I know I'm screwed just want to confirm what is screwing me

guavaball

guavaball

Extremely [H]
Joined
Apr 16, 2008
Messages
44,518
Ok so a few days ago I had a movie freeze running off a RAID 5 on my main system through Kodi. The computer looked ok but I decided to reboot. No dice. Had to do a hard reboot. Got the Hit F1 to enter bios message. I turned the clock back to normal on the cpu and went in. Decided to run Prime 95. Massive lock up and BSOD about 15 seconds in. Rebooted to another OS, both win7 64 bit. Ran Prime 95 again. Boom another BSOD and lockup. So while at work I ran meme test and below is a screenshot.

Already gotten some good advice here but just want to get a little more confirmation.

I'm guessing the ram is ok but I've got an issue with the CPU or mobo. Do you guys agree? I would hate dropping a shitton of a cash on the wrong problem.

Prime95 screenshot

BLdWuh.jpg


Memetest results

sZsUhR.jpg


H4Qe42.jpg



Anyone who can help me pin down the bad hardware I'd appreciate it.
 
CPUs and motherboards usually don't go bad, but I have seen it happen before. I assume you've reflashed the bios and set everything to default settings? Have you also tried to reinstall windows on a separate drive just so see?

If your CPU was overclocked it's possible that something was damaged. Good luck.
 
Sure looks as if the cpu is unstable for what ever reason.....double check temps and looking at any voltage settings in the bios related to cpu stability. Could also look at it from a memory perspective and take out most of the ram and recheck. I would put the cpu and memory to relaxed speed and clocks an see what effect it has. For all i know the cpu might be over heating as you didnt give this information. your power supply could be going out, but in most those cases the system would just loose power.
 
primetime said:
Sure looks as if the cpu is unstable for what ever reason.....double check temps and looking at any voltage settings in the bios related to cpu stability. Could also look at it from a memory perspective and take out most of the ram and recheck. I would put the cpu and memory to relaxed speed and clocks an see what effect it has. For all i know the cpu might be over heating as you didnt give this information. your power supply could be going out, but in most those cases the system would just loose power.
Click to expand...

Yeah I reset to default speeds and prime still did a belly up on 2 separate OSs. One vanilla and my regular.

Running a dual 140mm corsair cpu enclosed cooler. The temps in prime never went above 60C. regular the cores are in the mid 30s

The memetest rules out the ram right?
 
The Prime95 error is pretty interesting. Usually that's overclock related though. Make sure you aren't actually overclocking - a lot of motherboards "Default" includes some auto-overclock stuff.

The one DIMM at a time thing is a good idea, and a pretty easy test to run.

Another test to run may be disabling HTing, and then going to individual cores, and seeing if you still see the Prime failure. I've not seen a CPU fail internally and still POST/Boot/run stuff, but there's a first for everything. Asus AI Suite will also usually let you set a power limit on the CPU - for whatever reason, my CPU will fail on a load test if I don't have that enabled and set to 100% TDP, it will hard reboot my system (probably some auto-overclock crap I haven't locked down yet - I gave up looking and just live with the fix I found that works)

I would probably swap out PSUs if you have one available (or can buy one for testing and return it w/i 30 days) - if for no other reason than weird things are acting up, those are really hard to test beyond "Yup it turns on", and if everything else is still good and it ~is~ your PSU, it will fry something else eventually in the process of testing.
 
Also, I've seen memtest pass, but the motherboard have issues with DIMM sockets when running Windows. I think it's more of a motherboard or driver thing in those cases than DIMM related, as swapping DIMMs around works fine, it's usually particular sockets on the mobo.
 
I owe you guys. And thanks Brian for pushing me to recheck the bois settings. I moved it from normal to energy saver and its passing prime now.

nENk5r.jpg


Not sure if I'm out of the woods but hopefully it will last a little longer.
 
guavaball said:
I moved it from normal to energy saver and its passing prime now.
Not sure if I'm out of the woods but hopefully it will last a little longer.
Click to expand...
Be wary of new motherboards and adaptive voltages, and power save modes, i usually will set my voltages to fixed, with an offset. i have heard too many stories of people on bone stock everything getting crashes on load, and the whoel reason for it is either voltage droop (VRRIN) or not stable enough vcore (vCore doesnt go high enough to match freq) usually what i will do when troubleshooting, is set the voltage fixed to the default, or +.025 above and test. That rules out the PWM system of your board.
 
Since the system stabilizes with the "Energy Saver" setting, that may indicate that the PSU is failing under load.

I had a system like that a couple years ago. Everything was fine while I was using the system, but it would reset
when I was away from the computer for a while.

Figured out that the 3D screen saver was kicking in and stressing the PSU with the increased power draw from
the video card.

Ran a stress test under AIDA64 and could see the 5v drop off under load.

Replaced the PSU and no more problems.

.
 
Spartacus said:
Since the system stabilizes with the "Energy Saver" setting, that may indicate that the PSU is failing under load.

I had a system like that a couple years ago. Everything was fine while I was using the system, but it would reset
when I was away from the computer for a while.

Figured out that the 3D screen saver was kicking in and stressing the PSU with the increased power draw from
the video card.

Ran a stress test under AIDA64 and could see the 5v drop off under load.

Replaced the PSU and no more problems.

.
Click to expand...

Thanks for the tip. Would a failing psu also explain why even basic games like fate and defense grid are now choppy?
 
Just to add on to this thread, depending on how long you have had the chip, if you have been overclocking or even if the system has been doing its own overclock on it, it may start to develop instabilities. In the past when I have had this problem, I just had to up the volts a bit until it was stable again. Sometimes resetting it to the default will bring the stress down enough that the normal voltage will work, but not all the time. Sometimes you need to bring the clock down and up still increase the default voltage. Also in the past that usually lead to inconsistencies with the RAM when they shared the same voltage as the CPU. Now they should each be separately controlled.
 
agree with everything already stated and I will also toss in recommending checking the cmos battery. I've had dying batteries cause weird issues before and recently had issues on a z97 system with a dying battery.
 
pendragon1 said:
agree with everything already stated and I will also toss in recommending checking the cmos battery. I've had dying batteries cause weird issues before and recently had issues on a z97 system with a dying battery.
Click to expand...

Wow. I never thought of that in years. Good call.
 
Lots of good suggestions. :)

I would say if memtest passes your memory and your memory controller are probably just fine. The fixed voltage thing might be a good idea to try as well, as someone else suggested.
 
I did just get through a 3dmark test with no issues if that means anything.
 
guavaball said:
Thanks for the tip. Would a failing psu also explain why even basic games like fate and defense grid are now choppy?
Click to expand...

Hard to say.

I suppose the GPU might start to have performance problems if it's starved for power, but I would think it would just lock up or cause restarts in that case.
Certainly the Energy Saver setting decreases performance of the system and will cause loss of performance in games.

You can see a lot of weird system behavior with any power issues though, the system needs clean power.

.
 
What board and cpu do you have? Gpu and power supply? Is your gpu running at its full clock speeds during benches and staying?
 
5820k 3.3Ghz
Asus GTX 970
ASUS X99 A 3.1 Motherboard
EVGA 1300@ 1000W

CPU and GPU are stock speeds now.

I decided to replace it all except the video card to be safe.

I'll sell it all very cheap since I have no idea what was giving me trouble. I did get it stable only under econ mode but it wouldn't be fair to ask for much since I have no idea what gave me the problem in the first place.
 
guavaball said:
5820k 3.3Ghz
Asus GTX 970
ASUS X99 A 3.1 Motherboard
EVGA 1300@ 1000W

CPU and GPU are stock speeds now.

I decided to replace it all except t he video card to be safe.
Click to expand...
Well that sucks. Still more then capable rig. Something was causing it... but GLW new build.
 
Testing the old PSU independently is easy enough, at this point I wouldn't put any major effort into troubleshooting if it's the mobo or CPU that failed and would probably just recycle both.
 
guavaball said:
5820k 3.3Ghz
Asus GTX 970
ASUS X99 A 3.1 Motherboard
EVGA 1300@ 1000W

CPU and GPU are stock speeds now.

I decided to replace it all except the video card to be safe.

I'll sell it all very cheap since I have no idea what was giving me trouble. I did get it stable only under econ mode but it wouldn't be fair to ask for much since I have no idea what gave me the problem in the first place.
Click to expand...

I can have it, depending on how you define "cheap" =D

PM me if interested.
 
You must log in or register to reply here.
Back
Top