A7V333 BIOS Settings for Barton
Steve
I have an A7V333 rev 2.0 fitted with a Barton 2500+ CPU and running the
1017 BIOS flash.
The system seems to be working fine, but SiSoft Sandra reports a warning
that the PCI and AGP buses are being overclocked. I think that's because
I'm running at a 1/4 multiplier rather than 1/5?
What should by BIOS settings be for this CPU to run without overclocking
the buses? I have never manually configured the CPU settings in the past,
and have just left the system to do it for me. Perhaps this is no longer
possible?
My settings currently are:
Frequency: 1833MHz
Multiplier: 11x
CPU/PCI Frequency: 166/33
Egil Solberg
"Steve" <st...@nospam-goofer.co.uk> wrote in message
news:bc4kuq$42h$1...@bananasplit.info...
> Hi,
>
> I have an A7V333 rev 2.0 fitted with a Barton 2500+ CPU and running the
> 1017 BIOS flash.
>
> The system seems to be working fine, but SiSoft Sandra reports a warning
> that the PCI and AGP buses are being overclocked. I think that's because
> I'm running at a 1/4 multiplier rather than 1/5?
>
> What should by BIOS settings be for this CPU to run without overclocking
> the buses?
My bet is: Sandra is wrong, as always. She probably only knows of the 1.xx
rev boards without the 1/5 multiplier.
Try another utility.
... et al.
Probably.
I don't know enough about how the hardware works to know if it's
possible, but one would have hoped that such tools would poll the
hardware for it's speeds, sizes and functions and report that, or else
tell you _very clearly_ that what it's showing you is just a presumed
value.
If his 2.0 board indeed has a 1/5 (CPU/PCI bus-speed) divider and, with
his CPU, it uses it like it should but Sandra reports it wrong, it will
be the basis for a lot of confusion and misinformation. Not a good
thing, and will make it difficult to find out when Asustek introduced
functional 1/5 dividers in the A7V333.
Egil, do you know of a software that does the job properly?
Steve, (even if it's just that Sandra has been informed about the 2.0
version, and thus presumes that you have a 1/5 divider) you could see if
there is a newer version Sandra and see if she still shows you the same
thing. If you test this and there is a difference, please post the
sandra versions.
--
Please followup in newsgroup.
E-mail address is invalid due to spam-control.
Zax
Thanks for the feedback.
Sandra is the current version as far as I can tell: 2003.3.9.44
Interestingly, I've been reliably informed via other sources, that Sandra is
unable to independently determine the clock speed, multiplier or divider.
Sandra just queries the BIOS for this information.
If this is true, then something is certainly not telling the truth. The
BIOS shows 166/33 which is clearly 5/1, whilst Sandra’s query of the BIOS
shows a 4/1. I get the feeling that something isn’t quite as it should be.
Egil Solberg
"Zax" <fle...@bananasplit.info> wrote in message
news:bcb13t$mc4$1...@bananasplit.info...
> .. et al. wrote:
> Sandra is the current version as far as I can tell: 2003.3.9.44
>
> Interestingly, I've been reliably informed via other sources, that Sandra
is
> unable to independently determine the clock speed, multiplier or divider.
> Sandra just queries the BIOS for this information.
>
> If this is true, then something is certainly not telling the truth. The
> BIOS shows 166/33 which is clearly 5/1, whilst Sandra's query of the BIOS
> shows a 4/1. I get the feeling that something isn't quite as it should
be.
I know there is a utility called CPU-Z, I don´t know if it works on this
issue.
Safest bet is to get a diagnostic PCI-card that can be inserted into a
PCI-slot. From those you can get good readings. But it would be a waste to
buy one just for this task.
I would just believe Asus on this one and assume all was correct.
Eventually, one could dig up a review with someone testing this mobo with
the abovementioned PCI-card, but I guess it will be a hard job to find that
kind of review.
Gerald Dehn
> If his 2.0 board indeed has a 1/5 (CPU/PCI bus-speed) divider and,
> with his CPU, it uses it like it should but Sandra reports it wrong,
> it will be the basis for a lot of confusion and misinformation. Not
> a good thing, and will make it difficult to find out when Asustek
> introduced functional 1/5 dividers in the A7V333.
> Egil, do you know of a software that does the job properly?
Hi,
I have found the same fault with my A7V333, Rev. 1.04 tested by Sandra.
Then I tried AIDA and saw that the divider was correct 1/5.
Greetings
Gerald
--
Die Know-How-Site: www.dehn-online.de
Die Wahrheit liegt in der Tat zwischen zwei Extremen, aber nicht in der
Mitte. - Moritz Heimann (1868-1925)