odd restart/bootup problem - pentium d 805 nforce4

[farrow099]

A couple months back I built a pentium d 805 based system and i've had an odd problem that persists even after installing a replacement motherboard and psu...

The system runs flawlessly until I attempt to restart it... half the time it just refuses to restart (and/or come out of hibernation/standby) without physically restarting the system via the reset button - which also resets the bios clock for some reason?

Has anyone encountered anything resembling this?

I thought I had it nailed when I realized spread spectrum was enabled on the sata bus - which (presumably) was the cause of both my harddrives refuseing to post upon restart quite a few times)

components are as follows:
Pentium D 805 @ 1.4v 3.66Ghz (I ran it at 2.9ghz for a long time and it didn't effect anything (not even the operating temp, haha), so I went back up to 3.66ghz)

ECS C19-A nforce 4 SLI XE
Wintec AmpX ddr2-667 (666mhz, 1.9v, 4,6,6,16,19 - command rate 2t)
Dual Seagate 7200.9 80gb drives
580watt 24 + 4 pin powersupply
Coolermaster R120 watercooler
Radeon 1800gto and an Audigy

 

[gomerpile]

are you sure you have the negitive on the negitive for the reset button. The reset will still work if wired backwards but the thing is evertime you use the reset button you are adding a + charge to the - and this can lead to bad effects like you are experiencing cant say that is

 

[farrow099]

I'm fairly certain its correct, but I'm going to double check it just incase.

I recall making sure the colored wire was on the + pin...

** UPDATE **

Checked the switch, its on correctly...
And I also came across an article that claims reset switches have no polarity; Which makes sense if you consider that the switch itself recieves no power... You are simply jumping the 2 pins on the motherboard.

 

[gomerpile]

Its not the switch that has power its the pins in the board and it does have power my muitl meter says so also, I did say, 'cant say that is so' in the earler post. Maybe the clock batt it going that could explain the bios reset. I was looking at your specks I thought the volts were 1.45 for that processor 1.9 seems two high for that processor.

 

[razvanstroiu]

I was looking at your specks I thought the volts were 1.45 for that processor 1.9 seems two high for that processor.

The 1.9 volt is, as I read it, DDR voltage, wich is OK.

 

[farrow099]

Yeah the ddr2 is at 1.9volts and the processor is at 1.4v and 3.33ghz (at the moment)

** Also, I wasn't saying that the reset button doesn't have a positive pin... What I was saying is it doesn't have polarity... As in no matter how you hook up the button, you are still going to make the connection in the same direction.

For instance you could jump the 2 pins (+, -) on the motherboard with a screwdriver and get the same result... Does it matter if the screwdriver is face up or face down?

But Thankyou for the advice, ANY advice would help at this point!

 

[gomerpile]

ya that is true I've done that too, just was thinking of different things for that is a rare thing, have you done a search at all. I did get simiular when I turn my volts too high but the system just shuts down than I have to take out the battry to reset it. I have the 805 and dont get that but than I'm not using ddr2 just a crappy board

 

[farrow099]

i might have been better off with some good old tried and true ddr, haha.... or my old AMD based system. I've never had issues like this... Generally the system either runs or it doesn't (with the exception of lockups that is).

Switched to Intel? Sounded good in theory.

I did search google, scour the internet through there.... Not much help - though this topic did sound familiar:
http://forum.osx86project.org/lofiversion/index.php/t12605.html
Unfortunately I've tried a ton of different setting to no avail.

 

[testbenchdude]

I posted about this very thing yesterday in another forum here... I'm running the same chip on a different motherboard, and I bumped up the voltage to 1.325 and it seems to be stable now.

 

[farrow099]

Gah, Intel....lol

 

[testbenchdude]

Gah, Intel....lol

Gah, helpful....lol

 

[jimytheassassin]

It resets your clock..physical time? in your bios?? Nothing else gets reset i assume unless you're referring to clock as in..Front Side Bus and Megahertz. Double check your jumper is correct if so..but i figure you've checked that by now. Voltage on your processor might still be low and causing poor start ups and restarts, hangs in hibernation perhaps. Out of curiosity do you have anything connected via USB. I had a monitor sharing hub that occasionally ran haywire disabling my mouse and keyboard. It gave the illusion that my comp had locked up. Pulling the usb on it briefly and replugging hot-fixed my keyboard and mouse.

 

[farrow099]

I do not, but I do have some front panel usb sockets that could potentially be hooked up incorrectly...

And by reset the clock i mean it resets the bios time and date. (and nothing else!)

 

[Tattysnuc]

Forgive me if I've misread your post, and have got the wrong end of the stick. Are you suggesting that motherboard headers "redirect charge"?

If this is the case, then please enlighten me as I am fascinated how a DC switch can put charge onto the wrong terminal.

Please elaborate. I'll explain why I'd like you to.....

The motherboard in the main is a DC component. The "headers" that feed to the front of the case, either have push to make, or push to break switches, with an LED.

Wire it the wrong way round, and the LED doesn't work, if it's a HD/Power LED.
Wire it the wrong way round and its a power/reset switch, it makes NO ODDS. The action of pushing to break, still breaks the connection, or making a push to make, makes the connection.

There is no "redirection of current flow", or "switching of polarity", but simply a switch that makes or breaks a connection.

Now, as I said, forgive me if I've got the wrong end of the stick, but surely you can see why I'm now a little confused.....

 

[gomerpile]

like I said in the post "I don't know if that is so"?What I was thinking was if a - is on a ground and that negitive is hooked up to a positive a shorting could happen but I don't know if that is so with a switch on a computer, I remember one time I hook up a switch for my heater and when I turned on the stat, sparks flew everywhere and that is where I got the idea but like I said I dont know if that is so in this case.

 

[jimytheassassin]

Extreme tech shows a quick what's up with polarity, in summary
"Note that the power switch, reset switch and speaker connectors can be installed without worrying about the polarity. The LED jumpers, however, must be installed with the correct polarity. This usually means that the (+) pin is connected to the non-black connector."

And now back to your clock reseting..that's weird..i'm clueless now

 

[Tattysnuc]

If the bios/clock is resetting, to me that indicates either:

1. A bios battery that is flat
2. A faulty bios or a bios that is not working correctly.

Some things to check:

Do you have the reset bios jumper set in the operating position?

Is the memory functioning 100% correctly? If the computer is dumping everything to ram, and the RAM is not functioning correctly, then that will obviously cause some problems. Not sure about resetting the clock though.

What happens when you completely remove power to the board? Does it remember the bios settings? 10 mins? 1 hour? 12 hours?

Have you tried updating/reflashing the bios? Bootign with the "Hat" chip on? On AMD motherboards, ECS provide a chip that piggybacks the bios chip so that you can boot. Do you have this functionality?

 

[farrow099]

Yeah the ddr2 is at 1.9volts and the processor is at 1.4v and 3.33ghz (at the moment)

** Also, I wasn't saying that the reset button doesn't have a positive pin... What I was saying is it doesn't have polarity... As in no matter how you hook up the button, you are still going to make the connection in the same direction.

For instance you could jump the 2 pins (+, -) on the motherboard with a screwdriver and get the same result... Does it matter if the screwdriver is face up or face down?

But Thankyou for the advice, ANY advice would help at this point!

I believe i mentioned that.

 

[farrow099]

And Yes it works fine (and keeps normal time) except for when it locks up...

Sometimes it restarts fine, sometimes it doesn't!

For the time being I am assuming (from process of elimination more than anything) that it is a RAM problem.

For now, I'm just not restarting it very often, lol.