A7N8X reboots

Nope, that isn't checking the power supply. Look in hte bios at the
5v. If it is below about 4.95, get a different power supply and try
again.

That is too tight a spec. The power supply itself is allowed +/- 5%.
There is a slight loss on the motherboard, and there can be some
measurement error. Seeing somewhere between 4.5 and 4.75V on the
+5V, might be cause for concern.

The 4.95 number would only allow 1% error, and you could easily
lose 0.05V in the cables themselves. That would result in most
power supplies being rejected.

Paul

Yea.. I am gonna back paul up on this one .. Just so the op doesnt trash
a good supply..
Besides I have a feeling peter suggested the ops fix .. Do not reboot on
errors.. If that's not it then it's the ram..

I'll check that with the power supply later, I'm at work now. I think
it's ram related though. I pulled 1 stick out, so I'm at 512 instead
of 1GB. I've had no problems, and even went to a game night a full
bore. I haven't tested the stick I pulled to see if that one has a
problem yet, but that is on my todo list here. I had the sticks in the
1 & 3 spots, and I pulled the 3. I'll stick that one in 1, and test
that for a bit. If that works, then I'll put both in 2 & 3 and see how
that goes. Thanks for all the help guys.

On Mon, 08 Nov 2004 02:11:21 -0400, (Paul) wrote:


That is too tight a spec. The power supply itself is allowed +/- 5%.
There is a slight loss on the motherboard, and there can be some
measurement error. Seeing somewhere between 4.5 and 4.75V on the
+5V, might be cause for concern.

It is more than cause for concern, it'll cause your hd to reset once
in a while. Maybe 4.95 is too tight, but under 4.75 measured with a
meter on the 5v of a drive connector means you have a problem waiting
to happen.

In article . com,
" wrote:

I'll check that with the power supply later, I'm at work now. I think
it's ram related though. I pulled 1 stick out, so I'm at 512 instead
of 1GB. I've had no problems, and even went to a game night a full
bore. I haven't tested the stick I pulled to see if that one has a
problem yet, but that is on my todo list here. I had the sticks in the
1 & 3 spots, and I pulled the 3. I'll stick that one in 1, and test
that for a bit. If that works, then I'll put both in 2 & 3 and see how
that goes. Thanks for all the help guys.

I've been all through this a couple weeks ago with an A7N8X-E and
a Mobile 2600+ overclocked to 11x200. I had some cheap RAM that has
been sitting in my P4 system for about a year, and I never suspected
there was a problem with the RAM.

My symptoms were - run in dual channel (slot 2 and 3), I can only
get the FSB to 180 and have a reasonably error free system. If I
move the RAM to slot 1 and slot 2 (single channel mode), I can
push the FSB to 200, with about the same error rate. (A third
thing you could try, is get a hacked BIOS from nforcershq.com .
A "CPC off" BIOS, runs with command rate 2T, and will also
stabilize bad RAM, at the expense of a big drop in memory
bandwidth. It makes your clock rate look more impressive, but
doesn't solve the problem, so forget it as a solution. You have
to raise the FSB another 30MHz, to compensate for lost performance
with command rate 2T.)

You can try memtest86 in those two conditions, and maybe it will
be error free in an overnight test. Things got a lot more
interesting when I tried Prime95 from mersenne.org . I got
reproducible errors from the same region of memory, and usually
I could get an error in about 30 minutes of running multiple copies
of Prime95 in Linux.

I've got two 512MB sticks of Ballistix in there now, and it is
solid as a rock at 200MHz. I've even run with three sticks, still
at 200MHz, still rock solid. I haven't pushed it further, because
I don't really want a system that runs hot. Works at 2-2-2-6.

There are many reports of the Nforce2 being picky about RAM,
and so far, I don't see a common thread to what works and what
doesn't work. You can read these threads, to see what people have
tried. I think buying RAM that is used to make the current
generation of CAS2 memory (the latest rev of Micron parts, or
the Samsung TCCD) might be a start. The trick is to find some
cheap memory, where the manufacturer doesn't know the potential
of those chips.

http://nforcershq.com/forum/viewtopi...light=poll+ram
http://nforcershq.com/forum/viewtopic.php?t=48020

Note that a number of the people in the above threads, don't
have their boards set up for best performance, so even if they
claim stability while running some kind of async settings, that
is not what you should be aiming for.

HTH,
Paul

Chris Hill wrote in
:

On Mon, 08 Nov 2004 02:11:21 -0400, (Paul) wrote:


That is too tight a spec. The power supply itself is allowed +/- 5%.
There is a slight loss on the motherboard, and there can be some
measurement error. Seeing somewhere between 4.5 and 4.75V on the
+5V, might be cause for concern.

It is more than cause for concern, it'll cause your hd to reset once
in a while. Maybe 4.95 is too tight, but under 4.75 measured with a
meter on the 5v of a drive connector means you have a problem waiting
to happen.


I had a thread last week on this same subject, my 5v was 4.49 - 4.52v on
MBM. Either memory, power supply, hard drive, or MB problem depending.


I have a new hard drive coming in from WD, but I finally wised up and
checked the voltage on the underside of the hard drive at the circuit
board contacts under the connector. Dead on needle at 4.5 volts. So the
lovely Antec 430 I just bought is kerflooey. I ordered a 470 Silencer
from PC Power and Cooling and will eBay the replacement warranty Antec as
I can't be without a computer and EQ2 hitting in the next day or two!

Unk

" wrote in message roups.com...
I have a new system here that reboots when it starts to get taxed.
Here are the specs:
A7N8X MBoard
AthonXP 3200+ (I think, I have thoughts, but that is what I bought)
2x512 3200 DDR Kingston ValueRam (I think the 3200 is right)
30GB, 40GB IDE drives
SB Live
ATI Radon 9600

It seems to reboot more when I have a game running, so it might be
related to graphics card, but I have all the newest drivers, and it did
it with the out of the box drivers as well, so I'm less inclined to
think that. I have the ram at 200 Mhz, and right now it seems to be
going fine. I'm leaning towards a processor problem.
Right now I have it clocked at 133 Mhz (bus I'b guessing, I havn't read
all the details yet). It seems to run fine at 133, or 100 Mhz. If I
up it to 166, or 200 (which it should be at) it reboots when I'm doing
more taxing things. I've had it do it while browsing as well (movie
dnlding and something else I think). If anyone can give me some ideas
as to what I should do, that would be great.
Thanks.

A marginal power supply can be the cause of the symptoms you describe.
If one of the power rails collapses, or goes into current limit mode,
the computer will reboot.

arnie

Ok, I finally checked the voltage in the BIOS, and switched the RAM
chips to check that. With20 min of playing time in the WOW open beta,
I had no problems, but sometimes it takes a bit. I'll test more later
when I get time to get something running.
As for the BIOS, it says 4.94. I'm going to guess that is the problem
possibly from what everyone is saying?

Ok, I finally tested with both sticks in again, this time in the 2 & 3
spots, and it crashed again. It looks like my power supply is fine, so
I'm going to try to see what I can do about getting new memory here.
Unless someone else has another option.
Actually, what do some people use in this board?

" wrote in message roups.com...
Ok, I finally tested with both sticks in again, this time in the 2 & 3
spots, and it crashed again. It looks like my power supply is fine, so
I'm going to try to see what I can do about getting new memory here.
Unless someone else has another option.
Actually, what do some people use in this board?


You might not be able to see it with a voltmeter because the response
time is too slow. You might only be able to see it with an
oscilloscope. A voltmeter would never see a voltage collapse lasting a
few milliseconds, but the processor wouls because the switching power
supply circuits can't store a lot of energy, that's why they operate
at 10 KHz, switching frequencies.

arnie