Asus Sabertooth x79 Boot Issues

  • Thread starter Deleted member 2266436
  • Start date
D

Deleted member 2266436

Guest
Update 1:

System was unable to boot for roughly a day or so. Whilst I was passing it I accidentally gave the side of the case a thud with my laptop bag (not a regular event), decided to press the power button and it sprang to life. I got a blue screen for a few seconds with a message which read "PAGE_FAULT_IN_NONPAGED_AREA". The system re-booted and started running a C drive repair shortly after re-booting which is strange because it's the first time I've received that blue screen error.

Original Post:

Hi,

I’ve recently been having an issue with my desktop computer and believe I have tracked down the issue to the Motherboard or possibly the CPU (please find system details below).

The majority of the time the system runs fine but can randomly cut-out and require a reboot. Sometimes it’ll reboot fine and sometimes it’ll get 'stuck' in the process and need to be shut-off by hitting the power button; when this happens the hardware LED will stay on and if I leave the computer long enough it’ll give power to the connected peripherals but not boot up. When the system tries to boot and gets ‘stuck’ the monitor is receiving no signal.

More recently it has been cutting out without warning then failing to boot at all, the system will come on for a brief second then cut out again, during this brief second of power the CPU LED comes on. This will go on indefinitely until power is cut via the PSU or directly from the power outlet. Sometimes the system will boot first time after the power is cut from the PSU or power outlet, sometimes it’ll just not work at all and take a half day to start working again.

I’ve tried almost everything I know short of replacing the motherboard / CPU. I’ve tested each GPU independently for 7 days and encountered the same issues. I’ve tried running a different set of GTX 980s and independently but, same issues. I’ve replaced the SLI Bridge. Each RAM stick has been tested individually and are running at the correct speeds, etc. The CPU has been tested thoroughly (power consumption, temperature, clock speeds, cache, etc - all normal). I’ve tried resetting, downgrading and updating the BIOS with no effect. The system has been highly maintained with correct cleaning practices - could rule out debris shorting the mobo. All of the connections are secure and everything is updated. The CMOS battery is apparently working fine but I’ve got a replacement coming anyways.

After reading over the forums I've checked my capacitors and they seem to all be in order, no bulging/discharge.

Possibly damaged dimms or a PCI-e lane?

Main:
Motherboard: Asus Sabertooth X79
RAM: 32GB HyperX 1866MHz DDR3 – Quad channel
CPU: Intel i7-3970X - No OC
Cooler: Corsair H110i GTX – Noctua NF-A14 PPC 2000 PWM
GPU: GTX 980 STRIX x2 (SLI) - No OC
PSU: Corsair 800W Gaming PSU

Storage:
Samsung 850 EVO 250GB SSD – x2 in RAID 0 configuration.
Corsair Force 480GB SSD – application storage.
Seagate 2TB 7200RPM 64MB 6GB/s Internal – x2, for storage and backups.

External:
Case: Cooler Master Mastercase Pro 5
Case Fans: Corsair Air Series AF140 Quiet Edition 140mm – x4 + LED strip x2.
Monitor: Asus PB279Q - DisplayPort connected
Keyboard: Corsair Vengeance K90
Mouse: Corsair Vengeance M90
Audio in/out: Focusrite Scarlett 2i2
Microphone: Electro-voice RE320
Speakers: Logitech Z333 Multimedia Speakers
Headset: Astro A40

 
Solution
PAGE_FAULT_IN_NONPAGED_AREA usually means RAM issues.
It might be anything from overclocking, controller(CPU), MB, dead sticks and even PSU.
Some of the below might be looking unnecessary with test you've done, but they are not:
Since for some odd reason you are not overclocking, turn of the computer, remove all 4 memory modules, use eraser to clean the contacts on the sticks (even if the look clean). Put the memory back.
If you keep having this issue, try to use different memory slots. Splitting them into two dual channel might also help if the issue is with CPU - there are very few scenarios that you'd notice the difference between dual and quad channel RAM config.
If the above didn't help, than start removing RAM sticks. If removing...
PAGE_FAULT_IN_NONPAGED_AREA usually means RAM issues.
It might be anything from overclocking, controller(CPU), MB, dead sticks and even PSU.
Some of the below might be looking unnecessary with test you've done, but they are not:
Since for some odd reason you are not overclocking, turn of the computer, remove all 4 memory modules, use eraser to clean the contacts on the sticks (even if the look clean). Put the memory back.
If you keep having this issue, try to use different memory slots. Splitting them into two dual channel might also help if the issue is with CPU - there are very few scenarios that you'd notice the difference between dual and quad channel RAM config.
If the above didn't help, than start removing RAM sticks. If removing solves the issue, try to use the "removed" sticks as well. If there is no issue, the problem might be the CPU (memory controller) or the MB.

P.S.
Faulty PSUs have proven to be the reason for many problems. You have not tried to replace yours - kinda easy thing and seems like you have a spare one ;)
 
Solution
D

Deleted member 2266436

Guest


Before my system began having these issues I was running a stable 4.9GHz overclock on my CPU and 1500MHz on my GPUs. I lost the CPU overclock when I removed the CMOS battery to clear the bios and just haven't gotten around to setting it up again. The GPUs just require the profile to be changed to resume their overclock but it's just not necessary to be running overclocks with an unstable system.

I've already tried several configurations with the RAM without any luck (I wasn't clear enough in my original post, apologies), I'll try cleaning the contacts, hopefully this'll help.

I think I'm going to swap out the PSU later this week, thanks for your advice.



 
No problem.
Have you tried the "load optimized defaults" option in BIOS after the reset ?
Also, check that RAM settings in BIOS are correct - XMP profile in use.
I had some RAM issue that was resolved by reflashing the latest BIOS.

P.S.
Lucky you with 4.9GHz. I can't go beyond 4.5 :( 4.5 is rock solid at 1.266v. But to do the 4.6 i need to go beyond 1.35v and that's the point where VRM overheats with my mini ITX :(
 
D

Deleted member 2266436

Guest
Yes, the only thing that did was change RAID back to AHCI. The RAM looks fine in the BIOS, correct settings, etc.

Tried removing all the RAM just to force a post beep but the system isn't able to boot atm, sometimes it'll start but after 30 mins - 1 hour it'll cut out, etc...

I'm really leaning towards a failing PSU, the system is essentially running at 800w according to the manufacturers specifications ( GPUs 600W, MOBO 60W and CPU 150W) and it's only an 800W PSU in the system (didn't upgrade when I got the second 980, ran without any issues for just over a year), so accounting for RAM, drives and fans, that's what... an extra 5-10W? So any type of failure in the PSU could possibly be causing some if not all of the issues I'm having - what do you think?

I've ordered a Corsair HX1000i so hopefully that'll solve the issues.

That's a sweet lookin' setup you've got! I'm rocking the black and white theme myself.

I've had the CPU at 5GHz before but the temps were uncomfortably high... I think I managed to get away with quite a low voltage at that frequency too, although it was over a year ago and I didn't run it for more than a day so I doubt it was very stable at all.



 
Thank you. that system was very tricky to build.
Your GPUs are ~200watt each under full load.
so your system is about 600-650w total power consumption under full load. While Crosair's GS series is not the best, it's decent. Most probably, you will not run SLI with your next GPU upgrade (it does not make any sense anymore), so even 800w is a bit overkill not to mention HX1000i.
I'd try it with other PSU before ordering new one -(especially 200$ one)
 
D

Deleted member 2266436

Guest
Looks like I misread the cable's output as the cards power draw...

Over the last few days I've wasn't able to get the system to boot at all; it would just power up and immediately power down again without any POST codes, so, I decided to go ahead and order the HX1000i on Friday - installed on Saturday. I went for the HX1000i because of the outstanding reviews, low noise, monitoring abilities and it wasn't much more than the HX750/850i.

The system initially had some issues booting but after clearing the RTC there were no problems booting. Just to be thorough I booted the system using each individual stick of RAM (8 total), system booted fine each time. I tried running the Windows Memory Diagnostic Tool last night but it just got stuck at 21%(was at 21% for 8 hours, it said there were no issues detected) on the first pass - planned on running 10... I'm going to try running MemTest today instead.

So for now it looks like we're back up and running!