Mysterious A7V333 no POST problems
esc...@bellnospamatlantic.net
Here's an A7V333 mystery which I hope someone in this group can solve.
My new system:
A7V333 RAID, 1.04, Bios 1015
2100 Athlon XP Palomino
Titan all copper heatsink with Titan 7500 RPM fan
512K 2700 DDR RAM
400 W POWMAX power supply
ATI All in Wonder Pro Video Card
Quantum Fireball 20.5 G HD
Generic floppy
Windows 2000
When I power on with the above, the green board light (and my NIC card
light) go on, but no POST, no video. The CPU fan acts as if it wants
to start, starts to spin, then quits. Nothing happens.
If I leave everything the same except substitute an older 350 PS for
the 400, I get POST, everything boots up normally, and Windows opens
fine. But even then, when I shut down, if I leave it off for more than
45 minutes, I have to reset everything in order to boot and POST. (I
get that stupid CPU failure message even though CPU and settings work
fine at 1733 MHz, but I have to restart to boot.) Then everything is
fine and stable, unless I leave it off too long.
Now onto the video card part of this melodrama. That seemingly "dead"
brand-new 400 watt power supply worked fine in my old Pentium II, so I
thought, well what would happen if I tried another video card in the
Athlon? I did that with a generic AGP card, and suddenly the bad power
supply works, and everything is again fine.
I have not yet spent the day attempting to see how long I have to wait
for the "refusal-to-boot-after-leaving-it-off" problem kicks in with
the 400 Watt PS, and I don't know how relevant that is.
I thought ASUS motherboards were supposed to be the best, which is why
I bought this one. Should I send it back for another one? Or is there
something easy I am overlooking? I really like the configuration of
this board, but I fear that the number of variables makes
troubleshooting a bit tricky.
I attempted the ROMSIP jumper trick, to no avail. I also switched the
BIOS card setting from PCI to AGP, with no effect.
Any advice greatly appreciated!
Eric
Paul
esc...@bellnospamatlantic.net wrote:
"Some monitors feed a leakage voltage back down the cable though
you usually get a post voice message saying cpu not onstalled
try another monitor or unplug the AC power cord and the monitor
cable from the VGA card
- press power button
- re-connect the power cord and monitor cable to VGA card
- press power button to check if there is still
"no CPU installed" warning
give this a try . maybe it works....
Tony
"
I think what he means is:
- switch power off on monitor and switch off power to the computer
via the switch on the back (in your case, wait for more than 45 minutes)
- disconnect monitor from video card
(at this point, there is no leakage into the computer, so it should
recover from the problems caused by the leakage...)
- press power switch on back of computer case (to ON)
(computer is now waiting for you to press the power button on the front
of the case)
- reconnect monitor to VGA card
- press power switch on front of computer case, to start the computer
- check for "no CPU installed" warning
or something like that.
At least that will verify whether it is the leakage problem or not.
HTH,
Paul
esc...@bellnospamatlantic.net
You are right that there is a leakage issue with the A7V333!
According to one posting to this group (from Andrew Markland
(andrew....@ntlworld.com), "there is apparently a 'known' issue
with the Asus A7V333, whereby "leakage down the VGA lead from some
monitors" causes boot failures.
The problem is, I don't even get to that point with the ATI All in
Wonder Pro card -- at least not with the 400W PS. That is because even
if the ATI card is so much as plugged in, the power supply refuses to
work! I don't even have to plug the monitor into the card to test
that, as it won't work at all. That video card works perfectly in the
Pentium II, and it also works in the Athlon with the 350 PS, once I
get past the "cold boot" problem.
I have been testing the system with the generic AGP card all day, and
I cannot get it to misbehave. No cold boot problem, no operating
problem, everything fine.
But I want it to work with the ATI, because that has a TV tuner.
I did a little more research and found that there is apparently a
voltage issue with some of the older video cards. According to one
Ron Miller<mill...@osu.edu>(posting to this group):
"The original AGP cards ran on higher > voltages than the AGP 2 cards.
>Some new 1.5v cards have onboard regulators > that allow them to work
>in older 3v slots, but putting an old 3v card in an > AGP slot that
>supplies only 1.5v probably will not work. (I may not have > the
>exact voltages stated correctly, but the reduced AGP-slot voltages on
> newer motherboards is probably what is keeping this "working" card
>from > working in your new A7V333."
There may be a voltage issue with my card, but the problem with that
theory is that it does not explain why it will work with the 350 PS
(with the cold boot problem), but stops the 400 PS dead.
Maybe I have both a partial monitor "leakage" issue and a "voltage"
issue when I use the ATI card -- both separately and in the
alternative.
For the life of me, I don't understand why. The logic escapes me.
Eric
(Still baffled.)
Kvon...@ubernet.net
falscher (fehlender) Kodierung für 3.3V, wie z.B. 3dfx Voodoo-Serie,
Matrox G200, die ersten Matrox G400, ATI Rage Fury, viele Karten mit
TNT oder TNT2 von nVidia und Intel i740, können Mainboard und
Grafikkarte zerstören!)
On Sat, 15 Feb 2003 23:22:20 GMT, esc...@bellnospamatlantic.net
wrote:
Paul
Kvon...@ubernet.net wrote:
> Nur AGP-Karten für 1.5V verwendbar (VORSICHT: Alte AGP-2X Karten mit
> falscher (fehlender) Kodierung für 3.3V, wie z.B. 3dfx Voodoo-Serie,
> Matrox G200, die ersten Matrox G400, ATI Rage Fury, viele Karten mit
> TNT oder TNT2 von nVidia und Intel i740, können Mainboard und
> Grafikkarte zerstören!)
I didn't even think about the AGP card compatibility, because I think
the Fury Maxx is the only ATI card to watch out for. Anything else would
be stopped by the keying in the AGP slot.
http://mirror.ati.com/support/faq/agpchart.html
Maybe the 400W power supply fan is not spinning fast enough, and you
have plugged in the power supply fan tachometer wire to a header on
the motherboard ? The motherboard decides to shutdown because the
fan speed is out of whack. It is likely the 400W uses a smart fan,
whereas the older supply uses a dumb fan that spins at a constant
speed.
Other than the leakage issue, thermal/fan issues are the only other
thing that comes to mind. On my motherboard, my slow PS fan just
causes me to get dumped into the BIOS, and by the time I exit, the
PS fan speed is above the threshold.
The leakage issue is something that has existed in electronics since
the invention of Complementary Metal Oxide Semiconductors (CMOS). I
saw it on my first project, where a system had two boards, and when the
power was turned off on one board, power flowed from the board that
was still powered, through the common logic signals they shared, and
into the second card. Enough power flowed, to get the rail voltage
on the unpowered card to 3.6v, and there was even enough power to
make the card fully operational, and light the LEDs on the card.
In the motherboard leakage issue, there is a serial interface on
the monitor (whose name escapes me right now), that gives info on
the monitor characteristics, so that the max resolution of the
monitor is not exceeded. What is happening, is that the monitor
remains powered, and current flows down the monitor cable into
the motherboard. There is apparently enough current to prevent
a key circuit on the board recognizing that the motherboard is
powered down. As a consequence, the motherboard doesn't properly
recognize the reapplication of power (or something to that effect).
It seems not all motherboard models suffer from this leakage issue.
I suppose the closest thing to an A7V333 is a A7V8X ? I don't know
if you would like the RAID setup on that thing.
Paul
esc...@bellnospamatlantic.net
were AGP card issues. I read the manual very carefully and it gives no
warning about incompatible AGP cards. My ATI card went right into the
slot, too! It is the old style All in Wonder Pro, the chip being the
Rage Pro Turbo. The page you mention states that it is a 3.3V card.
But if it isn't supposed to fit in the slot, why does it fit in the
slot? Now I am worrying about the generic AGP card I stuck in there as
a replacement. I find that after waiting overnight I cannot cold boot
with the "strike spacebar" feature or I get the CPU problem voice and
have to power off and restart with BIOS settings. To avoid this I have
to manually depress the on button.
I suppose I can haul this box upstairs and try the other monitor, but
that introduces yet another variable.
Earlier on this group, a posting from obtuse (no...@nooway.com) asked
about this same problem:
there are supposedly some older 3.0 volt AGP cards that should
not fit in the AGP pro (1.5 volt only) slot BUT DO . . and will
burn out the motherboard. That's why I need to
find out if the ATI AIW Pro (older rage pro card) which is
a AGP 2.0 card has BOTH 1.5 and 3 volt signaling.
After reading the above, I went to ATI's website, which says this
about my card:
· Incorporates comprehensive support for Intel's AGP including 2x
mode with Sideband addressing (3.3V only)
Hope I haven't damaged (or am continuing to damage) the motherboard! I
think my ATI card is sneaky -- because it should not fit but does!
Why don't they warn customers about this? (Maybe they'd rather make
the sale and deal with a possible later RMA than scare buyers
away....)
THANK YOU AGAIN PAUL!
Eric
Paul
esc...@bellnospamatlantic.net wrote:
Don't get too many issues into the fire at the same time :-)
Here are the theories to work on right now -
1) Video card plus 350W supply works, but has a cold boot problem.
Workaround - disconnect video cable to stop leakage.
Workaround - use a KVM switch ?
Permanent solution - find another monitor or replace whatever doesn't
work right at the computer end. Based on your symptoms, the video
card is involved some how, but I don't see how. I'm not sure
that getting another A7V333 is going to fix it, as I haven't read
any reports of someone with a problem solving it that way.
And no change to the BIOS is going to fix this, as it is a hardware
problem of some kind.
2) Video card doesn't even work with 400W supply.
The issue here is, if you connect the power supply fan connector to
the motherboard, the BIOS might be refusing to start because it
thinks the power supply fan is not going fast enough. Try just not
connecting the PS fan plug to the motherboard monitoring connector.
3) The AGP slot issue. Not an issue for you, because you did get your
system to work.
If you look at the ATI web page, it shows six combinations of AGP
cards and slots. The user discovers the compatibility issue, when
they cannot plug the card in, due to the use of keys in the AGP slot.
Now, that would normally be the end of the issue, but for a small
problem. A few brands/models of video cards had the wrong combination
of slots cut into them, so they are declaring the wrong thing.
The brand/models have been listed before, so I won't bother to dig
up that list again (especially as I get chewed out for trying).
The issue is, if a card that runs at 3.3V has the 1.5V slot cut in
it, it can be plugged into a motherboard that only operates the
AGP slot at 1.5V. The issue first showed up with an Intel Northbridge.
If you plugged in some old Sis video card, the Northbridge was
"burned out" as the video card applied 3.3V signals to a 1.5V only
Northbridge. Asus has since responded to this issue, by placing the
red AGP LED on the motherboards that can suffer a burnout if a
mis-keyed card is plugged in. So the window of opportunity is small,
it takes a motherboard just released around the Intel 1.5V
Northbridge era, plus an old video card. If you have a newer
motherboard, plus an old video card, the red LED protects you by
shutting off the motherboard in the event of mis-keying.
In the A7V333 user manual, I don't see any keys in the AGP slot
in the motherboard picture. (You can verify this, by comparing
your AGP slot to the pictures on the ATI webpage.) I guess this
means the KT333 chipset can handle a 3.3v or 1.5v card. (So, no
burnout issue and no need for the red LED circuit.) It is up to
the video card, to tell a universal motherboard, which voltage
it wants via the TYPEDEF signal. In your case, I guess the ATI
card will say "make mine +3.3 please" to your motherboard. If it
does this, then you won't be running at AGP 4x (only 1x and 2x
rates are allowed at 3.3V).
So only someone buying old stuff off Ebay and not doing their
homework is going to burn out a motherboard.
HTH,
Paul
esc...@bellnospamatlantic.net
Before reading it, I decided to err on the side of complete safety, so
I pulled my 1 year old ATI Radeon out of another machine, which was
accepted without hesitation. However, I also got the little voice
going, because I was wondering why I never heard any POST feedbacks.
Now that I have, I am more confused than before.
Regardless of whether there was a 3.3 volt video card issue, now there
definitely is not, and I am still experiencing the cold boot problem.
If I leave the machine off overnight, upon booting I get first a CPU
overclocking warning, followed by interminable memory error warnings,
and no POST. There are many ways to fix the problem: switch off a few
times, or remove and reattach the network cable, monitor cord, or
remove the RAM and replace it. Any of these actions seems to "reset"
the computer so that it will boot properly. I have to simply hit F-10
in the bios, answer yes and then everything boots up perfectly. When I
power off, I have about two hours, and then I get slammed by the
little voice.
I saw some posting about capacitors, and I wonder whether that might
be the issue, or maybe the monitor. But why would the monitor only do
this after the passage of two hours or more?
By the way, there is nothing wrong with my RAM (2700 DDR CAS 2.5T),
which I tested for a couple of hours with memtest86.com. Nor is the
CPU overclocked or problematic; I ran it through all SiSoftware Sandra
tests, and let it burn in for eight hours.
A posting suggested switching the RAM to number 3 position. I did that
and it booted fine warm (as it does if I merely pull it out and put it
back in its same position no matter where). But nothing changes the
cold boot problem. I don't think the RAM or CPU warnings are valid.
Thanks again; I still remain optimistic, and my number one complaint
is that this expensive CPU fan sounds like someone is running a vacuum
cleaner!
Eric
esc...@bellnospamatlantic.net
power supply fan connector to the motherboard. I never did that,
either. (Asus Probe has to be turned off because it read 0 for the PS
fan.) I can't see a connector which would go anywhere anyway.
Thanks!
Eric
2) Video card doesn't even work with 400W supply.
The issue here is, if you connect the power supply fan connector to
the motherboard, the BIOS might be refusing to start because it
thinks the power supply fan is not going fast enough. Try just not
connecting the PS fan plug to the motherboard monitoring connector.
On Tue, 18 Feb 2003 01:33:27 GMT, esc...@bellnospamatlantic.net
wrote: