BSOD while gaming, please assist!

bwidlus

Honorable
Aug 1, 2013
8
0
10,510
Hey guys,
I started playing Assassin's Creed 3 a few days ago, and every time I play my computer ends up getting a blue screen (30 or 45 minutes in... I'm not exactly sure).

Here are the reports from Blue Screen View:

==================================================
Dump File : 080113-13930-01.dmp
Crash Time : 8/1/2013 12:15:45 PM
Bug Check String :
Bug Check Code : 0x00000124
Parameter 1 : 00000000`00000000
Parameter 2 : fffffa80`0a344028
Parameter 3 : 00000000`be000000
Parameter 4 : 00000000`00800400
Caused By Driver : hal.dll
Caused By Address : hal.dll+12a3b
File Description :
Product Name :
Company :
File Version :
Processor : x64
Crash Address : ntoskrnl.exe+75c00
Stack Address 1 :
Stack Address 2 :
Stack Address 3 :
Computer Name :
Full Path : C:\Windows\Minidump\080113-13930-01.dmp
Processors Count : 8
Major Version : 15
Minor Version : 7601
Dump File Size : 293,536
==================================================

==================================================
Dump File : 073113-14102-01.dmp
Crash Time : 7/31/2013 10:22:11 PM
Bug Check String :
Bug Check Code : 0x00000124
Parameter 1 : 00000000`00000000
Parameter 2 : fffffa80`0a3f6028
Parameter 3 : 00000000`be000000
Parameter 4 : 00000000`00800400
Caused By Driver : hal.dll
Caused By Address : hal.dll+12a3b
File Description :
Product Name :
Company :
File Version :
Processor : x64
Crash Address : ntoskrnl.exe+75c00
Stack Address 1 :
Stack Address 2 :
Stack Address 3 :
Computer Name :
Full Path : C:\Windows\Minidump\073113-14102-01.dmp
Processors Count : 8
Major Version : 15
Minor Version : 7601
Dump File Size : 293,536
==================================================

==================================================
Dump File : 073013-14102-01.dmp
Crash Time : 7/30/2013 3:30:09 PM
Bug Check String :
Bug Check Code : 0x00000124
Parameter 1 : 00000000`00000000
Parameter 2 : fffffa80`0a3f6028
Parameter 3 : 00000000`be000000
Parameter 4 : 00000000`00800400
Caused By Driver : hal.dll
Caused By Address : hal.dll+12a3b
File Description :
Product Name :
Company :
File Version :
Processor : x64
Crash Address : ntoskrnl.exe+75c00
Stack Address 1 :
Stack Address 2 :
Stack Address 3 :
Computer Name :
Full Path : C:\Windows\Minidump\073013-14102-01.dmp
Processors Count : 8
Major Version : 15
Minor Version : 7601
Dump File Size : 293,536
==================================================

==================================================
Dump File : 072113-14274-01.dmp
Crash Time : 7/21/2013 11:05:43 PM
Bug Check String :
Bug Check Code : 0x00000124
Parameter 1 : 00000000`00000000
Parameter 2 : fffffa80`0a3ee028
Parameter 3 : 00000000`be000000
Parameter 4 : 00000000`00800400
Caused By Driver : hal.dll
Caused By Address : hal.dll+12a3b
File Description :
Product Name :
Company :
File Version :
Processor : x64
Crash Address : ntoskrnl.exe+75c00
Stack Address 1 :
Stack Address 2 :
Stack Address 3 :
Computer Name :
Full Path : C:\Windows\Minidump\072113-14274-01.dmp
Processors Count : 8
Major Version : 15
Minor Version : 7601
Dump File Size : 293,496
==================================================

Here's some of my system information:
Nvidia GTA 580 Graphics Card
Intel Core i7-2600k @ 3.40GHz (8 CPUs)
Two 4096 DDR3 modules (for a total of 8192) RAM from G Skill Intl
ASUSTeK P8P67 Motherboard

After the first time, I updated my motherboard's BIOS, and my graphics card's driver.

(Nothing is overclocked, as far as I know.)

I have no idea what to do.
I just ran a memory test, that didn't find any problems.
Please help!

 
Solution


AISuite II is fine because it reads the values from the DSPs on the motherboard, but if you installed the one that came with the disk it's probably quite out of date. Uninstall it and download the latest version from Asus' website.
have a read of this. I have not gotten to the GPU stability tests yet but this should help you narrow it down to what is causing the issue.

Guide to stability testing for Windows OS

First: you need programs to to stability test and these you should have before building a new system. I like to keep them on a flash drive so once the Operating system is installed I can install them without having to down load them.
Second: all of these steps will work if you are overclocking your system. When overclocking remember to stability test after each change in the settings.
Third: Stability testing is a necessary evil of building and overclocking a PC and will take hours to finish. Plan to have something else to do in the same room to keep an eye on temps and programs as they run. Most programs do not need constant supervision but there are some that produce a great amount of heat and without some supervision could hurt your system.

Program list for stability testing
Intel Burn test - to test CPU and Ram.
Prime 95 - To test CPU and Ram
Memtest86+ - To test Just Ram, Should be run every time of initial instal of OS and change of FSB/HT/Base Clock
Furmark - to test GPU
OCCT - To test GPU and hole system as a hole.

Program list for monitoring
Hardware Monitorboard. - Every motherboard wil read different sensors sor you will have to determine what sencore reading apply to your board.
Speed Fan - Every motherboard wil read different sensors sor you will have to determine what sencore reading apply to your board.
CPU-Z
GPU-Z

Use these programs at your own risk. I will not be held liable for you not using the temp monitoring programs properly. Some of these test can hurt your components if not monitored or used correctly.

Memory Stability Testing

After a fresh build I like to start the machine and set the bios to the correct settings and restart. Next I run the program MEMTEST86+. This program gets run from the PC's boot menu and runs from a disk. No Operating System is required in order to run this program. This is the reason I like to run it right off the start. This is a program that can be left alone to run.

MEMTEST86+ Operation and Use
Once Memtest86+ is run from the drive boot menu from the PC, the program will automaticly scan the systems hardware and start testing.

What does Memtest86+ do?

The program tests the system memory (RAM) to make sure there are no errors. If the memory errors then it can cause corruption of data and potentially crash your newly built/Overclocked system. Errors can occur for a few reason. Those being, but not limited to: Wrong Bios Settings, Bad Stick/s of Ram, 2 or more Incompatible Sticks Of Ram, Ram Incompatible With Motherboard.

How do I know if the Ram Errors?

When the program is running it displays information at the top half of the screen. If the program encounters an error, the program will use RED writing in the bottom half of the screen informing you there was an error encountered. The program uses a blue background so the red writing really sticks out. There is really no way to miss it.
If there is no errors the bottom of the screen will be clear of Red writing.

How long should I run the program?

You should run the program for a minimum of 12 Passes. In the information on the top half of the screen, it will tell you how many passes it has completed. The time it takes to run a pass depends on the amount of Ram installed. The more ram installed the longer it will take to complete a pass.

Where do I look to see how many passes the program has gone through?

Towards the middle of the page you will see this

WallTime_ Cached_ RsvdMem_ MemMap_ Cache_ ECC_ Test_ Pass_ Errors_ EccErrors
__00:00 __3565 _____192KB __ +65+9_ 5347 _ 0 _ 546 _ HERE___0_____0___

Where it says Pass and right under neath it where it says HERE are where you will find how many passes have been completed. The program will replace Here with a number and that number will be how many passes are complete. The numbers I have entered will be different, I just used Random numbers to give a visual representation so you have an idea what your looking for and at.

If you have not had any errors and have not already installed the Operating System (OS) do so along with fully updating it. Make sure to have your monitoring programs installed at this point as the next programs will cause heat issues. The rest of these programs require the Windows OS to be installed to run.


CPU Stability Testing

Next I run Intel Burn Test. (IBT)
Intel Burn Test Operation and Use

This program will generate the most heat from the CPU out of them all. Be sure to watch the CPU temps during the testing. The temps will get higher as the program runs then will level off. It is normal for the CPU to cool off just before the program produces a result from its pass. The heat will raise again right after the test starts a new pass.

Because of the heat generated by this program it is a good indicator of what your Max temps will be. My personal PC's and gaming PC's have never hit the high temp that was generated by this program. This is why I feel it is a good indication of what your Max temps will be.

This program tests two components on your PC. The first is the CPU, this is its main function. The second is Ram. If the Ram errors, the system will fail the test but you will not know which one it is. This is why I run Memtest86+ first. This eliminates faulty Ram with its settings from being the issue and leaves us with CPU being tested.

Once you start this program it is best to just let the PC sit without being used during testing.

What does IBT do?

Intel Burn Test uses a linpack designed by Intel to test there CPU's in the quality control process. This Does not mean that it only works on Intel CPU's. It Will work on AMD CPU's just as well. It uses your systems Ram and CPU to do mathematical equations. If the result is wrong then the system is unstable and the test will stop. If the system fails it can be for many reasons. They can be, But not limited to: Faulty CPU or Settings, Faulty Ram or Settings, Faulty Motherboard or Settings. I realize these are a broad description of issues but there are so many different bios settings that can cause a fail that for me to list them would be a major undertaking.

Why can the motherboard cause a fail when it's a CPU test?

This is simply because the motherboard connects everything together. If for some reason a part of the motherboard is unstable it causes the CPU to not function properly.

How do I know if the system fails the test?

The test will stop, play a noise, and put a popup box on your screen that says that the system has been found to be unstable. Alternatively you can look in the results column. If the results are not the same then the system is unstable. The Speed and GFLOPS will vary, this is normal and nothing to be concerned about.

What should I set the settings to?

32bit systems: you will not be able to use more than 2 GB of Ram. The program will not start if you set it to use more than 2 GB. For the stress level set to MAX if you have 2 GB or less ram installed. If you have more than 2 GB ram installed set it to custom and enter in 2000 MB. Yes I'm aware that 2 GB = 2048 MB But the program wont start with a 32 bit OS and much more than this.

64bit systems: Set the Stress Level to maximum

32 and 64 bit systems: Set Times to run to 40. The instructions say for the max to set at 20 but I have had systems fail at pass 25 and later.

32 and 64 bit systems: set Threads to auto. The program will use all threads on its own.

Now just click start and run the program keeping an eye on the temps. This program needs to be watched!

Prime 95 Operation and Use

Prime95 stress tests two different parts of your system, the Ram and the CPU. It does this by trying to find the highest prime number. This program will generate grate amounts of heat from your CPU as well so you will need to keep an eye on the temps while using this program.
This program tests two components on your PC. The first is the CPU, this is its main function. The second is Ram. If the Ram errors, the system will fail the test but you will not know which one it is. This is why I run Memtest86+ first. This eliminates the Ram and its settings from being the issue and leaves us with CPU being tested.

How do I know if the system fails the test?

If the system is found to be unstable or errors are found, the program will do one of three things. Either stop the core that had the error, Stop the hole program or the system will crash and restart. If either the core or hole program stops the green boxes next to each core will turn red. The icon in the system tray will turn red if you have the program minimized.
It is better to let this program run with out using your PC.

How do I set up the program?

When you first start the program it will ask you what you want to do. You will want to select "Just Stress Testing" and click OK. Make sure that all cores have a worker assigned to them. Meaning if you have a 6 core CPU that there are 6 workers working.

How long should I run the program?

You should run this program for 12 to 24 Hours before you can say it is stable. This is NOT a program where you can just let it run and never check on it. Because the CPU temp will raise you need to make sure you will not over heat during this test. Once you are sure it wont over heat you can walk away for a short time but being sure to look once in a while to see the progress and temps.

GPU Stability Testing
 


The error is being caused by hal.dll which is the Windows hardware abstraction layer. Specification, the 0x124 stop code is a hardware complaint.

Since it's caused by a piece of hardware complaining it's going to be very hard to diagnose. Fortunately the parameters appear to be the same, which indicates that it may be the same fault repeated over and over again.

The first thing that I would do is make sure that everything is adequately cooled. I would expect artifacts to appear before a crash if the GPU was overheating, but check it anyway.

The second thing that I would do is check and see if it's deterministic with respect to Assassin's Creed 3, or affects the whole system. Driver bugs can cause this kind of behaviour, but the crash will usually be in the driver rather than in the HAL; the driver could however cause the card to misbehave in such a way as to cause the error. Try running Furmark for a bit, or even another game (a demanding one) and see if that crases as well.

EDIT: if it was a result of general instability in your memory or CPU the errors would not be so well correlated. Don't bother with Memtest86, you're just wasting your time.
 

bwidlus

Honorable
Aug 1, 2013
8
0
10,510
I just looked at the BlueScreenView app again and found that I had this exact same BSOD 3 months ago (before I bought Assassins' Creed 3), so I don't think it's specific to that game.

When I play, I do monitor the heat levels of my GPU via EVGA Precision X.
I believe the card reaches just about 70 degrees celsius while I'm playing.
 


70 degrees should be a safe temperature, how hot is your CPU getting? Make sure to remove all over clocks and pretty much follow what the previous poster listed.
 


eek!

Uninstall any third party monitoring tools such as EVGA Precision X, MSI Afterburner, etc... these can cause the exact problems that you describe. Some of them come with kernel modules that clash with the video drivers, or even with each other.

MSI Afterburner can be put into user-mode which is a bit more stable, but I'd still remove it. There's no need to run those unless you're overclocking, and if you're overclocking and experiencing BSoDs that's the first thing that you should stop.
 

bwidlus

Honorable
Aug 1, 2013
8
0
10,510
So I uninstalled EVGA Precision X and tried again.

My CPU got to 49 degrees celsius.

After 30 minutes of playing, something different happened.
Instead of getting a blue screen, like I have the last 3 times, AC3 simply crashed.
The program froze and the sound started to stutter, but instead of staying on that screen and stuttering until I hard-restarted it, or going to blue screen, the stuttering cut off and the program crashed.

Does this mean anything/help at all?

Should I step back and run one of those tests you guys recommended? Furmark and/or Prime 95?
 


That's an improvement. Did you restart your PC after uninstalling all that junk? Make sure that you also remove any other monitoring tools such as coretemp, speedfan, etc... all of them have to go.
 

bwidlus

Honorable
Aug 1, 2013
8
0
10,510
I haven't restarted yet.
I don't have any other monitoring tools, that I know of, other than the ones included in the AI Suite II that came with my motherboard.
Those include: Turbo Evo, DIGI + VRM, EPU, Fan Xpert, Probe II, Sensor Recorder, and BT GO!
I don't know if I can uninstall any of those individually, and AI Suite II is what I use[d] to update the BIOS on my motherboard (it came in a disc with the board).
Do I need to uninstall that?
 


AISuite II is fine because it reads the values from the DSPs on the motherboard, but if you installed the one that came with the disk it's probably quite out of date. Uninstall it and download the latest version from Asus' website.
 
Solution
If you follow my stability guide, you will be able to find out if it is a hardware issue or a software issue. There was a reason I posted it for you. Some programs do not like Monitoring programs. Most monitoring programs have an option to turn off the monitoring for specific programs and do not need to be uninstalled. EVGA Precision is one of those along with MSI After burner, HIS Iturbo and riva tuner. These are all made in conjunction with Riva Tuner and all work the same way with just a different skin.

If you run all the tests and no crashes then run a game and crash it is a software conflict issue.
 

TRENDING THREADS