<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 Steve,<br><br>> > I don't think that the current algo is completely wrong.<br><div>> > I think it is better to start with the complete data and divide it into<br>> > smaller parts as that algo keeps the information of<br>> > the location of elements, while your approach is likely<br>> > to compare many bboxes which are (too) far away from each other.<br>> <br>> Yes, implementation speed might require some more complexity - but it<br>> seems that it should produce a division split where all of them are<br>> no bigger than required without further splitting the elements.<br><br>Maybe, but I don't think that this would be optimal,<br>as we may get too many sub divs and thus a much larger img <br>file. I don't know how the device determines the<br>relevant sub divs, but even a good algo will suffer<br>if we produce 10000 instead of 250 sub divs.<br><br>> <br>> If that is the case, and it still makes no difference in speed, then<br>> maybe we are pretty much as good as we are going to get.<br><br>I thought about writing an algo to test this, but that seems to be<br>even more complicated. I'd try to simulate the rendering <br>for each element in a grid, and compare the number<br>of rendered elements (caused by overlaps) which <br>are not visible in the grid element.<br>This could be done for each level, and maybe one<br>has to take into account that a device has to render<br>a rectangle which heads to north-west or west.<br><br>Gerd<br></div> </div></body>
</html>