[GRASS-user] v vol rst, segmentation and overshoot

[Francois Chartier]

Hi

I am running multiple 3d interpolation using v vol rst on a large data set
of 12k points.  i have noticed that segmentation is observable by showing
up as Panels in the 3d raster when looking at the data using paraview.  i
have tried numerous setups and even using a segment as large as the dataset
which took 36 hrs to run the interpolation....
is there a way to avoid this segmentation effect? by keeping segment
smaller and increasing npmin and npmax.  any rule of thumb or advice that
can avoid running the jnterpolation over 30 hrs ( resolution is quite
coarse: 500x500x3m).

my second question is depending on the Tension used 40, 80 etc there tends
to be very large or small number in the resultjng 3d raster.  my data set
ranges from 0.002 to 300 so 6 order of magnitude.  when viewing the raster,
the values go up to 10 to the power of 7 as the max and 7 decimal points as
the min. how can i control this so that the interpolation stays in the
range?

thanks
F.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.osgeo.org/pipermail/grass-user/attachments/20190417/46ab5c9d/attachment.html>