Re: [PHP] Re: Question on code profiling

2009-07-24 Thread Andrew Ballard
On Fri, Jul 24, 2009 at 2:27 AM, Lupus Michaelis wrote: > Andrew Ballard a écrit : >> >> I'm trying to profile a site on our development server to see why it >> takes around 4 seconds to generate a pretty basic page. > >  Last time I seen this is when I did validate DOM Document without DTD on > lo

[PHP] Re: Question on code profiling

2009-07-23 Thread Lupus Michaelis
Andrew Ballard a écrit : I'm trying to profile a site on our development server to see why it takes around 4 seconds to generate a pretty basic page. Last time I seen this is when I did validate DOM Document without DTD on local disk :D Can you put somewhere the essential code that take ti

Re: [PHP] Re: Question on code profiling

2009-07-23 Thread Jonathan Tapicer
Just an idea: try using the (microtime(true) - $start) approach in portions of code to try isolate the portion that is taking more time. Sometimes that helps me to find "the" function that is slowing everything down. Jonathan On Thu, Jul 23, 2009 at 6:18 PM, Andrew Ballard wrote: > On Thu, Jul 23

Re: [PHP] Re: Question on code profiling

2009-07-23 Thread Andrew Ballard
On Thu, Jul 23, 2009 at 5:10 PM, Ben Dunlap wrote: >> significant (around 46%), it says they only account for 193ms. What >> could account for that much difference between what xdebug calculates >> versus the total elapsed time? > > Are you counting "total elapsed time" from the perspective of the

[PHP] Re: Question on code profiling

2009-07-23 Thread Ben Dunlap
> significant (around 46%), it says they only account for 193ms. What > could account for that much difference between what xdebug calculates > versus the total elapsed time? Are you counting "total elapsed time" from the perspective of the web browser? If so, YSlow might be helpful: http://devel