Computer restarts unexpectedly

Status
Not open for further replies.

Narey

Posts: 70   +0
Firstly I am not entirely sure if I am in the write section so sorry if this has to be moved.

My system specs are in my profile.

I went to uni without my computer, I come back to get it for this term and suddenly it is acting up. Possibly due to siblings messing about on it.

The problem is this: The computer is restarting under different conditions however usually bought about by the computer being on for a while or doing something testing like playing a game. Untill recently it could play low requirements games for hours or indefinitely before crashing but lasting in some high req ones only for a matter of mins (All games which the system should easily be able to handle and has done in the past). More recently it has been crashing on very low req games after not very long atall.

I have checked all temperatures and all seem to be normal. I have also tried running the pc with only 1 of the sticks of ram and then swapping it. On one the games crashed and on the other the pc reset. The pc seems to last the longest with both sticks in. (From this I sort of assumed it was unlikely that both sticks had suddenly stopped working? They are only just a year old)

I will attatch my most recent mini dumps and it would be great if someone could have a look at them.

1 2 and 3 are from today and the other 1 is from a few weeks back but it is the same problem.

Thanks a lot guys for any help
 
You system spec is not complete!

How many optical drives?
Make and model of the power supply? (If it is generic no name brand, the max Amp from the sticker for the +3.3V, +5.0V and the +12.0V rails)

Let us know
 
Oppsy Daisy my mistake, Optical drives are currently only a CD RW at best, it was originally a DVD RW but now refuses to read DVDs as of over a year ago.

PSU is an Arctic 700
+3.3 - 30
+5 - 38
+12 - 22
 
Three of your four minidumps are 0x8E errors and these are almost always caused by hardware though sometimes they can be caused by drivers. In all three they cited nv4_disp.dll which is an Nvidia display driver.

However, the fourth error was an 0xA and these are another strong indication of hardware issues. In this one it specifically cited memory corruption. Corrupted memory will cause system crashes anytime.

So before you attempt to update your Nvidia drivers I suggest you first run Memtes on your RAM. If you have never done this before just follow these steps:

1. Go to www.memtest.org and download the latest ISO version. It is free and perfectly safe.

2. Burn to a CD.

3. Place CD in your drive and reboot with CD in drive. The test will take over.

4. Let it run for a LONG time. The rule is a minimum of 7 Passes. There are 8 individual tests per Pass. Many people will start this test before going to bed and check it the next day.

5. If you have errors you have corrupted memory and it needs to be replaced.

6. Also, with errors you need to run this test per stick of RAM. Take out one and run the test. Then take that one out and put the other in and run the test. If you start getting errors before 7 Passes you know that stick is corrupted and you don’t need to run the test any further on that stick.
 
I would try the 8 Steps here just so you're sure the system is clean.

https://www.techspot.com/community/...lware-removal-preliminary-instructions.58138/.

A memtest probably is due as well, memtest's results on a "sick" system is not always true though.

One thing I noticed about you power supply is the fact that it uses 4pin molex to 6 pin PCI-E adapter, The Arctic 700 I found also has 2 +12.0V rails each with 22A max which is a good thing.

With multiple +12V rails, some times the yellow wire (+12V) has distinguishing marks, so you can tell which rail it is coming from, might be a good idea to try and plug the adapter for the video card to another 4pin molex by itself and test again.

In the Hardware Monitor section of the BIOS, can you check the actual readings for the same rails, want ot make sure they are within + %5, a voltmeter is best if you feel comfortable, yellow is +12V, orange is +3.3V and red is +5.0V.

While you're at it, I would also check for a BIOS update and run the defaults for starters.

If the system was left for few months and not used by anyone else, depending on where it is pluged to with how many other systems (power bar perhaps) I have seen power supplies take a hit once turned on specially if a power bar with few other electronics plugged in to it in the ON position was turned ON at once.

I have the minidump viewer on a system which I am going to be at later on, will check it our for you.

Just to be sure, there are no artifacts on the screen, are there? Artifacts, point to the video card's memory and crashes point to the GPU. This could be power, hardware or software related.

Let's start by making sure your system is clean, up to date with all the firmware, BIOS and drivers.
 
Ok thanks guys I will do everything there to make sure the system is clean. The one thing that might be a problem is the mem test as I am not entirely sure the CD RW will actually write cds it has been acting up. I'll get hold of one and see though.

One more question. When using singular sticks of ram does it matter which slot you put them in the pc? I have 4 slots 2 orange 2 yellow so I assumed they are for pairing sticks but when using 1 I was not sure if it mattered atall?

EDIT: Answering the artifact question. I think artifacts are when things to not appear correctly? Like green squares appear and the like? That has happened once and only when the game crashed and I persisted in trying to reopen the game without restarting and it would only reach the first screen for a second then crash again. Eventually it would not open atall untill I restarted the system.
 
What does your system manual say about memory configuration? I'm sure, without looking, putting a stick in the first slot should be fine, but I am only guessing. Also, do you have another system whereby you can burn a CD like a friend's?
 
I have got a friend bringing a cd over later on, if it does not burn i'll run round to a friends. One thing which has just been bought to my attention that might be important. (Not sure if it is related maybe another problem. GOD!) Is my virus scanner AVG is picking up these everytime:

kernel32.dll
wsock32.dll
user32.dll
shell32.dll
ntoskrnl.exe

Thanks

EDIT: Having done a run around on the net apparently AVG is just telling me the files have changed (AVG does report them as "changed") which is a reuslt of a windows update. So I think that is alright?
 
Ok thanks guys I will do everything there to make sure the system is clean. The one thing that might be a problem is the mem test as I am not entirely sure the CD RW will actually write cds it has been acting up. I'll get hold of one and see though.

One more question. When using singular sticks of ram does it matter which slot you put them in the pc? I have 4 slots 2 orange 2 yellow so I assumed they are for pairing sticks but when using 1 I was not sure if it mattered atall?

EDIT: Answering the artifact question. I think artifacts are when things to not appear correctly? Like green squares appear and the like? That has happened once and only when the game crashed and I persisted in trying to reopen the game without restarting and it would only reach the first screen for a second then crash again. Eventually it would not open atall untill I restarted the system.
Your board is Dual Channel capable, if you have 2 preferbly identical modules, you should put them both in either the yellow or orange sockets, if you put one in yellow and the other in orange, you will be running in single channel mode (64bit data path rather than 2 X 64bits = 128b)

When you use memtest, I would disconnect as much power and data cables from anything that is not being used. All you need is the processor, heat sink and fan, video card, keyboard, that's it, I would disconnect all hard drives and even pull add-in cards out, this is to ensure, the RAM, CPU and the board are getting enough juice. When you are trying to test RAM, it is a good idea to try one stick in each socket and then the second stick in all sockets one at a time, I know it is 8 passes in your case.

Some boards and chipsets depending on the module, chipset and the memory controller, like the first sockets to be populated, some don't care, this is for you after the tests, to decide where the modules are going to live.

I would check the +3.3V, +5.0V and the +12.0V rails in the BIOS to see what they are at (hardware monitor).

Artifacts are dots and lines and boxes on the screen which is caused by the video card RAM running hot, out of spec or failing. Crashes are mainly due to the GPU.

The files you posted, does point to infection /s, you might even have to run LSP-Fix to repair Winsock 2 settings after cleaning.

It could also be data corruption but infection is most likely, if your XP is not patched (I know you have SP3) with all the updates from MS, you can still get infected, it is not AVG's fault (not saying it is a good AV or not) make sure after cleaning, you do Windows update not auto, select custom rather than express, if you have MS Office or other MS software, use Microsoft Update rather than Windows Update.

Keep us posted

EDIT: just read your addition, wonder why AVG never prompted me after updates, is yours a retail version or the free one?! Still a good idea to make sure the system is clean, thanks for the update.
 
UPDATE:

I ran through all 8 steps. The guide says to post results for 3 but 2 didn't come up with anything and looking at the log it just seems to confirm that nothing is wrong and not give any other information really so I thought they were not important. Here is the log for hijack this.

One other thing that happend was that my Java was not the right version, I updated then rechecked and it now tells me that it is correct. This is odd because Java usually tells me from the desktop that I need an update and I allow it. Intrestingly recently a website www.teamtalk.com was causing me to crash. That page to my knowledge uses a lot of java. I don't know.

In response to the artifacts the only time that occured was in the occasion I described when the game stopped running entirely and I kept trying to open it. If it helps atall I will list the games:

World of Warcraft - Runs for about 10 mins on full settings perfectly then either the pc restarts or the game crashes and takes me back to windows.

Warcraft 3 - Never crashes although I have not tried playing it in the last few days these crashes have been going on much longer then that so the game seems pretty much unaffected.

Football Manger 2009 - The game would run for hours on end and then the pc would reset. More recently the pc would reset sooner.

One more thing. When attempting to get into bios I could not. In the past I would press DEL to enter but pressing the keyboard got no response. Trying to reach the bootmenu or setup using F keys got no respose either. Could this be becuase I am using a USB keyboard?
 
Thanks for the update.

It is good that your system is clean, Logs are only needed in case of infections, we just wanted to make sure your are clean and patched.

I can only think of Video Card heating up or not getting enough power (did you try using another 4pin molex?) Same thing with the CPU and the system.

Your USB keyboard might not be initializing fast enough, you either need to use a PS2 and re-set BIOS settings, or reset the BIOS using the jumper or the old battery removal trick and try again.

If heat is not an issue and your memory test comes out good, we have to try another power supply prefebly one with a PCI-E connector and not an adapter, then we can say it is the vid card.

I'm assuming you have the latest DirectX, Video drivers and are not running the games with settings too high!
 
Bad news, as I thought might be the case what was once a DVD RW is now a CD R. So memtest is not working. I am going to give a friend a call and try and get him to write it for me. Another problem would be that even if I get it onto cd I will not be able to get the pc to boot from it as the keyboard does not seem to start running fast enough to make any decisions. Problems problems problems. (In the past the keyboard did manage to enter bios I am almost sure I have been in bios in the last year (age of the keyboard)).
 
In cases like this, I like to drain the motherboard by removing all power and data cables and as much as possible any add-in card, then remove the battery for a at least a minute, then put things back in, I even remove the board but I have the space and the work benches :)

For doing memtest this might be a good idea and also get you back in the BIOS,

Something doesn't sound right here though!

Keep us posted.
 
I have to go stay with my girlfriend this weekend so will not be able to continue solving the puzzle that is my computer untill monday. If I manage to sneak away at some point to go home I can get a spare PSU and bring it back to uni with me. Thanks for all the help so far guys, see you on monday.
 
Status
Not open for further replies.
Back