Orban OPTIMOD-FM 8600S, Operating Manual

Page: 51 / 354

OPTIMOD-FM DIGITAL INTRODUCTION
1-25
Monitoring on Loudspeakers and Headphones
In live operations, highly processed audio often causes a problem with the DJ or
presenter’s headphones. When an “MX” preset is active, the delay through the
8600S can be as much as 270 milliseconds. This delay will be audible as a distinct
echo that almost no one can tolerate in his or her headphones while speaking.
The 8600S offers two main solutions to this problem:

Using a non-MX preset, which provides the same algorithms and performance as
Optimod-FM 8500

Using the 8600S’s low-delay “monitor” output to drive talent headphones
Using the monitor output is appropriate for both HD Radio and non-HD Radio op-
erations, while using a non-MX preset on-air is only useful in non-HD Radio opera-
tions where no diversity delay has been applied to the analog FM transmission.
Because of their higher performance, we recommend using MX presets in HD Radio
operations. In this context, the preset delay becomes part of the HD Radio diversity
delay and the 8600S automatically compensates for the preset delay to ensure that
the diversity delay always agrees with the value you specified when setting the
8600S up.
With non-MX presets, the normal delay through the 8600S (from input to FM out-
puts) is about 18 ms when
H
ARD
or
M
EDIUM
bass clipping is selected, as it is in all non-
MX factory presets other than those with “LL” (“low latency”) or “UL” (“ultra-low
latency”) in their names. An 18 ms delay is workable for most talent (although it
may require some acclimatization) because 18 ms is below the psychoacoustic “echo
fusion threshold,” which means that talent will not hear discrete slap echoes in their
headphones. This means that they can monitor comfortably off-air without being
distracted or confused. Moreover, off-air cueing of remote talent is routine.
Some talent moving from an analog processing chain will require a learning period
to become accustomed to the voice coloration caused by “bone-conduction” comb
filtering. This is caused by the delayed headphone sound’s mixing with the live voice
sound, which introduces notches in the spectrum that the talent hears when he or
she talks. All digital processors induce this coloration to a greater or lesser extent.
Fortunately, it does not cause confusion or hesitation in the talent’s performance un-
less the delay is above the psychoacoustic “echo fusion” (Haas) threshold of ap-
proximately 20-25 ms, where the talent starts to hear slap echo in addition to fre-
quency response colorations.
Two lower-delay options are available. “Low latency” reduces input-to-FM-output
delay to 13 ms and “ultra-low latency” reduces delay to about 3.7 ms. The trade-off
for this reduction is approximately 1 dB decrease in loudness compared to the
8600S’s full look-ahead processing for low latency and about 2.5 dB loudness de-
crease for ultra-low latency.

When using a non-MX preset, you can invoke the low latency mode by setting