I've done some quick research into this, and it turns out that while it's possible to obtain a timestamp in Windows to 100-nanosecond accuracy, the value returned is only updated once each timer tick[1], or about once every 16 milliseconds.
It'd seem that that update mechanism is the "magic barrier" you see. Cambridge University offer a solution to this, written in C++ - but whether that'd help you any I don't know :)
-- Foxcub
#include www.liquidfusion.org.uk
1. http://www.cl.cam.ac.uk/users/kw217/useful/win32time.html
In reply to Re^3: Microseconds in Per
by Tanalis
in thread Microseconds in Per
by longliveAJ
| For: | Use: | ||
| & | & | ||
| < | < | ||
| > | > | ||
| [ | [ | ||
| ] | ] |