When I was the R&D teaboy at Wharfedale in the late 70's, I tried stereo in our anechoic chamber; expecting great things w/o pesky room reflections. The results were terrible; poor stereo sound as well as formal localisation which was the reason for the experiment. Both much poorer than in a normal listening room.
Peter Fryer told me that BBC Research had tried the same thing with the same results. They found putting a plywood floor down improved the sound and localisation.[1] MAG knew of this experiment and always advocated speakers close to room boundaries rather than well away as the pseudo prophets would have it. His reasoning was that if early reflections came from the same general direction, they would not confuse. I should also point out that simulating localisation w/o "some" reflections is likely to be inaccurate.[3] There are several reports of poor Ambi demos in Anechoics; including the false prophet Floyd who repeats this ad nauseum. However, he may not have been using a proper Ambisonic Decoder.[2] I'm not convinced of the need for "semi(?) anechoic" listening rooms, ambisonic or otherwise ... except for Joern's "... very convincing renderings of semi-anechoic or free-field conditions. " However, I'd urge that any purpose built listening room or studio have wonky walls, floor & ceiling. Perfectly parallel surfaces have terrible flutter echoes and you end up applying so much treatment to deal with these that your room becomes too dead. Prof Peter Lord, Salford recommended at least 2 degrees wonkiness. Julian Wright, Celestion finally managed to simulate and confirm this in the late 90's when Patrick Macy, PAFEC developed the 1st successful acoustic boundary element (we were Beta testers). Wonky walls give you much more flexibility to choose your room reverb profile > [Richard, I know what you are thinking....yet again, I couldn't help but > mention the Hankel functions...sorry :-) ] Duu...uh! Professor, does " ... substituting the spherical Bessel/Hankel (j_n / h_n) functions with traditional Bessel/Hankel functions (J_n / H_n) ... " for NFC with "line arrays" give a simple Minimum Phase 3dB/8ve filter? My calculator has no big B or H button. [1] I've been trying to find the BBC Research report w/o success. If anyone knows it, please post. It might be mentioned in "Stereophonic image sharpness" - Harwood, WW vol74 1968 p207-11 The Kingswood Warren anechoic was rather bigger than ours. [2] "Is my decoder Ambisonic?" - Heller et al, preprint 7753, AES San Francisco 08. This is BLaH3 and includes "Designing Classic Ambi Decoders for Dummies". [3] "Loudspeaker and the Stereo Image" - Millward, HFN 29, nov84 _______________________________________________ Sursound mailing list Sursound@music.vt.edu https://mail.music.vt.edu/mailman/listinfo/sursound
