1 <HTML><HEAD><TITLE>xiph.org: Ogg Vorbis documentation</TITLE>
2 <BODY bgcolor="#ffffff" text="#202020" link="#006666" vlink="#000000">
3 <nobr><img src="white-ogg.png"><img src="vorbisword2.png"></nobr><p>
5 <h1><font color=#000070>
6 Ogg Vorbis I format specification: codec setup and packet decode
9 <em>Last update to this document: October 15, 2002</em><br>
13 This document serves as the top-level reference document for the
14 bit-by-bit decode specification of Vorbis I. This document assumes a
15 high-level understanding of the Vorbis decode process, which is
16 provided in the document <a href="vorbis-spec-intro.html">Ogg Vorbis I
17 format specification: introduction and description</a>. <a
18 href="vorbis-spec-bitpack.html">Ogg Vorbis I format specification:
19 bitpacking convention</a> covers reading and writing bit fields from
20 and to bitstream packets.<p>
22 <h1>Header decode and decode setup</h1>
24 A Vorbis bitstream begins with three header packets. The header
25 packets are, in order, the identification header, the comments header,
26 and the setup header. All are required for decode compliance. An
27 end-of-packet condition during decoding the first or third header
28 packet renders the stream undecodable. End-of-packet decoding the
29 comment header is a non-fatal error condition.<p>
31 <h2>Common header decode</h2>
33 Each header packet begins with the same header fields
36 1) [packet_type] : 8 bit value
37 2) 0x76, 0x6f, 0x72, 0x62, 0x69, 0x73: the characters 'v','o','r','b','i','s' as six octets
40 Decode continues according to packet type; the identification header
41 is type 1, the comment header type 3 and the setup header type 5
42 (these types are all odd as a packet with a leading single bit of '0'
43 is an audio packet). The packets must occur in the order of
44 identification, comment, setup.
46 <h2>Identification Header</h2>
48 The identification header is a short header of only a few fields used
49 to declare the stream definitively as Vorbis, and provide a few externally
50 relevant pieces of information about the audio stream. The
51 identification header is coded as follows:<p>
54 1) [vorbis_version] = read 32 bits as unsigned integer
55 2) [audio_channels] = read 8 bit integer as unsigned
56 3) [audio_sample_rate] = read 32 bits as unsigned integer
57 4) [bitrate_maximum] = read 32 bits as signed integer
58 5) [bitrate_nominal] = read 32 bits as signed integer
59 6) [bitrate_minimum] = read 32 bits as signed integer
60 7) [blocksize_0] = 2 exponent (read 4 bits as unsigned integer)
61 8) [blocksize_1] = 2 exponent (read 4 bits as unsigned integer)
62 9) [framing_flag] = read one bit
65 <tt>[vorbis_version]</tt> is to read '0' in order to be compatible
66 with this document. Both <tt>[audio_channels]</tt> and
67 <tt>[audio_sample_rate]</tt> must read greater than zero. Allowed final
68 blocksize values are 64, 128, 256, 512, 1024, 2048, 4096 and 8192 in
69 Vorbis I. <tt>[blocksize_0]</tt> must be less than or equal to
70 <tt>[blocksize_1]</tt>. The framing bit must be nonzero. Failure to
71 meet any of these conditions renders a stream undecodable.<p>
73 The bitrate fields above are used only as hints. The nominal bitrate
74 field especially may be considerably off in purely VBR streams. The
75 fields are meaningful only when greater than zero.<p>
76 <ul><li>All three fields set to the same value implies a fixed rate, or tightly bounded, nearly fixed-rate bitstream
77 <li>Only nominal set implies a VBR or ABR stream that averages the nominal bitrate
78 <li>Maximum and or minimum set implies a VBR bitstream that obeys the bitrate limits
79 <li>None set indicates the encoder does not care to speculate.
83 <h2>Comment Header</h2>
85 Comment header decode and data specification is covered in <a
86 href="v-comment.html">Ogg Vorbis I format specification: comment field
87 and header specification</a>.
91 Vorbis codec setup is configurable to an extreme degree:<p>
93 <img src="components.png"><p>
95 The setup header contains the bulk of the codec setup information
96 needed for decode. The setup header contains, in order, the lists of
97 codebook configurations, time-domain transform configurations
98 (placeholders in Vorbis I), floor configurations, residue
99 configurations, channel mapping configurations and mode
100 configurations. It finishes with a framing bit of '1'. Header decode
101 proceeds in the following order:<p>
106 <li><tt>[vorbis_codebook_count]</tt> = read eight bits as unsigned integer and add one
107 <li>Decode <tt>[vorbis_codebook_count]</tt> codebooks in order as defined
108 in <a href="vorbis-spec-codebook.html">the codebook specification
109 document</a>. Save each configuration, in order, in an array of
110 codebook configurations <tt>[vorbis_codebook_configurations]</tt>.
113 <h3>time domain transforms</h3>
115 These hooks are placeholders in Vorbis I. Nevertheless, the
116 configuration placeholder values must be read to maintain bitstream
120 <li><tt>[vorbis_time_count]</tt> = read 6 bits as unsigned integer and add one
121 <li>read <tt>[vorbis_time_count]</tt> 16 bit values; each value should be zero. If any value is nonzero, this is an error condition and the stream is undecodable.
126 Vorbis uses two floor types; header decode is handed to the decode
127 abstraction of the appropriate type.
130 <li><tt>[vorbis_floor_count]</tt> = read 6 bits as unsigned integer and add one
131 <li>For each <tt>[i]</tt> of <tt>[vorbis_floor_count]</tt> floor numbers:
133 <li>read the floor type: vector <tt>[vorbis_floor_types]</tt> element <tt>[i]</tt> = read 16 bits as unsigned integer
134 <li>If the floor type is zero, decode the floor configuration as defined in <a href="vorbis-spec-floor0.html">the floor type 0 specification document</a>; save this configuration in slot <tt>[i]</tt> of the floor configuration array <tt>[vorbis_floor_configurations]</tt>.
135 <li>If the floor type is one, decode the floor configuration as defined in <a href="vorbis-spec-floor1.html">the floor type 1 specification document</a>; save this configuration in slot <tt>[i]</tt> of the floor configuration array <tt>[vorbis_floor_configurations]</tt>.
136 <li>If the the floor type is greater than one, this stream is undecodable; ERROR CONDITION
142 Vorbis uses three residue types; header decode of each type is identical.
145 <li><tt>[vorbis_residue_count]</tt> = read 6 bits as unsigned integer and add one
146 <li>For each of <tt>[vorbis_residue_count]</tt> residue numbers:
148 <li>read the residue type; vector <tt>[vorbis_residue_types]</tt> element <tt>[i]</tt> = read 16 bits as unsigned integer
149 <li>If the residue type is zero, one or two, decode the residue configuration as defined in <a href="vorbis-spec-res.html">the residue specification document</a>; save this configuration in slot <tt>[i]</tt> of the residue configuration array <tt>[vorbis_residue_configurations]</tt>.
150 <li>If the the residue type is greater than two, this stream is undecodable; ERROR CONDITION
156 Mappings are used to set up specific pipelines for encoding
157 multichannel audio with varying channel mapping applications. Vorbis I
158 uses a single mapping type (0), with implicit PCM channel mappings.<p>
161 <li><tt>[vorbis_mapping_count]</tt> = read 6 bits as unsigned integer and add one<p>
162 <li>For each <tt>[i]</tt> of <tt>[vorbis_mapping_count]</tt> mapping numbers:<p>
164 <li>read the mapping type: 16 bits as unsigned integer. There's no reason to save the mapping type in Vorbis I.<p>
165 <li>If the mapping type is nonzero, the stream is undecodable<p>
166 <li>If the mapping type is zero:<p>
167 <ol> <li>read 1 bit as a boolean flag<p>
168 <ol><li>if set, <tt>[vorbis_mapping_submaps]</tt> = read 4 bits as unsigned integer and add one<p>
169 <li>if unset, <tt>[vorbis_mapping_submaps]</tt> = 1<p>
171 <li>read 1 bit as a boolean flag<p>
172 <ol><li>if set, square polar channel mapping is in use:<p>
173 <ol><li><tt>[vorbis_mapping_coupling_steps]</tt> = read 8 bits as unsigned integer and add one<p>
174 <li>for <tt>[j]</tt> each of <tt>[vorbis_mapping_coupling_steps]</tt> steps:<p>
176 <li>vector <tt>[vorbis_mapping_magnitude]</tt> element <tt>[j]</tt>= read <a href="helper.html#ilog">ilog</a>([audio_channels] - 1) bits as unsigned integer<p>
177 <li>vector <tt>[vorbis_mapping_angle]</tt> element <tt>[j]</tt>= read <a href="helper.html#ilog">ilog</a>([audio_channels] - 1) bits as unsigned integer<p>
178 <li>the numbers read in the above two steps are channel numbers representing the channel to treat as magnitude and the channel to treat as angle, respectively. If for any coupling step the angle channel number equals the magnitude channel number, the magnitude channel number is greater than <tt>[audio_channels]</tt>-1, or the angle channel is greater than <tt>[audio_channels]</tt>-1, the stream is undecodable.<p>
181 <li>if unset, <tt>[vorbis_mapping_coupling_steps]</tt> = 0
183 <li>read 2 bits (reserved field); if the value is nonzero, the stream is undecodable<p>
184 <li>if <tt>[vorbis_mapping_submaps]</tt> is greater than one, we read channel multiplex settings. For each <tt>[j]</tt> of <tt>[audio_channels]</tt> channels:<p>
185 <ol><li>vector <tt>[vorbis_mapping_mux]</tt> element <tt>[j]</tt> = read 4 bits as unsigned integer<p>
186 <li>if the value is greater than the highest numbered submap (<tt>[vorbis_mapping_submaps]</tt> - 1), this in an error condition rendering the stream undecodable<p>
188 <li>for each submap <tt>[j]</tt> of <tt>[vorbis_mapping_submaps]</tt> submaps, read the floor and residue numbers for use in decoding that submap:
189 <ol><li>read and discard 8 bits (the unused time configuration placeholder)<p>
190 <li>read 8 bits as unsigned integer for the floor number; save in vector <tt>[vorbis_mapping_submap_floor]</tt> element <tt>[j]</tt><p>
191 <li>verify the floor number is not greater than the highest number floor configured for the bitstream. If it is, the bitstream is undecodable<p>
192 <li>read 8 bits as unsigned integer for the residue number; save in vector <tt>[vorbis_mapping_submap_residue]</tt> element <tt>[j]</tt><p>
193 <li>verify the residue number is not greater than the highest number residue configured for the bitstream. If it is, the bitstream is undecodable<p>
197 <li>save this mapping configuration in slot <tt>[i]</tt> of the mapping configuration array <tt>[vorbis_mapping_configurations]</tt>.
206 <li><tt>[vorbis_mode_count]</tt> = read 6 bits as unsigned integer and add one<p>
207 <li>For each of <tt>[vorbis_mode_count]</tt> mode numbers:<p>
209 <li><tt>[vorbis_mode_blockflag]</tt> = read 1 bit<p>
210 <li><tt>[vorbis_mode_windowtype]</tt> = read 16 bits as unsigned integer<p>
211 <li><tt>[vorbis_mode_transformtype]</tt> = read 16 bits as unsigned integer<p>
212 <li><tt>[vorbis_mode_mapping]</tt> = read 8 bits as unsigned integer<p>
213 <li>verify ranges; zero is the only legal value in Vorbis I for <tt>[vorbis_mode_windowtype]</tt> and <tt>[vorbis_mode_transformtype]</tt>. <tt>[vorbis_mode_mapping]</tt> must not be greater than the highest number mapping in use. Any illegal values render the stream undecodable.<p>
214 <li>save this mode configuration in slot <tt>[i]</tt> of the mode configuration array <tt>[vorbis_mode_configurations]</tt>.<p>
217 <li>read 1 bit as a framing flag. If unset, a framing error occurred and the stream is not decodable.
221 After reading mode descriptions, setup header decode is complete.<p>
223 <h1>Audio packet decode and synthesis</h1>
225 Following the three header packets, all packets in a Vorbis I stream
226 are audio. The first step of audio packet decode is to read and
227 verify the packet type; <em>a non-audio packet when audio is expected
228 indicates stream corruption or a non-compliant stream. The decoder
229 must ignore the packet and not attempt decoding it to audio</em>.
231 <h2>packet type, mode and window decode</h2>
234 <li>read 1 bit <tt>[packet_type]</tt>; check that packet type is 0 (audio)<p>
235 <li>read <a href="helper.html#ilog">ilog</a>([vorbis_mode_count]-1) bits <tt>[mode_number]</tt><p>
236 <li>decode blocksize <tt>[n]</tt> is equal to <tt>[blocksize_0]</tt> if <tt>[vorbis_mode_blockflag]</tt> is 0, else <tt>[n]</tt> is equal to <tt>[blocksize_1]</tt><p.
237 <li>perform window selection and setup; this window is used later by the inverse MDCT:<p>
238 <ol><li>if this is a long window (the <tt>[vorbis_mode_blockflag]</tt> flag of this mode is set):<p>
240 <li>read 1 bit for <tt>[previous_window_flag]</tt><p>
241 <li>read 1 bit for <tt>[next_window_flag]</tt><p>
243 <li>if <tt>[previous_window_flag]</tt> is not set, the left half
244 of the window will be a hybrid window for lapping with a
245 short block. See <a href="vorbis-spec-intro.html#window">the
246 'Window' subheading of the specification introduction
247 document</a> for an illustration of overlapping dissimilar
248 windows. Else, the left half window will have normal long
251 <li>if <tt>[next_window_flag]</tt> is not set, the right half of
252 the window will be a hybrid window for lapping with a short
253 block. See <a href="vorbis-spec-intro.html#window">the
254 'Window' subheading of the specification introduction
255 document</a> for an illustration of overlapping dissimilar
256 windows. Else, the left right window will have normal long
259 <li> if this is a short window, the window is always the same
260 short-window shape.<p>
265 Vorbis windows all use the slope function y=sin( .5 * PI * sin^2( (x+.5) /
266 n * PI) ) where n is window size and x ranges 0...n-1, but dissimilar
267 lapping requirements can affect overall shape. Window generation
268 proceeds as follows:<p>
271 <li> <tt>[window_center]</tt> = <tt>[n]</tt> / 2
272 <li> <tt>[left_window_start]</tt>
273 <li> if (<tt>[vorbis_mode_blockflag]</tt> is set and <tt>[previous_window_flag]</tt> is not set) then
274 <ol><li><tt>[left_window_start]</tt> = <tt>[n]</tt>/4 - <tt>[blocksize_0]</tt>/4
275 <li><tt>[left_window_end]</tt> = <tt>[n]</tt>/4 + <tt>[blocksize_0]</tt>/4
276 <li><tt>[left_n]</tt> = <tt>[blocksize_0]</tt>/2
279 <ol><li><tt>[left_window_start]</tt> = 0
280 <li><tt>[left_window_end]</tt> = <tt>[window_center]</tt>
281 <li><tt>[left_n]</tt> = <tt>[n]</tt>/2
284 <li> if (<tt>[vorbis_mode_blockflag]</tt> is set and <tt>[next_window_flag]</tt> is not set) then
285 <ol><li><tt>[right_window_start]</tt> = <tt>[n]*3</tt>/4 - <tt>[blocksize_0]</tt>/4
286 <li><tt>[right_window_end]</tt> = <tt>[n]*3</tt>/4 + <tt>[blocksize_0]</tt>/4
287 <li><tt>[right_n]</tt> = <tt>[blocksize_0]</tt>/2
290 <ol><li><tt>[right_window_start]</tt> = <tt>[window_center]</tt>
291 <li><tt>[right_window_end]</tt> = <tt>[n]</tt>
292 <li><tt>[right_n]</tt> = <tt>[n]</tt>/2
294 <li> window from range 0 ... <tt>[left_window_start]</tt>-1 inclusive is zero
296 <li> for <tt>[i]</tt> in range <tt>[left_window_start]</tt> ... <tt>[left_window_end]</tt>-1, window(<tt>[i]</tt>) = sin(.5 * PI * sin^2( (<tt>[i]</tt>-<tt>[left_window_start]</tt>+.5) / <tt>[left_n]</tt> * .5 * PI) )
299 <li> window from range <tt>[left_window_end]</tt> ... <tt>[right_window_start]</tt>-1 inclusive is one
301 <li> for <tt>[i]</tt> in range <tt>[right_window_start]</tt> ... <tt>[right_window_end]</tt>-1, window(<tt>[i]</tt>) = sin(.5 * PI * sin^2( (<tt>[i]</tt>-<tt>[right_window_start]</tt>+.5) / <tt>[right_n]</tt> * .5 * PI/2. + .5 * PI) )
303 <li> window from range <tt>[rigth_window_start]</tt> ... <tt>[n]</tt>-1 is zero
307 An end-of-packet condition up to this point should be considered an
308 error that discards this packet from the stream. An end of packet
309 condition past this point is to be considered a possible nominal
313 <h2>floor curve decode</h2>
315 From this point on, we assume out decode context is using mode number
316 <tt>[mode_number]</tt> from configuration array
317 <tt>[vorbis_mode_configurations]</tt> and the map number
318 <tt>[vorbis_mode_mapping]</tt> (specified by the current mode) taken
319 from the mapping configuration array
320 <tt>[vorbis_mapping_configurations]</tt>.<p>
322 Floor curves are decoded one-by-one in channel order.<p>
324 For each floor <tt>[i]</tt> of <tt>[audio_channels]</tt>
325 <ol><li><tt>[submap_number]</tt> = element <tt>[i]</tt> of vector [vorbis_mapping_mux] <p>
327 <li><tt>[floor_number]</tt> = element <tt>[submap_number]</tt> of vector [vorbis_submap_floor]<p>
328 <li>if the floor type of this floor (vector <tt>[vorbis_floor_types]</tt> element <tt>[floor_number]</tt>) is zero then decode the floor for channel <tt>[i]</tt> according to the <a href="vorbis-spec-floor0.html#decode">floor 0 decode algorithm</a><p>
329 <li>if the type of this floor is one then decode the floor for channel <tt>[i]</tt> according to the <a href="vorbis-spec-floor1.html#decode">floor 1 decode algorithm</a><p>
330 <li>save the needed decoded floor information for channel for later synthesis<p>
331 <li>if the decoded floor returned 'unused', set vector <tt>[no_residue]</tt> element <tt>[i]</tt> to true, else set vector <tt>[no_residue]</tt> element <tt>[i]</tt> to false<p>
334 An end-of-packet condition during floor decode shall result in packet
335 decode zeroing all channel output vectors and skipping to the
336 add/overlap output stage.<p>
338 <h2>nonzero vector propagate</h2>
340 A possible result of floor decode is that a specific vector is marked
341 'unused' which indicates that that final output vector is all-zero
342 values (and the floor is zero). The residue for that vector is not
343 coded in the stream, save for one complication. If some vectors are
344 used and some are not, channel coupling could result in mixing a
345 zeroed and nonzeroed vector to produce two nonzeroed vectors.<p>
347 for each <tt>[i]</tt> from 0 ... <tt>[vorbis_mapping_coupling_steps]</tt>-1
349 <ol><li>if either <tt>[no_residue]</tt> entry for channel
350 (<tt>[vorbis_mapping_magnitude]</tt> element <tt>[i]</tt>) or (channel
351 <tt>[vorbis_mapping_angle]</tt> element <tt>[i]</tt>) are set to false, then both
352 must be set to false. Note that an 'unused' floor has no decoded floor
353 information; it is important that this is remembered at floor curve
357 <h2>residue decode</h2>
359 Unlike floors, which are decoded in channel order, the residue vectors
360 are decoded in submap order.<p>
362 for each submap <tt>[i]</tt> in order from 0 ... <tt>[vorbis_mapping_submaps]</tt>-1<p>
363 <ol><li><tt>[ch]</tt> = 0<p>
364 <li>for each channel <tt>[j]</tt> in order from 0 ... <tt>[audio_channels]</tt><p>
365 <ol><li>if channel <tt>[j]</tt> is in submap <tt>[i]</tt> (vector <tt>[vorbis_mapping_mux]</tt> element <tt>[j]</tt> is equal to <tt>[i]</tt>)<p>
366 <ol><li>if vector <tt>[no_residue]</tt> element <tt>[j]</tt> is true<p>
367 <ol><li>vector <tt>[do_not_decode_flag]</tt> element <tt>[channels_in_bundle]</tt> is set<p>
368 </ol>else<ol><li>vector <tt>[do_not_decode_flag]</tt> element <tt>[channels_in_bundle]</tt> is unset<p>
370 <li>increment <tt>[ch]</tt><p>
373 <li><tt>[residue_number]</tt> = vector <tt>[vorbis_mapping_submap_residue]</tt> element <tt>[i]</tt><p>
375 <li><tt>[residue_type]</tt> = vector <tt>[vorbis_residue_types]</tt> element <tt>[residue_number]</tt><p>
376 <li>decode <tt>[ch]</tt> vectors using residue <tt>[residue_number]</tt>, according to type <tt>[residue_type]</tt>, also passing vector <tt>[do_not_decode_flag]</tt> to indicate which vectors in the bundle should not be decoded. Correct per-vector decode length is <tt>[n]</tt>/2.<p>
378 <li><tt>[ch]</tt> = 0<p>
379 <li>for each channel <tt>[j]</tt> in order from 0 ... <tt>[audio_channels]</tt><p>
380 <ol><li>if channel <tt>[j]</tt> is in submap <tt>[i]</tt> (vector <tt>[vorbis_mapping_mux]</tt> element <tt>[j]</tt> is equal to <tt>[i]</tt>)<p>
381 <ol><li>residue vector for channel <tt>[j]</tt> is set to decoded residue vector <tt>[ch]</tt><p>
382 <li>increment <tt>[ch]</tt>
387 <h2>inverse coupling</h2>
389 for each <tt>[i]</tt> from <tt>[vorbis_mapping_coupling_steps]</tt>-1 descending to 0
393 <li><tt>[magnitude_vector]</tt> = the residue vector for channel
394 (vector <tt>[vorbis_mapping_magnitude]</tt> element <tt>[i]</tt>)
396 <li><tt>[angle_vector]</tt> = the residue vector for channel (vector
397 <tt>[vorbis_mapping_angle]</tt> element <tt>[i]</tt>)
399 <li>for each scalar value <tt>[M]</tt> in vector <tt>[magnitude_vector]</tt> and the corresponding scalar value <tt>[A]</tt> in vector <tt>[angle_vector]</tt>:
400 <ol><li>if (<tt>[M]</tt> is greater than zero)
401 <ol><li>if (<tt>[A]</tt> is greater than zero)
403 <li><tt>[new_M]</tt> = <tt>[M]</tt>
404 <li><tt>[new_A]</tt> = <tt>[M]</tt>-<tt>[A]</tt>
408 <li><tt>[new_A]</tt> = <tt>[M]</tt>
409 <li><tt>[new_M]</tt> = <tt>[M]</tt>+<tt>[A]</tt>
413 <ol><li>if (<tt>[A]</tt> is greater than zero)
415 <li><tt>[new_M]</tt> = <tt>[M]</tt>
416 <li><tt>[new_A]</tt> = <tt>[M]</tt>+<tt>[A]</tt>
420 <li><tt>[new_A]</tt> = <tt>[M]</tt>
421 <li><tt>[new_M]</tt> = <tt>[M]</tt>-<tt>[A]</tt>
425 <li>set scalar value <tt>[M]</tt> in vector <tt>[magnitude_vector]</tt> to <tt>[new_M]</tt>
426 <li>set scalar value <tt>[A]</tt> in vector <tt>[angle_vector]</tt> to <tt>[new_A]</tt>
432 For each channel, synthesize the floor curve from the decoded floor
433 information, according to packet type. Note that the vector synthesis
434 length for floor computation is <tt>[n]</tt>/2.<p>
436 For each channel, multiply each element of the floor curve by each
437 element of that channel's residue vector. The result is the dot
438 product the floor and residue vectors for each channel; the produced
439 vectors are the length <tt>[n]</tt>/2 audio spectrum for each
442 One point is worth mentioning about this dot product; a common mistake
443 in a fixed point implementation might be to assume that a 32 bit
444 fixed-point representation for floor and residue and direct
445 multiplication of the vectors is sufficient for acceptable spectral
446 depth in all cases because it happens to mostly work with the current
447 Xiph.Org reference encoder. <p>
449 However, floor vector values can span ~140dB (~24 bits unsigned), and
450 the audio spectrum vector should represent a minimum of 120dB (~21
451 bits with sign), even when output is to a 16 bit PCM device. For the
452 residue vector to represent full scale if the floor is nailed to
453 -140dB, it must be able to span 0 to +140dB. For the residue vector
454 to reach full scale if the floor is nailed at 0dB, it must be able to
455 represent -140dB to +0dB. Thus, in order to handle full range
456 dynamics, a residue vector may span -140dB to +140dB entirely within
457 spec. A 280dB range is approximately 48 bits with sign; thus the
458 residue vector must be able to represent a 48 bit range and the dot
459 product must be able to handle an effective 48 bit times 24 bit
460 multiplication. This range may be achieved using large (64 bit or
461 larger) integers, or implementing a movable binary point
464 <h2>inverse MDCT</h2>
466 Convert the audio spectrum vector of each channel back into time
467 domain PCM audio via an inverse Modified Discrete Cosine Transform
468 (MDCT). A detailed description of the MDCT is available in the paper
470 href="http://www.iocon.com/resource/docs/ps/eusipco_corrected.ps">_The
471 use of multirate filter banks for coding of high quality digital
472 audio_</a>, by T. Sporer, K. Brandenburg and B. Edler. The window
473 function used for the MDCT is the window determined earlier.<p>
477 Windowed MDCT output is overlapped and added with the right hand data
478 of the previous window such that the 3/4 point of the previous window
479 is aligned with the 1/4 point of the current window (as illustrated in
480 <a href="vorbis-spec-intro.html#window">the 'Window' portion of the
481 specification introduction document</a>. The overlapped portion
482 produced from overlapping the previous and current frame data is
483 finished data to be returned by the decoder. This data spans from the
484 center of the previous window to the center of the current window. In
485 the case of same-sized windows, the amount of data to return is
486 one-half block consisting of and only of the overlapped portions. When
487 overlapping a short and long window, much of the returned range is not
488 actually overlap. This does not damage transform orthogonality. Pay
489 attention however to returning the correct data range; the amount of
490 data to be returned is:<p>
491 <tt>window_blocksize(previous_window)/4+window_blocksize(current_window)/4</tt>
492 from the center (element windowsize/2) of the previous window to the
493 center (element windowsize/2-1, inclusive) of the current window.<p>
495 Data is not returned from the first frame; it must be used to 'prime'
496 the decode engine. The encoder accounts for this priming when
497 calculating PCM offsets; after the first frame, the proper PCM output
498 offset is '0' (as no data has been returned yet).<p>
500 <h2>output channel order</h2>
502 Vorbis I specifies only a channel mapping type 0. In mapping type 0,
503 channel mapping is implicitly defined as follows for standard audio
507 <dt>one channel:<dd> the stream is monophonic
508 <dt>two channels:<dd> the stream is stereo. channel order: left, right
509 <dt>three channels:<dd> the stream is a 1d-surround encoding. channel order: left, center, right
510 <dt>four channels:<dd> the stream is quadraphonic surround. channel order: front left, front right, rear left, rear right
511 <dt>five channels:<dd> the stream is five-channel surround. channel order: front left, front center, front right, rear left, rear right
512 <dt>six channels:<dd> the stream is 5,1 surround. channel order: front left, front center, front right, rear left, rear right, LFE
513 <dt>greater than six channels:<dd> channel use and order is defined by the application
516 Applications using Vorbis for dedicated purposes may define channel
517 mapping as seen fit. Future channel mappings (such as three and four
518 channel <a href="http://www.ambisonic.net">Ambisonics</a>) will make
519 use of channel mappings other than mapping 0.<p>
522 <a href="http://www.xiph.org/">
523 <img src="white-xifish.png" align=left border=0>
525 <font size=-2 color=#505050>
527 Ogg is a <a href="http://www.xiph.org">Xiph.org Foundation</a> effort
528 to protect essential tenets of Internet multimedia from corporate
529 hostage-taking; Open Source is the net's greatest tool to keep
530 everyone honest. See <a href="http://www.xiph.org/about.html">About
531 the Xiph.org Foundation</a> for details.
534 Ogg Vorbis is the first Ogg audio CODEC. Anyone may freely use and
535 distribute the Ogg and Vorbis specification, whether in a private,
536 public or corporate capacity. However, the Xiph.org Foundation and
537 the Ogg project (xiph.org) reserve the right to set the Ogg Vorbis
538 specification and certify specification compliance.<p>
540 Xiph.org's Vorbis software CODEC implementation is distributed under a
541 BSD-like license. This does not restrict third parties from
542 distributing independent implementations of Vorbis software under
545 Ogg, Vorbis, Xiph.org Foundation and their logos are trademarks (tm)
546 of the <a href="http://www.xiph.org/">Xiph.org Foundation</a>. These
547 pages are copyright (C) 1994-2002 Xiph.org Foundation. All rights