Theatre Design & Technology - Winter 1987 - 9

Soun~
Beginning with this issue, the Sound
Column, formerly written by Rollins Brook, will be written by
Charlie Richmond. Mr. Richmond is
a theatre sound designer and consultant, President of Richmond Sound
Design of Vancouver, and USITT
Sound Design Commissioner.
By the time you read this, you'll
probably be hearing quite a bit about
R-DAT, primarily from the people
trying to sell it to you. There's a good
reason for this; a lot of money has
been spent developing R-DAT and
they want to be reimbursed. There
are also some good reasons why theatre sound departments should think
twice before purchasing one, no matter how great R-DAT may seem.
First, a little background: R-DAT
is the apparent winner in the lconsumer) Digital Audio Tape recording
format battle. It utilizes a spinning
rotary head (the "R-") to record the
large amount of digital information
needed to encode high-quality stereo
sound onto a cassette similar to the
8mm video format. The losing format is S-DAT which uses a stationary head, consumes more tape (but is
gentler to it), is more sensitive to
dirty heads, and whose development
fell behind because of expensive
thin-film head technology (and because the video market helped develop the rotary head design). None
of this has anything to do with the
current professional digi tal audio
recording formats DASH or PRODIGITAL, neither of which will
likely drop to theatre budget levels
before they become obsolete.
I should qualify this negative attitude from the start by explaining
that R-DAT does have a number of
positive attributes. Its overriding advantage over all previous consumer
tape formats is sound quality. Even
though it incorpora tes numerous
recording and playback modes, they
are all digi tal and promise low noise,
adequate bandwidth, flat response,
and low distortion. TOt only that,
but extensive use is made of error
correction, sub-codes, indexing, pro-

grammability, and contents management in order to give the format capabilities similar to CDs.
Techniques have been developed
to enable high-speed duplication of
prerecorded material on R-DAT cassettes, instantly making the production cost of commercial products
lower than for CDs. It is very possible
that consumers weaned on regular
cassettes will find the R-DAT format
satisfyingly familiar with the ability
to both record and play back in a similar fashion. R-DAT has a generally
longer overall playing time using one
side only (depending on tape thickness) in a smaller cassette, making
random access slightly more difficult and effectively trading convenience for quality.
But that'S where the positive aspects begin to wane. In order to obtain long recording times, thin and
exotic tapes and coatings are being
developed who e permanence and
reliability have yet to be established.
You are likely to see tapes and coatings with half the thickness of current analog audio cassette tapes.
Combine this fragility with an automated loading system similar to a
video cassette, a rotary dmm spinning at 2000 rpm and potential manual handling problems, and I become
extremely skeptical of the advisability of depending on R-DAT for
good sound reproduction.
For comparison, I'd like to explain
what it has been like working with a
similar format. At our recording studio we have been making digital
master recordings using PCM processors in combination with high
quali ty Ih" and '/0" video recorders
for several years now. We have implemented numerous modifications to
make the format better, including
eliminating the drop-out compensation circuit (only works properly
with video signalsl and adding an external sync input for ynchronizing
audio to video. Yet the format remains extremely fragile. We have to
be extremely careful not to touch the
tape inside the cassette; if the video

deck is left in pause even for a short
time we can almost guarantee a dropout will occur at that point, even
upon re-recording. I consistently recommend that the engineers use 14bit recording rather than 16-bit because of its superior error correction
even tl:ough it means slightly lower
sound quality.
R-DAT utilizes the higher 16-bit
accuracy for its standard recording
and playback modes but, although
superior error correction is required,
the CD standard Reed-Solomon code
with 37.5% redundancy is used and
has to contend with an entire range
of potential problems. First, the thin
tape wraps around a 90° angle of the
rotary dmm, yet only two heads are
mounted on the drum, 1 0° apart. An
elaborate tracking correction system
is employed but approximately half
the time the tape is moving without
either head monitoring its progress.
Actual track width is less than 13
microns, somewhat smaller than
any video format, and the signal
must switch between the tWO heads
more frequently than with video formats. Consequently, mechanical
alignment is absolutely critical. It is
not a question of whether the error
correction can contend with the natural system variations but rather JUSt
how much head and tape wear, dirt,
and damage will be tolerated before
errors can no longer be corrected.
R-DAT also remains a longitudinal format that requires lengthy
searches to reach any de ired location even though such searches may
be programmable and absolutely accurate. Digital editing is only possible with at least twO separate R-DAT
machines, one of which IS specially
designed for gapless drop-ins. A separate digital edit controller and servo
controlled source machine may also
be required. All of these will be quite
expen ive when and if they become
available; at the moment nothing
has suggested that these applications
will be addressed. Until that time,
the usefulness of R-DAT in the thecontinued on p. 46
TD&T· WI TER 1987

Table of Contents for the Digital Edition of Theatre Design & Technology - Winter 1987