Hi Felix, the polygon-size-limit also matters as it is used to determine which islands will not be visible. I used 2 for all levels, but I only looked at the overview map for now as the detail maps are not changed by the --improve-overview option.
I suggest to experiment with a small set of tiles. australia-oceania is probably good. Gerd ________________________________________ Von: mkgmap-dev <[email protected]> im Auftrag von Felix Hartmann <[email protected]> Gesendet: Freitag, 11. Juni 2021 11:25 An: Development list for mkgmap Betreff: Re: [mkgmap-dev] Proof of concept for better sea in overview map yes I was using: --x-simplify-filter-line-errors=23:2.6,22:4.2,21:5.4,20:6,19:7,18:7.5,16:8,14:9 --x-simplify-filter-polygon-errors=23:3.6,22:7,21:6,20:9 --polygon-size-limits="24:12, 23:14, 22:14, 21:20, 20:20, 19:20, 18:20, 17:20, 16:20, 15:20, 14:20, 13:20, 12:20, 11:20, 10:20" levels = 0:24, 1:22, 2:21, 3:20, 4:19, 5:18 overview-levels = 6:17, 7:16, 8:15, 9:14, 10:13 As this only affects the overview map - I guess I should change it to: --x-simplify-filter-polygon-errors=23:3.6,22:7,21:6,20:9,17:2.6 ?? The polygon-size-limits should not matter as I have the skip-size filter set for sea. I had worked quite a long time to optimize those settings with the old behaviour. On Fri, 11 Jun 2021 at 11:50, Gerd Petermann <[email protected]<mailto:[email protected]>> wrote: Hi Felix, Maybe you are using large values for the Douglas-Peucker filter at low resolutions? That's probably a bad idea with the new option. I always tested with the default 2.6 for all levels. The effect of the new option is that DP really can do its work. Gerd ________________________________________ Von: mkgmap-dev <[email protected]<mailto:[email protected]>> im Auftrag von Felix Hartmann <[email protected]<mailto:[email protected]>> Gesendet: Freitag, 11. Juni 2021 10:23 An: Development list for mkgmap Betreff: Re: [mkgmap-dev] Proof of concept for better sea in overview map --generate-sea --precomp-sea=C:\openmtbmap\maps\sea.zip --order-by-decreasing-area --allow-reverse-merge I thought using precomp-sea is fine (I did not update the precomp-sea in between) On Fri, 11 Jun 2021 at 11:10, Felix Hartmann <[email protected]<mailto:[email protected]><mailto:[email protected]<mailto:[email protected]>>> wrote: compilation time includes splitting, and some other stuff. Loads of countries (9:34 / 9:21 (sorry dumb error 9:19) for all european single countries and a few continents (but not Europe continent) in hours. On Fri, 11 Jun 2021 at 11:08, Felix Hartmann <[email protected]<mailto:[email protected]><mailto:[email protected]<mailto:[email protected]>>> wrote: I just compiled the australia-oceania map with that option, and I think I must be missing something. It got substantially worse (it is identical until the overview map kicks in) - then I get some huge squares of sea missing. Compile time for all maps until that point before Compilation time is not up a lot: 0:12 - 9:46 = 9:34 one week before (except older mapdata no changes in compilation procedure and older mkgmap) 20:17 - 5:38 = 9:19 Anything I could be missing? I use the newest low-res branch build and added --improve-overview to the compilation process. in my polygons file I have: natural=sea { add mkgmap:skipSizeFilter=true; set mkgmap:drawLevel=2 } [0x10f1d resolution 10] On Fri, 11 Jun 2021 at 07:12, Gerd Petermann <[email protected]<mailto:[email protected]><mailto:[email protected]<mailto:[email protected]>>> wrote: Hi all, I hoped for some positive feedback on this but got none so far. Did I miss something important? Gerd ________________________________________ Von: mkgmap-dev <[email protected]<mailto:[email protected]><mailto:[email protected]<mailto:[email protected]>>> im Auftrag von Gerd Petermann <[email protected]<mailto:[email protected]><mailto:[email protected]<mailto:[email protected]>>> Gesendet: Montag, 7. Juni 2021 21:05 An: Development list for mkgmap Betreff: Re: [mkgmap-dev] Proof of concept for better sea in overview map Hi Felix, the map contained was without routing or index, so for a normal map the difference should be even smaller. There is no need to change sea.zip. You just have to use --improve-overview for now. Gerd ________________________________________ Von: mkgmap-dev <[email protected]<mailto:[email protected]><mailto:[email protected]<mailto:[email protected]>>> im Auftrag von Felix Hartmann <[email protected]<mailto:[email protected]><mailto:[email protected]<mailto:[email protected]>>> Gesendet: Montag, 7. Juni 2021 21:02 An: Development list for mkgmap Betreff: Re: [mkgmap-dev] Proof of concept for better sea in overview map I guess that is a time for a full Norway map based on default style - so that is quite okay. Not a sea only Norway map... Well I hope that Thorsten Kukuk can adapt his sea files in the near future then... On Mon, 7 Jun 2021 at 19:37, Gerd Petermann <[email protected]<mailto:[email protected]><mailto:[email protected]<mailto:[email protected]>><mailto:[email protected]<mailto:[email protected]><mailto:[email protected]<mailto:[email protected]>>>> wrote: OK, I think found a good solution. Speed is quite OK, I see 9 min 48 secs for map of norway and instead of 8 min 44 secs with r4756, and I consider Norway to be a worst case. See https://www.mkgmap.org.uk/websvn/revision.php?repname=mkgmap&rev=4761 for the details. Gerd ________________________________________ Von: mkgmap-dev <[email protected]<mailto:[email protected]><mailto:[email protected]<mailto:[email protected]>><mailto:[email protected]<mailto:[email protected]><mailto:[email protected]<mailto:[email protected]>>>> im Auftrag von Gerd Petermann <[email protected]<mailto:[email protected]><mailto:[email protected]<mailto:[email protected]>><mailto:[email protected]<mailto:[email protected]><mailto:[email protected]<mailto:[email protected]>>>> Gesendet: Montag, 7. Juni 2021 12:11 An: Development list for mkgmap Betreff: Re: [mkgmap-dev] Proof of concept for better sea in overview map With the provided patch the speed is very poor for areas like Norway, probably twice the time. It is much slower because it does the complex multipolygon cutting for each level in the overview map. Probably too slow for complex coastal areas. This time will be the same for precomp-sea unless we can store sea polygons for each resolution. Performance will be no problem if I find a way to use Douglas-Peucker or similar before cutting. That was my original idea but DP produces self intersecting polygons and the MultipolygonCutter cannot cope with that. Gerd ________________________________________ Von: mkgmap-dev <[email protected]<mailto:[email protected]><mailto:[email protected]<mailto:[email protected]>><mailto:[email protected]<mailto:[email protected]><mailto:[email protected]<mailto:[email protected]>>>> im Auftrag von Felix Hartmann <[email protected]<mailto:[email protected]><mailto:[email protected]<mailto:[email protected]>><mailto:[email protected]<mailto:[email protected]><mailto:[email protected]<mailto:[email protected]>>>> Gesendet: Montag, 7. Juni 2021 11:55 An: Development list for mkgmap Betreff: Re: [mkgmap-dev] Proof of concept for better sea in overview map Is it only much slower because of not using precomp sea? Or will it be much slower in general? And what is much slower for say Europe continent map? If a modern 4core/8thread processor needs 30 minutes more I would prefer the old way knowing it is worse (if the time difference is negligible with precomp-sea then that would be great). On Mon, 7 Jun 2021 at 12:26, Ticker Berkin <[email protected]<mailto:[email protected]><mailto:[email protected]<mailto:[email protected]>><mailto:[email protected]<mailto:[email protected]><mailto:[email protected]<mailto:[email protected]>>><mailto:[email protected]<mailto:[email protected]><mailto:[email protected]<mailto:[email protected]>><mailto:[email protected]<mailto:[email protected]><mailto:[email protected]<mailto:[email protected]>>>>> wrote: Hi Gerd This is going to take some studying to work out the implications. Can't do much for the next few days however, but will look carefully at the end of the week. Ticker On Sun, 2021-06-06 at 14:02 +0000, Gerd Petermann wrote: > Hi, > > the attached patch improves the overview map, but so far only when > precomp-sea is NOT used. > I tested it with --generate-sea=multipolygon,floodblocker so that > mkgmap really has a multipolygon with the natual=sea data. > > For each level in the overview map it uses the original multipolygon > data to compute the rings which will might visible at the given > resolution. This requires more time compared to the current code but > the result is much better and some fine tuning is possible. > > To be able to use this also with --precomp-sea we need some changes > in the code which generates sea.zip so that one multipolygon relation > for each tile is stored. > > I've uploaded the results here: > https://files.mkgmap.org.uk/download/511/compare.7z > What do you think? > > Gerd > > > _______________________________________________ > mkgmap-dev mailing list > [email protected]<mailto:[email protected]><mailto:[email protected]<mailto:[email protected]>><mailto:[email protected]<mailto:[email protected]><mailto:[email protected]<mailto:[email protected]>>><mailto:[email protected]<mailto:[email protected]><mailto:[email protected]<mailto:[email protected]>><mailto:[email protected]<mailto:[email protected]><mailto:[email protected]<mailto:[email protected]>>>> > https://www.mkgmap.org.uk/mailman/listinfo/mkgmap-dev _______________________________________________ mkgmap-dev mailing list [email protected]<mailto:[email protected]><mailto:[email protected]<mailto:[email protected]>><mailto:[email protected]<mailto:[email protected]><mailto:[email protected]<mailto:[email protected]>>><mailto:[email protected]<mailto:[email protected]><mailto:[email protected]<mailto:[email protected]>><mailto:[email protected]<mailto:[email protected]><mailto:[email protected]<mailto:[email protected]>>>> https://www.mkgmap.org.uk/mailman/listinfo/mkgmap-dev -- Felix Hartman - Openmtbmap.org & VeloMap.org _______________________________________________ mkgmap-dev mailing list [email protected]<mailto:[email protected]><mailto:[email protected]<mailto:[email protected]>><mailto:[email protected]<mailto:[email protected]><mailto:[email protected]<mailto:[email protected]>>> https://www.mkgmap.org.uk/mailman/listinfo/mkgmap-dev _______________________________________________ mkgmap-dev mailing list [email protected]<mailto:[email protected]><mailto:[email protected]<mailto:[email protected]>><mailto:[email protected]<mailto:[email protected]><mailto:[email protected]<mailto:[email protected]>>> https://www.mkgmap.org.uk/mailman/listinfo/mkgmap-dev -- Felix Hartman - Openmtbmap.org & VeloMap.org _______________________________________________ mkgmap-dev mailing list [email protected]<mailto:[email protected]><mailto:[email protected]<mailto:[email protected]>> https://www.mkgmap.org.uk/mailman/listinfo/mkgmap-dev _______________________________________________ mkgmap-dev mailing list [email protected]<mailto:[email protected]><mailto:[email protected]<mailto:[email protected]>> https://www.mkgmap.org.uk/mailman/listinfo/mkgmap-dev -- Felix Hartman - Openmtbmap.org & VeloMap.org -- Felix Hartman - Openmtbmap.org & VeloMap.org -- Felix Hartman - Openmtbmap.org & VeloMap.org _______________________________________________ mkgmap-dev mailing list [email protected]<mailto:[email protected]> https://www.mkgmap.org.uk/mailman/listinfo/mkgmap-dev -- Felix Hartman - Openmtbmap.org & VeloMap.org _______________________________________________ mkgmap-dev mailing list [email protected] https://www.mkgmap.org.uk/mailman/listinfo/mkgmap-dev
