[Xastir] A good time for APRS at 9600?

[Guy Story KC5GOI]

If you look at the overall time involved from the time PTT is keyed up to
the end of the transmission, you do not save that much time for a POSIT.  I
doubt transmission times are half that of 1200 baud.  If the transmit delay
on both 1200 and 9600 are set at the same, say 180ms, and the 1200 baud
posit is 200ms, you have used 380 ms.  At 9600 baud, the tx delay at 180ms
and the posit takes say(rough guess) 50ms, you are at 230 ms. That is for a
small posit.  There is not a direct speed improvement of 8 times faster.
Then you factor in the time that every tnc calculates dwait, slottime and
persist timer in an attempt to avoid transmissions, the benefit of 9600
baud starts to drop out.  That excludes the timing added or subtracted from
a single digipeater.  Only the power users attempt to reduce their
txdelay.  The D700 and D7a (which I own both) defaults to 500ms for the
txdelay.  I do not know about the D710.

I say all that but there is something else to include in all of that.  This
effect is cumulative.  The more stations on 9600 baud, the overall time
freed up is not trivial.  150ms adds up over time and is measurable in just
an hour.

I can see better odds at flutter on mobile stations being reduced. 150ms is
not chump change.

A big improvement is the fact stations are not in 144.390 and somewhere
else.

I never said in 15 years of personal APRS activity including setting up
digipeaters and operating the second igate (aprsD) in the Dallas/Ft Worth
area for better than 12 years that 9600 baud was a bad idea.  All I said
was that this has been talked about and the pitfalls discussed.  I know
about the experiments to a limited extent and I encourage it.  I just do
not have time to start it in Denton Co.

73

Guy



On Sat, Nov 12, 2011 at 12:01 PM, Tom Hayward <esarfl at gmail.com> wrote:


> 9600 baud data is 8 times as fast as 1200 baud. When you factor in the
> txdelay and length of an average APRS packet, 9600 baud APRS comes out
> to about twice as fast as 1200 baud. To me, that seems significant
> enough to be worthwhile.
>
> In the Pacific Northwest, there are a number of people experimenting
> with 9600 baud APRS. They have a reasonable network of i-gates and
> digipeaters. They've shown a couple of things: 9600 baud works for
> APRS; it can be more reliable for mobiles (less chance flutter will
> corrupt the packet, because the data burst is so short); and it ends
> up doing more for the network than halving the bandwidth because it
> moves a lot of stations off 144.39 and onto an alternate channel. Most
> of that network is Kenwood D700's, D710's, D7's and D72's.
>
> Tom KD7LXL
> _______________________________________________
> Xastir mailing list
> Xastir at lists.xastir.org
> http://lists.xastir.org/cgi-bin/mailman/listinfo/xastir
>



-- 
73

Guy Story KC5GOI
kc5goi at gmail.com