Hi WanMil, I've tried again. Reg. the performance improvements I see only small differences, I guess that's because I use splitter with default overlap of 2000. I assume the greater the overlap the greater is the improvement of the UnusedElementsRemoverHook ? reg. different result: I've uploaded the tile (sorry, it is very big): http://files.mkgmap.org.uk/download/65/63240022.osm.pbf I can reproduce the problem with trunk (r2263) (and my identical_output.patch) using the following parms: java -Xmx1600m -Xms1600m -jar mkgmap.jar --remove-short-arcs --route --preserve-element-order 63240022.osm.pbf If I comment the call of UnusedElementsRemoverHook I get a different output file. I hope you can reproduce it? Gerd WanMil wrote
Please, can you review if the UnusedElementsRemoverHook is still useful? With my test data, it is slowing down mkgmap a little bit and I also see a different result for one tile in the UK when I disable it.
Gerd
Gerd,
I cannot reproduce that the UnusedElementsRemoverHook does not improve the speed of mkgmap. Can you please try again?
If you see a different result it should be analysed. So please post your tile and your mkgmap parameters and all the details so that I can check that.
WanMil _______________________________________________ mkgmap-dev mailing list mkgmap-dev@.org http://www.mkgmap.org.uk/mailman/listinfo/mkgmap-dev
-- View this message in context: http://gis.19327.n5.nabble.com/Small-holes-in-boundary-coverage-tp5569161p56... Sent from the Mkgmap Development mailing list archive at Nabble.com.