Richard Lee wrote:> But there is a caveat. Does Vorbis coupling preserve "phase" relations? Any references explaining this in simple detail?Sebastian :>"coupling" is a rather loose term in the sense that it doesn't exactlyspecify HOW it's done. It can be understood as the opposite of coding channels independently. Obviously there are many ways to do coupling.>Both a "lossless" variant and a "lossy" is possible within the Vorbisspecification. The former is supposed to exploit redundancies between channels only while the latter is also a kind of irrelevancy reduction (like ignoring phase relations for intensity stereo).>If you really want to know the gritty details you should take a look atthe Vorbis specification and the Vorbis Stereo modes HTML page. Thank you for Sebastien. I am new to compressed audio coding so please correct any stupid assumptions I make ... 8>D>From the Vorbis spec and Stereo mode page, my tentative recommendations as a (pseudo) Soundfield and Ambi guru are as follows.- The Ambisonic B-format WXYZ channels should be should be lossless coupled together over most of the frequency range. This will give the most gain as W "usually*" has everything that XYZ has. A perfect B-format signal has scaled versions of XYZ in W.* - We allow "4 phase coupling" at higher frequencies. A perfect B-format set of WXYZ should only have real relation to each other*. The deviations from perfection can take 2 forms. a) Minimum Phase deviations of XYZ from W. These start happening above about 6kHz with present Soundfields. b) Random Phase deviations of XYZ from W. These occur at High Frequencies when a real Soundfield Mike no longer has flat frequency response in all directions. Above 10kHz with present Soundfields. - B-format from Ambisonically panned material (as opposed to natural Soundfield recordings) will be perfectly in phase so this recommendation will not result in loss. This needs investigating. Presently no Ambisonic effects units use -- No virus found in this outgoing message. Checked by AVG Free Edition. Version: 7.5.432 / Virus Database: 268.17.20/664 - Release Date: 2/02/07 15:42
The message appears to have been truncated. This is the important bit. Vorbis coupling is described on the Vorbis-Stereo page ______________>From the Vorbis spec and Stereo mode page, my tentative recommendations as a (pseudo) Soundfield and Ambi guru are as follows.- The Ambisonic B-format WXYZ channels should be should be lossless coupled together over most of the frequency range. This will give the most gain as W "usually*" has everything that XYZ has. A perfect B-format signal has scaled versions of XYZ in W.* - We allow "4 phase coupling" at higher frequencies. A perfect B-format set of WXYZ should only have real relation to each other*. The deviations from perfection can take 2 forms. a) Minimum Phase deviations of XYZ from W. These start happening above about 6kHz with present Soundfields. b) Random Phase deviations of XYZ from W. These occur at High Frequencies when a real Soundfield Mike no longer has flat frequency response in all directions. Above 10kHz with present Soundfields. - B-format from Ambisonically panned material (as opposed to natural Soundfield recordings) will be perfectly in phase so this recommendation will not result in loss. This needs investigating. Presently no Ambisonic effects units use "phase" at High Frequencies but this should not be ruled out. - Allowing "4 phase" above 18kHz, like default Dolby Digital at 448kb/s, is likely to be undetectable on present Soundfield recordings. Allowing "4 phase" down to 10kHz might force the random phase deviations of a real Soundfield mike to be real and MAY result in better results. This needs investigating. - "Point phase" coupling should be avoided Any one have any comments? I have a few questions. Is there a document which details the frequency banding strategy used by Vorbis? Is it more usual to adopt eg "8 phase" at a lower frequency band before going to "4 phase"? Do the terms "8 phase", "4 phase", "point" still make sense when we are coupling 4 channels? Richard * This falls down at LF and close sources and sometimes when you are recording organ pedal notes. But lossless coupling at LF is cheap. This also falls down for diffuse field. It is possible that the pseudo random differences in response of a real Soundfield in different directions above 10khz may preserve enough of the "random" nature of reverb to be acceptable without needing full random phase. Any Ambisonic artificial reverb units out there? Mr Glasgal, as Ambiophonics uses impulse responses usually derived from a Soundfield mike or cruder copies, it is covered by the "real Soundfield mike" discussion. -- No virus found in this outgoing message. Checked by AVG Free Edition. Version: 7.5.432 / Virus Database: 268.17.25/669 - Release Date: 4/02/07 21:58
>- The Ambisonic B-format WXYZ channels should be should be lossless coupled together over most of the frequency range. This will give the most gain as W "usually*" has everything that XYZ has. A perfect B-format signal has scaled versions of XYZ in W.*>- We allow "4 phase coupling" at higher frequencies. A perfect B-format set of WXYZ should only have real relation to each other*. The deviations from perfection can take 2 forms.I have a few questions. Is there a document which details the frequency banding strategy used by Vorbis? Is it more usual to adopt eg "8 phase" at a lower frequency band before going to "4 phase"? Do the terms "8 phase", "4 phase", "point" still make sense when we are coupling 4 channels? -- No virus found in this outgoing message. Checked by AVG Free Edition. Version: 7.5.441 / Virus Database: 268.17.35/680 - Release Date: 10/02/07 21:15
>> Is there a document which details the frequency banding strategy used by Vorbis?>Only as part of the spec. In short, it's 'flexible'; you can set uppretty much any number of subbands of any size>> Is it more usual to adopt eg "8 phase" at a lower frequency band before going to "4 phase"?>No. It was 'once upon a time', but the modern encoders use onlylossless and point.>> Do the terms "8 phase", "4 phase", "point" still make sense when we are coupling 4 channels?>Yes, but only if you squint and look at one plane at a time.Many thanks Monty. This simplifies my recommendations for Vorbis Ambisonics. If "4 phase" and "8 phase" do not gain much in efficiency ... - "Lossless Coupling" of 3 .. 9, 11, 16 channels will give substantial efficiencies; more than 2x compared to no coupling, with no cons. - "Lossy Coupling" is Verbotten! -- No virus found in this outgoing message. Checked by AVG Free Edition. Version: 7.5.441 / Virus Database: 268.18.4/705 - Release Date: 27/02/07 15:24