A8N-SLI PC Giving Me Probs(AGAIN!)

[PsychoMadMan]

Hello,

For starters here is my current config:

AMD 3800+ x2 Not OC'd (With a Thermatake HSF)
Asus A8N-SLi Deluxe Bios 1014
2 GB of Corsair Value Select Ram
2x BFG 6800 GTs Not OC'd
Enermax Noisetaker 600W PSU
SB Live! 5.1 Sound Card
2x 80 WD ATA/100 IDE Drives RAID 0
2x 250 Maxtor ATA/133 IDE Drives RAID 0
Silicon Image 0680r PCI ATA/133 RAID Controller Card
I/O Magic 16x DVD±RW Drive
Windows XP Pro SP2

I've had problems with this pc ever since I built it a 2 years ago. I have to run the PC with the case open and a desk fan blowing into it to keep it from overheating when playing anything that is graphic intensive. (ie Oblivion, WOW, BF 2...etc) And now I'm haveing a problems cold booting the system. It takes 20 maybe 25 times to boot the system becasue the bottom card in the SLi setup typically refuses to boot. The red light on the card will come on for split second and turn off along with the HSF not turning on either. It runs fine after it boots as long as I keep a fan blowing on it. The other day I was playing Hitman: Blood Money and the pc froze. So I rebooted it and then all of the stuff on the POST screen including the BIOS Screen were completely scrambled the were all random characters like a whole line of ú lined up. So I put in just one card and took out the ohter and it worked fine put in the other card again and put the sli bridge on it and it went back to functioning like it usually does.

I went to boot today and now XP doesn't boot. It goes through the loading screen then the monitor turns off and then turns back on it again and the monitor then goes in a cycle looking for the input Digital/Analog meaning that its not seeing the signal from my vga card. It then after about 20 or so seconds goes to a BSOD which states...a whole bunch stuff no with no error message and the bottom it says: "nv4_disp" Which I know is my nvidia graphics driver. So I am able to boot in to safe mode I unstalled the drivers completely with driver cleaner and then boot up again in to windows install the newest drivers from nvidia. It goes through installs then I reboot it boots I enable the check box for sli the screen flickers, pc locks up, then reboots itself and goes back to what it was doing before. Another thing is that I did RMA both of the Video cards to BFG and returned the mobo when I first had problems about 2 years ago. (No help)

I test the PSU and it is ok reads 12.13DCV on the 12v lead 5.12DCV on the 5v lead and 3.34DCV on the 3.3v lead. All tested with a multimeter. I swapped positions on the cards top one in the the bottom PCI-e slot and the bottom one in the top PCI-e slot. Does the same thing bottom card doesn't boot. The PC runs and works fine with just one card(in either slot)...either one of my 6800GTs and my friend's MSI 6600LE.

I've gone through all the troubleshooting ideas that I have and nothing has helped my pc takes for ever to get started and now windows will not boot because of the nv4 drivers...So do you guys have any ideas? You think its a dud mobo, graphics cards just had it, maybe another piece of hardware not functioning correctly? Do you think the problem with the drivers related to the reason why the cards aren't booting?

I've been scrating my head over this for 2 years now. My pc was fine and happy and had no problems for about 5 months then ALL of these problems just randomly resurface a few weeks ago. I've been working with PCs for about 10 years now and I still cant figure this one out...Anyone else had this sort of problem with their rig? Or seen anythign else like it?

Please forgive me for spelling/grammer issues I do not have office on this pc so I cant run in through spell/gram check.

Thanks in advanced for any comments.

 

[PsychoMadMan]

Bump...anyone???

 

[pol II]

Just a few guesses: I would go on the assumption that the vid cards are okay seeing that youRMA'd them both, and either card is fine alone.

What is the model number of your PSU? I have an enermax 600W also; it is one of the first enermax noisetaker series with dual 12v rails. Not a good PSU IMO; I experienced frequent shutdowns and graphic corruption with it. I think it was rated at 17A 12v1 and 16A 12v2...I am of the opinion that the 12v1 line could not handle cpu, fans connected to mobo, and wattage delivered to GPUs through the PCIe slots. ALthough, theoretically, the current on the 12v1 *should* have been enough, I am not sure it was. So, if you have an early enermax noisetaker with dual 12v rails, you may want to have a look.

Also: I checked my rails with a digital multimeter and fluctuations were well within the +/- 5% ATX spec.

A full reinstall may be necessary; obviously, with the minimum number of peripherals installed. It sounds like a number of files have been corrupted. If you can swap PSUs with someone else and try again, maybe you could rule out (or identify) the PSU as the problem.

 

[PsychoMadMan]

Thanks for the info...

My Enermax has 18A on both 12v rails

I am currently able to boot into windows perfectly now with one card. I've tried playing games with just the one card and it crashes to the desktop after about 20min. The error message saying:

"To protect your hardware from potential damage or causing a potential system lockup, the graphics processor has lowered its performance to a level that allows continued safe operations."

So I ran rivatuner in the background whilst playing oblivion. From looking at the graph, the GPU core temperature was at idle 65C and while playing the game was ~83C. Then at a certain point(the point that the error pops up) the Core VID dropped from 6.00 to 2.00 instantly. Could the PSU be the cause of my GPU running insanely hot, or was the insanely hot temps causing my core vid to drop suddenly? I have no other PSUs that I can currently test my system on. If it seems like the PSU I might just go out and buy a PC P&C PSU that has one 12v rail. But I don't want to spend the money if its not gonna fix my prob. You guys got any other ideas?

 

[pol II]

The GPU temps are well within specs actually, even at load. The 6800GT max operating temp is 120-degrees C, so you're good to go there.

I'd like to get back to the PSU, since it seems that you've looked at almost everything else. In my system, I had replaced every single component except the CPU and the power supply; still, I had crashes. Switching to a PCP & C 510 watt PSU completely abrogated the problems. Check out this article for a little bit on PSU rail load-balancing:

http://www.extremetech.com/article2/0,1697,1932947,00.asp

I totally understand your reluctance to put any more money into your comp. Switching PSU's solved my problem, and I suggest you might consider doing the same. Good luck and keep us posted.

 

[PsychoMadMan]

Good info thanks again...

I'm pretty conviced that it is the power supply. But last night I had the idea of opening the case up and putting a fan in front of it...no reboots, no crashes, no errors. Now to me it sound like overheating. haveing a fan in front of it is really annoying, so I would still like to fix the problem. Does this still sound like a PSU issue?

 

[Kindjal]

I was searching the forum this morning and came across your thread.

I would inspect the chipset fan that came with the MB. Make sure it is working. I believe this chipset has both the northbridge and southbridge combined in one chip - thus it runs a bit warmer. I installed a Swiftech MCX159-CU chipset cooler before even using this MB (I have the non-deluxe version). I noticed a lot of concerns over this chipset fan - noise and failure rates. There is a great thread over at PC Perspective on this board.

I ordered the cooler from PerformancePCs.com and I got the version that has the fan with RPM monitoring.

This board also ate my brand new Antec 550w SLi PSU - so it's worth checking out the PSU too.

Good Luck.