| |
ZL3AI > APRDIG 20.09.03 10:17l 304 Lines 10227 Bytes #999 (0) @ WW
BID : 1955-ZL3AI
Read: GUEST
Subj: TAPR Digest, Sep 17, 2/7
Path: DB0FHN<DB0RGB<DB0FSG<DB0PV<DB0MRW<OK0PKL<OK0PPL<DB0RES<ON0BEL<CX1AAO<
EA2RCF<W4JAX<VK4TUB<ZL2BAU<ZL3VML
Sent: 030919/1343Z @:ZL3VML.#80.NZL.OC #:8663 [Chch-NZ] FBB7.00i $:1955-ZL3AI
From: ZL3AI@ZL3VML.#80.NZL.OC
To : APRDIG@WW
Subject: Re: Paths
From: Bob Bruninga <bruninga@usna.edu>
Date: Wed, 17 Sep 2003 08:56:52 -0400 (EDT)
X-Message-Number: 5
On Tue, 16 Sep 2003, Germino wrote:
>Bob: Is WIDE,WIDE1-1 better than WIDE,WIDE?
Someone correct me if I am wrong, but I think they will work identically
in most digipeaters. The WIDEn-N has a better dupe elimination algorithm,
but only after it has been digipeated. And since WIDE1-1 is only one hop,
there are no more dupes to eliminate.
In fact, I am slowly concluding that the great "WIDEn-N" concept is past
its prime. Back in 1994 when I came up with it, it had advantages over
long paths. (But this was before we had Steve Dimse's APRS-IS.). In
today's world, WIDEn-N shows little advantage for small numbers of N
(which we all agree are what we should be using).
So in my mind a self identifying 2 hop path begining with WIDE is really
the norm in developed areas. Or for brevity without caring about the
point of entry, then WIDE2-2...
Bob
----------------------------------------------------------------------
Subject: RE: Proper DIGI ID interleaving
From: Bob Bruninga <bruninga@usna.edu>
Date: Wed, 17 Sep 2003 09:23:11 -0400 (EDT)
X-Message-Number: 6
On Tue, 16 Sep 2003, Cap Pennell wrote about their way in Calif:
>Something like this .txt file for the hypothetical KPC-3Plus:
>BLT 1 EVERY 01:00:00 START 00:00:00
>BLT 2 EVERY 00:30:00 START 00:10:00
>BLT 3 EVERY 00:30:00 START 00:20:00
>BLT 4 EVERY 01:00:00 START 00:30:00
>LTP 1 APN383 VIA WIDE2-2
>LTP 2 APN383
>LTP 3 APN383
>LTP 4 APN383 VIA WIDE
Yes, that's a great one for the 10 min direct, 30 min 1 hop and 1 hour
2 hop. Good idea as shown below:
2 0 0 1 0 0 2 0 0 1 0 0 2 0 0 1 0 0 2 0 0 1 0 0 2 0 0 1 0 0 2 0 0 1
1 0 1 0 1 0 1 0 1 0 1 0
0 0 0 0 0 0
compared to my 10 min direct, 20 min 1 hop, 40 min 2 hop, 160 min 3 hop:
> 0 1 0 2 0 1 0 3 0 1 0 2 0 1 0 . 0 1 0 2 0 1 0 3 0 1 0 2 0 1 0 . 0 1
> 0 1 0 2 0 1 0 0 1 0 2 0 1 0 0
> 0 1 0 0 1 0
> 0 0
> DIRECT every 20 starting at 00
> WIDE every 40 starting at 10
> WIDE2-2 every 80 starting at 30
> WIDE3-3 every 160 starting at 70
I bet there is a better way than mine to work in the 3 hour or so
3-3 without the slightly higher local density... I gotta prepare for the
hurricane, so cant think much more about this today, though...
Bob, WB4APR
----------------------------------------------------------------------
Subject: Re: Paths
From: "Spider" <spider@rivcom.net>
Date: Wed, 17 Sep 2003 06:24:29 -0700
X-Message-Number: 7
----- Original Message -----
From: "Bob Bruninga" <bruninga@usna.edu>
To: "TAPR APRS Special Interest Group" <aprssig@lists.tapr.org>
Cc: "TAPR APRS Special Interest Group" <aprssig@lists.tapr.org>
Sent: Wednesday, September 17, 2003 5:56 AM
Subject: [aprssig] Re: Paths
>On Tue, 16 Sep 2003, Germino wrote:
>
>>Bob: Is WIDE,WIDE1-1 better than WIDE,WIDE?
>
>Someone correct me if I am wrong, but I think they will work identically
>in most digipeaters. The WIDEn-N has a better dupe elimination algorithm,
>but only after it has been digipeated. And since WIDE1-1 is only one hop,
>there are no more dupes to eliminate.
WIDE,WIDE causes the fisrt digi to insert it's call and then the second
digi inserts it's call on the second WIDE.
So, you'll see a packet with two calls inserted if it goes thru two digis.
Jim
----------------------------------------------------------------------
Subject: Re: Paths
From: Bob Bruninga <bruninga@usna.edu>
Date: Wed, 17 Sep 2003 09:32:04 -0400 (EDT)
X-Message-Number: 8
On Wed, 17 Sep 2003, Spider wrote:
>>>Bob: Is WIDE,WIDE1-1 better than WIDE,WIDE?
>
>WIDE,WIDE causes the fisrt digi to insert it's call and then the second
>digi inserts it's call on the second WIDE. So, you'll see a packet with
>two calls inserted if it goes thru two digis.
Good point Jim! And the more I think about it, that means automactic
tracing of all paths just like we had back at the beginning back before
we got off onto the WIDE3-3, 4-4, 5-5, 6-6 and 7-7 concept (back before
the APRS-IS).
If we are cutting back to WIDE2-2 in this new paradigm shift because of
the APRS-IS, and the realization that 2 hops gets you everywhewre you
need to go (within your LOCAL area (as defined as that area surrounding
you with 50 or so other stations)) then maybe WIDE,WIDE is the right way
to go?
It sure conveys a lot more information to users when packets arrive as via
W3ABC,W4XYZ than having them always arrive as WIDE2-0! And the
cost of 7 bytes is small compared to the value... Hummh...
de WB4APR@amsat.org, Bob
----------------------------------------------------------------------
Subject: Re: Paths
From: Mark Cheavens <mcheavens@usa.net>
Date: Wed, 17 Sep 2003 08:59:03 -0500
X-Message-Number: 9
Bob and others.
The long standing DUPE problem is associated with KPC-3.
DUPE checking on a KPC-3 ONLY occurs in n-n packets.
If you have a path of Relay, Wide2-2 and a KPC-3 hears the packet first and
digipeats the packet based upon the RELAY, When the next digi hears the
packet and acts upon the Wide2-2 and digipeats the packet as Wide2-1 the
first digipeater will digipeat the packet again. This is why relay and wide
are acted upon in some areas with KPC-3's.
NOTE: I have never checked for this bug in a KPC-3+ (I don't own one to
test with)
UiDigi stores ALL packets in it's dupe checking buffer.
Mark Cheavens
KC5EVE
----------------------------------------------------------------------
Subject: RELAY,WIDE,WIDE
From: Mark Earle <wa2mct@mearle.com>
Date: Wed, 17 Sep 2003 08:58:10 -0500
X-Message-Number: 10
OK.. what areas of the country REQUIRE relay in the user path?
I've been running WIDE3-3 for some time, but an thinking of changing my two
profiles to:
WIDE,WIDE
RELAY,WIDE4-4 (for trips to the boonies)
Are there areas where the preceeding relay is either:
required, ie, no digipeat without it?
or
it breaks things, no digipeat becuase relay is there?
) ) de WA2MCT Mark
( (
) ) You will be assimilated... oooh, coffee!!
_|****| http://www.findu.com/cgi-bin/find.cgi?wa2mct-8 Home
( | | http://www.findu.com/cgi-bin/find.cgi?wa2mct-9 Mobile
`|____| wa2mct@mearle.com wa2mct@juno.com www.mearle.com/mark
----------------------------------------------------------------------
Subject: Re: Paths
From: Robbie <mwrobertson@comcast.net>
Date: Wed, 17 Sep 2003 09:13:47 -0500
X-Message-Number: 11
I thought one a the good features of WIDEn-N was that some of the WIDE
digipeaters dupe check with this, and if you have WIDE,WIDE,WIDE, some
digipeaters hear it coming back and transmit it again. I have seen this
many times in the Chicagoland area.. With callsign sub in this case,
anyone can see it happen..
I think WIDE,WIDE is appropriate, and anything beyond 2 hops, would be
WIDE,WIDEn-N
There has to be a better dupe checking process?
Robbie, wa9inf
----------------------------------------------------------------------
Subject: Re: RELAY,WIDE,WIDE
From: Mark Cheavens <mcheavens@usa.net>
Date: Wed, 17 Sep 2003 09:22:50 -0500
X-Message-Number: 12
In West Texas (North) where there are many KPC-3's Relay is NOT allowed.
(because of packet duplication bug, see my previous email).
I am NOT the sysop of any of those TNC's.
Mark
KC5EVE
----------------------------------------------------------------------
Subject: PCSat does it again.
From: Ken Collins <g4fip@btinternet.com>
Date: Wed, 17 Sep 2003 15:23:08 +0100
X-Message-Number: 13
Hopefully this morning I did make a two way contact via PCSat with
KB1GVR. I say hopefully because my reply to his message was digipeated
by PCSat just as the satellite footprint was leaving the US. Here are
the details:-
09:22:57R KB1GVR>APU16K,PCSAT-1* Port=1 <UI Len=28>:
:G4FIP-1 :GM de Mark 73{02
09:22:58T G4FIP-1>APU25H,PCSAT-1 Port=1 <UI Len=17>:
:KB1GVR :ack02
With little time to spare I could only send an ordinary unconnected
packet from the telemetry decoder program I was using, however, after
about 3 retries I saw the following:-
09:23:51T G4FIP-1>KB1GVR,W3ADO-1,SGATE Port=1 <UI Len >:
Hi there. 73 de Ken
09:23:53R G4FIP-1>KB1GVR,PCSAT-1*,SGATE Port=1 <UI Len >:
Hi there. 73 de Ken
This really is a fun satellite. Thanks to Bob and the crew who put it
together!!
--
Ken Collins - G4FIP
E-mail:- g4fip@btinternet.com
http://www.btinternet.com/~g4fip
----------------------------------------------------------------------
Subject: RE: Proper DIGI ID interleaving
From: Bob Bruninga <bruninga@usna.edu>
Date: Wed, 17 Sep 2003 11:28:34 -0400 (EDT)
X-Message-Number: 14
More ideas.
On Tue, 16 Sep 2003, Cap Pennell suggested for 10 min direct, 30 min 1
hop, 1 hour 2 hop:
>>WIDE2-2 EVERY 01:00:00 START 00:00:00
>>DIRECT EVERY 00:30:00 START 00:10:00
>>DIRECT EVERY 00:30:00 START 00:20:00
>>WIDE EVERY 01:00:00 START 00:30:00
>
> 2001002001002001002001002001002001
> 1 0 1 0 1 0 1 0 1 0 1 0
> 0 0 0 0 0 0
>
>compared to a 10 min direct, 20 min 1 hop, 40 min 2 hop, 160 min 3 hop:
>
>> 0 1 0 2 0 1 0 3 0 1 0 2 0 1 0 . 0 1 0 2 0 1 0 3 0 1 0 2 0 1 0 . 0 1
>> 0 1 0 2 0 1 0 0 1 0 2 0 1 0 0
>> 0 1 0 0 1 0
>> 0 0
>> DIRECT every 20 starting at 00
>> WIDE every 40 starting at 10
>> WIDE2-2 every 80 starting at 30
>> WIDE3-3 every 160 starting at 70
Here is another one based on Cap's with slightly longer legs and
slightly less 2 hop QRM but one lost direct packet every 2 hours:
3 0 0 1 0 0 . 0 0 1 0 0 3 0 0 1 0 0 . 0 0 1 0 0 3 0 0 1 0 0 . 0 0 1
2 0 0 2 0 0 2 0 0
1 1 1
0 0 0
Just some options depending on local considerations.
Bob...
----------------------------------------------------------------------
Read previous mail | Read next mail
| |
