:55 PM, Loïc Reboursiere <loicreboursi...@gmail.com> > wrote: > > Hello Jerry, > > Thanks for your feedback. > > The crosstalk is ok but can become somehow problematic. > The pickup technology here has an effect. Roland guitar like the LGXT embeds > hexaphonic piezo pickups. You'll have of course sympathetic vibrations that > will happen but also some harmonics that are transferred through the bridge, > no matter how much you mute the strings you don't want to hear. > So it becomes problematic when mixing high gain distortion with delay for > exemple. If you have distortion on the low E and delay on the high e for > example, you might hear some resonances due to the delay even if you don't > play the high e. > My acoustical knowledge is pretty limited, so this remark is only the result > of experiments. > When it's only distortion, there's no problem, but when you're adding other > effects, yeah that need to be taken into account. Do you have any links on > the technique you mention to erase the crosstalk. Even if I understand the > logic I've never used such a thing. > > Are you talking about the hexfuzz demo on soundcloud ? > > As for the setup, I'm using a homemade preamp with § adjustable gains, RME > Fireface, Laptop. > The standalone hardware porting is in process right now. > > If you have further questions maybe we should continue off-list as it's not > Faust specific ? > > Cheers, > > Loïc > > Le dim. 6 déc. 2020 à 04:45, <list_em...@icloud.com > <mailto:list_em...@icloud.com>> a écrit : > >> On Nov 2, 2020, at 3:15 AM, Loïc Reboursiere <loicreboursi...@gmail.com >> <mailto:loicreboursi...@gmail.com>> wrote: >> >> Hi, >> >> I'm working on hexaphonic guitar (electrified guitar with an hexaphonic >> pickup, i.e, one pickup per string) and hexaphonic effects. >> By following one of the Faust lockdown webinar I manage to get an hexaphonic >> delay with the following code : >> bounce(d) = @(vslider("delay %d[style:knob]", 4410, 0, 44100, 441)) : >> *(vslider("feedback %d[style:knob]", 0.1, 0, 1, 0.01)); >> monoecho(d) = +~bounce(d); >> multiecho(N) = hgroup("hexdelay", par(i, N, monoecho(i))); >> >> process = multiecho(6); >> >> I'd like to embed the GUI in an HTML website, but when I try to export it in >> any of the available web formats and launch the generated HTML file, nothing >> appears except for the block size (I guess) menu. I've tried on different >> browsers but nothing changes. I've also tried with Faust examples and it >> doesn't work either. >> >> What am I missing ? Does it have to do with the "poly voices" option ? Or is >> it be being a newbie to javascript and how to test javascript code ? >> >> Thanks for your answers, > > Loïc: > > I just saw this post of yours and my questions are off-topic. > > I have been thinking for a long time of trying a hexaphonic pickup on a > guitar and using something like Faust to process the outputs. One area that > interests me is eliminating the intermodulation distortion between strings > when using a nonlinear effect so that non-major chords played through a > fuzz-type device don’t turn into nasty near-white noise. > > So the obvious question is: what is the crosstalk between the six channels, > or put in more audio-like terms, what is the channel separation between the > six? Also, is the pickup linear? If the pickup is linear then the crosstalk > could be described by a symmetric 6x6 matrix of transfer functions with ones > on the diagonal and it would be possible to build a crosstalk canceller to > reduce the crosstalk if it is too high. > > I listened to your HexFuzz demo and if I am interpreting what I hear > correctly the channel separation is pretty good since it seems that each > string’s fuzz distortion does not mix much with the other strings. The effect > that I hear is awesome! > > Also, what do you use for a sound card or external digitizer? Do you use a > computer or have you ported to standalone hardware? > > Thanks for any comments at all. > > Jerry > > > -- > ------------------------------------------------------------ > http://www.medicationtime.net <http://www.medicationtime.net/> > http://soundcloud.com/MedicationTime <http://soundcloud.com/MedicationTime> > http://www.vimeo.com/user2896549 <http://www.vimeo.com/user2896549> > http://vimeo.com/user2896549/videos/all > <http://vimeo.com/user2896549/videos/all> > > http://www.numediart.org <http://www.numediart.org/> > ------------------------------------------------------------
_______________________________________________ Faudiostream-users mailing list Faudiostream-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/faudiostream-users
