On 05/03/2011 03:04 AM, Aaron Heller wrote:
I believe the reasoning is that when there are many speakers producing
almost the same signal, you get more comb filtering effects when
moving out of the sweet spot. This is discussed in
Solvang, Audun. "Spectral Impairment of Two-Dimensional Higher Order
Ambisonics" JAES Volume 56 Issue 4 pp. 267-279; April 2008.
most interesting paper, thanks for the pointer! i won't pretend i
understand the math in detail, but the simulations are quite enlightening.
however the paper doesn't specify the decoder used. I've seen many
reports of comb-filter effects and then found that "matching" (aka
velocity, basic) decoders were being used at HF. (we discuss this in
BLaH3). In fact, the only time I've ever heard comb filtering in
Ambi playback is with incorrect decoders.
that could explain why the IEM cube performs particularly oddly with
first-order material, even though its HOA qualities are quite stunning.
btw, the captions to the plots in the solvang paper seem to indicate
"basic" decoding.
I will say that when Eric, Richard, and I set up a demo during the
2008 AES in San Francisco, using the 24-speaker hemisphere array at
"the Bubble", there was a distinct HF dip at the sweet spot, but we
didn't have precise info about the locations of the speakers and
didn't have a lot of time to make measurements or tweak the decoder
before the demo. Moving 10-20cm away from the sweet spot in any
direction restored the HF response, but there were no comb filtering
effects. The configuration was 3 rings of eight speakers, with some
of the speaker behind a projection screen.
as mentioned before, the one time i worked on stacked rings, i was quite
amazed at the perfectly smooth behaviour - no phasing, and pretty stable
timbre when moving.
--
Jörn Nettingsmeier
Lortzingstr. 11, 45128 Essen, Tel. +49 177 7937487
Meister für Veranstaltungstechnik (Bühne/Studio)
Tonmeister VDT
http://stackingdwarves.net
_______________________________________________
Sursound mailing list
[email protected]
https://mail.music.vt.edu/mailman/listinfo/sursound
