Mark:
> I have a bunch of points (4+ million) with one outlier that needs
> removed.
perhaps use r.mapcalc or r.reclass to filter out the bad point?
r.mapcalc "clean = if(map > 999999, null(), map)"
or
r.reclass
999999 thru 9999999 = NULL
* = *
and finally "r.null setnull=bad_value" may be the simplest.
... or d.rast.edit ...
Hamish