Crystal Vision
Data packet management
TANDEM-200 User Manual R1.7
19
13/02/2008
2.3 Vertical switching
When upstream input switching occurs or when input drop-out occurs the embedded
digital audio can suffer degradation and loss of synchronisation. The result could produce
audible defects.
Repeating samples from a buffer during the disturbance can ease the problem or in the
case of upstream switching, lines around the standard switching point can simply be
avoided and not used for digital audio.
There are a number of methods in use for avoiding the switching point suggested in
SMPTE recommendation RP168. SMPTE standards 272M and 291M suggest the use of a
single line gap after the switch point, but larger gaps around the switch point are
sometimes required to provide sufficient robustness. In addition, there is equipment
already in the market place manufactured by Sony before the SMPTE recommendation
was implemented that embeds on all lines. For these reason both SMPTE and proprietary
embedded formats need to be supported.
The card edge display uses the following text to distinguish between the two basic
embedded audio formats:
•
SMPTE: no data on lines around the preferred switching point e.g. lines 5,6,7,8 in
625 and lines 9,10,11,12 in 525
•
SONY: data on every line
TANDEM modules provide a range of embedding and de-embedding formats that
provide a trade off between robustness and processing delay.
The de-embedding and embedding formats do not have to be the same. De-embedding
can be set to look at all lines for embedded audio whilst embedding can be done
according to SMPTE recommendations with a switch-point gap.
This allows TANDEM-200 to convert material from SONY formats to SMPTE. It is also
possible to convert SMPTE to SONY format, however, in this case the de-embed mode
should be placed in the special setting of SONY/SMPTE or ‘ALL’ rather than SONY.
