Path: utzoo!utgpu!water!watmath!clyde!att!osu-cis!tut.cis.ohio-state.edu!mailrus!uwmcsd1!nic.MR.NET!umn-cs!wsmith
From: wsmith@umn-cs.cs.umn.edu (Warren Smith [Randy])
Newsgroups: comp.dcom.lans
Subject: Re: 802.3 a panacea ?
Message-ID: <6896@umn-cs.cs.umn.edu>
Date: 23 Aug 88 17:03:27 GMT
References: <10400002@osiris.cso.uiuc.edu> <4942@cos.com> <1988Aug18.165532.26169@utzoo.uucp> <20447@watmath.waterloo.edu> <1988Aug22.170009.4743@utzoo.uucp>
Reply-To: wsmith@umn-cs.UUCP (Warren Smith [Randy])
Organization: University of Minnesota
Lines: 25

<In article <1988Aug22.170009.4743@utzoo.uucp> henry@utzoo.uucp (Henry Spencer) writes:
<>In article <20447@watmath.waterloo.edu> egisin@watmath.waterloo.edu (Eric Gisin) writes:
<>>> ... the 802.3
<>>> distance constraints shrink if you raise the bit rate, meaning that it's
<>>> not going to work well at FDDI speeds. 
<>>
<>>Is this true? I thought you could either decrease the physical length
<>>OR increase the minimum packet size when increasing the bit rate.
<>
<>That's correct; as I recall it, the constraint is basically that the time
<>needed to transmit the shortest possible packet must exceed the worst-case
<>round-trip time of the network.  The underlying requirement is that a
<>collision must be a network-wide phenomenon that all nodes involved with
<>a packet will agree on.  ...

Partially correct.  However, you must remember that the increased propagation
delay results in an increased likelihood of collisions due to the
enlarged time window between the time a transmission is started and before
it is heard by all other stations on the network.  Here the actual distribution
of stations along the network begins to play a part.

-- 
Randy Smith
wsmith@umn-csw.cs.umn.edu
...!rutgers!umn-cs!wsmith