Thank you for being a valued part of the CNET community. As of December 1, 2020, the forums are in read-only format. In early 2021, CNET Forums will no longer be available. We are grateful for the participation and advice you have provided to one another over the years.

Thanks,

CNET Support

Question

Confusing Hardware Malfunction

Feb 4, 2013 8:51AM PST

Hello

For a few days I've been struggling to find out what went wrong with my desktop. My config is:

Pentium E5300 Dual Core 2.6GHz
2 GB DDR2 800 mhz dual channel in 2 Syncron dimms of 1GB each
Geforce 9600 (I think it's GT) graphics card
Asus P5N73-CM with nForce 630i chipset
Western Digital HDD 200 GB with Windows XP and another WD 1000 GB with Windows 7 that depends on the first one to boot

My PC hadn't caused me any problems for months and a few days ago froze in the middle of loading something in a game (in win 7) and could not load windows since then.

But it's more than that, it seems to freeze on every type of system load, including Safe Mode, Win XP or Win 7. Most of the time it freezes as soon as I pick an OS after I start it but sometimes it starts loading win XP or 7 and freezes right before the screen should switch to desktop while other times it doesn't even make it to the choose OS menu.

It even froze in bios a couple of times or bootable software I tried on my USB stick such as MemTest and even gave me a few graphical artifacts but only in those situations.

Here is what I have investigated so far.

I tested the memory sticks one by one in each slot with no result bu then tested them again the same way in a different computer and they both ran just fine. I also tried a Kingston 2gb DDR2 800 from that pc on mine and still got no result. Also, if I'm not mistaking, MemTest found errors on all 3 sticks even though they all worked on the other pc. As I said MemTest barely ran and after finding a few errors it would freeze but one time it ran for an hour and found about 9 million errors on one of my sticks...

I tested my 200Gb drive on the other pc and to my huge surprise it worked without a problem and even booted my xp. I could have bet the old timer was a goner. I didn't test the 1000Gb drive in the other pc because I figured since it can't boot by it's self it wasn't that relevant and just left it aside unconnected.

I ruled out the graphics card because after switching my monitor to my integrated motherboard nvidia chip everything was still happening the same way. I didn't insist on reproducing the graphical errors with the onboard GPU mainly because I had also updated my Bios. Also, as I mentioned, I saw no graphical errors other then when the pc froze in Bios and MemTest. I tried resetting the Bios to factory settings too.

I even tested my computer with a different power supply and still saw no change. I tried a different monitor, mouse and keyboard aswell.

My conclusion is it could only be the motherboard or CPU and from all I've heard and seen CPUs tend to go out with a bang rather than degrade partially. Also I noticed the radiator on my motherboard's chipset was a bit loose. I detached it and found the conductor paste wasn't exactly evenly spread across the chip so I bought some, cleaned the chip and radiator best as I could and applied a new coat of paste that in a few hours was solid enough and the radiator didn't feel loose anymore. Of course that brought me no results either but it did lead me to think maybe that was the whole problem and the chip might be partially fried now.

The CPU fan seems to be working and bios tells me it's at 1200 rpm constantly and around 2000 during work loads, which was it's usual speed it seems. I have to admit I didn't keep it as clean as I should have though.

So right now I'm trying to find out if I need to buy a motherboard or a CPU. Of course, I could be completely wrong and it could be something else entirely.

So, if you have any ideas about this whole situation please feel free to take a jab at it, I would greatly appreciate any help.

Discussion is locked

- Collapse -
Answer
The moment I read 9600.
Feb 4, 2013 8:56AM PST

Those old video cards are not faring well. Borrow any other video card to see if that's it. Also, your post is short. No mention of PSU rating. I find many old machines that should have died long ago as the owner pushed the PSU hard. Sort of like driving your car around the indy track at 120MPH. Won't run for long.
Bob

- Collapse -
Could be
Feb 4, 2013 9:09AM PST

During my years of using gaming PCs I've learnt the graphics cards tend to be the most fragile (that and cheap power supplies...) but I did try using my other graphics GPU, the integrated nForce, and nothing changed. Now I know that one is even worse but it hasn't been used at all and I find it unlikely that both the integrated GPU and the graphic card are defective and causing the same problem. I would have clearly tested my graphics card elsewhere or a different graphics card on mine but didn't have any solutions at hand. My other computer doesn't even have an agp slot, nevermind pci express and there is no friend nearby that could land me a pci express card for testing.

- Collapse -
Then is it worth fixing?
Feb 4, 2013 9:12AM PST

Without spares you will be a little lost. Today, unless the part bursts into flames it rarely tells you it's that part. So we have shop cards and drives to swap in. I'm guessing no spares so you have to choose what to do next.
Bob

- Collapse -
PSU Rating
Feb 4, 2013 9:19AM PST

My PSU is crap I guess. It's an unknown brand of 450 (theoretical) watts and it worked fine for 2 years now. But I ruled that one out by testing my pc with the PSU from the other pc which happens to be newer and claims to provide around the same value but more importantly is properly working as I type this.

- Collapse -
Answer
Heat stress
Feb 4, 2013 1:41PM PST

You got a lot going on and it maybe showing its age or wear&tear. You got dual OSes, dual HD(old and new), cheaper PSU and a 9600 GPU. Since, the problem crops before you even boot into the OS, then that suggests either the PSU or HD. I suspect more that the 200gb HD has seen better days. Even though it may not showing hard errors it maybe starting to be wonky. If I suspect these 2-HDs are mounted one on top of the other, there's a hot spot that has "stressed" one drive or the other. I think the older HD is lacking. So, while you can backup any critical data before it really fails. heat stress tends to make difficult to pin-point the cause as i stated it gets wonky. Since, you mentioned gaming that too if prolonged play and compromised cooling or constant strain, then you find the weakest link. It maybe too late is heat stress is cause, because it's so hard to narrow it down other than swapping parts and re-verifing cooling and heatsink pastes, etc. are all well. Open case and place household fan to better cool. Next, clean the dickens out of it, *everywhere*. You also mentioned, the PC hasn't caused problems for months, has this been recently build from used parts? If yes, then some old part is giving up the ghost.

tada -----Willy Happy

- Collapse -
A lot going on indeed
Feb 4, 2013 4:35PM PST

My hard-drives were placed separately so they wouldn't heat each-other up. I haven't had overheating problems that I can remember because my case has an enter and exit vent but I suppose both the CPU and the motherboard chip could have developed such issues since my cpu was boxed so the fan is not one of the best and the chip had the paste problem.

Also I removed the new HDD from the system because the old 200Gb proved to be working still when I tested it in my other PC. I don't keep any important data on it anymore but when I bought the new HDD I installed win 7 on it from win XP and now the old HDD is the "master". I was sure the old thing had given in but it seems it's still working so that only leaves me with a more complicated issue.

As for the cleaning, there was plenty of dust in my case but luckily for the motherboard hardly any settled on it since it's vertical. I cleaned it up but I will look for bits of dust hiding around.

- Collapse -
Test and test again
Feb 5, 2013 5:51AM PST

If indeed there was alot of dust, then you need to check the PSU and frt. panel again. There unknown to many users are coated or caked vents or parts. You need to open PSu and check it. Once open verify for "bad cpas" problem, google for pix examples. The frt. panel vents once removed allows complete access. I have found many cases apparently clean then remove frt. panel and a sheet of dust falls. That of course clogs or blocks "in" vents in most cases. Thus, you need to truly check everything out and if you have to, test on another PC. Don't take anything for granted because it worked X-yrs. or gave no problems before. This is the reason for tests and diags.. There are HD, memory and I use PCWizard 2012 for full details on a PC. Checks fan speed, heat ranges, etc., too. Use the Win7 memory test upon boot-up menu area, or popular memtest.

tada -----Willy Happy

- Collapse -
Answer
Intermittent failures
Feb 5, 2013 4:57AM PST

I've seen a lot of whacky and intermittent faults that stemmed from a CMOS battery failure. The most obvious clue might be the time keeping but in most cases, the BIOS gets whacked and it starts not being able to verify known installed hardware. Them progresses in any number of directions from memory, video, audio and other peripherals malfunctioning or non-responsive. Like a patient listening to a doctor ordering a bunch of tests, a complete examination of the patient is required before any real prognosis can be rendered. Renewing the thermopaste on any detachable heatsinks and a cleaning of all boards, vents, fans and heatsinks will go a long way to at least have a square one to start with. Replacing the coin battery cell on the MOBO may at least get everything back to a factory default.state.

- Collapse -
Answer
Memory Issues and power source
Feb 7, 2013 3:58AM PST

It seems the problem has been solved. I took the pc to a computer shop and before I bought anything they tested my parts one by one. It didn't boot until both my PSU and memory sticks were replaced. I ended up buying a Sirtec 500w SPU with a large, silent fan and a 2GB Kingston memory stick (I decided not to care about dual channel since it's effects seem to be marginal).

The main problem was that my Asus motherboard is a bit picky about memory brands and my old Sycron sticks clearly deteriorated at some point enough so that they became unusable by the motherboard. They still work on my other PC but they do cause strange lagg at times. The Kingmax 2GB stick I had as a backup was also deemed unworthy by mu Asus although I'm pretty sure there's nothing wrong with it.

Currently the PC is running smoothly (I'm writing on it) and even though I'm not particularly happy about spending money only to have the same performance I had before I do consider the PSU a good investment and really enjoy the silence. Apparently no other parts suffered any damage.

Thank you to everyone who posted. I learned from this that my PC needs a good "scrub" every once in a while and that it's important to have a reliable PSU.