[OZAPRS] APRS on 30 metres HF (Robert Bruninga)
Dion Bramich
vk7ybi at yahoo.com.au
Fri Feb 9 10:00:58 EST 2007
On Fri, 09 Feb 2007 09:12:00 +1100, Robert Bruninga <bruninga at usna.edu>
wrote:
>> Yes, I understand the importance of the path length limit,
>> but why WIDE1-1? Why not WIDE2-1? ...GATE,WIDE1-1 is
>> the same as GATE,RELAY in the old setup. GATE,WIDE2-1 is
>> the same as GATE,WIDE.
>
> You are entirely correct.
Thanks, it helps to know that..
>
>> ... a frame is addressed to GATE,WIDE1-1 will activate 2
>> extra local fill-ins unnecessarily, as well as all the
>> WIDEn-N's, doing nothing at all to help the frame get to
>> an IGate, and wasting valuable air time.
Thanks for the edit!
>
> You are correct here as well. And correct that either solution
> has advantages and disadvantages. So let me try to explain my
> thinking. Either answer might be right, but I think we need one
> answer so that things remain simple for all. I'll number the
> points to keep track of the ideas. And you are right that most
> of these poitns can go either way too...
>
> 1) Since this is HF, the path recommendation is not a local
> thing, but will affect everyone everywhere. Thus, we don't want
> multiple regional recommendations. We want -one- that works
> everwyhere
>
I believe the main men of APRS in Australia are writing guidelines at the
moment, so I thouhgt I'd raise the issue early.
> 2) Your comaprison of GATE,WIDE1-1 to GATE,RELAY is correct,
> except for a subtle change that has evolved. Back when APRS was
> first introduced, -every- home station was expected to be a
> RELAY, but now with the change to WIDE1-1 for fill-in-digi's
> only, it is not every home station that is impacted, but only
> those special fill-in digis which are much-much less numerous
> than the old RELAY concept. In your example, only 2 are
> affected.
True.
>
> 3) I am surprised that your local GATE can hit all of the local
> WIDEn-N and both of the 2 WIDE1-1 digis directly. In that case,
> the local fix is to reduce the power of the HF gate so that it
> doesn't hit the WIDE1-1's, and hits only the WIDEn-N's. A local
> fix to a local problem.
Not all all of the time, but most some of the time. Propagation
dependant.
>
> 3) I suggested GATE,WIDE1-1, because it will go thorugh any
> remote GATE that may only have a local WIDE1-1 to get it to the
> nearest Igate. Whereas using GATE,WIDE2-1 would not make it
> through those few wilderness Igates.
I agree with that too.. I think better network planning can fix that
problem, however.
>
> 4) Another reason is that I get tired of always having to
> explain that WIDE2-1 is the same as WIDE1-1, and so by
> receommending GATE,WIDE1-1, everyone can easily see that it is
> only 1 hop on VHF. See reason #1
>
When there is a 1 there, it cant be mistaken for anything else, however,
see pt. 5
> 5) Packets originated as GATE,WIDE2-1 arrive as GATE,DIGI,WIDE2*
> and this implies that the originator took two hops on VHF in
> violation of the receommendations. When in fact he sent WIDE2-1
> properly. But the two are indistinguishible on receipt, so why
> put everyone at risk of self-policing doo-gooders. If we
> recommend GATE,WIDE1-1 then they all arrive as GATE,DIGI,WIDE1*
> and everyone is happy and can prove they did it right.
>
Two hops would look like GATE,DIGI,DIGI,WIDE2*?
One hops would look like GATE,DIGI,WIDE2* or
GATE,DIGI,WIDE1*
3 different paths, two different results and I can see how they all
arrived. Obviously only works with New Paradigm, which we'll hopefully
have nation wide soon, so no real problem? Do-gooders are a problem,
their time would be better spent educating those doing it wrong than
telling them they are and what they need to do. If we pick it up early
and get it into the guidelines, there will be no trouble? A HF mobile op
uses a non VHF path, and if hes new to the HF thing he needs to read about
it somewhere? There need to be a definitive set of guidelines posted
right at aprs.net.au
> So when I added up all the pro's and con's, I came up with the
> conclusion that GATE,WIDE1-1 looked the best for recommending
> universally.. Mostly due to the unresolvable problem of #5
>
What weight do you give problem 5? If we can reduce its weight through
proper user education the scales might tip the other way?
>> So, if keeping traffic low is the number one rule,
>> then why are we advertising a path that increases it
>> needlessly?
>
> Actually, it does not increase traffic:
>
> A) all digipeats of a packet should all occur at the same time
> so that no additional time slots are used no matter whether it
> is a WIDE1-1 or a WIDE2-1.
>
I have one fill-in built on MFJ firmware, its faster than the WIDEn-N
digis, therefore uses an extra time slot as the network is CSMA
> B) Letting all HF packets arrive on VHF as GATE,WIDE2* even
> though only 1 hop was used might be easily missinterpreted as
> GATE,WIDE2-2, and so it migh tactually encourage the use of the
> wrong path and cause increased load unintentionally...
Understood, but as I pointed out in pt. 5, if a person is properly
educated and is actually reading what theyre looking at in the path then
how can it be mistaken?
Dion.
>
>
> Just my thoughts.
> Bob, Wb4APR
>
>> On Fri, 09 Feb 2007 01:29:55 +1100, Robert Bruninga
>> <bruninga at usna.edu>
>> wrote:
>>
>> >> Why WIDE1-1?
>> >
>> > See HF.TXT.
>> >
>> > Because all HF packets in the world using the path of VIA
>> > GATE,WIDE1-1, have the potential to enter ALL local VHF
> networks
>> > all over the world. This is OK, when they get gated into a
>> > local
>> > network *locally*.
>> >
>> > But if those same HF operators use the path of VIA
>> > GATE,WIDE2-2, then there are two BIG problems: 1) they
> would be
>> > multiplying their potential presence in *every* local area
> by a
>> > factor of 4 to 5 or more everywhere because they would be
>> > flooding everywhere. And 2) the AREA being QRM'ed is no
> longer
>> > "local" to the GATE doing the flooding, and hence the areas
>> > being QRM'ed have no local control over the flood in their
> area.
>> >
>> > It is the number 1 rule in APRS. Do unto others as you
> would
>> > have them do unto you.
>> >
>> > Due to the low bandwidth of the 1200 baud APRS channel, it
> is
>> > best when local sources of outside area QRM are kept under
> Local
>> > control. Hence, the courteous HF operator uses a path no
> longer
>> > than GATE,WIDE1-1 to keep his packets *local* to the Distant
> HF
>> > gate that is injecting them into his -local- area.
>> >
>> > Just makes it courteous to everyone else...
>> >
>> > Bob, WB4APR
>> >
>> >
>> >
>> >> On Thu, 08 Feb 2007 23:55:41 +1100, Radio
> <radio at dodo.com.au>
>> > wrote:
>> >>
>> >> > Well, I based my original setup on the pages that Tony,
>> >> VK5AH quoted in
>> >> > good
>> >> > faith. Today, I phoned Ross, VK3MY and he suggested GATE,
>> >> WIDE1-1 was the
>> >> > correct path. I guess there are now conflicting settings
>> >> that need to be
>> >> > adjusted, to suit a common agreed path setting for HF
>> >> stations on 30
>> >> > metres, that are mobiles etc.
>> >> >
>> >> > Jack VK4JRC QG56FJ
>> >> >
>> >> > Tony wrote....
>> >> > Snip>
>> >> >
>> >> > Just isnt true.. Recomending anything with WIDE1-1 isnt
>> > allways
>> >> > guaranteed
>> >> > to get very far either
>> >> >
>> >> > That might all work nice for you but are we really
> expecting
>> > the
>> >> > Ionosphere
>> >> > to work that well that we can make regular use of your HF
>> >> gates over
>> >> > there
>> >> > ?? I think not. By the way we have our own HF gates over
>> > here and
>> >> > sometimes
>> >> > they pick up posit beacons transmitted from your gateways
> ..
>> >> >
>> >> > Maybee for the best solution to the question in VK land
>> >> have a look at
>> >> > this
>> >> > web page which isnt all that dated.
>> >> > http://www.aprs.net.au/hf_aprs_information.htm
>> >> >
>> >> >
>> >> > Tony Hunt VK5AH
>> >> >
>> >> >>
>> >> >> Message: 2
>> >> >> Date: Wed, 7 Feb 2007 18:39:58 -0500
>> >> >> From: "Robert Bruninga" <bruninga at usna.edu>
>> >> >> Subject: Re: [OZAPRS] APRS on 30 metres HF
>> >> >> To: "'Radio'" <radio at dodo.com.au>, "'VK / ZL APRS
> Users'"
>> >> >> <ozaprs at aprs.net.au>
>> >> >> Message-ID: <026e01c74b11$473c7460$42577a83 at SGSbb>
>> >> >> Content-Type: text/plain; charset="us-ascii"
>> >> >>
>> >> >> > I have tried several HF path variations like GATE,
>> >> >> > WIDE and did not make it to www.findu.com
>> >> >> >
>> >> >> > What is the correct path statement that should be
> used?
>> >> >>
>> >> >> The path that has always been recommended since we first
>> > began
>> >> >> HF APRS on that freequency back in 1993, that is, VIA
>> >> >> GATE,WIDE1-1. (it used to be VIA GATE,WIDE, but the
> WIDE
>> > part
>> >> >> of that is now obsolete and will not work). See the
>> > original
>> >> >> HF.TXT file. See the web page:
>> >> >>
>> >> >> http://www.ew.usna.edu/~bruninga/gates.html
>> >> >>
>> >> >> Do not use ECHO. The page explained above tells why...
>> >> >>
>> >> >> Thanks
>> >> >>
>> >> >> Bob, Wb4APR
>> >> >>
>> >> >
>> >> > _______________________________________________
>> >> > Ozaprs mailing list
>> >> > Ozaprs at aprs.net.au
>> >> > http://aprs.net.au/mailman/listinfo/ozaprs
>> >> > ________________________________________________
>> >> >
>> >> >
>> >> > Dodo - an Official
>> >> > Sponsor of the
>> >> > 2006 FORMULA 1 (tm)
>> >> > Foster's Australian Grand Prix
>> >
>> > _______________________________________________
>> > Ozaprs mailing list
>> > Ozaprs at aprs.net.au
>> > http://aprs.net.au/mailman/listinfo/ozaprs
>>
>>
>>
>> --
>> Using Opera's revolutionary e-mail client:
> http://www.opera.com/mail/
>> Send instant messages to your online friends
>> http://au.messenger.yahoo.com
>>
>
>
>
--
Using Opera's revolutionary e-mail client: http://www.opera.com/mail/
Send instant messages to your online friends http://au.messenger.yahoo.com
_______________________________________________
Ozaprs mailing list
Ozaprs at aprs.net.au
http://aprs.net.au/mailman/listinfo/ozaprs
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://second.aprs.net.au/pipermail/ozaprs/attachments/20070209/88f11b82/attachment.htm
More information about the Ozaprs
mailing list