On Tue, Jun 10, 2014 at 07:40:14AM +0100, Curtis Alcock wrote: > 1. Just out of interest, when you "upsample" to Third Order Ambisonics, > does that mean simulating the missing information?
In a sense, yes. What happens is that the input signal is divided in a large number of narrow frequency bands, and for each of these the algorithm tries to find out the directions. Then each band is panned to either to a set of discrete speakers or re-enconded in TOA. The way is this works is by solving a set of equations that describe how two sources are encoded into first order AMB. For more than two sources the equations become ambiguous - there is no unique solution. The algorithm works well if you have one or two clearly defined sources in each frequency band, but it fails in complex ways in case this condition is not satisfied. If your studio is to be used for hearing research you should probably ask yourself if you want this sort of processing - it sort of 'interpretes' the spatial information in a way that is not at all related to how our brains do it. > Is it possible to record directly in TOA? It is sort of possible using the 32-channel EigenMic and some processing. But the resulting TOA is not complete: the higher order signals have a rather limited frequency range. Which in turn means that a normal decoder will not handle them correctly. It is possible to correct for this up to some point, and in that case this method can produce usable results. Some of my collegues here used it to produce a TOA sounscape of the city of Parma last year - the result is quite impressive. Normally HOA is recorded by panning individual sources and adding AMB encoded reverb or room acoustics. > 2. Richard, you mention using the TOA Harpex VST plugin to create > full 3D 16 channel TOA. Excuse my ignorance, but does 16 channels > equate to 16 speakers? You need more than 16 speakers for full 3D third order. Good layouts (without preferred directions or gaps) have 20-25 speakers. > 4. Are AmbDec and Rapture 3D Advanced equivalent? As far as I know, no. Ambdec is a full-featured decoder app for (currently) up to 3rd order and 36 speakers. Full-featured means dual-band and having near-field compensation. But it is *only* a decoder, you have to give it the decoding matrices - it will not compute them for you. I can provide hand-optimised decoder matrices for Ambdec users. If I'm not mistaken 3D-Advanced includes code to compute decoding matrices for any speaker layout. Note that computing these matrices for arbitrary non-regular speaker layouts is still some form of 'black art'. There are a number of algorithms that can produce reasonable results, but none of them produce optimal decoders. Ciao, -- FA A world of exhaustive, reliable metadata would be an utopia. It's also a pipe-dream, founded on self-delusion, nerd hubris and hysterically inflated market opportunities. (Cory Doctorow) _______________________________________________ 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.
