NOTICE: This version of the NSF Unidata web site (archive.unidata.ucar.edu) is no longer being updated.
Current content can be found at unidata.ucar.edu.
To learn about what's going on, see About the Archive Site.
Hello Alexandre, > 1>I'm using an Irregular3DSet to define the domain_set of a FlatField. I > need double precision for my datasets. Why can't we use double with an > Irregular3DSet ? Is there any other way to create a 3D domain_set based > on non gridded double values (w/o using Irregular3DSet) ? I would like > to avoid rescaling and recasting to float the datasets to get the right > information on the scale axis. Originally, the sample locations in Sets were all doubles, but that used so much memory that we changed them to floats. We added the class Gridded1DDoubleSet to support unevenly spaced samplings of RealType.Time, which need doubles to encode dates and times in a single number with adequate resolution. It would be possible to add a class Irregular3DDoubleSet, but it would take some real work. If your data don't really need double precision, converting doubles to floats will be a lot easier than creating a new class Irregular3DDoubleSet. > 2> Is there any way to make an axis display [-1000;1000] for let's say > data that have previously been rescaled to [-1.0;1.0] ? If you've done: ScalarMap xmap = new ScalarMap(x, Display.XAxis); xmap.setRange(-1.0, 1.0); then later you can do: xmap.setRange(-1000.0, 1000.0); and that should change the axis scale. Cheers, Bill ---------------------------------------------------------- Bill Hibbard, SSEC, 1225 W. Dayton St., Madison, WI 53706 hibbard@xxxxxxxxxxxxxxxxx 608-263-4427 fax: 608-263-6738 http://www.ssec.wisc.edu/~billh/vis.html
visad
archives: