Hi Paolo Thanks for your opinions on this topic, however we don't think the polygon topology approach is such a good idea. All the other answers in you text: > As far as I know, -32768 is "no data available" marker and does not indicate > water. We use preprocessed, void-filled SRTM data where NODATA means exactly water as defined by the supplier... See our mapgen source code for more information. >> 1) use the old way and have everything below 0m as water > Does not work. 0m or -10m does not indicate water. Netherlands is an > example, but not the only one. > England and Italy, and probably many other areas and most lakes do not have > 0m altitude. > Not using SRTM data at least. It does not work, exactly... >> 2) use an additional bitmap just for the water areas (Max's favorite) > Not derived from SRTM I suppose. Due to the statement I gave above, we would actually derive that from the SRTM data. >> 3) lift the entire source image by 1000m, then define -31768m = 0m to >> solve the interpolation problems and subtract 1000m from all elevations >> read in XCSoar. > See (1) Sorry, but I don't see how this is related to (1) ... It solves exactly the problem that (1) has... >> 4) use more then just one JPEG2000 image file. Those including coastline >> with lossless compression. > See (1) This is also not related to (1), since the basic problem is the lossy compression and not the the areas below 0m. >> 5) use lossless compression for all maps > See (1). Actually... See (4) > There are only two possible solutions for this: > 1) use a separated raster terrain data for water That is exactly what I tried to describe in the second solution > 2) use topology vectorial data for water This means drawing and clipping a very large coastline polygon which is potentially very calculation-intensive. It is a solution, as you have demonstrated but as you know we are perfectionists and we don't think a huge polygon is a good idea. > I have solved water coverage problem long ago using the second approach. > Netherlands, england, italy and many cost areas previously under water are > now correctly painted. > All other solutions will not work. See the source code of LK to understand > why and how the topology solution is working. If I was able to find my way through the code I might give this a try, but I don't see why you are so sure that the other solutions won't work. I am actually pretty sure that some of them would...
Turbo ------------------------------------------------------------------------------ Learn how Oracle Real Application Clusters (RAC) One Node allows customers to consolidate database storage, standardize their database environment, and, should the need arise, upgrade to a full multi-node Oracle RAC database without downtime or disruption http://p.sf.net/sfu/oracle-sfdevnl _______________________________________________ Xcsoar-user mailing list Xcsoar-user@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/xcsoar-user
