Computer restarting randomly

Pink

Posts: 123   +1
Ok, so i get no BSOD, no warnings, no nothing. it just restarts.
it doesn't save any minidumps either. what do i do?
 
You might open the computer and look at the capacitor on the bourd where the processor is at.

They have to be flat, but if they are pointed, they are faulty. You can replace them, but that is a thricky thing to do. When your are replacing one, then replace them all.
 
All flat, i checked all of them. the ones near the processor and the ones on the bottom of the mobo...
 
Could you talk us through what happens when you press the power button please? Do you see BIOS (the back screen before the Windows logo appears)? Does the computer make any audible beeps?

If the computer is freezing at a set point, can you describe what is happening on the screen?

Have you made any changes to the hardware recently? Opened the case and cleaned it, or tampered with it in any way whatsoever?

If you can remember, what were you doing immediately before you started experiencing problems? Did you install an update? Or install an item of software?

Sorry for all the questions. Its important we establish what led to the incident happening, and whether anything has been done prior to it that would suggest remedies worth trying.

Lastly, how old is the computer, and what is the specification of it?
 
Ok, well. computer starts up normally, it doesn't freeze ever, even when it resets, it jut resets without warning, or any kinds of freezes. there isnt even a bsod.

The bios appears normally, it opens windows normally, and while using windows, it runs normally.

I cleaned it about a month ago, and it was pretty superficial. just the fans, a quick brush around the mobo. the capacitors are fine. aside from that, i haven't installed new hardware in a while.

hmmm i really can't remember if i installed something that caused this... if i did, i'd have removed it to check. so i have no idea. i'm pretty sure it's software though, and something i installed. or maybe some windows service, since in safe mode it doesn't crash.

Well, some parts of my pc are old, some are new, and some are in the middle. i have a Amd Athlon 64 x2 5600, 2gb of ram ddr2, radeon hd 4670 512vram, what else. mobo is a m2n4 if i recall correctly, i can't remember, that's the oldest part of my pc. i have 3 hds, one is 160gb, one is 250, one is 500.
 
My main suspect is your power supply. Try upgrading it to at least 600w. If you are at or above 600w it could be failing. Its a good place to start.
 
@Circuit
That system will run absolutely fine with a 350w PSU. 600w is well and truly overkill.

I'm not convinced its the power supply either. In my experience they usually work, but fail when stressed (under bigger loads) or don't work at all if they're playing up.

@ Pink
Does the computer show signs of slowing down before it just randomly reboots?

Can you confirm that all fans are indeed working inside the case when its running?

You say its not BSOD'ing, but are there any logs in the following folder: c:\windows\minidump -- files using have the .dmp file extension if there is.

Also, it would be worth downloading HWMonitor, and running it, and then recording temperatures as well. Just so we can confirm that heat isn't the issue.
 
@Leeky
I disagree! I have seen power supplies fail due the capacitors slowly heating up and changing value causing it to fail randomly. lol and yes 350w can power that, but most Graphic card manufacturers recommend 450w or better for a card of 512mb or larger. So I say 600w strictly based on pricing. I may be a neophyte on these boards, but I do know a lil bit about computers
 
As far as I am aware the memory footprint has nothing to do with the required power rating of a GPU, it is just a measurement of the on-board dedicated memory capacity.

Though I do agree that pricing will have some bearing on the size of the psu, as its a false economy to spend $5 less and get something with a lot less wattage. However, you did not clearly state that, which is why I made my comment.

As a FYI, a HD4670 is listed as requiring a 400w PSU, which in itself is very conservative considering 350w is more than enough.

If it was a power supply issue it would be rebooting randomly in safe mode -- that alone is a good indicator of the issue being something else.
 
As you can see by my number of posts I'm not a major poster, so i don't always describe down to the fine points of what I mean.
I post what I believe will help and hope it was of use. And you know as well as I do that the PSU debate can go on forever.

Hopefully with help, Pink can get this issue fixed, I will love to know what it takes to fix it. (power supply he says in a cough)
 
Ok,

Circuit, i respect your opinion a thousand %, but i don't think its the PSU. the pc runs forever in safe mode and won't restart. plus i have to be doing something for the pc to restart. i have a 500W real wattage psu. it was pretty expensive too lol..

Leeky, nope, pc will run fine and out of the blue, restart. no minidumps, no log, i even check the event viewer, but theres nothing related there... fans are all working fine

about temps, speedfan is reporting these temps:

Temp1: 49c
Temp2: 38c
Temp3: 25c
HD0: 46c
HD1: 36c
HD2: 49c
Temp1: 40c
CPU: 47c
MB: 38c
 
Try this: Turn off Auto restart...if your running Win 7.....click start/right click computer/properties/advanced system settings/Advanced/Startup & recovery/turn off Auto restart. Lets see if it will bsod
 
it's already turned off, i turned it off when i formatted my pc. and i'm running winxp 32bit
 
Temps look okay, but it depends how it reacts when you push it that will dictate if its rebooting due to overheating.

Your hard disks are a little warm for my liking, which would suggest to me that airflow is not massive in your case (certainly not across your disks anyway).

When you say you have to be doing something for it to reboot, can you give us a few examples of what you would be doing? So I can get an idea of load etc.

You can monitor temps and check the CPU by hitting it with some Prime 95. It will really stress the CPU so have the temps window in Speedfan open and keep an eye on it.

Also, can you download Memtest, and burn it to a disc, then run that overnight. You need at least 7 complete passes in order to rule out RAM as the issue.
 
well thats all of my ideas. If it were me I would reinstall. aka Punt. maybe someone else has other ideas. I will keep an eye on this thread, to see how it goes.
 
Yeah my hds suffer.... i leave one of the "walls" of my pc open for airflow though (not completely open, but leaning against the pc itself.)

Hmm, it can go all the way from Playing some pc game, to opening a notepad file. so i doubt it's due to heavy loads or overheating.

i still think it's software.... or some windows service.

Thanks anyway Circuit, i appreciate your time
 
Ok, i just wanted to write a quick update.

My pc didn't restart yet, two days now. I dunno if i fixed it, but what i did was disable O&O Defrag software from booting with windows. (both the service and the startup app)
It's also an app i installed not long ago, maybe a couple of weeks.

Let's see if it solved it, i'll report back periodically
 
A good way of testing that would be to run a stress test using something like Prime 95 or another for an hour or so.

If there is any problems they will always show up under heavy load.
 
Ok, new symptom.

Now the pc would turn on, show me the bios screen and then the monitor would go out of sync, and stay there.
I unplugged the pc and re connected it, now it works, but whenever i go into some heavy load app (like any game) the monitor would go out of sync.

So, what is it?
 
Hmm, well i got a Samsung T220 monitor. when it "goes out of sync", the monitor will go black, and the DIGITAL, ANALOG thingy on the top left corner will appear (when it's trying to sync with my digital or analog plugs, it alternates between digital and analog, like when i boot my pc and it tries to find my connection. what i think it is, is that the psu stops powering some parts of my pc. if it was my gpu, wouldn't it prevent me from running it at all?). i got no idea why it happens, i already ruled out RAM (i switched between them, nothing doing) i already ruled out my GPU (temps are low, and when i go ingame, sometime it'll actually go ingame and then go out of sync, and without any artifacts, and the capacitors on my gpu are good as new). i'm shooting for the PSU.... what do you think?
 
It is possible it is, but it could also be the graphics card. The PSU isn't really under any load from the GPU in 2D (e.g. when turning the computer on, or on the desktop) but running games it will be under load.

Given your previous problems it would appear that maybe the PSU is worthy of suggesting its the culprit, but it could also be the GPU.

I'd try PSU first.
 
Yeah, but i'm running the pc normally now. it runs great.
is there any way to stress test my pc (and choosing which parts to test?) i'm getting Burnin test now, maybe that'll lead me to the culprit. i'd hate to waste money replacing something that didn't need replacing..
 
Back