To LUGNET HomepageTo LUGNET News HomepageTo LUGNET Guide Homepage
 Help on Searching
 
Post new message to lugnet.roboticsOpen lugnet.robotics in your NNTP NewsreaderTo LUGNET News Traffic PageSign In (Members)
 Robotics / 11396
11395  |  11397
Subject: 
RE: Duplex Comms with RCX
Newsgroups: 
lugnet.robotics
Date: 
Fri, 28 Apr 2000 18:12:06 GMT
Original-From: 
Jim Thomas <Jim.Thomas@trw[antispam].com>
Viewed: 
666 times
  
One problem that I can see with Chris' token ring arrangement is that all RCX's must be able to see the broadcasts of every other RCX (not likely if they are all pointed at the tower) in order to know if it can use the slot.  This certainly can be fixed and I think some of the options he presented might eliminate this problem.  JB's (aka PinkTail) suggestion of polling would also solve this problem.  In general the drawback of either of these schemes (and single master networks in general) is high latency for a low overhead implementation or high overhead for a low latency implementation.  Since this is all on IR and not wires, I suppose latency cannot be prevented because I expect the error rate is high anyway.

To solve the multiple PC problem in JB's polling arrangement you could always have the PC's request and then exchange the master role.  The polling packets could contain a master ID.  Now you have a multi-master situation and you only have to worry about whether the RCXs can see both towers.  You could extend this to allow an RCX to take over as master but then this compounds the problem of line-of-sight networking.  This would all probably work much better with an RF link instead.

JT



1 Message in This Thread:

Entire Thread on One Page:
Nested:  All | Brief | Compact | Dots
Linear:  All | Brief | Compact
    

Custom Search

©2005 LUGNET. All rights reserved. - hosted by steinbruch.info GbR