so if anyone would like to play the diagnosing game, i think i got a good one here. i built this computer awhile ago ('09 or '10, i think... it was whenever windows 7 came out) and i don't really have the funds to build a whole new one, so finding the specific problem is paramount. besides, when it's working, it's still good enough for my current needs (which, at the most, is medium-level gaming ... think tf2 and football manager). over the past couple of weeks, my computer has an issue where it freezes and locks up. the only thing that works is hitting the reset button on the tower. it used to do it maybe once a day a week or two ago, but now it seems like the rate is increasing. what makes it difficult is that it appears to be completely random. sometimes it freezes a few minutes after startup, sometimes it'll make it through a whole day. sometimes it'll freeze in the middle of an intense game, sometimes it'll freeze when i'm doing nothing more than browsing bigsoccer. basically, there doesn't appear to be a specific trigger. i used to think that it did it more often after i had put it in sleep mode, but it's done it so many times after a fresh reboot that i don't think there's any significance to that anymore. before i go into what i've done, i'll tell you my specs (yes, i know it's an old computer): e6300 pentium dual core 2.8ghz 4gb ddr2 800 corsair xms2 ecs g41t-m2 ver. 1.0 motherboard ati radeon hd 4670 video card 1gb ddr3 400w corsair psu windows 7 - 64bit now, we can rule out anything to do with a virus or malware because it's happened outside of windows while i was doing memtest86. before that, though, i'd done virus scans and malware scans with no issues presenting. i've also checked the registry for errors and cleared out any bad entries with ccleaner. i've run my primary hdd through chkdsk and it's fine and passes s.m.a.r.t. with no issues. so i don't think the issue is with my hard drive. speaking of memtest86, i ran it while both sticks of ram were in. it passed the first passthrough with 0 errors, but i figured i'd just let it keep going as i understand it may take a couple passthroughs. i left the room and came back many hours later. apparently, after around 3 and a half hours, it froze and locked up during the test. it was on its 6th passthrough and had 0 errors. i've since ran it on each individual stick, all reporting 0 errors. i've also had the computer freeze while each stick of ram was installed by itself, and also in each slot, so i don't think i have a bad stick of memory. and no, i don't have any other sticks of memory to test with. i ran a diagnostic on my video card (i think it was called vmt or something like that) and it came back after multiple passthroughs with 0 errors. i also removed it from my case and it's locked up multiple times since then, so i don't think it's my video card. so next, you might think maybe my cpu is overheating or something. well, i've run realtemp and my cpu temp never goes over 58c.... well, it may be higher when it's locked up during a game, but i can't very well tab out to see what the current temp was when it froze. regardless, all the other times it's locked up when i haven't been in a game, the temps are around 38c-49c for both cores. all my fans are working and there's plenty of space and good airflow in my case, so i don't think it's an issue of my cpu overheating and glitching out. i've run intel's diagnostic tool on my cpu multiple times and it's passed them all, including the stress tests. now, out of all the times it's locked up, there's been two times where it was different. once, a couple days ago, instead of freezing, it just restarted itself. i was doing nothing more than browsing the internet with no other windows or programs running (i may have been signed into steam). that happened before i'd removed any hardware. it hasn't happened since. the other time it did something different was just a little bit ago. i'd removed one stick of ram and my video card (so my monitor was plugged into my mobo). it froze, but it shut my monitor off. you know when your display goes into standby mode? it was like that. it went dark, and said the typical "no signal present" or whatever it always says when it goes to sleep, and then shut off. when it's in standby mode, though, it has a light that flashes on and off. no flashing light this time. i clicked the physical power button on the monitor on and off and there was no response. everything was frozen. i rebooted and it all came back on like normal. that was two freezes ago. so that was weird. based on what i've tested and ruled out, i think it has to be my mobo or maybe my power supply, right? as far as i know, there isn't really a way to test your mobo. you just kind of have to rule everything else out, right? are there any other tests i should try? i know this area of bigsoccer isn't a high traffic area, but i thought i'd throw this out there anyway, and i'd appreciate any help anyone might be able to give. if i have to replace my mobo, i'd like to be 100% sure that it's the culprit. because if i replace that, then it wouldn't make sense to not also get a better cpu (which would at least double the cost) ... and then there's the issue of my oem version of windows not validating on a new mobo and having to deal with that hassle. uggh. oh yeah, i've done two other things. i've gone into my bios and reset it back to default. i'm 99% sure i have the latest driver they put out for it as well since that was back 4-5 years ago, and i don't think they even support it anymore as it was already a generation old when i got it. the other thing i did was i read about a trick to clear your temporary memory. i shut it down and unplugged the power source, and then i held the power button for 30 seconds and then i plugged the power back in and booted it back up. near as i can tell, that did nothing. so anyway, sorry about the length, but i wanted to be as thorough as i can be about the issue and what steps i've taken so far. thanks again in advance if anyone chooses to take pity on me and wade into this.
This is similar to an issue I experienced with a computer I built in 2004. By 2009/2010 it was randomly freezing and re-booting every 15 to 30 minutes - more reboots than freezes. In the original build, I used a case that came with a cheap power supply. I replaced it with a $30 Rosewill 350w power supply from Newegg. That solved the problems and the darn thing still runs today (PCLinuxOS using LXDE), though it's more of a relic that I try to keep running than a useful day-to-day computer. I would suggest buying an inexpensive PS from Amazon, Newegg, or other dealer with a good return policy. See if it solves the problem. If not, return it for a refund.
thanks for replying. i went ahead and used this as an excuse to upgrade to a more modern system. i replaced the mobo, cpu (went to an amd chip this time), and ram (doubled the size and speed). i kept the same power supply, hdd's, and video card. i'll probably "finish" the build later in the summer by getting an ssd, a new video card, and probably, a higher wattage psu. two things to note. the first being that my OEM copy of windows, for whatever reason, had no objections to me activating it on a new mobo. as far as i understand, it should've been tied to my old system and then refused to activate with the presence of a new board. this was a big relief, as i wasn't prepared to plunk down the money for another windows license. the second thing is that it's been a week now and it hasn't frozen once.
Great choice on the AMD platform! I think this is where your problem was. Not the sticks themselves necessarily, but the slots on your board. Running a low level program like memtest and having it crash can only mean that either the sticks are faulty or the slots on the motherboard are faulty. That's something difficult to troubleshoot since it involves swapping the board, but you could quickly diagnose this kind of problem by running the same tests in a different machine with those sticks. If it passed, your problem is most likely with the slots. They could use some cleaning out, or worse, they could just be done. ECS didn't make the most reliable boards at the time, more of an OEM manufacturer. I had an E6300 back when I was more into overclocking and I managed to run it for 2 years at 3.8GHz which made it a very capable processor well into 2011-2012.