Re: A8V randoom reboot problem
- From: nospam@xxxxxxxxxx (Paul)
- Date: Sun, 14 May 2006 10:47:47 GMT
In article <u30e6219pq6protak9sbebe351uggffq6k@xxxxxxx>, Wizzard
<wizzard@xxxxxxxxxxxxx> wrote:
I have read some posts here but cannot see a solution so far. My
system is
A8V with AMD 3700+
Kingston DDR400 2 x 512mb memory (2.6v Standard 64M X 64 Non-ECC
400MHz 184-pin Unbuffered DIMM (SDRAM-DDR, 2.6V, CL2.5, Gold)
XP Prof
2 IDE drives, DVD writer and a DVD player
550W power supply
AIW 9800 pro graphics card
PC Probe shows temperature around 36c. I noticed that it was 41c when
the reboot issue started. Probe temperature limit is set at default
72c.
PC is ok for a few hours and then randomly reboots with no specific
application running. I thought about the memory power - the memory
needs 2.6v and how can I check that it is getting this please?
Also what is the max CPU temperature for this board please?
The max CPU temperature is defined by the processor.
On PDF page 8 here, the max temperature is specified as a character
in the part number. The value is somewhere between 65C and 70C,
according to the table. That is a case temperature, as measured on
the lid of the processor.
http://www.amd.com/us-en/assets/content_type/white_papers_and_tech_docs/30430.pdf
The actual THERMTRIP mechanism on the processor, monitors a
second diode on the processor die. The AMD doc 31411, section 6.3,
says THERMTRIP stops the processor at 125C diode temperature.
If the CPU overheats, the AMD document says you need to press
RESET to recover from that, so the motherboard shouldn't be
able to boot if a THERMTRIP happened.
You cannot check the Vdimm, unless you have a multimeter and
can probe the DIMM regulator near the DIMM sockets. There
are a number of "interesting" voltages on a motherboard that
are not connected to the hardware monitor. The hardware monitor
only shows you half of what is there.
Start with the basics.
1) Run memtest86+ (memtest.org). Normally I'd suggest running it
for two full passes, but I'd be very curious whether it will
run for many hours without the computer rebooting.
2) While booted into Windows, use the Prime95 Torture Test with
blended FFT setting, to test both the processor and memory
together. Prime95 (mersenne.org) does a calculation with a
known answer and can detect problems that memtest cannot find.
AIW cards tend to be sensitive to the 3.3V rail. On some of
those cards, it helps if the power supply runs at 3.4V instead
of 3.3V. Of course, you have no control over that :-) That is
all that comes to mind for your particular video card.
Replacing the card and trying a different brand/model of card,
just to see if the problem stops, is another possible test.
Remove the ATI video card driver, before shutting down and
changing the video card.
You can also try increasing Vagp a bit - your board has an option
to go from 1.5V to 1.6V. It probably won't make any difference
to stability, but you can try it anyway.
You can also try disabling Cool N' Quiet if you are using it.
Or check that you are using the latest driver from AMD (1.2.2).
http://www.amd.com/us-en/Processors/TechnicalResources/0,,30_182_871_9706,00.html
Really, it is hard to know where to start, as there is more
than one thing that can cause random reboots. Stopping
automatic reboots, so the computer will stop with a blue
screen, is one thing people try. If the computer still reboots,
that means there is something a bit more low level in the
hardware that is at fault. So even if you cannot force the
computer to give you a blue screen, at least it tells you
something (more likely to be hardware than software).
Paul
.
- Follow-Ups:
- Re: A8V randoom reboot problem
- From: Wizzard
- Re: A8V randoom reboot problem
- References:
- A8V randoom reboot problem
- From: Wizzard
- A8V randoom reboot problem
- Prev by Date: Re: I must be going NUTS by now
- Next by Date: Re: "HD Audio" or "AC97" front panel?
- Previous by thread: A8V randoom reboot problem
- Next by thread: Re: A8V randoom reboot problem
- Index(es):
Relevant Pages
|
