----- Original Message -----
From: Wikston, James <[EMAIL PROTECTED]>
To: <[EMAIL PROTECTED]>; <[EMAIL PROTECTED]>
Sent: Thursday, August 24, 2000 12:29 PM
Subject: RE: [rtl] clock drifts
> Use NTP and you do not have to worry about the drift. The clock will be
> adjust to account for it.
>
> http://www.cis.udel.edu/~ntp
>
> -----Original Message-----
> From: Pedro Miguel Rebelo Martins [mailto:[EMAIL PROTECTED]]
> Sent: Wednesday, August 23, 2000 2:48 PM
> To: [EMAIL PROTECTED]
> Subject: [rtl] clock drifts
>
>
>
> I need to measure an interval betwen two instants given by the local
> clock which is implemented using the Pentium timestamp counter. I want
> to know how much the clock may drift in the intervall, or better the
> drift depends on what and how can i get such value, to account the drift
> on the measure. Are CPU cycles strictly periodic ?
>
> Thanks in advance and sorry my bad english
>
> Pedro Martins
>
>
>
>
> -- [rtl] ---
> To unsubscribe:
> echo "unsubscribe rtl" | mail [EMAIL PROTECTED] OR
> echo "unsubscribe rtl <Your_email>" | mail [EMAIL PROTECTED]
> ---
> For more information on Real-Time Linux see:
> http://www.rtlinux.org/rtlinux/
>
>
> EOM
>
> NOTICE - This message is the property of HATCH. It may also be
confidential
> and/or privileged. If you are not the intended recipient of this message
> you are hereby notified that you must not disseminate, copy or take any
> action with respect to it. If you have received this message in error
> please notify HATCH immediately via mailto:[EMAIL PROTECTED].
>
>
I think that NTP has an high granularity timing for my needs. My interval
measurements (upper bounds) may be in order of a few microseconds and I
expected clock drifts in order of less than one nanosecond. The Pentium
timestamp counter is incremented at every processor clock cycle. If the
processor clock cycles been strictly periodic, the clock wouldn't drift
towards a real-time reference (assuming a correct CPU frequency). My
questions are: the CPU clock cycles have jitter or are periodic without
jitter? If have jitter, depends only on the system crystal oscillator? How
can I get a feasible value for the crystal oscillation accuracy (i.e.
maximum drift towards a real-time reference)? It is specified on the
motherboard manual?
sorry the question again
Pedro Martins
-- [rtl] ---
To unsubscribe:
echo "unsubscribe rtl" | mail [EMAIL PROTECTED] OR
echo "unsubscribe rtl <Your_email>" | mail [EMAIL PROTECTED]
---
For more information on Real-Time Linux see:
http://www.rtlinux.org/rtlinux/
