<!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 4.01 Transitional//EN" "http://www.w3.org/TR/html4/loose.dtd">
<html><head>
<meta http-equiv="Content-Type" content="text/html; charset=utf-8">
<meta name="generator" content="Osso Notes">
<title></title></head>
<body>
<p>----- Original message -----
<br>> <a href="mailto:kate.alhola@nokia.com">kate.alhola@nokia.com</a> wrote:
<br>> > I don't know hat mappero (ex maemo mapper) does internaly but least,
<br>> it
<br>> > can eat all cpu power when doing "processing maps" so that only way
<br>> out
<br>> > is remove battery from device.
<br>>
<br>> I've been told that this happens when downloading large amounts of map
<br>> tiles, so
<br>> it has nothing to do with the map widget itself. That one is based on
<br>> clutter,
<br>> and seems to be quite efficient.
<br>
<br>Situation is when lot of map tiles are loaded. I dont't know what internaly happens. Some memory leak ? For application performance and cpu usage also time needed for tile pre-processing need to be counted.
<br>
<br>Does it some scaling for tiles after loading them ?
<br>
<br>Finally time used for scaling tiles is one key issues. Do we scale them or download every size. How much cpu power is used for scaling or do we leave it all for GPU ?
<br>
<br>For mappero case i don't know how much cpu is used for actual scaling and how much by exhausting virtual memory ?
<br>
<br>
<br>> (off topic: if you can find some way to reproduce it reliably, please
<br>> file a bug
<br>> -- although if you really need to take out the battery, that cannot be a
<br>> bug in
<br>> the application itself)
<br>
<br>I think, without any actual debug or analysis that it consumes all virtual memory and there won't be any memory for system UI even  kill application when everything is swapped out.
<br>
<br>
<br>Kate
<br>>
<br>> Ciao,
<br>>  Alberto
<br>>
<br>> --
<br>> <a href="http://blog.mardy.it">http://blog.mardy.it</a> <-- geek in un lingua international!
<br>>
<br><br></p>
</body>
</html>