The FIX to the RELAY-DUPLICATION Problem 2 April 2005 -------------------------------------------------------------------------- Revised 20 March 2006. The original APRS path method of RELAY is obsolete in the USA and its legacy causes 2 to 5 times as many duplicated packets as required for those stations that use it in areas where KPC-3 TNC's are predominantly used. Until he New-N Paradigm, there had been no workable subsitute that would eliminate the dupes and still work seamlessly with WIDEn-N and still use existing hardware. But on the suggestion of WA8LMF and refined by discussion on the SIG, we found the solution in April, 2005. Here is the overall APRS USER recommendation for most situations: WIDE1-1 One HOP works everywhere and via HOME stations WIDE2-2 General mobile recommendation and metro areas WIDE3-3 Rural areas with no mountains for hundreds of miles But for mobiles that live in an area where FILL-IN digis are common, (these used to be called RELAY digis in the USA) then there are two new paths that will work the way the old RELAY,WIDE2-2 paths used to work: RELAY,WIDE2-2 The OBSOLETE way which causes dupes and 3 hops WIDE1-1,WIDE2-1 New TWO hop path which is safe to use everywhere WIDE1-1,WIDE2-2 New 3 hop path that may be used in remote areas This solution eliminates the really bad DUPE problem caused by the old RELAY,WIDE... paths. All it does is substitute the exact "WIDE1-1" path for "RELAY". This is not needed in Europe where most digis eliminate dupes between RELAY's and WIDEN-N's. Here is how it works: Both DIGIS and the fill-in home digis will respond to WIDE1-1, but only for one hop. This is just like RELAY used to be. But the real advantage is that once it is digipeated by any digi, even a home station, that same digi will never digipeat it again, because it always looks like a WIDEn-N packet and the WIDEn-N system has the perfect DUPE-ELIMINATION algorithm. WARNING: However, like anything, the setting can be abused. Just like RELAY, users should never use WIDE1-1 anywhere except the first hop in a path. Just like RELAY was never used beyond the first hop, because it keys up all of the fill-in-digis at once. Now that this solution solves the RELAY-DUPE problem in the USA but still allows us to to continue with the ability to have fill-in digis that still will help the mobiles in fringe areas, we can proceed full speed with implementing the New n-N Paradigm. That is: SUMMARY: Everything is now WIDEn-N with small values of N (typically 2 in most areas). RELAY, WIDE, TRACE, TRACEn-N and SS are all obsolete in the USA, though RELAY may still be used in Europe or other continents without the RELAY problem. For occassional state-wide nets, the alternate of SSn-N is also supported in most states. Done! de WB4APR, Bob