On Thu, Apr 25, 2013 at 7:50 PM, Sampo Syreeni <de...@iki.fi> wrote: > On 2013-04-25, Robert Greene wrote > > How does anyone think that this is enough to record a soundfield in the >> neighborhodd of a point? >> > > It necessarily is if you think purely about the pressure field. There the > pointwise pressure plus three velocity components always dictate the > close-by pressure gradient as well. But while they do so, they don't > dictate the velocity field. That can be freely added on subject to the > condition that it agrees at the coincident point where you chose to put > your Soundfield in. If you do the math, that leaves you three full degrees > of freedom in velocity undetermined outside of your measuring point, and > the classical ambisonic decoding solution then takes full and unabashed > advantage of that latitude. >
In slightly more layman's terms, recall that above 400Hz or so, Ambisonics switches from attempting to recreate the pressure and velocity, to simply concentrating the energy flux (and hence transients) in the direction of the source -- a 'velocity' decode vs. a 'max-rE' decode. In fact, if you don't adjust the decoding in this way, you get nasty comb filtering artifacts and confusing localization cues as you move your head. This switchover corresponds to frequency regime where ITD cues become unreliable and we start to use ILD cues. Is it as good as in natural hearing? No, but its arguably the best you can do with four channels, if you want all directions to be treated equally. (and I acknowledge that not everyone feels that is desirable) In BLaH5 aka "Why Ambisonics Works" we looked at how well (or not) Ambisonics reproduces ITD and ILD cues. (get it here: http://www.ai.sri.com/~heller/ambisonics/) At first order, the best you can do is concentrate the energy is as much as you can into the hemisphere centered around the source direction, so from a virtual microphone perspective a supercardiod (max front-hemisphere/back-hemisphere ratio). There's still a back lobe, but it is much smaller. One of the reasons Ambisonics starts working much better at third order is that the side lobes get much smaller and the energy is more focused in the correct direction, roughly correlating with our directional acuity. Zotter and Frank have a nice graphical interpretation of rE in Fig. 7 of their JAES paper. (It's worth looking at, regardless of what you think about hybrid ambi/vbap schemes). One other thing, Sampo mentioned that four speakers is the maximum that should be used for first-order horizontal playback. Most people who have done the listening tests, acknowledge that a six-speaker hexagon is an improvement over a square. Some people have reported good results with 8, others report spectral contamination. Mathematical analysis supports the latter. Aaron Heller Menlo Park, CA US -------------- next part -------------- An HTML attachment was scrubbed... URL: <https://mail.music.vt.edu/mailman/private/sursound/attachments/20130426/39ef0da0/attachment.html> _______________________________________________ Sursound mailing list Sursound@music.vt.edu https://mail.music.vt.edu/mailman/listinfo/sursound
