<html>
<head>
<style><!--
.hmmessage P
{
margin:0px;
padding:0px
}
body.hmmessage
{
font-size: 12pt;
font-family:Calibri
}
--></style></head>
<body class='hmmessage'><div dir='ltr'>Hi all,<br><br>I've finally found a way to test a lot more possible solutions<br>in splitter in a reasonable time.<br>As a result, splitting a single country like Germany is now done <br>with a full search, means, all posible splits are tested and the<br>best is chosen. <br>For large files (e.g. Europe), the program can't search all<br>possible solutions, but it still finds better ones than r389.<br><br>This takes a bit more time than before, but<br>is likely to find really good solutions. If needed, I can implement <br>new options to allow stopping the search if e.g. a solution is found<br>that has at least 85% of max-nodes in each tile.<br><br>A few numbers: <br>Split Germany with max-nodes=1200000<br>r 389: Solution is nice. Can't find a better solution: 164 tile(s) and a rating of 2756. The smallest node count is 903153, the worst aspect ratio is near 3.99<br>patched: Solution is nice. Can't find a better solution: 150 tile(s). The smallest node count is 1081130 (90 %), the worst aspect ratio is near 3.83<br><br>Split Germany with max-nodes=1000000<br>r389: Solution is nice. Can't find a better solution: 192 tile(s) and a rating of 3014. The smallest node count is 742452, the worst aspect ratio is near 3.9<br>patched: Solution is nice. Can't find a better solution: 181 tile(s). The smallest node count is 894244 (89 %), the worst aspect ratio is near 3.93<br><br>Please try with your workload and let me know the results.<br>If you are not happy with the new result, please send the densities_out.txt and <br>the splitter.log.<br><br>A compiled binary is available here:<br><code><a href="http://files.mkgmap.org.uk/download/208/splitter.jar" target="_blank">http://files.mkgmap.org.uk/download/208/splitter.jar</a><br><br>Gerd<br></code><br> </div></body>
</html>