Failed reboots with OC'd E8500

[Brutality]

So I've overclocked my E8500, and for some reason (tried to troubleshoot everything, can't find the problem) the system fails to reboot after 2+ hours of gameplay (Elder Scrolls IV) or 20min+ stress tests in Furmark - You may have read my others topics but I recently found out it wasn't a System (motherboard) temp problem when overclocked... I also got reboot problems with stock GPU settings (non-overclocked), come to think of it... I claimed that there were no problems during the LAN party... but I never rebooted, I've only been playing for many hours, then shut down computer, actually no reboots...

 

So this rebooting problem can be described as follows:

The system seems to shut down the OS properly, but as it gets to the point where it's supposed to reboot, the screen shuts off (or rather the motherboard shutting off connection from PCI to Monitor - that'll be by guess) and nothing happens, the system keeps running but it will never reboot, forcing me to hit the power button, and once I start the system again (doesn't matter if I wait for hours, if it had been a overheating problem) it turns on for a couple of seconds, and then shuts off, remains shut off for a second or two, then powers it self up again and I will be met with American Megatrends message: "Overclocking Failed..."

 

So basically I get this if I don't wait for my system to cool down before rebooting, I haven't figure out what temperature this depends on, since I have tried overclocking my gpu to 890/1280 and I get the reboot problem if the System (motherboard) temp is above 38*C, but if I don't OC the gpu, it reboots fine at 40*C (haven't tested where the line is, but I doubt it's exactly the system (motherboard temperature)) but at higher temps (I didn't get to see) I get the reboot problem as well with a non-overclocked gpu :/

 

So these are my BIOS settings:

 

Multiplier: 9.5

FSB Frequency: 413

PCIE Frequency: 105

DRAM Frequency: DDR2-1036MHz

Timings: 5-5-5-15

CPU Voltage: 1.35

CPU PLL Voltage: 1.5

FSB Termination Voltage: 1.3

DRAM Voltage: 2.2

Load-Line Calibration: Enabled

Spectrums, EIST, C1E and C3-6: Disabled

Rest on Auto

 

CPU Frequency = 3923.5MHz - Max temp: 49*C through 24 hours hours Prime95 Blend test

 

I haven't tried setting it all back to defaults, but I have on the feeling the problem has to be solved in here.

 

Please allow me to apologize for spamming these forums, I know there are (at the moment) lots of topics made by me which might be pointless, but I am just so confused about this reboot problem

 

Here are the max temps during gameplay without OC'd GPU, if needed:

 

MSI Afterburner:

GPU temp: 78*C

 

Core Temp:

Core #0: 43*C | Core #1: 39*C

 

HWMonitor:

System: 46*C

CPU: 31*C

AUX: 38*C

Core #0: 44*C

Core #1: 37*C

Assembly (HDD): 23*C

Air Flow (HDD): 22*C

Radeon HD 5870: 78*C | 65*C (I guess it's core/mem temps)

 

Oh and here's picture of my system, someone mentioned airflow, but never told me whether it was fine or not... (the cables are just pushed up in the tray.... area, meh forgot the word for it)

 

 

As always - Many thanks in advance

Edited December 4, 2010 by Brutality

[sack_patrol]

Hello, don't worry about posting about any sort of issues you might have. That's what the forums are for after all right.

 

Seeing those temps, I can't really tell if it's an overheating problem, though I'd suggest you reset the CMOS which is done with a jumper on the motherboard. You can find which one it is by reading the Motherboard manual. It's a fundamental information so it won't be hard to find.

 

Since it's rebooting, that means that it isn't stable so I suggest your start overclocking it again from scratch, throughly testing the system after each modification you make so that you're sure that it's stable at all times.

 

Also, make sure you run Memtest86+ from the MS-DOS if you want to overclock the RAM timings so much. Make a bootable CD with Memtest. There should be lots of sites with instructions on the Internet. Basically you need to get a CD, and then using a program like NERO, burn the .iso file on the CD and that should be it. Then just run the program for 30 minutes after each little overclock, and preferably 12 hours when the overclock is done. That's aside from the 12 hour test you'll need to run on the CPU after the final overclock. If you do not do those tests, you'll probably end up with another crash and you won't know what the hell is going on.

 

gl.

[Brutality]

Hello, don't worry about posting about any sort of issues you might have. That's what the forums are for after all right.

 

Seeing those temps, I can't really tell if it's an overheating problem, though I'd suggest you reset the CMOS which is done with a jumper on the motherboard. You can find which one it is by reading the Motherboard manual. It's a fundamental information so it won't be hard to find.

 

Since it's rebooting, that means that it isn't stable so I suggest your start overclocking it again from scratch, throughly testing the system after each modification you make so that you're sure that it's stable at all times.

 

Also, make sure you run Memtest86+ from the MS-DOS if you want to overclock the RAM timings so much. Make a bootable CD with Memtest. There should be lots of sites with instructions on the Internet. Basically you need to get a CD, and then using a program like NERO, burn the .iso file on the CD and that should be it. Then just run the program for 30 minutes after each little overclock, and preferably 12 hours when the overclock is done. That's aside from the 12 hour test you'll need to run on the CPU after the final overclock. If you do not do those tests, you'll probably end up with another crash and you won't know what the hell is going on.

 

gl.

 

Thanks, I'll give it a go

 

But I haven't overclocked the RAM, 5-5-5-15 are stock timings at 2.2V

 

http://www.valueram.com/datasheets/KHX8500D2K2_4G.pdf

 

I reset the CMOS before I started overclocking (back when I got my Noctua NH-D14 - couple weeks ago), should I do it again?

[sack_patrol]

Nice ram.

 

As for the CMOS - Yes. Every time something goes wrong with your overclock or you just wanna start over, you can reset the CMOS. You can use it as much as you want. That's what it's there for.

[medbor]

Overclocking failed is because you killed it with the powerswitch before the bios is fully loaded (i.e. when it hangs at reboot)

The main weird setting is the PCI overclock, WHY?! i've heard that a thats a dumb idea, bu i don't know why (could be causing troubles)

The other thing i usually think about (you said it was stable in P95 so that is probably alright) if all else fails is a bad psu (which is nearly impossible to test). If a psu gets more than some years old they lose performace in Watts and Ripple suppression, so that could be a problem (though i see that it's a good corsiar with power to spare so it shouldnt be that).

Might also be the bios battery is out of juice?

 

It's a very low level problem, either Overclock problem (which it shouldn't be) or HW problem on either MB, CPU, GPU, PSU, RAM. Check the capacitors on the motherboard:

[Brutality]

Overclocking failed is because you killed it with the powerswitch before the bios is fully loaded (i.e. when it hangs at reboot)

The main weird setting is the PCI overclock, WHY?! i've heard that a thats a dumb idea, bu i don't know why (could be causing troubles)

The other thing i usually think about (you said it was stable in P95 so that is probably alright) if all else fails is a bad psu (which is nearly impossible to test). If a psu gets more than some years old they lose performace in Watts and Ripple suppression, so that could be a problem (though i see that it's a good corsiar with power to spare so it shouldnt be that).

Might also be the bios battery is out of juice?

 

It's a very low level problem, either Overclock problem (which it shouldn't be) or HW problem on either MB, CPU, GPU, PSU, RAM. Check the capacitors on the motherboard:

 

I see

 

after a failed overclock message I decided to let it reset everything to default values (just gonna try to see if it's stable at this or not), but will definately check out the capacitors soon (gtg soon)

 

If you by PCI overclock mean the 105 I just read somewhere (will find link soon) that it doesn't hurt anything, only improve gpu performance

 

Anyways thanks alot I will check things out, now I'm gonna play some Elder Scrolls IV to see if the system is stable at even default settings...

[sack_patrol]

Thw PCI improving gpu performance is a myth. It's not set in stone and even if there's some kind of improvement, it's worthless so no use modifying the value. All you could get out of it is a Corrupt HDD and USB ports.

[medbor]

Thw PCI improving gpu performance is a myth. It's not set in stone and even if there's some kind of improvement, it's worthless so no use modifying the value. All you could get out of it is a Corrupt HDD and USB ports.

That sound like what i heard, and also PCIe 2.0 cant really bottleneck anything (almost atlest). there should be bandwidth enough to not even notice the 5% increase

[Brutality]

I've played some games with default bios values, and the system seems stable (no reboot problems) so luckily (I guess) it's not hardware issues...

 

Come to think of it, it could very much sound the like PCI overclock could be the issue... I've actually never had my headphone on during "reboot" failure.. Couldn't it just be the system failing to enable the PCI (graphics card -> monitor) because of the OC, and then the monitor obviously wouldn't turn on...

 

Soon I'll begin from scratch with overclocking and obviously won't touch PCI Frequency and see how it goes, anyways at least the system is stable at stock / default settings =)

 

Damn I was so worried I had damaged my motherboard during the mounting of the giant cooler aka. Noctua NH-D14

 

Cheers Brutality

 

PS. someone asked following question on http://www.computerlounge.co.nz/forum/Default.aspx?g=posts&t=50

 

"PCIE Frequency: 105 - The default is 100. Why is this set to 105?"

 

And he got answer:

"I normally set my PCI @ 100 or 101 just to make sure the motherboard doesn't change it itself. Not an overly important setting unless do some heavy graphics overclocking. Can ignore and leave at 101/105"

 

What do you guys think? How would the MB suddenlty change itself?

Edited December 6, 2010 by Brutality

[Brutality]

How do I delete a post? had multiple pages open and clicked add reply on the wrong one lol

Edited December 6, 2010 by Brutality

[dihartnell]

 

What do you guys think? How would the MB suddenlty change itself?

 

I don't think it would. I've only ever kept mine at 100.

[Brutality]

I don't think it would. I've only ever kept mine at 100.

 

Okay I'll leave mine at 100 too then

 

Now this gave me a lot of disbelief in the cpu OC "guide" (not really a guide, just some bios settings) at http://www.computerlounge.co.nz/forum/Default.aspx?g=posts&t=50

 

What changes in the bios settings do you suggest (obivously I will start with a low frequency and slowly build up, not gonna just put in the settings and hope o.O)

 

As a matter of fact I miss a real cpu overclocking guide, like the one I found for overclocking my gpu (http://tech.icrontic.com/articles/overclocking-the-radeon-hd-5850/), which gave me a better understanding of GPU overclocking (I know it's for the 5850, but I guess the procedure is the same for the 5870, just not same frequencies)

 

It would also be nice to know how far you should stay away from an instable clock (like with the gpu OC guide, 10 MHz - now I know it didn't state that explicitly, but more like "when you reach instability, reduce the frequency by 10 and bla bla bla"), it would be nice to know something similar with the cpu

 

Cheers Brutality