Hi all, I have the following old rig - CPU - Intel Core2Duo E4500 2.2GHz(Oc'ed to 2.5GHz) Motherboard - Gigabyte P35 S3G Ram - Corsair DDR2 2x2GB DDR2 800 1.8V Graphics card - MSI 8800GT factory OCed HDD - 1x 320GB 7200RPM with Linux MInt 17.2 64bit 1x TB 7200RPM with Win7 32bit PSU - Corsair VX450W Case - CoolerMaster RC330 I have been having random computer restarts since a few months. Even after getting the motherboard repaired twice, the problem still occurs. Thinking it was a problem with the electrics in my house, I shifted the computer to a friend's place. I ran the blend test for 24 -35 hours with the following configs and it never restarted. 320GB HDD only, 1TB HDD only, Both HDDs but under linux, Both HDDs but under Win7, 'System Voltage Control' in BIOS to Auto, OC'ed to 2.5GHz Auto Voltages. Then I ran the full power consumption test with the cpu OC'ed to 2.5GHz and auto voltages under Windows 7. The system restarted after about 12 hours. I then ran the same test on stock speed with auto voltages and it restarted again after about 9 hours. I cant quite figure out what the problem is. Any suggestions? I have attached a screenshot of HWmonitor when the ssytem was on stock clocks and auto voltage setting. http://i.imgur.com/hBVx3MA.jpg Thanks!
I'd probably try a memtest on the RAM to make sure that's not borked first, possibly one stick at a time.
Does that VCore look a little low to anyone else? I thought the E4500's VCore should be around the 1.35v mark.
I had tried a memtest a few weeks ago before trying everything else. I'll give it a try again for 24 hours with stock clocks and all voltages on auto and power saving functions enabled like C1E enabled with both sticks.
Not sure. The intel website says that its within range that is 0.8500V-1.5V - http://ark.intel.com/products/32242/Intel-Core2-Duo-Processor-E4600-2M-Cache-2_40-GHz-800-MHz-FSB The screenshot was taken with all voltages on auto and on stock clocks.
Sorry. Its a Gigabyte P35-S3G. When I run the Prim95 full power consumption test, then no. It switched off after approx. 9 hours. When I run the blend test then yes.
I agree it's within range, the lower end being 0.8500V when it's down-clocking, idling, and other types of power saving modes, and 1.5V being the maximum where there's a risk of toasting the CPU. The thing that has me questioning the voltage is (afaik) Intel sets the voltage for each chip based on the minimum voltage needed for it to remain stable at standard clocks and then add a little, as you've OC the chip in the past for a good few years that minimum voltage needed to remain stable at standard clock speeds could well have gone up. Most of the time with stability issues using Prime95 (imo) it would be down to not enough voltage, although it would be nice if someone else could either confirm or debunk that.
Well the stability issue was not just with Prime95. In the past few months, it has been the case with normal usage as well, under varying loads like simply reading text on the browser to playing games.The repair guy said the past few times that it was a problem with the motherboard that he got fixed, the last time being the BIOS chip. But he gave up when I called him the day after he dropped off the machine saying that the problem had returned. So I am left on my own now. The blend test, on Prime95, oddly enough didn't pose any problems, whether the CPU was OC'ed or not. It ran a full 24 hours. Its the full power consumption test that's posing this problem.
If your motherboard has been repaired twice then that is the likely culprit. But your chip is old and has been overclocked I'm assuming for a large period of its life, so that may also be degraded. There may be other sources of the issue such as the power supply stability. Or it could be any combination of any of these and others. The only way you are going to get to the bottom of this is to switch out parts. You could undervolt the cpu and down clock it, and check for stability. I'm not sure what that's going to tell you though as it doesn't really eliminate any of the possible issues.
So I ran Memtest and left it for 22 hours. When I came back it had the message "Pass complete, no errors, press Esc to exit". Strangely though when I pressed the escape today it didnt respond. It was as if it was hung but on the top left corner the 'plus' sign, next to the test'Memtest86" was blinking. Was that because it was in the middle of a test or because of something else? Ok, something odd just happened. While reading an article on Opera on Windows 7 the computer just hung. No beeps or other odd behaviour. Just a freeze and no response on the keyboard or mouse.
I moved the system back to my place to try it out. Within an hour of using it, I got random freezes while browsing the internet. I found another repair guy and have given the motherboard to him. From a quick preliminary check, he found that the northbridge and southbridge chips were problematic and needed to be replaced. He says that finding replacements will be hard so I should look into buying a new motherboard. The problem here is that I am not sure if its my power supply that's damaging the motherboard or the board itself was never repaired properly in the first place. If its the power supply, it will end up damaging any new equipment. Is there any way to get the power supply checked thoroughly? Any advice would be much appreciated.
