1	History and current state of the war between the two main approaches to surround sound: discrete speaker feeds versus hierarchical matrix Angelo Farina Dipartimento di Ingegneria Industriale, Università di Parma, Via delle Scienze 181/A Parma, 43100 ITALY HTTP://pcfarina.eng.unipr.it - mail: angelo.farina@unipr.it
2	Who is where ?
3	“They are here”
4	“You are there”
5	“They are here”
6	Discrete vs. Matrix Discrete speaker feeds Quad-8 ---------------------- OCT-Surround Williams MMA Fukada-Tree Hamasaki-Square INA-5 Decca Tree
7	Discrete vs. Matrix Discrete speaker feeds
8	Discrete vs. Matrix Discrete speaker feeds
9	But more often the approach is this:
10	Surround Panning Laws
11	Surround Panning Laws
12	Which type of surround is being emulated by the spot mikes + panning method? At a first glance, it is discrete, as the signals are usually delivered ready for the loudspeakers But, as the panner usually adjustes amplitude only, the signals are in reality much more similar to those created by a hierarchical matrix
13	So, what is the real difference between discrete and matrix? Whatever the method employed for recording or synthesizing the surround material, every loudspeaker can always be thought to be fed with a signal coming from one or more microphones (virtual or real)
14	So, what is the real difference between the two systems? In a Hierarchical Matrix system, the virtual microphones are always COINCIDENT, albeit some types of “Logic” decoders can add delays, which emulates virtual microphones more far from the source In a Discrete system, the microphones can be either real or virtual, and they can be SPACED providing some potential perceptual advantages (and consequent risks)
15	Historical review Here we show some milestones in the development of surround sound production Both discrete systems and hierarchically matrixed methods are described, as they interleaved during the years
16	Timeline Fantasia – Fantasound (4 ch.)
17	Timeline
18	Timeline
19	Solid Music (1934)
20	Fantasia - Fantasound (1940)
21	Fantasia - Fantasound (1940)
22	Cinerama (1953) – Cinemascope (1954)
23	Quadraphony (1970)
24	Quad matrixed QS, SQ etc. (1972)
25	Quad matrixed SQ (1972)
26	Quad matrixed SQ (1972)
27	Quad matrixed SQ (1972)
28	Ambisonics (1973)
29	Ambisonics theory
30	Ambisonics theory
31	Psychoacoustic criterions At low frequency (<500 Hz), the gains are adjusted so that the pressure and particle velocity signals are carefully reconstructed at the center of the array, making use of “push-pull” effects obtained generating out-of-phase signals At higher frequencies, the gains are adjusted so that the Sound Intensity vector recreated at the center of the array is as close as possible to the original one Shelf filters are employed for making the gains frequency-dependent
32	Capturing the Ambisonics signals A tetrahedrical microphone probe was developed by Gerzon and Craven, originating the Soundfield microphone
33	Soundfield microphones
34	Synthetic Ambisonics signals
35	Dolby Stereo / Surround (1975)
36	Dolby Surround
37	Dolby Surround
38	Dolby Pro Logic (1987)
39	Dolby Digital – AC3 (1992)
40	DTS (1993) An optical timecode on the film synchronizes a CD player loaded with a DTS-encoded audio CD
41	Discrete microphone arrays The new discrete digital formats allowed for true discrete microphone systems
42	Discrete microphone arrays Several “standard” setups were developed
43	Discrete microphone arrays Several “standard” setups were developed
44	Software for array design Image Assistant by H. Wittek
45	Software for array design Image Assistant by H. Wittek
46	Software for array design Image Assistant by H. Wittek
47	Software for array design Image Assistant by H. Wittek
48	Software for array design Image Assistant by H. Wittek
49	Software for array design SOS MMAD by Mike Williams
50	Software for array design SOS MMAD by Mike Williams
51	Software for array design SOS MMAD by Mike Williams
52	Software for array design SOS MMAD by Mike Williams
53	Software for array design SOS MMAD by Mike Williams
54	A compact self-contained array Holophone “pear-shaped” arrays
55
56
57	Ambiophonics 3D (2001) Layout
58	Ambiophonics (2001)
59	The Ambiophonics Institute
60	The Ambiophonic Institute
61	THX 10.2 Surround (2001)
62	NHK 22.2 Surround (2005)
63	NHK 22.2 Surround (2005)
64	The 2+2+2 system (2007)
65	Schoeps Double MS (2007)
66	Schoeps Double MS (2007)
67	Schoeps Double MS (2007)
68	Schoeps Double MS (2007) In practice, the double-MS system is another way of recording the horizontal components of the Ambisonics signals WXY (B-format), and decoding them with the “virtual microphone” approach Schoeps does not advertises this method as Ambisonics based, and relies on the superior quality of their capsules for beating the concurrent Soundfield microphone system
69	High Order Ambisonics (2010)
70	Software for 4th-order Ambisonics 32 ch. VST plugins developed by France Telecom
71	Software for 4th-order Ambisonics 32 ch. VST plugins developed by France Telecom
72
73
74	Comparative experiments We present here some sound samples coming from three experiments The first was the ORF Seminar, Wien 2001 The 2^nd was Verdi Projekt at TU Ilmenau in 2002 The third is currently undergoing at RAI Research Center in Turin, Italy
75	The ORF seminar (Vienna)
76	The ORF seminar
77	The ORF Seminar
78	The ORF seminar
79	The ORF seminar
80	The ORF seminar
81	The ORF seminar
82	The ORF seminar
83	The ORF seminar
84	The ORF seminar
85	The ORF seminar
86	The ORF seminar
87	The ORF seminar
88	The ORF seminar
89	The ORF seminar
90	The Verdi Projekt
91	The Verdi Projekt
92	The Verdi Projekt
93	The Verdi Projekt
94	The Verdi Projekt
95	The Verdi Projekt: Soundfield
96	The RAI – CRIT project
97	The RAI – CRIT project
98	The RAI – CRIT project
99	The RAI – CRIT project
100	DEMO sound samples