Re: [R-sig-Geo] R vs GSLIB

2007-02-16 Thread Thomas Adams
Thierry I have used used both GSTAT and GSLIB a lot and from a usability point of view, I much prefer GSTAT because of it's ability to interface with with GRASS GIS. This is particularly true of the standalone version of GSTAT. I began using GSLIB, and while I continue to make use of it, I fin

Re: [R-sig-Geo] R vs GSLIB

2007-02-16 Thread Paulo Justiniano Ribeiro Jr
Thierry Indeed many algorithms between GSLIB and some geostats packages in R (have a look at the Spatial CRAN task view) are similar, although they can produce sligtly different results in some cases. This is specially valid for the "core" goestats algorithms (SK, OK, KTE, UK) However, there are

Re: [R-sig-Geo] R vs GSLIB

2007-02-16 Thread Edzer J. Pebesma
Hi Thierry, I think spatstat covers point pattern analysis, which is mostly not covered by gslib. I have seen either no or small differences between gstat and gslib, but never differences that I could not explain. They may for instance relate to how you integrate covariances over a block; ther

[R-sig-Geo] R vs GSLIB

2007-02-16 Thread ONKELINX, Thierry
Dear useRs, Currently I'm taking a course on spatial statistics. We use GSLIB in the exercises. Maybe I could convince the professor to let the R aficionados do their homework with R. So I was wondering if someone had already compared the algorithms used in R packages like spatstat and gstat ve