| |
ZL3AI > APRDIG 14.02.04 05:06l 246 Lines 9249 Bytes #999 (0) @ WW
BID : 2839-ZL3AI
Read: GUEST
Subj: TAPR Digest, Feb 11, 3/3
Path: DB0FHN<DB0RGB<OK0PPL<DB0RES<ON0AR<7M3TJZ<ZL3VML
Sent: 040214/0241Z @:ZL3VML.#80.NZL.OC #:19064 [Chch-NZ] FBB7.00i $:2839-ZL3AI
From: ZL3AI@ZL3VML.#80.NZL.OC
To : APRDIG@WW
Subject: question about 'timestamping' with IGATEd traffic to APRS-IS
From: James Washer <washer@trlp.com>
Date: Wed, 11 Feb 2004 16:00:04 -0800
X-Message-Number: 22
When a local Igate sends a packet up to APRS-IS ( aprswest for example )..
and that data is later made available through findu ( for example )...
Where does the timestamp come from?
For example, the following URL
http://www.findu.com/cgi-bin/rawposit.cgi?call=kg7hh&time=1&start=1&length=1
Will yield
20040211231857,KG7HH>APX130,TCPIP*,qAC,KG7HH:=3901.83N/11944.89WyXastir/Linux
7/24 iGate
20040211234858,KG7HH>APX130,TCPIP*,qAC,KG7HH:=3901.83N/11944.89WyXastir/Linux
7/24 iGate
What I want to know is when/how does that timestamp at the beginning of the
line come to be.. Does APRS-IS generate it at the time the data is
received?
thanks
- jim
p.s. The reason I'm asking, is there are two igates near here, and one of
them tends to report the data 5 to 10 minutes late.. which causes object to
appear jump back and forth as they move through the area.
----------------------------------------------------------------------
Subject: Re: VX-2R Tiny HT HF SSB?
From: Dale Blanchard <wa7ixk@arrl.net>
Date: Wed, 11 Feb 2004 16:28:31 -0800
X-Message-Number: 23
That was done in the old days with tubes. Some times just by putting
another tunable radio next to it. I can do it with my MJF ant analizer.
It should be possible to build something the size of a match book for
very little these days.
Bob Bruninga wrote:
>The teeney-tiny VX-2R HT has HF receive but AM only.
>
>What we need is somebody to make a little 450 KHz
>external BFO oscillator on a watch battery in something
>the size of a matchbook. Then just hold it on the
>back of the radio and you get a BFO for tuning in
>CW, SSB and PSK-31.
>
>I just tried a sig gen using 10 mW at 450 KHz into a
>3 turn coil near the base of the radio. It works
>fine. Great little market for someone to
>make 'em and sell em!
>
>de WB4APR, Bob
----------------------------------------------------------------------
Subject: Re: multi emails?
From: Steve Dimse <k4hg@tapr.org>
Date: Wed, 11 Feb 2004 19:40:31 -0500
X-Message-Number: 24
On 2/11/04 at 5:47 PM blitz <blitz@macronet.net> sent:
>Not everyone and not all the time, just every once in a while, like Steves
>last email?
Most of the messages that pertain to findU's status I send on both the APRS
SIG and APRS NEWS lists, perhaps you are subscribed to both?
Steve K4HG
----------------------------------------------------------------------
Subject: Re: question about 'timestamping' with IGATEd traffic to APRS-IS
From: Steve Dimse <k4hg@tapr.org>
Date: Wed, 11 Feb 2004 19:42:10 -0500
X-Message-Number: 25
On 2/11/04 at 4:00 PM James Washer <washer@trlp.com> sent:
>What I want to know is when/how does that timestamp at the beginning of the
line
>come to be.. Does APRS-IS generate it at the time the data is received?
The timestamp is the UTC when the packet was received by findU. I ignore the
timestamps within packets, as findU is more accurate than almost all of them
(both servers are synced with NTP to a level one time server).
Steve K4HG
----------------------------------------------------------------------
Subject: Re: question about 'timestamping' with IGATEd traffic to APRS-IS
From: "Scott Miller" <scott@3xf.com>
Date: Wed, 11 Feb 2004 17:59:26 -0800
X-Message-Number: 26
>The timestamp is the UTC when the packet was received by findU. I ignore the
>timestamps within packets, as findU is more accurate than almost all of them
>(both servers are synced with NTP to a level one time server).
Say what? How do you figure that? When I transmit a timestamp it's because
I want to know when that fix was taken, not when it managed to make its way
into the network. For weather stations that don't have a GPS and are
possibly relying on an internal RTC I can maybe understand it, but trackers
have very accurate timestamps provided along with the position from the GPS.
And even if we ARE talking about weather stations, what if I want to take
readings at 2-minute intervals but only transmit them in a batch every 10
minutes? You're saying that the timestamps are completely ignored and that
they won't plot correctly? I can't see how it would ever be a good thing to
second-guess information that's explicitly provided in the packet.
Scott
N1VG
----------------------------------------------------------------------
Subject: Re: VX-2R Tiny HT HF SSB?
From: "Stephen H. Smith" <WA8LMF2@aol.com>
Date: Wed, 11 Feb 2004 18:05:08 -0800
X-Message-Number: 27
Bob Bruninga wrote on 2/11/2004, 3:41 PM:
>The teeney-tiny VX-2R HT has HF receive but AM only.
>
>What we need is somebody to make a little 450 KHz
>external BFO oscillator on a watch battery in something
>the size of a matchbook. Then just hold it on the
>back of the radio and you get a BFO for tuning in
>CW, SSB and PSK-31.
>
>I just tried a sig gen using 10 mW at 450 KHz into a
>3 turn coil near the base of the radio. It works
>fine. Great little market for someone to
>make 'em and sell em!
Almost certainly will be unusable on PSK31. Simple synthesizers for FM
and AM applications usually have considerable "wobble" and phase
jitter. This has essentially no effect on AM or FM, but is completely
disasterous on PSK31 where the frequency stabilty has to be within a few
Hz and the short-term PHASE stability must stay within 40-50 degrees.
Even if the external "BFO" is crystal-stable and noise-free, it will be
beating against a signal down-converted by a wobbly, jittery synthesizer.
Try the BFO scheme you described with a clean stable RF gen at 450 KHz.
Put a second clean stable generator at the operating frequency on HF
as a test "station". Tune the radio so that a low beat note is
produced. I suspect that you will hear not a pure clean note, but
rather a sort of growling, "gargling" sound due to the short-term
instability of the synthesizer.
Stephen H. Smith wa8lmf (at) aol.com
Home Page: http://wa8lmf.com
Ham Radio/Mobile SSTV page: http://members.aol.com/wa8lmf/ham
APRS Stuff
http://members.aol.com/wa8lmf/aprs
----------------------------------------------------------------------
Subject: Re: question about 'timestamping' with IGATEd traffic to APRS-IS
From: James Jefferson <jj@aprsworld.net>
Date: Wed, 11 Feb 2004 20:07:34 -0600
X-Message-Number: 28
Steve Dimse said:
>>(both servers are synced with NTP to a level one time server).
Scott Miller said:
>Say what? How do you figure that? When I transmit a timestamp it's
>because I want to know when that fix was taken, not when it managed to make
>its way into the network. For weather stations that don't have a GPS and
aprsworld behaves the same way as findu and timestamps when the packet
enters the database. Many timestamps are garbage and there isn't a good way
to determine which ones are and which are not. So just think of it as a
the database timestamp, not the packet time stamp.
I did some statistics a few years ago on this topic. Of 1 million NMEA
packets that included a time stamp the average time between timestamp and
getting into the aprsworld database was approximately 2.3 seconds. Pretty
good considering this is when the GPS actually spit out the sentence to
when it is received on the other side of the world.
-Jim
----------------------------------------------------------------------
Subject: Weater Station Question Thanks
From: kb2scs@optonline.net
Date: Wed, 11 Feb 2004 21:13:06 -0500
X-Message-Number: 29
Hi All
Thank you all for your very prompt and informative answers to my question.
I have forward them on to my friend.
Thanks again.
Let us hope we never witness the "Silence Of The Hams"
73 DE John KB2SCS
E-Mail: kb2scs@arrl.net
----------------------------------------------------------------------
Subject: Re: question about 'timestamping' with IGATEd traffic to APRS-IS
From: Mike Galgano <mgalgano@nanosecond.com>
Date: Wed, 11 Feb 2004 19:14:27 -0800
X-Message-Number: 30
Yes, what you have is a database stamp, not a time stamp for the origin of
the record. What prompted my interest in this is that I am 'the other
Igate' in J. Washer's query. We have seen some 13 minutes between when he
logs the data, and when it hits my machine and is logged. Interestingly, my
reading of the APRS specs say both findu and aprsworld are dead on spec.
'Real time' is when the report is received, not when it was sent. I
disagree with it, but there it is. Multiple reports of a single record to a
web server, some of them timestamped (within the data) prior to another
record logged in the database, should be ignored if at all possible if this
'real time' spec is to be as accurate as is possible within the protocol
limitations. I know cycles count, and this would cost cycles. It just seems
silly for us to quite deliberately tag the record 'I was here at noon',
only to have it overwritten as current with 'I was over there at 11:47'
simply because the genuinely earlier record took a lunch break along the
way.
Mike Galgano
KB7PUX (-7 when mobile)
---
END OF DIGEST
Read previous mail | Read next mail
| |
