early rgdal example
Yes, I think the issue was size and Roger was right to not include a 54MB file in the package! I had it there originally simply because I thought it was cool that you could easily display a 54MB file in R. (GDAL is good at subsampling images.) BTW, I can report a recent rgdal success. We've been working with global MODIS 1/2 kilometer datasets which are huge. My student was using ArcGrid, but it sputters and dies just thinking about files that large (its amazing to me that someone can charge those prices for code that can't catch out-of-memory exceptions!) Five lines of code in R was enough to row-by-row multiply a few gigs of data. Took a couple of minutes. Less time than restarting our lone XP box after ArcGIS took it down. T.
On Mon, 2004-08-23 at 01:27, Roger Bivand wrote:
On Mon, 23 Aug 2004, Hisaji ONO wrote:
Hi. Timothy H. Keitt's web page for early rgdal had tutorial using DEM of Madagascal and its surroundings. I remembered this example was better than current rlogo examples because gdal is originally used for geo-spatial data. However I couldn't find this "original " web page. Can I see this page again?
Unless someone has a copy, no. It isn't on sourceforge, as far as I can see. The underlying files were E020S10.*, the DEM being 54MB, so a bit big to distribute with the package. "Small file" and "remoe sensing" seem to be contradictory. Shall I add a new Madagascar page to the sourceforge page - since the page needs editing to add a link to Prof. Ripley's VC++ Windows binary of rgdal (http://www.stats.ox.ac.uk/pub/RWin/rgdal_0.2-7.zip) anyway? By the way, has anybody tried to build Rmap against the VC++ gdal12.dll in Prof. Ripley's build of rgdal? Contact me for details of his approach. Roger
Regards.
_______________________________________________ R-sig-Geo mailing list R-sig-Geo at stat.math.ethz.ch https://stat.ethz.ch/mailman/listinfo/r-sig-geo
Timothy H. Keitt Section of Integrative Biology University of Texas at Austin http://www.keittlab.org/