Path: utzoo!utgpu!news-server.csri.toronto.edu!cs.utexas.edu!swrinde!zaphod.mps.ohio-state.edu!pacific.mps.ohio-state.edu!linac!att!ucbvax!NSCULTRIX1.NETWORK.COM!dotytr
From: dotytr@NSCULTRIX1.NETWORK.COM (Ted R. Doty)
Newsgroups: comp.protocols.tcp-ip
Subject: Re:  Compressed Video Transport Requirements
Message-ID: <9102181526.AA06447@nscultrix1.network.com>
Date: 18 Feb 91 15:26:51 GMT
Sender: daemon@ucbvax.BERKELEY.EDU
Organization: The Internet
Lines: 26

I think that there are two problems with *REAL TIME* video: variation of
delay and synchronization.  If the delay between packets varries too much,
the picture would jump (like the strobe lights at your old disco parties)
like crazy, causing Excedrin headache #423.  if you have synchronization
problems (voice and video arriving out of sync), it looks like a badly
dubbed movie (remember the Hercules skitt on Saturday Night Live?).  Either
way you irritate the users.

Note that ethernet doesn't look good for problem #1.  Maybe nothing is (FDDI
II?  Now how much would you pay ...).

I think these problems are not so severe if you're not going real-time.

I don't have references about video, but I do have some performance modeling
results of FDDI synchronous token stuff.  Send me your U.S. Postal address if
you want 'em (I only have hard copy).

- Ted

-----------------------------------------------------------------------------
Ted Doty, Network Systems Corporation  |  phone:      +1 301 596-2270
8965 Guilford Rd./Suite 250            |  fax:        +1 301 381-3320
Columbia, MD 21046  USA                |  voice mail: (800) 233-1485 x4436
-----------------------------------------------------------------------------
"These opinions are my own, and not necessarially those of Network Systems
Corporation."