Topband
[Top] [All Lists]

Re: Topband: DSP and Latency

To: <topband@contesting.com>
Subject: Re: Topband: DSP and Latency
From: "Larry" <lknain@nc.rr.com>
Reply-to: Larry <w6nws@arrl.net>
Date: Thu, 19 Mar 2015 07:16:29 -0400
List-post: <topband@contesting.com">mailto:topband@contesting.com>
It would be rare for an internet connection to be a single hop. I forget the command (something like tracert) but you can see the nodes where the connection goes. The number of hops is usually minimized as much as possible. The NYC to KP4 connection is a minimum of 3 hops most likely: NYC PC to internet provider (NYC), internet provider (NYC) to internet provider (KP4), and internet provider (KP4) to KP4 PC. If I recall the return route may be different.

In your TV example, just the difference between HD and non-HD will show latency even if the TV's are connected the same way otherwise. Perhaps the delay is due to converting HD to non-HD at the source although it seems like HD usually trails non-HD.

73, Larry  W6NWS

-----Original Message----- From: D Rodman MD
Sent: Thursday, March 19, 2015 6:41 AM
To: topband@contesting.com
Subject: Re: Topband: DSP and Latency

Everything takes time.  I have a great example of audio latency.
Kitchen has television with direct cable input and family room has
digital cable box with HDML to large screen LED set.  When watching
events, like superbowl, both sets might be on.  It is highly obvious
which television displays the video and has audio first.  It is the
kitchen one.

I am not sure how we got a 120msec latency from NYC to NP2.  It is more
than pinging the route one way.  I have been doing this for over 10
years.  Latency, for me, is really the total time between the station on
one end hears a signal, transmits and the station on the other end hears
the audio.  I can ping within my own network and get times of around
10msec.  I think 120msec is going to be a one way figure, isn't it?  It
needs to be doubled to account for the two way internet passage of
signals both ways and in order to get the true latency add in the
processing time of the computer or radio at either end (if applicable).

The real killer here can be variation in the internet latency, either at
the local network or international link.  That variation can negatively
influence communications and be frustrating.

Moreover, I submit in some circumstances a quarter of a second is plenty
of time for another station to be first in.  We have all seen that.

--
David J Rodman MD
Assistant Clinical Professor
Department of Ophthalmology
SUNY/Buffalo

Office 716-857-8654
_________________
Topband Reflector Archives - http://www.contesting.com/_topband
_________________
Topband Reflector Archives - http://www.contesting.com/_topband

<Prev in Thread] Current Thread [Next in Thread>