Glossary - Prism Sound ADA-8 Operation Manual

Prism Sound ADA-8 Multi-channel A/D D/A Converter

9 Glossary

ADAT
AES3
AES10
AES11
AES/EBU
DARS
Dither
DRE
DSD
Interface Jitter
Jitter
Lightpipe
MADI
MR-X
Noise Shaping
© Prism Media Products Limited, 2002
An optical eight-channel digital audio interface format first used in
Alesis ADAT recorders.
A two-channel digital audio interface format complying with the AES3
standard.
See MADI.
A Reference Sync format complying with the AES11 standard, similar
to an AES3 carrier. Also known as DARS (Digital Audio Reference
Signal).
See AES3.
See AES11.
A technique to eliminate truncation distortion in A/D converters or
where digital wordlength is reduced. Dither involves adding noise
prior to truncation in order to randomize transfer-function thresholds
and thus linearize the process, resulting in analogue-like low-signal
behaviour. See also Truncation, Noise Shaping.
'Dynamic Range Enhancement'. A Prism Sound proprietary
encoding scheme whereby extended wordlengths can be recorded or
transmitted on shorter-wordlength media without sacrificing channels.
'Direct Stream Digital' – a technique wherein digital audio signals are
recorded, transmitted or processed in an over-sampled, noise
shaped one-bit format.
Jitter on a digital audio interface. This might not be problematic until
it is sufficiently bad to cause data to be lost on the interface.
However, most A/D and D/A converters cannot reject interface jitter
from their Reference Sync, and so the interface jitter survives as
sampling jitter at the conversion point, resulting in distortion of the
converted audio.
Instability or phase-noise in a digital audio interface or in the clock of
an A/D or D/A converter. See also Interface Jitter, Sampling Jitter.
See ADAT.
'Multichannel Audio Digital Interface' – a 56-channel digital audio
interface format on coaxial or fibre-optic cable complying with the
AES10 standard.
A Prism Sound proprietary encoding scheme whereby extended
wordlengths can be recorded or transmitted on shorter-wordlength
multi-channel media by 'word mapping' or 'bit splitting'. Channels are
sacrificed to hold the extended parts of the audio samples.
A sophisticated dithering technique whereby the requantization error
is filtered and subtracted from the input data. The filter is designed
so that the resulting noisefloor is optimally shaped in frequency so as
to be as inaudible as possible. See also Truncation, Dither.
Operation Manual - Revision 1.00
Page 1.38