Due to the current gap in continued funding from the U.S. National Science Foundation (NSF), the NSF Unidata Program Center has temporarily paused most operations. See NSF Unidata Pause in Most Operations for details.
Hi Tom, The undersampling happens because VisAD uses texture mapping by default for effciency. You can disable it by: GraphicsModeControl mode = display.getGraphicsModeControl(); mode.setTextureEnable(false); You can also limit the degree of undersampling with texture maps by: mode.setCurvedSize(int curved_size); Good luck, Bill p.s., visad-list-digest is not working, but our system admins are working on fixing it. ---------------------------------------------------------- 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 On Mon, 20 May 2002, Tom Nelson wrote: > Greetings, > > I have just started using VisAD. I want to visualize > a 'terrain' of data samples, say 100x100 in longitude > and latitude, with various height values. > > I modified P4_01.java from the tutorial to do this. > > I also added option buttons to allow me to set PolygonAttributes > to POLYGON_FILL, POLYGON_LINE, and POLYGON_POINT, and to > also setPointMode(true) or (false) on the GraphicsModeControl. > > When I setPointMode(true), I see all of my data samples. > When I attempt to show a 'terrain' (POLYGON_FILL) or a grid > (POLYGON_LINE) there is an undersampling of my data points > leading to an incorrect display. > > I see the same effect with the original P4_01.java code > when I add my components to change the PolygonAttributes > and/or the PointMode. The 50x50 samples are reduced to 25x25 > in the POLYGON_LINE and POLYGON_POINT mode. > > I've been digging into the source code to track down just where > my samples are converted to Java3D objects, and I've also > looked in this list archive for previous question about this. > > Ideally, I'd like to have control over whether my data is undersampled > or not. I just haven't stumbled on the obvious way to fix that! > > Tom Nelson > > > >
visad
archives: