[aprssig] The final WIDE2-1,WIDE1-1 solution!

Greg Noneman greg at clubnet.net
Sat Apr 2 13:44:03 EST 2005


Bob,

Before we get into SoCal specifics, I want to be clear on what the plan 
is for RELAY support on the digis.  Are we talking about immediately 
(or as fast as can be done) removing support for RELAY at all 
digipeaters or is there some transition period?

Back to the LA situation.  Yes, we may be somewhat unique.  I don't 
know if anyone else is employing a one-hop network, but it sure is 
effective.  Try taking a look at Phil's UI-Traffic Monitor Page 
(http://aprsca.net:8080/uiview/trafficlive.htm).  As of today, with 
preemptive digipeating employed on two of the four digis, RELAY is 
perfectly supported and isn't causing extra packets.  The same can be 
done with the proposed WIDE1-1,WIDE2-1 fill-in fix, but for us to have 
the same operation (no extra hops generated), it will either mean that 
the remaining two digis will have to go to UIDIGI firmware (with 
preemptive digipeating) or (if still running KPC3s) they will have to 
have WIDE1-1 removed from their UIDIGI path list.

Right now, nothing much is required on our metro digis to support the 
change. For the two UIDIGI firmware units removing WIDE1-1 from the 
PREEMPTCALLS list will help.  I think this will take time to flow into 
everyday operation. We can massage the digi network to best accommodate 
the situation.  I think the tough thing will be getting those same 
people whose never removed RELAY from the trackers or never turned-on 
callsign substitution on their home RELAY digis to make the changes.  
As a  result, we have some fixed percentage of bandwidth chewed-up by 
people using useless paths.


73,
Greg
WB6ZSU


On Apr 2, 2005, at 6:30 AM, Robert Bruninga wrote:
>
> True.  But in LA you simpily continue to promote WIDE2-2
> only even for mobiles and cut off RELAY entirely.   Yes an
> occasional visitor may come in and get WIDE1-1,WIDE2-1
> double hop, but it should not be a problem.  Also since
> it is only 4 digis that are currently enforcing the one-hop
> scenario, if worst comes to worse, we simply "fix" those
> digis with software if.
>
> Yes LA is unique.  Maybe we can think of other tweaks to
> it to help make this fantastic WIDE1-1,WIDE2-1 idea work
> (where needed).  Still mobiles should use not use it unless
> the area requires fill-in digis...
>





More information about the aprssig mailing list