On Fri,12/4/2015 12:04 PM, David Pruett wrote:
During 30+ years in the automotive industry, we had a rule of thumb
that anything under 100 mS couldn't be "noticed" by the human operator.
Human tolerance of delay is VERY dependent on the application. With
AUDIO, an echo greater than about 35 msec becomes increasingly
problematic; even delays as short at 30 msec can be fatiguing. Increase
the delay to 80 msec, and the talker will slow down and stop talking. In
my live sound work, I've witnessed VERY professional announcers turn to
silly putty when hearing themselves coming through a ballpark sound
system 100 msec after their speech.
The problem occurs with the talker hears him/herself coming through
distant loudspeakers, producing the delay. There are two very effective
fixes. 1) Prevent them from hearing themselves on delay; or 2) give them
their voice UN-delayed in headphones.
I have not seen studies, nor have I experienced, latency beyond about 10
msec or so ham applications, but I would expect anything much beyond 100
msec to get tricky in a contesting application, simply because you're
giving away that much reaction time. Also, remember that our systems are
two-way, so TX and RX delays are additive.
I'd love to hear a thoughtful, analytical presentation/discussion by
some top contesters who have dealt with latency in high rate contesting.
I think it would VERY educational.
73, Jim K9YC
73, Jim K9YC
_______________________________________________
CQ-Contest mailing list
CQ-Contest@contesting.com
http://lists.contesting.com/mailman/listinfo/cq-contest
|