[OZAPRS] Help with digi paths

[David and Ann Fraser]

On Mon, 14 Mar 2005 14:53:56 +1100, Richard Hoskin wrote:

> Hi Dave,

Hi Dick,

>> -----Original Message-----
>> I can quite understand the responses to my posting, as it flies in the
>> face of conventional logic for connected mode packet.  But APRS is a
>> one-to-many broadcast scheme.

> Not at all. The local network design and tuning is based on local
conditions
> and requirements. If you have a low density of ARPS stations you can run
> your network in a different fashion and to the best advantage of the
locals
> who use the network. (The definition of local in this case is the local
> State)

How can triplicates of every packet be an advantage?

>> Appended is a relevant posting made by WB4APR, and extracts from
>> UIDIGI and DIGI_NED configuration files (used worldwide, not just the
>> USA).

> Again these comments are based on the USA experience where they have a
large
> number of ARPS operators, Operators with excessive paths of greater than
2
> hops and legacy digipeaters (Digis that run KPC or Pacomm firmware).

No.  They are general statements as a result of debate around September
2003 about APRS networks.  They apply to any APRS network and improve
performance no matter what the load, number of users, paths used, etc.
One-to-many APRS broadcasts are totally different to connected mode
operation.  They should radiate cleanly out from the sender, not bounce
around in a random manner cluttering up the channel.

> Why build a Billion dollar high speed freeway when you only need a
Million
> dollar highway with scenic rest spots along the way.

But why litter it with speed bumps every 300 metres?

The original poster complained about seeing triplicates.  I'd be grumpy
too, if I was trying to get reliable performance out of APRS in your
area.  We had mere duplicates coming from a local Igate and even that
caused some grief on our lightly loaded channel, I'd hate to think what
triplicates would have done!  No wonder the first poster complained.

73 David.

_______________________________________________
ozaprs mailing list
ozaprs at marconi.ics.mq.edu.au
http://marconi.ics.mq.edu.au/cgi-bin/mailman/listinfo/ozaprs