On Wed, 07 Jun 2006 20:26:22 +0200, Gennaro Prota <gennaro_prota@yahoo.com> wrote:

Hi guys,

for the reasons noted in my previous message about boost::timer, I've updated it to report a custom exception in case it cannot (re)start the timer. If there are no objections I'll change the docs to reflect this. The modification is on the trunk only, of course.

http://boost.cvs.sourceforge.net/boost/boost/boost/timer.hpp?r1=1.7&r2=1.8

Hi again, after a careful analysis of the code and of the C standard I have further comments and proposals for this class. However there's one point I'd like to understand, first. The code has a long comment saying: "It is recommended that implementations measure wall clock rather than CPU time since the intended use is performance measurement on systems where total elapsed time is more important than just process or CPU time." I wonder what does that mean. Of course std::clock() is required to measure process time, not wall clock time (though it measures wall clock time, instead, on some non-conforming platforms; consider for instance the VC++ run-time library). --Gennaro.