Windows 10 released version + Geforce Titan X generating random BSODs in games

[champe]

Hi all,

I have a new build with the following specs:

MB: Asus Rampage V Extreme/USB3.1
CPU: Intel i7-5930k, currently not OC'ed
CPU Cooler: Corsair Hydro Series H100i GTX
RAM: Corsair Vengeance LPX 32GB (4 x 8GB) DDR4 DRAM 2666MHz C16
PSU: EVGA SuperNOVA 1000 P2 80+ PLATINUM, 1000W
GPU: EVGA GeForce GTX TITAN X (EVGA Overclocked version)
SSD (Win 10 boot drive): Intel 750 Series PCIe AIC 1.2TB Internal SSD
HDD: Western Digital Red Pro 4TB NAS
Keyboard: Corsair K95 RGB Mechanical
Mouse: Logitech G502 Proteus Core

I installed Windows 10 Pro on the Intel SSD, installation went smoothly. I updated all of the drivers from the Asus web site, and am
currently running those. Everything runs fine, mostly web browing with Chrome, MS Office apps and Quicken. I have sleep turned
off, and system is stable.

The only problem is when I try to run the two games I've got installed, GuildWars 2 and Skyforge, they both get blue screens after
approximately 1 to 5 minutes from starting them up. This is happens 100% of the time. I've tried many, many things to isolate the problem, but so far, no dice.

Some of the things I've done, that haven't helped:

1. Reseated the Titan X in the 1st PCI slot.
2. Removed the memory modules and installed them in reverse order in the same red slots.
3. Manually checked all connections on the motherboard (and elsewhere) to make sure they're attached well,
4. Inspected front and back of motherboard for possible contact with any loose wires or with other components that might
cause shorts.
5. Ran memory and stress tests on system using Asus PC Diagnotics program.
6. Tried turning down the overclocking on the Titan X using the Asus Precision X program. I don't know if this does anything
because the GPU board may be "hard-wired" to the overclocked speed.

There's one anomaly that I see in the UEFI Bios. It recognizes my RAM (all 32Gb) but it says the clock speed on the RAM
is 2133hz. I don't know if this is a problem, or if the Bios, by default, doesn't recognize anything over 2133 hz, which I thought
I read somewhere.

Anyway, if any of you are using a high-end graphics card on this motherboard with Windows 10, and have had seemingly random
BSODs when running games only, and have any solutions, recommendations, or additional information about this setup, I'd really
appreciate your response.

Since many of you are like me and have spent a small fortune to build a system for gaming primarily, I'm sure you can understand
that this is really frustrating.

Thanks in advance for your help.

[champe]

Hi Hopper64, it was a program that came with my motherboard called ASUS PC Diagnostics. The memory test ran for over an hour and I think it tested at least half of my installed memory, although I'm not sure I was using it right. I want to try Memtest86+, if I can figure out how to get it booted from the memory stick I've got it installed on.

[mdzcpa]

Is that memory on the QVL?

[champe]

Hi mdzcpa, the memory I've got installed is not on the QVL list for the motherboard, but it's sister product was on the list, and they only differ by the latency specs. I thought it would be ok. Maybe I'll have to get another set. At least it wasn't the most expensive component of my system, lol.

[jacks12345]

IRQL_NOT_LESS_OR_EQUAL error is usually caused by the CPU. It sounds like the CPU isn't getting enough voltage or your memory isn't setup properly. It's also possible it's coming from the operating system itself, though unlikely.

By the way, did you do an upgrade or a fresh install of Win 10?

EDIT: I would reset the CMOS and start all over with your BIOS settings. Either set it to auto or manually enter in the settings and see what happens. You also may need to perform a clean install of Windows 10. I've heard nothing good coming from performing an upgrade versus a clean install.

[champe]

@ jacks12345, Thanks for giving this some thought. I did a fresh install of Windows 10, because everything is new in this system. Also, I haven't even tried any overclocking or messed with any voltage settings in the bios.But I did go in and change the RAM scheme from XMP to DC, because I heard that someone had trouble using the XMP scheme.

But I will try what you suggest and clear the CMOS and make sure things are back to motherboard defaults, and test the games again.

[Chino]

With DDR4, the two programs I mentioned earlier are more effective than memtest86+.

[champe]

Well I ran HMI Memtest as Chino recommended, and got a lot of memory errors from it. So for now, I guess I'll be doing more memory testing.

Also, I did the CMOS clear and it didn't appear to affect anything.

Thanks again, all of you for your great suggestions! I'll return after my memory testing to let you know the results.

[Chino]

If you look in my DDR4-3000 guide, there is a section on how to run HCI Memtest effectively. I would suggest you clear your CMOS first to bring everything back to stock defaults. If errors are found at stock clocks, then RMA your memory kit.

[champe]

I removed 2 of my 4 RAM modules, and I've got what appears to be a stable system. When I cleared CMOS and tested this new setup with both HCI Memtest (using Chino's recommended method, I let it run for several passes and got no errors. I also ran Memtest86+ for 4 full cycles and got no errors. Then I ran the RealBench Stress Test on the system for 30 minutes with no problems and played my games for a couple of hours each with no problems.

I know I've still got some work to do to sort out which of the removed modules might be bad, which I intend to do. I also know that I have more testing to do to eliminate other possible causes or contributing causes, like for example, one or more of the RAM slots in my motherboard not working properly, or PSU isn't supplying enough stable voltage to handle all 4 DRAM modules installed at the same time. I'll be working this for some time to try to convince myself that I know what the cause of the problem was.

I think I can close this thread now, and go on with using my system and testing. Thanks so much to everybody who contributed suggestions or comments.

[Chino]

To find the defective module, you need to test them individually. Then RMA the entire kit unfortunately.