Win7 64bit | DumpFile | 16GB Ram

[the7thcolumn]

Ok I have been running fine until tonight it crashed during BF3 I think it might be a heat issue on the CPU.

I have attached a photo of me playing BF3 with temps (see the high, it had dopped after closing the game) My high was actually 71c. I also attached photos from the UEFI. I cant seem to find the auto shutoff temp in the UEFI.
View attachment 186972

View attachment 186973

 

[pominator]

If your problem is not sorted yet, I have encounter BSOD caused by Avast not that long ago on a customer PC. Uninstalling the software resolved the problem. It may worth a shot.

Also, a quick test to sort out temp issue, you may open your case and point inside a big fan that will cool down the whole PC component. If the BSOD stop with the fan, verify/change your current CPU/GPU cooling solution or maybe add a small fan on your chipset.

Do not want to hijack others help, just to add my 2 cents. Good luck.

Edit : To modify auto shutoff temp for you CPU, in the Monitor tab in UEFI, set CPU Fan Profile to Manual, then set CPU Upper Temperature higher than 70. I would not recommand setting the limit much higher though.

I would compare your CPU temp with your original 2x4gb RAM and if it's still pretty high and close to 70c, verify your CPU cooling assembly (clamp, thermal paste, fan).

 

[Dave76]

Those CPU core temps are high, you should check the CPU cooler and replace the thermal paste.

You can post those pictures the same way you posted the CPUz shots.

The last picture is where you change the VCCIO voltage.

View attachment 187032
What did you set your CPU offset voltage to?

From you motherboard manual, Asus P8P67 Pro

View attachment 187033

Your current RAM voltage is 1.662v on Auto, this is too high a voltage.

Your CPUZ SPD tab says all JEDEC and XMP voltages should be 1.50v.

Let's lower the DRAM voltage, on the EFI page above change the 'DRAM Voltage' to 'Manual' and set it at 1.550v.
The voltage is due to your 16GB RAM, with 4GB RAM I would recommend 1.50v.

Then change the VCCIO voltage to 'Manual' and set it at 1.120v.

Then check you system for stability, keep a constant watch on the CPU core temps, don't let it go over 75°C.

Let us know if there are any stability issues.


We don't see many issues with Avast, most BSoD analysts here recommend it, I use it myself.
Anything is possible, though.
Lets look at the CPU temps and the RAM settings before you look at software, for now.

 

[the7thcolumn]

If your problem is not sorted yet, I have encounter BSOD caused by Avast not that long ago on a customer PC. Uninstalling the software resolved the problem. It may worth a shot.

Thanks for the info but I have only gotten one BSOD, it makes me think its the heat not the memory. It just normally shuts off.

 

[the7thcolumn]

Those CPU core temps are high, you should check the CPU cooler and replace the thermal paste.

I feel like such a moron, I think I know what it is. I have never pasted a CPU before and I was having trouble getting the heat sink to lock down (it is locked down 100% after multiple attempts a long time ago) but took a few tries. I also swapped mobos a few months ago. So it has been on and off on and off. I never repasted as to me it looked ok. Apparently not. I am going to buy more paste today to do it properly.

 

[Dave76]

You should replace the thermal paste every time you take the CPU cooler off.

You really should change the voltages on the DRAM and VCCIO, the high DRAM voltage can damage your CPU.

 

[pominator]

Those CPU core temps are high, you should check the CPU cooler and replace the thermal paste.

I feel like such a moron, I think I know what it is. I have never pasted a CPU before and I was having trouble getting the heat sink to lock down (it is locked down 100% after multiple attempts a long time ago) but took a few tries. I also swapped mobos a few months ago. So it has been on and off on and off. I never repasted as to me it looked ok. Apparently not. I am going to buy more paste today to do it properly.

Any news ?

 

[the7thcolumn]

Sorry I have been busy, had a accident this week with my car so been a little tied up with that. And no way to go out and get paste since I live in the boonies. Ill consider this thread solved but will reply in about a week.

Im not abandoning it, I just don't have the time this week to fix it so Im leaving my PC off till I can re-paste.

 

[Dave76]

Thanks for the update.

Sorry to hear about the accident, hope no one was hurt.

Let us know the results after the thermal paste application, when you have time.

 

[the7thcolumn]

Ok got the thermal paste! Did it twice with cleaning until I found I had the correct amount on it. Running Prime95 for about 2-4 mins I still creep up to 80c with a max of 81c ambient case temp does not go above 40c. The heatsink is seated properly,

I also want to note that pressing down on the heatsink during the test did not make a difference in temp. Been running for about 8 mins, 100% CPU load and at 84c, it idles at 36c.

 

[pominator]

Pretty weird. You only need a little bit of thermal paste. I usually spread it with a credit card to only leave a thin layer. Other put a little amount in the center of the cpu and let the heatsink spread it evenly.

You need to clean the heatsink and cpu completely with alcool/clean rag (and credit card if old thermal paste is hard) before applying the new one.

The only other things I see is bad seating of the cooler and defect CPU. You can take a look at this video to see how to seat CPU and heatsink properly with core i7 (advance to 2 min) :

installing intel core i7 processor to gigabyte motherboard - YouTube

If everything is ok, you may have a bad cpu, I would return it with the case if you bought it locally and tell them about the problem for them to look by themself. If you bought it online, they will probably agree to proceed to an exange when you'll tell them about the problem.

Maybe others have good tips too, good luck !

 

[the7thcolumn]

Ok installed a new heatsink! Temps are within norms now.

Idle 24c | Load 58c max during a burn test with Prime95. Feeling good I got o play BF3. Play for about 20 mins and the PC reboots again.

So its not the temp/heatsink of the CPU. Any other ideas? I could run in singlefire and see if that makes a diff.

 

[pominator]

You may test your graphic card cooling with FurMark.

 

[pominator]

A bit more information, I would start over and load optimized default in the bios, then test in order : Ram, CPU, GPU, Hard disk, and eventually power supply if you have a tool or a spare power supply.

To test the RAM, use memtest86+. Boot using ultimate boot cd or a bootable ISO/CD from memtest86+ website. Ensure you have latest version. Let the test run for 5 pass, at least. More is better, 5 is usually enough though.
-> If their is error, probably bad memory or bad auto setting in bios.

To test CPU, use Prime95 latest version in Windows. Use the In-Place Large FFTs test, run it for 8 hours and it will be enough. Monitor temp with RealTemp to ensure it won't go higher than ~70c.
-> If their is error, probably bad CPU or still bad cooling.

To test the GPU, as you have a crossfire system, I would finally start by removing one graphic card and to test battlefield 3 with one card only, lower the setting to a playable framerate and play a lot, ~8 hours. If the system is stable that way, remove the current card and install the other one alone, and do the same test. If the system isn't working with 1 card in particular, that card is probably faulty. If the system is stable with both card when used alone in the system, it's probably a driver issue. Try to update the driver. If it dosn't solve the problem, if you can, do a complete format and reinstall the latest driver and try again with crossfire setup. It may also be a problem with your motherboard bios who aren't supporting crossfire properly and a bios update may resolve the problem, if one is available. Before upgrading bios, assuming that your system is stable with one graphic card only, remove one graphic card to run a single GPU setup then do the update according to the manufacturer instruction. Hopefully using a flash drive and the EzFlash bios utility.

To test hard disk, boot using the latest ultimate boot cd disk and do a complete hitachi drive fitness test. One test is enough. It probably won't have any error but the test is easy and quick to do. If you have multiple drive do the test on every drive.

To test the power supply, the best test is to use another know to work, and powerfull enough (600 watts is more than enough) power supply. If you don't have one on hand, you can still use a power supply tester, something like this :

Antec.com - Product: Digital PC Power Tester

But it will only detect basic problem.

If you can do all those test and don't find any problem, disconnect other device if you can in the case, like DVD drive, unnecessary hard disk, extension card, etc. Also ensure that every cable are well connected. Another test could be to install a big fan pointing toward the inside of the case, if the problem stop, install additional RAM cooling and chipset cooling. I'm pretty sure you'll find the problem before this step though.

Also, I have found another thread with a problem similar to yours. It may be your thread though. If not, keep an eye on it you may have good solution there too.

Battlefield 3 Crash Issues. Need help determining if its my CPU or GPU - ATI - Graphic-Displays

Good luck with this.

 

[the7thcolumn]

You may test your graphic card cooling with FurMark.

tested with Furmark with default (unforced) fanspeed on the GPU. normally when I play games I pump both cards up to 50% for lower and 55% for upper to keep them around 60c (I know GPUs can get hotter but I want them in range even if I kill my fans).

No reboot when running Furmark in crossfire.

Maybe I should run FurMark and Prime95 at the same time stress the CPU and the GPU out?

Im thinking PSU you guys?

 

[the7thcolumn]

A bit more information, I would start over and load optimized default in the bios, then test in order : Ram, CPU, GPU, Hard disk, and eventually power supply if you have a tool or a spare power supply.

Ill load it back tonight.

To test the RAM, use memtest86+. Boot using ultimate boot cd or a bootable ISO/CD from memtest86+ website. Ensure you have latest version. Let the test run for 5 pass, at least. More is better, 5 is usually enough though.
-> If their is error, probably bad memory or bad auto setting in bios.

Ill run the newest version, I think I have it but Ill check their site and report back

To test CPU, use Prime95 latest version in Windows. Use the In-Place Large FFTs test, run it for 8 hours and it will be enough. Monitor temp with RealTemp to ensure it won't go higher than ~70c.
-> If their is error, probably bad CPU or still bad cooling.

Did not feel comfortable running prime95 that long until I got the heatsink Ill try running it that long and see what happens

To test the GPU, as you have a crossfire system, I would finally start by removing one graphic card and to test battlefield 3 with one card only, lower the setting to a playable framerate and play a lot, ~8 hours. If the system is stable that way, remove the current card and install the other one alone, and do the same test. If the system isn't working with 1 card in particular, that card is probably faulty. If the system is stable with both card when used alone in the system, it's probably a driver issue. Try to update the driver. If it dosn't solve the problem, if you can, do a complete format and reinstall the latest driver and try again with crossfire setup. It may also be a problem with your motherboard bios who aren't supporting crossfire properly and a bios update may resolve the problem, if one is available. Before upgrading bios, assuming that your system is stable with one graphic card only, remove one graphic card to run a single GPU setup then do the update according to the manufacturer instruction. Hopefully using a flash drive and the EzFlash bios utility.

Both cards have a double lifetime warranty, so if this is the case ill RMA it. I feel I may be stressing the cards by running eyefinity all the time but the FPS seems decent ~55fps so I dont think Im killing them under load.

To test hard disk, boot using the latest ultimate boot cd disk and do a complete hitachi drive fitness test. One test is enough. It probably won't have any error but the test is easy and quick to do. If you have multiple drive do the test on every drive.

I have a shitty WD Green drive until the price of SSD's drop. I wondered if the rates were killing it, but I doubt it.

To test the power supply, the best test is to use another know to work, and powerfull enough (600 watts is more than enough) power supply. If you don't have one on hand, you can still use a power supply tester, something like this :

Antec.com - Product: Digital PC Power Tester

I have a tester I use at work. I know how they work, someone has suggested that since my PSU is dual rails maybe I was overloading one rail. But I dont think I am.

But it will only detect basic problem.

Yeah tell me about it... I use it at work. basic as in boolean on or off basic.

If you can do all those test and don't find any problem, disconnect other device if you can in the case, like DVD drive, unnecessary hard disk, extension card, etc. Also ensure that every cable are well connected. Another test could be to install a big fan pointing toward the inside of the case, if the problem stop, install additional RAM cooling and chipset cooling. I'm pretty sure you'll find the problem before this step though.

No DVD, I mount over the network DVD drive.
No extra drives.
However I have a shit ton of frills. (Techno LEDancer, NZXT Sentury, 6 Fans, Cold Cathode, Killer Nic. All could be factors.)

Whats a good temp to have ram at?

Also, I have found another thread with a problem similar to yours. It may be your thread though. If not, keep an eye on it you may have good solution there too.

Ill check it out thanks, Im really baffled by this I thought it was temps for sure.

Battlefield 3 Crash Issues. Need help determining if its my CPU or GPU - ATI - Graphic-Displays

Good luck with this.

 

[the7thcolumn]

Ok; Im back! Unfortunately I am still having the same issue.

I have been running some new tests recently here are the results:

No longer testing with BF3, running Prime95 for under 10 mins gives the same result on a fresh boot.
I have stripped the PC down to narrow down anything else that may be causing it running single GPU's at a time; alternating them made no difference.
PSU Checks out all clear with tester.
New heatsink added temps under 100% load for 1Hr 40Mins are 58ish
I have been testing the ram in sets (old and new) even though they are the same model and series.

Tested stock settings (DRAM) Old ram would cause reboot at 10 min mark. New ram ran forever.
Tested with 1.55V Old ram ran for 1hr 40min then I stopped the test (didnt bother testing new ram since it passed the previous test)
Put both ram in Prime95'ed it and after about 15 mins rebooted.​

I mounted/pasted the CPU Properly.

So now Im lost from my results I clearly see its the old (first set) I bought as when its in the system it fails; however when I lower the voltage on it they wont crash, but when adding the new set to it I get a crash. I ran a Memtest86+ on the old ram over 18 hours and didnt get a single error. Windows works fine, its only when the PC is under full load; counterstrike and shit is fine as weel just BF3 or Prime95.

 

[writhziden]

Download and install CPU-Z and Upload screenshots of the CPU, Mainboard, Memory, and SPD tabs. In the SPD tab, upload an image of each slot.

Also, go into your BIOS and post all Voltages (CPU, RAM, NB, IMC, etc.) and all RAM settings (timings, frequency, etc.)

Thanks to Dave76 for help learning RAM Stability