I have a NForce 750i SLI FTW motherboard with an Intel Core 2 Quad processor (Q8200 2.33 GHz), and Nvidia PC2-6400 memory (2 - 2gb modules). The memory runs in a dual channel configuration. The video card is a GeForce 9600 GSO. Windows 7 is the OS.
A couple of days ago the computer started shutting down - no blue screen or warnings. One moment I would be reading email or browsing a web site and the next the machine would just be off. I ran memtest + version 4. The test crashed anywhere from test 4 to test 7. I removed one memory module and tested it separately and the test crashed (the computer shut off with no warning). I tested the other memory module and the same thing happened. I changed slots and still the computer simply shuts down with no warning. Then I tried another version of memtest - memtest 3.4. Same result.
I also tested the power supply with the Ultra power supply tester and the PS tested okay. I even replaced the graphics card with a PCI Radeon card, but the system still shuts down with no warning.
By the way, while I could get into Windows initially, now when the system starts up the GUI, it crashes. In other words it reboots. There is no blue screen.
So, I am thinking it is not the memory that is the problem. Any ideas?
You can’t depend on power supply testers to test power supplies properly they are simply a go-no-go indication that the power supply works. The only reliable test that an amateur can do is to substitute the power supply with a known good one.
If after checking that you have good airflow in the computer case and that the processor is not getting to hot, I would get a new power supply as the alternative is a new motherboard and the power supply replacement is the much cheaper choice.
Thanks for the replies. Nothing new hardware wise. The BIOS gives system temps and everything is within a safe range. All fans are working. That is one of the first things I checked.
Replacing the power supply sounds like a reasonable step and having an extra even if the current one checks out is not a bad idea. However, I wonder if that is really the problem. I ran diagnostics (Microscope 2000) for 24 hours, nothing failed and the system did not shut down. However, try to run Windows either from the hard drive or a DVD (trying to repair the OS) and it shuts down just before going into the GUI. Memtest fails after 5 minutes into the test suite. The system is getting stressed and failing. Is it possible that is the power supply or something else?
I would tend to suspect your power supply is the culprit. I had the EXACT same problem when I was using an Ultra X3 1000 Watt power supply. System would just turn off as if someone had just pulled the cord out of the wall. I replaced the power supply with a Corsair and haven't had the issue since then. That was 6 months ago.
Here is some additional info on the system. After reading EXT64's reply, I went into the BIOS and checked the CPU temp against Intel spec (below). The CPU temp on my system, in BIOS and at idle is 76°C. Additional Google searches showed numerous posts with CPU temps at 60°C or lower when idle and 71°C under load. MY CPU fan is spinning at 1985 RPM after starting at 1100 RPM. The motherboard temp is 41°C. I have the case side open so it is getting good air flow
Intel's Thermal Specification 71.4°C
So looks like the system has a CPU cooling issue after being up and running for 9 months. Would reviewers agree? How about the motherboard temp. Is it too high?
Thanks for everyone's input and help. Yes, the CPU was overheating. I checked the CPU fan and found it was unseated. That probably happened while moving it around. Actually, it probably was never seated properly. The CPU is still running too hot even after reseating the fan. (I downloaded a couple of CPU temperature monitors to keep track of the temps).
So, it's off to the computer store to buy a better CPU fan. The one that came with the processor is not doing the job.