On Nov 2, 2010, at 8:41 AM, Brady Callahan wrote: > David's concerns are definitely valid if there is any kind of difference in > the components of the Vertical CS other than units. In my particular case > this is not so, all is same-same. > > I don't know enough about how significant digits are processed in liblas, but > the correct conversion from meters to ift is exactly 0.3048 meters per ift. > The difference in error between 3.2808399 and .3048 would be invisible at > typical Z values scales.
You can control the significant digits by giving las2las a --scale parameter. > > For those who were interested here was the syntax I used for the conversion: > C:\OSGEO4w\las2las -i C:\temp\in\45123A5.las -o C:\temp\out\45123A5.las > --a_srs EPSG:26910 --t_srs EPSG:2992 After I implement --point-translate for 1.6.0b2, you could do >> C:\OSGEO4w\bin\las2las -i C:\temp\in\45123A5.las -o C:\temp\out\45123A5.las >> --a_srs EPSG:26910 --t_srs EPSG:2992 --point-translate "z*3.2808399" and the operation can happen in a single pass. --point-translate operations are going to happen *after* any reprojection that might occur. > > The las file was ~416mb and had 15228814 points. I used a WinXP SP2 machine > after installing OSGeo4W with GDAL and liblas. However, I can't seem to get > the same setup to work on a Win7 64-bit OS. OSGeo4W is definitely not win64 ready in many cases, and there are likely to be many issues. It's possible that 1.6.0b2 will solve some of them, but I do not test on win64. Howard_______________________________________________ Liblas-devel mailing list [email protected] http://lists.osgeo.org/mailman/listinfo/liblas-devel
