[opendtv] Re: tolerance Latency , Jitters & packet loss issues for MPEG-2 TS over IP netwo

  • From: "Ralph P. Manfredo" <rmanfredo@xxxxxxxx>
  • To: <opendtv@xxxxxxxxxxxxx>
  • Date: Sun, 27 Feb 2005 04:02:19 -0800

I agree John.  My racetrack example was just that.  One interesting bit of
info I received on that project was when the video production manager asked
me if I knew how long a horse was.  Of course I answered about 10 feet.  She
came back with 8 ms, which was an interesting answer.  BTW, we had latency
end-to-end  including going through ATM switches of 100 ms.  This as in a
private network, not a telco network.  I recently saw a demo of a HDTV
(1080i) with an encode to decode latency of approx. 30 ms.  Very impressive.

Ciao,

Ralph


Ralph P. Manfredo
President and CEO

rmanfredo@xxxxxxxx

************************************************************************
BroadBand Networks Corporation
2530 Berryessa Road, No. 237
San Jose, CA 95050

Phone:  408.988.2060
Fax:      408.988.2188
Cell:      559.289.2669

www.bbnc.com

Leaders in MPEG-2 and MPEG-4 video over ATM and IP Networks
************************************************************************


-----Original Message-----
From: opendtv-bounce@xxxxxxxxxxxxx [mailto:opendtv-bounce@xxxxxxxxxxxxx] On
Behalf Of John McClenny
Sent: Sunday, February 27, 2005 2:39 AM
To: opendtv@xxxxxxxxxxxxx
Subject: [opendtv] Re: tolerance Latency , Jitters & packet loss issues for
MPEG-2 TS over IP netwo

On Sat, 26 Feb 2005 15:43:34 -0800, Ralph P. Manfredo <rmanfredo@xxxxxxxx>
wrote:
> I disagree.  Latency is very important if you are doing interactive TV.
> Long latency with interactive game shows, movies, etc will be an issue 
> believe me.  Latency was an issue with a video system we did for a 
> race track, where there was a large video panel (billboard size) in 
> the infield and the horses are in the same field of view as the video 
> panel.  Think about it.  Would you be happy watching a race you just 
> bet on and the live horse was several feet in front o where he is shown on
he screen?  Question?
> Did I bet on a race that was already run.  Believe me, latency is an 
> issue with real-time video or interactive video.
> 

Sorry Ralph, I wasn't very clear.

Worst cast latency in any IPTV network is going to be under a 100 MS
from the network side.   Traditional cable network seem to have
latencies measured in weeks if my Time Warner Cable trick play control on
VOD is any indication.  The application level latencies can be much higher.
In particular, the video processing latency can be several seconds from
ingest to display on the TV because of transcoding, DRM/CA, and buffering in
the STB.

What I should have said is that the latencies on a IP network will be low
enough that it will not be an issue  and application level latencies will be
the dominant visible latency.

You race track example isn't the cable equivalent experience we were
discussing.  In your case, latency matters because the viewer has two sets
of inputs are are not matched in time - you are seeing two races out of
sync.

As long as both sets of visual inputs are the same, I can't tell the
difference sitting in my home.  The user has no way of knowing the total
delay from TV ingest to display on their TV, just how out-of-sync the
video/iTV experience is.  As long as they are in sync, how can I tell as the
viewer?
 
 
----------------------------------------------------------------------
You can UNSUBSCRIBE from the OpenDTV list in two ways:

- Using the UNSUBSCRIBE command in your user configuration settings at
FreeLists.org 

- By sending a message to: opendtv-request@xxxxxxxxxxxxx with the word
unsubscribe in the subject line.



 
 
----------------------------------------------------------------------
You can UNSUBSCRIBE from the OpenDTV list in two ways:

- Using the UNSUBSCRIBE command in your user configuration settings at 
FreeLists.org 

- By sending a message to: opendtv-request@xxxxxxxxxxxxx with the word 
unsubscribe in the subject line.

Other related posts:

  1. Home
  2. opendtv
  3. Archive
  4. 02-2005