Location, location

[Tony Langdon]

> DTED1 is a 3 sec arc which equates to about 90m on the ground. It is
> topographic only (it used to be/still is derived from people reading
> topographic maps and inputting the data). Buildings would 
> certainly be a

Hmm, where I live, this isn't accurate enough.  I ran off some path analyses
using software which has topographic data accurate to 100m.  It got some
paths totally wrong, saying it was LOS, when it wasn't (one only has to
stand outside and peer at the lump of dirt in the middle of the so called
"LOS path" ;) ).  I am talking about purely natural topographic features
here, not buildings or cuttings.

My problem is I'm on the side of a steep valley, which is a couple hundred
metres across at most, and the software didn't see the full depth of the
valley, making gradients in the east/west direction look a lot less than
they really are.  I think 20m data points are necessary around my area to
get a half decent picture of what's going on...

Let's say that around 60-70m to the east, you're on the plateau that is much
of the northern suburbs (Essendon a/p is a part of this one).  Less than
100m to the west, and you're in a BIG hole - only way out is aircraft
reflections (I've done that on VHF!) from planes around Tulla.  Go just a
little further west again, maybe 100m or less and you're back at my level,
or higher.

But these maps will still be good for most people.  Just posting my bit of
experience, to say sometimes, one needs better data than what is claimed to
be "good enough".