<!DOCTYPE html PUBLIC "-//W3C//DTD HTML 4.01 Transitional//EN">
<html>
<head>
<meta content="text/html;charset=ISO-8859-1" http-equiv="Content-Type">
<title></title>
</head>
<body bgcolor="#ffffff" text="#000000">
Thanks Mike !<br>
<div class="moz-signature"><br>
With kind regards,
<br>
Cav Edwards<br>
<br>
</div>
<br>
<br>
Mike Perry wrote:
<blockquote cite="mid:20100726200406.GB16798@fscked.org" type="cite">
<pre wrap="">Thus spake Cav (<a class="moz-txt-link-abbreviated" href="mailto:cav@gotadsl.co.uk">cav@gotadsl.co.uk</a>):
</pre>
<blockquote type="cite">
<pre wrap="">Hi Mike,
Thanks for a quick reply. Maybe its my warped mind that had me consider
that the management of timeouts (sliding the timeout up and down - as it
appears in the logs) would affect the distribution of timeouts and
therefore the pareto distribution around them.
I will admit to experimenting with smaller values of
CBT_DEFAULT_QUANTILE_CUTOFF. These lead to 'peaks' in the timeout
distribution. The smaller the value, the tighter these 'peaks' seem to be.
</pre>
</blockquote>
<pre wrap=""><!---->
This should not be the case in 0.2.2.14 and above. Changing the
QUANTILE_CUTOFF default should now only govern which circuits you
actually use. Circuits should continue to build until
CBT_DEFAULT_CLOSE_QUANTILE, so the distribution of recorded buildtimes
will be much more insenitive to changes in the timeout value.
</pre>
</blockquote>
</body>
</html>