[GRASS-dev] [GRASS GIS] #2820: v.surf.idw results seem seriously wrong and don't match r.surf.idw results

[GRASS GIS]

#2820: v.surf.idw results seem seriously wrong and don't match r.surf.idw results
----------------------+-------------------------------------------------
  Reporter:  lrntct   |      Owner:  grass-dev@…
      Type:  defect   |     Status:  new
  Priority:  blocker  |  Milestone:  7.0.3
 Component:  Vector   |    Version:  7.0.1
Resolution:           |   Keywords:  v.surf.idw r.surf.idw interpolation
       CPU:  x86-64   |   Platform:  Linux
----------------------+-------------------------------------------------

Comment (by mlennert):

 Replying to [comment:14 annakrat]:
 > I hopefully fixed the bug in the indexing part and the missing sqrt in
 r67211. Please test. Haven't looked at r.surf.idw yet.

 Beautiful, thank you very much !

 v.surf.idw results are now identical with and without the -n flag.

 For the NC precipitation example, setting the flag (i.e. not using the
 indexing method) actually make the module run faster. This merits some
 more testing. From what I can gather from the PhD thesis and from the
 code, it seems to me (but I'm really not sure) that the indexing is mostly
 useful for situations where you have several input points per raster cell.
 But in this case, why use interpolation and not aggregate statistics per
 cell (à la r.in.xyz) ?

 As for r.surf.idw, the NC precipitation data test gives almost identical
 results (r.covar -r between v.surf.idw and r.surf.idw results gives
 0.999883). There are some very localized differences that might merit a
 look, but they might just be slight differences in distances calculations
 and rounding.

 However, using the OP's testing data, I still see the issue with the
 r.surf.idw results.

 In my eyes we should backport the fix to v.surf.idw and lower the priority
 of this bug.

--
Ticket URL: <https://trac.osgeo.org/grass/ticket/2820#comment:16>
GRASS GIS <https://grass.osgeo.org>