From: Alexandre Naaman (naaman++at++laplace.engr.sgi.com)
Date: 08/12/2003 10:41:07
On Mon, 4 Aug 2003, Guillaume Millet wrote:
Bonjour Guillaume,
> Hello,
>
> I have a Dual Xeon 3.06 GHZ HP Worstation on which pfGetTime() return
> negative value, so all function depending on time have a strange
> behaviour (stats, motion, ...).
>
> I'm using Windows XP SP1 and OpenGL Performer 3.0.2
>
> By changing the PFCLOCKPERIOD environment variable to a value of less
> than 2300000000000000 pfGetTime return positive (but false) value.
> If think this value is corresponding to CPU frequency so my 3.06 GHz
> Xeon are over this value by default.
> I think this is a bug like the one found on Performer 2.5 with 2.4 GHz P4.
>
> Do you have a workaround for this and when to you expect to correct this ?
>
We use QueryPerformanceRate() on win32 to determine the clock period
but unfortunately, we were assigning the value of a 64 bit int to a 32 bit
int and using that to do some other internal computations. So, I've made
the change to a 64 bit int; hopefully that's all that needs to be done :)
Interested in testing things to see if this change fixes things for you?
A+,
Alex.
This archive was generated by hypermail 2b29 : Tue Aug 12 2003 - 10:41:51 PDT