home *** CD-ROM | disk | FTP | other *** search
- Newsgroups: comp.parallel
- Path: sparky!uunet!gatech!hubcap!fpst
- From: anurag@cc.gatech.edu (Anurag Gupta)
- Subject: Re:Synchronizing watches
- Message-ID: <1992Sep8.133720.29680@hubcap.clemson.edu>
- Sender: fpst@hubcap.clemson.edu (Steve Stevenson)
- Organization: Clemson University
- Date: Mon, 7 Sep 92 15:44:18 EDT
- Approved: parallel@hubcap.clemson.edu
- Lines: 34
-
- >Mike Nelson writes:
- >I must be missing something here, because this looks quite simple.
- >
- >-A & B already know this procedure.
- >
- >A: says "Bing" to B
- > B responds with "Bing" to A
- >
- >this is ABBA time - the total delay
- >
- >A:
- >1. says "Bong"
- >2. waits, say N minutes (N > ABBA)
- >3. says "Bong" again
- >
- >B already knows what N is, and can then determine
- >What AB is (Bong2 - Bong1 - N)
- >
- >B tells A what AB time is.
- >
- >A says "Bang" and waits AB time before zeroing
- >the watch
- >
- >When B hears "Bang", B immediately zero's the watch
- >
- >synchronized
- >
- >Michael Nelson
-
- The flaw is this: *Both* Bong2 and Bong1 will be delayed by AB and so
- Bong2 - Bong1 - N will be zero and not AB.
-
- Anurag Gupta
-
-