<html>
  <head>
    <meta content="text/html; charset=ISO-8859-1"
      http-equiv="Content-Type">
  </head>
  <body text="#000000" bgcolor="#FFFFFF">
    well - anything is better than the current crash. Maybe just a rerun
    with "safe" options added?<br>
    <br>
    Please also look on Antarctica extract by geofabrik. I didn't find
    time yet to test it again why it crashed... <br>
    <br>
    Or is search-limit that option what I am looking for? I so&nbsp; far
    didn't know about this new switch at all...<br>
    <div class="moz-cite-prefix">On 31.05.2014 08:33, Gerd Petermann
      wrote:<br>
    </div>
    <blockquote cite="mid:DUB112-W1463085225304807D60A5E9E260@phx.gbl"
      type="cite">
      <style><!--
.hmmessage P
{
margin:0px;
padding:0px
}
body.hmmessage
{
font-size: 12pt;
font-family:Calibri
}
--></style>
      <div dir="ltr">Hi Felix,<br>
        <br>
        it seems that the default --search-limit is too low for japan
        with sea data.<br>
        With --search-limit=5000000 I see a very good split after ~30
        seconds:<br>
        Solution is nice. Can't find a better solution: 86 tile(s). The
        smallest node count is 1111027 (92 %), the worst aspect ratio is
        near 3.94<br>
        <br>
        For many other input files, this will just increase run time a
        lot.<br>
        <br>
        I am not very happy with this option. I think I will replace it<br>
        with something more useful like<br>
        --wanted-fill-ratio=90<br>
        which means try to find a solution where the least populated<br>
        tile has 90% of max-nodes.<br>
        <br>
        The problem is that some areas do not allow such a good result,<br>
        so I am still looking for good criteria to stop the search,
        maybe<br>
        I'll use the overall search time in seconds:<br>
        --max-search-time=60<br>
        would use the best result found after 60 seconds.<br>
        <br>
        Any other ideas?<br>
        <br>
        Gerd<br>
        <br>
        <br>
        <div>
          <hr id="stopSpelling">From: <a class="moz-txt-link-abbreviated" href="mailto:gpetermann_muenchen@hotmail.com">gpetermann_muenchen@hotmail.com</a><br>
          To: <a class="moz-txt-link-abbreviated" href="mailto:mkgmap-dev@lists.mkgmap.org.uk">mkgmap-dev@lists.mkgmap.org.uk</a><br>
          Date: Sat, 31 May 2014 08:21:43 +0200<br>
          Subject: [mkgmap-dev] FW: Splitter defaulting on Japan<br>
          <br>
          <style><!--
.ExternalClass .ecxhmmessage P {
padding:0px;
}

.ExternalClass body.ecxhmmessage {
font-size:12pt;
font-family:Calibri;
}

--></style>
          <div dir="ltr"><br>
            <br>
            <div>
              <hr id="ecxstopSpelling">From:
              <a class="moz-txt-link-abbreviated" href="mailto:gpetermann_muenchen@hotmail.com">gpetermann_muenchen@hotmail.com</a><br>
              To: <a class="moz-txt-link-abbreviated" href="mailto:extremecarver@gmail.com">extremecarver@gmail.com</a><br>
              Subject: RE: [mkgmap-dev] Splitter defaulting on Japan<br>
              Date: Sat, 31 May 2014 08:17:31 +0200<br>
              <br>
              <style><!--
.ExternalClass .ecxhmmessage P {
padding:0px;
}

.ExternalClass body.ecxhmmessage {
font-size:12pt;
font-family:Calibri;
}


--></style>
              <div dir="ltr">Hi Felix,<br>
                <br>
                thanks for reporting. I can reproduce the problem. It
                disappears when<br>
                you remove the --precomp-sea option.<br>
                I'll try to find out why.<br>
                <br>
                &gt; Where is the log saved? Do I need to provide it?
                Splitter not fully up <br>
                This message is printed to stderr for the case that you
                pipe stdout to a file like this:<br>
                java -jar splitter ....&nbsp;&nbsp; &gt; splitter.log<br>
                <br>
                Gerd<br>
                <br>
                <div>&gt; Date: Sat, 31 May 2014 05:33:06 +0200<br>
                  &gt; From: <a class="moz-txt-link-abbreviated" href="mailto:extremecarver@gmail.com">extremecarver@gmail.com</a><br>
                  &gt; To: <a class="moz-txt-link-abbreviated" href="mailto:mkgmap-dev@lists.mkgmap.org.uk">mkgmap-dev@lists.mkgmap.org.uk</a><br>
                  &gt; Subject: [mkgmap-dev] Splitter defaulting on
                  Japan<br>
                  &gt; <br>
                  &gt; Failed to calculate areas. See log for details.<br>
                  &gt; <br>
                  &gt; Where is the log saved? Do I need to provide it?
                  Splitter not fully up <br>
                  &gt; to date - see compiled date (svn update just
                  before)...<br>
                  &gt; I'll retry with up to date splitter still...<br>
                  &gt; <br>
                  &gt; c:\OpenMTBMap\maps&gt;java -Xmx9600m -jar
                  c:\openmtbmap\splitter.jar <br>
                  &gt; --precomp-sea=c:\openmtbmap\maps\sea.zip
                  --max-nodes=1200000 <br>
                  &gt; --output=pbf "--keep-complete" --max-areas=255 <br>
                  &gt; --geonames-file=cities5000 --description=japan
                  --mapid=65590000 c:\OpenMTBMa<br>
                  &gt; p\osmpbf_geofabrik\japan.o5m<br>
                  &gt; Splitter version unknown compiled
                  2014-05-28T18:16:29+0200<br>
                  &gt; boundary-tags=use-exclude-list<br>
                  &gt; cache=<br>
                  &gt; description=japan<br>
                  &gt; geonames-file=cities5000<br>
                  &gt; keep-complete=true<br>
                  &gt; mapid=65590000<br>
                  &gt; max-areas=255<br>
                  &gt; max-nodes=1200000<br>
                  &gt; max-threads=8 (auto)<br>
                  &gt; mixed=false<br>
                  &gt; no-trim=false<br>
                  &gt; num-tiles=<br>
                  &gt; output=pbf<br>
                  &gt; output-dir=<br>
                  &gt; overlap=auto<br>
                  &gt; polygon-desc-file=<br>
                  &gt; polygon-file=<br>
                  &gt; precomp-sea=c:\openmtbmap\maps\sea.zip<br>
                  &gt; problem-file=<br>
                  &gt; problem-report=<br>
                  &gt; resolution=13<br>
                  &gt; search-limit=1000000<br>
                  &gt; split-file=<br>
                  &gt; status-freq=120<br>
                  &gt; stop-after=dist<br>
                  &gt; write-kml=<br>
                  &gt; Elapsed time: 0s Memory: Current 184MB (2MB used,
                  182MB free) Max 8533MB<br>
                  &gt; Time started: Sat May 31 02:40:37 CEST 2014<br>
                  &gt; Map is being split for resolution 13:<br>
                  &gt; - area boundaries are aligned to 0x800 map units
                  (0.0439453125 degrees)<br>
                  &gt; - areas are multiples of 0x800 map units wide and
                  high<br>
                  &gt; Processing
                  c:\OpenMTBMap\osmpbf_geofabrik\japan.o5m<br>
                  &gt; Bounding box 122.56070000000001 20.08228 154.4709
                  45.80245<br>
                  &gt; 10'000'000 nodes processed... id=752184948<br>
                  &gt; 20'000'000 nodes processed... id=922482748<br>
                  &gt; 30'000'000 nodes processed... id=1233696234<br>
                  &gt; 40'000'000 nodes processed... id=1315824764<br>
                  &gt; 50'000'000 nodes processed... id=1421002430<br>
                  &gt; 60'000'000 nodes processed... id=1497506798<br>
                  &gt; 70'000'000 nodes processed... id=1751094845<br>
                  &gt; 80'000'000 nodes processed... id=2040104150<br>
                  &gt; 90'000'000 nodes processed... id=2350328457<br>
                  &gt; in 1 file<br>
                  &gt; Time: Sat May 31 02:41:01 CEST 2014<br>
                  &gt; Counting nodes of precompiled sea data ...<br>
                  &gt; Bounding box 144.84375 19.6875 145.546875
                  20.390625<br>
                  &gt; Bounding box 135.703124999 20.390625
                  136.406249999 21.09375<br>
                  &gt; Bounding box 144.84375 20.390625 145.546875
                  21.09375<br>
                  &gt; Bounding box 122.34375000000001 23.90625
                  123.04687499900001 24.609375<br>
                  &gt; <br>
                  &gt; <br>
                  &gt; ........<br>
                  &gt; cut away here...<br>
                  &gt; .......<br>
                  &gt; Bounding box 142.734375 45.703125 143.4375
                  46.40625<br>
                  &gt; Bounding box 143.4375 45.703125 144.140625
                  46.40625<br>
                  &gt; Bounding box 149.0625 45.703125 149.765625
                  46.40625<br>
                  &gt; Bounding box 149.765625 45.703125 150.46875
                  46.40625<br>
                  &gt; Bounding box 150.46875 45.703125 151.171875
                  46.40625<br>
                  &gt; Added 835594 nodes from precompiled sea data.<br>
                  &gt; Precompiled sea data pass took 1797 ms<br>
                  &gt; Exact map coverage is
                  (20.08227825164795,122.56068706512451) to <br>
                  &gt; (45.80243110656738,154.47088479995728)<br>
                  &gt; Rounded map coverage is (20.0390625,122.51953125)
                  to <br>
                  &gt; (45.8349609375,154.51171875)<br>
                  &gt; Splitting nodes into areas containing a maximum
                  of 1'200'000 nodes each...<br>
                  &gt; Highest node count in a single grid element is
                  205'139<br>
                  &gt; Trying to find nice split for
                  (20.0390625,122.51953125) to <br>
                  &gt; (45.8349609375,154.51171875) with 98'482'055
                  nodes<br>
                  &gt; searching for split with min-nodes 12000, learned
                  0 good partial solutions<br>
                  &gt; Split was not yet succesfull. Trying to remove
                  large empty areas...<br>
                  &gt; Trying again with 1 trimmed partition(s), also
                  allowing big empty parts.<br>
                  &gt; Solving partition (23.818359375,122.6953125) to <br>
                  &gt; (45.8349609375,153.0615234375) with 98'482'055
                  nodes<br>
                  &gt; Trying to find nice split for
                  (23.818359375,122.6953125) to <br>
                  &gt; (45.8349609375,153.0615234375) with 98'482'055
                  nodes<br>
                  &gt; searching for split with min-nodes 12000, learned
                  0 good partial solutions<br>
                  &gt; Warning: No solution found for partition
                  (23.818359375,122.6953125) to <br>
                  &gt; (45.8349609375,153.0615234375) with 98'482'055
                  nodes<br>
                  &gt; Final solution has 0 tile(s). The smallest node
                  count is <br>
                  &gt; 9223372036854775807 (0 %), the worst aspect ratio
                  is near -1.0<br>
                  &gt; Failed to calculate areas. See log for details.<br>
                  &gt; Failed to calculate areas.<br>
                  &gt; Sorry. Cannot split the file without creating
                  huge, almost empty, tiles.<br>
                  &gt; Please specify a bounding polygon with the
                  --polygon-file parameter.<br>
                  &gt; Time finished: Sat May 31 02:41:04 CEST 2014<br>
                  &gt; _______________________________________________<br>
                  &gt; mkgmap-dev mailing list<br>
                  &gt; <a class="moz-txt-link-abbreviated" href="mailto:mkgmap-dev@lists.mkgmap.org.uk">mkgmap-dev@lists.mkgmap.org.uk</a><br>
                  &gt;
                  <a class="moz-txt-link-freetext" href="http://www.mkgmap.org.uk/mailman/listinfo/mkgmap-dev">http://www.mkgmap.org.uk/mailman/listinfo/mkgmap-dev</a><br>
                </div>
              </div>
            </div>
          </div>
          <br>
          _______________________________________________
          mkgmap-dev mailing list
          <a class="moz-txt-link-abbreviated" href="mailto:mkgmap-dev@lists.mkgmap.org.uk">mkgmap-dev@lists.mkgmap.org.uk</a>
          <a class="moz-txt-link-freetext" href="http://www.mkgmap.org.uk/mailman/listinfo/mkgmap-dev">http://www.mkgmap.org.uk/mailman/listinfo/mkgmap-dev</a></div>
      </div>
      <br>
      <fieldset class="mimeAttachmentHeader"></fieldset>
      <br>
      <pre wrap="">_______________________________________________
mkgmap-dev mailing list
<a class="moz-txt-link-abbreviated" href="mailto:mkgmap-dev@lists.mkgmap.org.uk">mkgmap-dev@lists.mkgmap.org.uk</a>
<a class="moz-txt-link-freetext" href="http://www.mkgmap.org.uk/mailman/listinfo/mkgmap-dev">http://www.mkgmap.org.uk/mailman/listinfo/mkgmap-dev</a></pre>
    </blockquote>
    <br>
  </body>
</html>