[aprssig] Is floating point telemetry allowed in APRS?

Robert Bruninga bruninga at usna.edu
Thu Mar 28 11:58:25 EDT 2013

Actually, the original APRSdos can take decimals, but again, the limit is
3 FIXED bytes.  So you can have 0.1 up to 9.9, but beyond 10, you cannot
use a decimal because it would make the field 4 bytes wide and violate the
spec.  Same goes for SPEED field.

Im not sure if the spec is that clear or not.  Don't have time to dig.
But the 3 bytes *is* the spec requirement, any positive number you can
stuff into 3 bytes should be OK.   I even allow up to 999 which is getting
close to 10 bit accuracy..


-----Original Message-----
From: aprssig-bounces at tapr.org [mailto:aprssig-bounces at tapr.org] On Behalf
Of Matti Aarnio
Sent: Thursday, March 28, 2013 11:10 AM
To: TAPR APRS Mailing List
Subject: Re: [aprssig] Is floating point telemetry allowed in APRS?

On Wed, Mar 27, 2013 at 04:09:12PM -0400, Steve Dimse wrote:
> The spec is not fluid that way. If it says telemetry should look like

The spec isn't fluid in many ways.  If you start to take a strict view
of what is carried in APRS-IS is only what aprs-1.01.pdf says, then most
of BoB's "errata" and "update" documents become irrelevant too -- and
frankly the whole APRS becomes an obsolete dinosaur starting with
prohibiting use of any characters outside ASCII.
(Tell that to Japanese, Greek, European in general...)

Telemetry is just one small example of things codifying what one
ancient microprocessor was capable of doing.  Strictly speaking it
reporting 10 or 12 or 16 bit ADC values, which has never stopped anybody
doing it anyway.

If receiver's display routines are converting integers to rational numbers
for displaying, or plotting, what would prevent from from parsing a
number to begin with?

Very long ago I was doing something where we considered for a moment about
using APRS for telemetry.  We needed something like 20 ADC data channels,
50 booleans...

73 de Matti, OH2MQK
aprssig mailing list
aprssig at tapr.org

More information about the aprssig mailing list