Some additional hints: I get this problem in 100% of the cases I tested, on ORGON 550. Did not matter what map was built. Bernd tried to reproduce the bug with a "bayern" map but did not get the bug neither on his OREGON 650 nor in BASECAMP. For faster testing I created a test-area called "xx", and with this area Bernd was able to reproduce the bug. He thinks it might be related to a tile boundary problem. Polygon of the test area: xx.poly <http://gis.19327.n5.nabble.com/file/n5800742/xx.poly> Tile-boundaries produced by splitter: xx.kml <http://gis.19327.n5.nabble.com/file/n5800742/xx.kml> Test areas.list: xx_areas.list <http://gis.19327.n5.nabble.com/file/n5800742/xx_areas.list> --- Somehow the notes for my screenshots got lost. 1st shows the starting point - does not really matter - just somewhere on the cycleway 2nd shows the target - does not really matter - just somewhere in Irsee or beyond 3rd shows Activity selection - IMPORTANT chosse Tourbike or Mountainbike 4th shows the calculation method - IMPORTANT choose "short distance" 5th shows the wrong route with the unmodified style 6th shows the correct route with the hacked style It is also important to go uphill - routing in the opposite direction works. -- View this message in context: http://gis.19327.n5.nabble.com/Finalize-section-not-working-as-expected-tp58... Sent from the Mkgmap Development mailing list archive at Nabble.com.