A8nsli deluxe

[TMI]

Ive been getting random reboots, and have had a friend anaylze the minidump file which is here:






*******************************************************************************

* *

* Bugcheck Analysis *

* *

*******************************************************************************



Use !analyze -v to get detailed debugging information.



BugCheck 1000000A, {9d066240, 2, 0, 8051ec1c}



Probably caused by : memory_corruption ( nt!MiDeletePte+198




looks like a memory corruption, so what do you recommend. I thought about pulling 512k of my 2 gb and checking for stability, then swapping until i can find the culprit. do you think that would work or could it be something else??

 

[GuitarDaddy]

Here are a couple of suggestions

1. This board tends to set the default ram volts lower than specs , check the specs on that Corsair and up the volts in bios accordingly

2. Check the volts on the 12v rail using ASUS probe and the amps with a multimeter(if you have one) this board has an unusually large requirement for amps on the 12v rail

3. Memory related errors and reboots can also be caused by VGA cards, if necessary go back to single card mode and test for stability.

Now a question for you

I see your using the XFX6600GT's how are they working? Did you get them at the same time?

I have RMA'd two of them trying to get SLI to work. I started with 1 card, a month later I added the second card and enabled SLI. Within 15 minutes the card in the primary slot failed(no signal) so I RMA'd it. A week later I get the replacement, pop it in. Same thing! 15mins in SLI operation and the card in slot 1 dies again, not completely dead this time, but causes random reboots and any games or graphics are virtually unviewable, so I RMA the second one

Very interested to see if yours works properly


*edit*

I just noticed 2gb ram in your sig. Do you have 2 x 1024 or 4 x 512? If you have 4 x 512 you must use 2T command rate in bios, and if you have 2 x 1024 you may need to loosen the timings to 3-4-4-8

Another thing to try. Remove the 20pin to 24pin adapter, I've read that they reduce electrical efficiancy and create heat . I have a 20pin PS plugged directly into the 24pin socket (the 4 pins nearest the chipset fan left open), and it has been rock solid for almost 2 months now. And of course you have to have a 4pin molex plugged into the EZplug.

 

[JMag]

GD:
Why do you think this board undervolts the memory?

I have tested it with a Multimeter and found out that it tends to overvolt.

I can't remember what BIOS I was using at the time but I know when I set the BIOS to 3.0v the reading off of the MOSFET was 3.18v

It might be different with BIOS changes, but that was my experience.

 

[JMag]

I don't mean to be completely useless here, so I agree with Guitardaddy's edit

You should use memtest to check if any errors are popin up due to memory or mem controller probs...

If I were you I would take out all but one dimm and run your machine for a while and see if you are still having probs.

Have you ever used any program or multimeter to watch you voltages coming out of your PSU? If not I would recommend doing that cause the PSU could cause mem errors (trust me on this one)

 

[GuitarDaddy]

Opps! used the wrong words. Sorry for the confusion, the motherboard is not overvolting. What I meant to say was the board detects the SPD volts to low, in my case it gave me 2.60 volts at default for my OCZ platinum TCCD ram that states its stock voltage to be 2.75. I'll fix my previous post

 

[JMag]

Oh ok... I understand now... HEH

I had seen you post this a few times and was really confused, I thought my board was the only one but then I asked Shamino over at VR and he confirmed that he too gets a higher voltage than what you set in the BIOS. I tried asking other people to test, but I think I am the only freaking nerd here who uses a MultiMeter.

Btw, I noticed the higher you set the VDIMM in the BIOS the greater it overvolts.

And i know Shamino knows what he is talking about because he easily has the craziest 3dmark scores on this board... But I guess if you put 3 phase change coolers on an A8N-SLI and get it up and running that tends to happen

Check It:

http://forums.vr-zone.com.sg/showthread.php?t=19597

 

[TMI]

Thanks, will try all. I am running 4x512 RAM of the corsair which I QVL'd on the ASUS site.

The dual XFX 6600gt's have been no problem. I bought them at the same time and installed them at the same time, and am running the 66.93 driver.


My friend that ran the debug test suggested it might be a bad paging file, so I reduced it to 2 mb, and the system has not crashed in the past 48 hours.

I was using the adapter because I called ASUS to ask if was necessary ( read a board review by Anandtech i think that said it wsnt needed). The tech said it would probably work withought, but you risk "frying" the board. But then again these same techs told me that the reason the Silicon image RAID manager software showed my drives as yellow was because one was dead....which it wasn't. ( had nothing to do with the fact that the drive was restoring redundacy I guess).

 

[TMI]

12.032v on the +12v w/ the 500w aspire PSU that came w/my case

 

[JMag]

Originally posted by: TMI
12.032v on the +12v w/ the 500w aspire PSU that came w/my case

Whered you get that # from? a MultiMeter?

Also, out of curiosity has your mahcine been stable with the smaller page file still?

 

[TMI]

started w/ a 2mb page file, but the task manager still showed commit limit at 4 gb . I then set No PAGE file, it shows a correct limit (1939600) available (1643004) and have been rock stable since. I think the culprit may have been a corrupt page file

 

[TMI]

asus probe like Guitar daddy sugested