"In both cases the sound was coming from seemingly random places, and a number of positions went practically silent." What is needed, not just for you, but for everyone, is a comprehensive set of test files. It may be that your loudspeakers aren't where the system thinks they are (wrong speaker assignments), or it may be that that the decoder is doing the wrong thing. I have more extensive versions of test files, including "with height" like the eight directions files on Ambisonia., featuring the voice of the lovely Haley Jo. I could upload those if anyone would like them. You can then use metering to determine if the specific sounds light up the speaker(s) that they should.
Sent from Yahoo Mail on Android On Tue, Mar 22, 2016 at 1:18 PM, Jörn Nettingsmeier<netti...@stackingdwarves.net> wrote: On 03/22/2016 07:49 PM, Martin Dupras wrote: > Today I tried playback sources in third order Ambisonics on a 8+6+1 > hemispheric speaker array using Reaper. It didn't quite work as > intended so I'm trying to figure out where I've gone wrong. > > I was using the Blue Ripple TOA-Core panner plugin to position the > sound. I understand that Blue Rippler plugins use the Furse-Malham > convention. > > The only decoders that I could find to decode to my specific array > (using coefficients that I calculated using the Ambisonics Decoder > Toolkit) were the Ambix Plug-ins and AmbDec. > > I tried Ambix first, which I understand uses the ACN ordering > convention. I tried re-ordering the channels based on information that > I found here: > https://en.wikipedia.org/wiki/Ambisonic_data_exchange_formats#ACN. > But that didn't really work. > > I then tried to run 16 outputs out of Reaper into Jack, and from Jack > into AmbDec, again using my ADT-calculated coefficients. I understand > that AmbDec uses the Furse-Malham convention, so I would have thought > it was compatible with the output of the Blue Rippler plugins. But > again, that didn't really work well at all. > > In both cases the sound was coming from seemingly random places, and a > number of positions went practically silent. To debug erratic panning behaviour, start with first order, verify, and work your way up from there. To make sure the error is not in your calculated coefficients, try to use a known-good decoding matrix that approximates what you have, before feeding in your optimized one. With Ambdec, weird things can happen if you connect several ins at the same time using some graphical client, because the order shown in for example qjackctl is lexical, whereas the internal order is different. So you will end up with garbled connections. You can feed an ACN signal into ambdec succesfully if you choose SN3D input scaling _and_ manually connect the inputs correctly. This wikipedia article https://en.wikipedia.org/wiki/Ambisonic_data_exchange_formats has some information on that, and other pitfalls when interfacing different formats. All best, Jörn -- Jörn Nettingsmeier Lortzingstr. 11, 45128 Essen, Tel. +49 177 7937487 Meister für Veranstaltungstechnik (Bühne/Studio) Tonmeister VDT http://stackingdwarves.net _______________________________________________ Sursound mailing list Sursound@music.vt.edu https://mail.music.vt.edu/mailman/listinfo/sursound - unsubscribe here, edit account or options, view archives and so on. -------------- next part -------------- An HTML attachment was scrubbed... URL: <https://mail.music.vt.edu/mailman/private/sursound/attachments/20160323/e78a01ce/attachment.html> _______________________________________________ Sursound mailing list Sursound@music.vt.edu https://mail.music.vt.edu/mailman/listinfo/sursound - unsubscribe here, edit account or options, view archives and so on.
