[aprssig] The Current Meaning of WIDEn-N (incorrect)

[Stephen H. Smith]

On 5/19/2014 1:47 PM, Robert Bruninga wrote:
> With apologies…, unfortunately the below information is wrong.
>
> ØToday, effectively the first "n" of WIDEn-N now indicates whether or not a
> home fill-in digi can potentially be used for the first relay hop of a
> multi-hop path.
>
> While that may be an “effective interpretation” of what people observe, it is
> not correct.
>
>


HOW is it "not correct"?   If you start the path with WIDE1-1, you will use a 
home WIDE1-1 digi, if one is available.  If there isn't one around, and there 
IS a true wide around, it will respond.    Or in the Tassajera / SLO case where 
most likely one of each are within range, BOTH will respond simultaneously.


>>If the first "n" is a "1", then *EITHER* a home fill-in digi --OR-- a
> high-level true wide can handle the hop.
>>If the first "n" is other than "1", then ONLY a true wide will respond to and
> handle the hop.
>
> Simply not true.  The first “n” is the original N hop when the packet was
> initiated.

So how does this invalidate my statement?

A "dumb" home digi TNC is going to respond exclusively to "WIDE1-1".  It will 
not respond to any path elements (a.k.a. "clauses" in my usage) that are 
anything other than exactly WIDE1-1.

On the other hand, true wides, with functional n-N decrementing, 
firmware/software will respond to any combination of "n" and "N" values in 
WIDEn-N.

Thus effectively today, WIDE1-1 says "let home digis be involved in the first 
hop digipeating", while any other value of "n" and/or "N" in WIDEn-N 
effectively says "home fill-digis stay out of this matter".


The exception (which causes all this confusion) is the  special
> “WIDE2-1” which is simply a cheating way to get ONE HOP without keying up all
> surrounding fill-in WIDE1-1 digis.

"Cheating" on the WIDE2-n path has nothing to do with home digis not responding 
to anything except "WIDE1-1".

For example, if I want to transmit a
> one-hop packet, but I don’t want to bring up my neighbors WIDE1-1 digi, and the
> one that might be in a neighboring car (heaven forbid), I would use the single
> hop path of WIDE2-1.  Which has always been a special case of a WIDEn-N packet
> only intended to go one hop and without bringing up all the old-fill-in-digis
> that can only operate on “WIDE1-1”.
>
>
>>Today the *total* number of digipeater relay hops requested by the user is the
> sum of the first N of the first clause (normally "1") and the first N of the
> second clause (typically "2"), minus any "pre-decrementing" indicated by the
> second "N" of a clause initially being smaller than the first.  I.e.
>
> Simply not true.


How is this "not true"????

If you use the standard WIDE1-1,WIDE2-1 path, you will (potentially, if the 
digis exist) one hop from the clause "WIDE1-1" and one hop from "WIDE2-1" for a 
total of two;   i.e. "n=1" + "n=2" minus the "pre-decrement" of WIDE2-1 = 2 
hops total.

WIDE1-1,WIDE2-2 would potentially yield THREE hops; i.e. "n=1" + "n=2" = 3


>
> ØThis is why WIDE1-1 is completely discouraged in the greater Los Angeles area.
>
> Should not be so.

You've been saying "NO" for 15-20 years now, but it is still SO.

>
> ØThe basin is completely rimmed with mountain-top "super-wides" that can hear
> and digipeat everyone everywhere on a single hop, without the help of home
> stations. The recommended path is "WIDE2-2" only which the local digis
> completely consume in a single hop.  When you travel out of the greater LA area
> (i.e. over the Tejon or Cajon passes or beyond SBA on the 101), you
> "automagically" start getting two hops from mountain tops that process
> "WIDE2-2" normally; i.e. as two hops.
>
> In SoCal and anywhere else where the Aloha circle is entirely contained within
> a single hop, it is the responsibility of the digi owners to implement 1-hop
> enforcement so that all mobiles don’t have to change their paths depending on
> where they dirve!


Which is exactly what they are effectively doing.   The WIDE2-2 path is trapped 
in the local digis in a manner similar to home WIDE1-1 fill-ins; i.e. a literal 
alias of "WIDE2-2" with no n-N decrementing stuff.   When you go "over the 
hill" (literally) to central CA or the Mohave and come into the range of digis 
that DO do the n-N decrementing thing, you get two hops.