Solved BSOD playing Hydrophobia and Portal 2, error 0x00000124

Hi, I'm looking for some help concerning these Blue Screens. I've recently finished building my pc and i'm very happy how it turned out. To celebrate and test my new i7 i figured i'd play so modern games to test the power. I launched Hydrophobia played for around 6 seconds and got a blue screen. After a reboot i tried again and again to the same result. I tried other less demanding games E.g. Minecraft and Killing Floor. I had no problem what so ever. Then downloaded Portal 2, this time i lasted about 2 minutes before crashing back to the blue screen. Being a fan of Crysis i also tried this game. I played for around about 10 minutes with no failure in sight. If there is one thing i have noticed the newer the game and the more demanding it is i seem to crash more. However when im on desktop i can leave it for hours and no crash at all. I considered that it would be the fact i have two HDD Steam on my second drive i thought that could be the problem but this was not the case. I'm at a loose end and i could really use some help. Thanks in advance.

inteppm.sys is crashing.. microsoft intel processor driver. You should probably check your temperatures with software like hwmonitor and download the latest chipset(inf) driver for your mobo and reinstall them.

Also, google intel processor diagnostic utility and download and run from intel website to check processor.

If you are overclocking you should throttle back to stock speeds until you can determine what issue is.

I saw two older crashdumps, one dated Feb 6 and the other Jan 7, that look unrelated. I take it these are unrelated to your current predicament?

Anyways, the rest of them are unanimous: WHEA errors caused by an internal CPU timer failure. There was one 0x101 bugcheck but one can't debug 0x101 bugchecks without at least a kernel dump. Still, 0x101 can and often do occur from CPU failure, especially when it involves timing errors like what you're evidently experiencing.

I have seen a few times motherboard software cause these problems. Motherboard software is extremely buggy and is very gimmicky, none of it is recommended, even temp monitors and OCing tools. If you need to overclock, it's best to do it straight from the BIOS/UEFI. Speaking of OCing, since we're dealing with a timing issue here, it's best that if you have OCed anything to reset everything to factory defaults, as event3horizon clearly pointed out. We're talking "safe settings" for your BIOS/UEFI, not some predefined performance settings or anything like that.

If things don't stabilize after resetting things, then go and uninstall any and all motherboard software, including USB "drivers". Only leave drivers for motherboard components (again, no USB drivers, as it's software posing as drivers) like LAN, Chipset, etc.

If that does not resolve your problem, then I'm afraid we're dealing with hardware failure here. Primary suspect is CPU, followed by RAM, PSU and Motherboard, in order of possibility. I have seen GPU failure also cause this, but it's very rare. If you wish, you can run some hardware tests as followed:


RAM: Memtest86+ - 7+ passes
CPU: Prime95 - Torture Test; Large FFTs; overnight (9+ hours)
GPU: MemtestCL - Run twice (if any of the tests work on your GPU; ATI cards will need to install the ATI APP SDK as it requires OpenCL)

All of these (excluding MemtestCL) are included in the UBCD if you prefer a Live CD environment (which is the best environment to test hardware on). Note that Prime95 currently does not work on the UBCD. Also, please provide us temps/voltages using HWInfo with Sensors only option checked. Log two 30-minute instances: one for idle, and one for high load.

@ Event3Horizon
I included the screenshot from hw monitor and i downloaded the inf files from Intel. I also used the diagnostic tool which said my CPU passed in all catergories. After doing all this i started up hydrophobia again, this time it lasted longer than normal but crashed again. Much faster this time however, it went straight to blue screen. The code was 0000007f, which is different to the normal 124. Oh, i forgot to mention about my CPU earlier. I have done no overclocking of any kind. The CPU is default at 3.5ghz and has a automatic turbo boost to 3.9ghz. Thanks.

thanks for posting the info. The CPU temps seem great. Obviously Vir Gnarus knows what he's talking about so follow his steps. You need to do some stress testing. Eventually maybe a slight increase in CPU voltage may be attempted to see if it fixes problem.

One thing that is weird is your +12V are very low??

Yeah, I saw that too, the rail voltage readings are just wacky. I think the software may not be reading the sensor data correctly. Try other monitor software like HwInfo (make sure to select Sensors only at startup for HWInfo) to see if they output similar values. Also go on your motherboard and check those voltage readings too if at all possible. Basically you want to look just at the +3.3v, +5v, and +12v outputs. They should be nearly identical to their names, with only minor decimal skews.

If even your motherboard is showing up problems, or all monitoring software has an issue reading them, then your motherboard may be bugged. Make sure to update BIOS and chipset drivers to ensure this is not the case.

I checked out HWinfo and the it displayed correctly. 3.3v 12v etc. So i assume the other program bugged out for me. I will continue with Vir Gnarus's suggestions and i hope i get some results from that. I also updated the bios and unistalled the none essential motherboard software. The error code has changed back to 124 now unfortunately. I hope doing these test will define the problem more. Thanks.

Just remember the tests aren't 100% definitive. For example, Prime95, while a solid CPU stress test, does not stress all CPU caches and utilizes a very small portion of the CPU instruction set. Blend testing on it will help in that it will add RAM-to-CPU talking into the mix, but it's still not a definite answer to discovering a hardware problem. I've also found that if all tests show up nil, then Motherboard and PSU are often the cause, as they commonly do not manifest their faults in specific hardware tests.

Hi again, so far i've taken the RAM test which passed, then i tried the GPU which passed both found 0 errors. Im trying to work out how to use Prime95 but i have no idea what to set it to. I believe you told me to set it to Large FFTs, right? When i do this though it says the completion date with me the 14th of June. Any advice on how to use this program would be a great help. Thanks. I'm also going to include all my latest blue screen reports. Which reminds me you asked earlier if the error reports from Feburary and another date were unrelated, which yes they are. That must have been from when my HDD's were in my old set up. I have one more question as well. After changing all the hardware apart from my HDD's in my pc, when i booted the pc it loaded Windows 7 and all my documents and files were still in place, i was told this was unusual could this effect the problem? Thanks.

Did you run Memtest86+ for over 7 passes? It runs indefinitely until you stop it. You need to have it run several passes in order to verify stability.

As for Prime95, when you first start it up it'll prompt you asking if you wanna join the Prime thing or you just wanna do Torture Test. Select Torture Test. Then it'll give you 4 basic options, Small FFTs, Large FFTs, Blend, and Custom. Start with Large FFTs first, since that will fill up your CPU L2 cache and just run almost exclusively on the CPU. Run for at least 9 hours. Prime95 will - like Memtest86+ - run indefinitely until you stop it. After 9 hours you should see if it's prompted any errors or the system has crashed while it was running. If you wish, after that run, you can do an additional 9+ hour run on Blend, as that'll mix in a lot of RAM to the equation as well.

As for Windows 7 being retained, you were just lucky. Sometimes hardware changes can be drastic enough to where Windows coughs on it. This is especially true of older versions of Windows like XP. Newer ones like 7 are far more likely to keep stable. I've personally transferred HDs from old computers to completely new ones and had no problems aside from needing to install the new drivers.

Right, reporting back again. I did a total of 9 passes on RAM test showed no errors at all. I ran Prime95 on Large FFTS for the best part of nine hours. The heat never went about 62 degrees and the fan cooled it great. Prime 95 also showed no errors what so ever. I searched on the Gigabyte website again, this time i found some even more up to date drivers, i installed all the relevant ones. Which yielded no change at all. Then in a last resort i did a clean install of windows 7 which also changed absolutely nothing. Got anymore ideas? In regards to it being down to my motherboard or psu, i'm not sure. I can tell you that they are both brand new and out the packet about 4 days ago so i doubt it could be there age. I'm sure the 650w psu is enough to keep up as well? Thanks.

NOTE: My motherboard has two sockets a PCIEX16 and a PCIEX4 normally i have my graphics card in the PCIEX16 but on a whim i decided to switch them around. When i rebooted the PC an launched the game it lasted for about 2 to 3 minutes before blue screening. However when in PCIEX16 its about 20 to 30 seconds. I don't know if this has any relevance but i thought I'd let you know. Thanks.

Ran Prime95 last night for around about 3 hours before crashing on blend mode here is the error log.

Okay, so i ran some more prime95 tests with some different ram its clock speed was 1800mhz where my mobo can only register 1333mhz so it auto underclocked it. Then i ran prime 95 through all tests small large and blend, which all crashed after about 20 minutes showing 0000007f. Then i combined both sets of RAM so at this point i had 16gb. I tried playing Skyrim, which since i wiped my pc is my new test. Which ran for longer than ever at one point it lasted about 15 minutes. At this point the case was open and the GPU was getting very, very hot. The next few crashes it went back to its regular 2 or 3 minutes. But now showed error message 00000101. I'll carry on updating what i find.

The latest crashdump listed showed an NPX_OVERRUN exception that triggered when Prime95 was the running process. This is a numeric coprocessor error, which means during numeric operations (like floating point) by Prime95 something bugged out. Again, another CPU failure. The rest appear to be more internal timer failures.

I'm sorry mate, your CPU is almost certainly bad. You'll want to get her replaced. Again, it could be the mobo or PSU, but I'll actually lean more towards CPU given the consistency of failure and the added NPX error to boot. CPU is far more likely to generate these failures than something initially triggered by the PSU or mobo.

Thanks, Vir Gnarus you've been a great help and i appreciate all the aid you've given me. Luckily i should still be within the month warranty and I've just emailed them to ask for a replacement. I'll let you know my result and I'll mark it as closed when i tell you my findings. Once again thanks, you've been a great help.

No prob, I certainly hope it's the CPU. There's a reason why I call the CPU, Mobo and PSU the "Trio of Trouble": they all are very difficult to diagnose issues with because they're so intertwined, and the Mobo/PSU has no software tests for them. Fortunately in your case it seems well consistent enough to blame the CPU.

Well you were right, they tested my CPU and gave me a new one. Works like a charm, thanks again Vir Gnarus you've been a great help!

Wonderful. Glad to hear everything's a-ok! If you're certain things have cleared up, mark this thread as solved for future reference. Thanks!