I am setting up a test environment and I need to run multiple virtual machiness in Virtual Box and as such I needed a small machine with plenty of memory.
I have bought:
- motherboard: Intel DQ77CP
- CPU: Intel i5 3470
- memory: Corsair Vengeance LP 32GB DDR3 1600MHz CL10 Quad Channel Kit
- disk: Intel 520 SSD of 240 GB
After struggling all day I have found that when using a memory module in the DIMM socket closest to the CPU (marked as DIMM3) and only in this socket the PC is crashing (under both Windows 7 Home Premium and Enterprise Linux 6). It does not happen when memory is installed in the other 3 DIMM sockets.
Also it does happen only in conjuction with video drivers, it does not happen when booting in Windows Safe Mode, but only when and after installing video drivers. As soon as Windows is reaching to the graphical interface the image becomes scrambled and the system reboots automatically.
Is this a known problem?
The http://downloadmirror.intel.com/21417/eng/dq77cp_TechProdSpec02.pdf TPS basically says not to populate the DIMM slot nearest the CPU alone.
If you are really concerned about it, run memtestx86 with all the slots populated and if you find errors on DIMM slot 3, shuffle memory around to see if the errors follow a particular stick of memory, or if it stays with the slot.
You also indicate that you struggled all day, what were you struggling with? What error or symptom were you attempting to abate?
It's the same. If I populate 1,2,3 or all 4 DIMM sockets. The crash is occurring only when DIMM3 is populated. I guess it is a defective motherboard. What is strange is that this issue seems to be related to video drivers.
So let me clarify again. I tried all combinations. 1, 2, 3, 4 DIMMS. If a DIMM is present in the socket numbered in the manual as DIMM 3 (the closest to the CPU) the computer is crashing. I tried rotating the 4 DIMMs too.
Also, Windows or Linux installers do not crash. Only when the OS is using drivers specific for this integrated video board (I do not have a discrete one so I guess it is using the one built-in the CPU, HD 2500) it is crashing.
There is no message. As soon as the video goes into a higher resolution (my monitor is 1200 x 900) and a DIMM is present in socket 3, then the display gets scrambled and the system reboots itself. The system basically keeps rebooting over and over unless I go into Safe Mode which has a lower resolution. The same happens in Linux too, it seems to be related to the resolution and the presence of a DIMM in socket 3. No other DIMM socket seems to create this problem. I am currently running with 3 DIMMs (24 GB of memory) and it is okay. It's very odd...
There should be an option on the F8 menu (press as the system boots) to disable automatic reboot when a BSOD happens. Assuming you have configured your system to create a minidump file.
Either that or you can boot into safe mode and fish out the minidump file and use http://www.nirsoft.net/utils/blue_screen_view.html BlueScreenView to read it. Dumpfiles should be at C:\Windows\Minidump.
Unfortunately I deinstalled Windows as the main goal is to run Linux on this machine and currently I am installing Oracle Linux 6.3 (should be very much alike RedHat). Linux is generating some errors with different call stacks. I will reproduce the error and copy/paste the messages if that is useful. But from what I have seen there is a different call stack (different functions) each time.