PC Freezes with Looping Sound

those settings could be why your PC crapped out. Have you tried running RO?

Not yet. I'm doing a Memory Test anyway, since I went through the trouble of burning it and what not anyway. It's passed on Test with no errors, it's doing another now. I think I'll let it do 1 or 2 more before I'm certain that it's fine.

MemTest86+ also tests the CPU's Memory Cache, so would that also clear the CPU?

Edit: Okay, after 2 and about half of a 3rd tests, my Memory all seems perfectly fine, so that rules out that potential problem.

In reference to using ATI Tools: Wouldn't that basically create a high-load situation, thus causing a crash? If so, that wouldn't exactly confirm my GPU is the problem. I should ask my brother if I can test my GPU in his computer, that would be a better test I think, since a high load crash may occur regardless during a GPU test because of how the ones I've seen work.

Unfortunately after a recent crash, the HWMonitor did not make a log, since it does it at the end (because that's wise), so I'm unsure of if it is a Voltage issue still. Since my Surge Protector is so old, I'll probably go out soon and replace both and hopefully that will remedy the situation. Still going to do a memtest tonight and I'll make sure I give the results just in case.

If memory was relevant, then you saw the error message either in the system (event) logs or in a BSOD screen. The numbers in either say everything. No error numbers? Then memory was eliminated from your list of suspects many days earlier.

Most of what you did (ie upgrade BIOS) put your hardware at great risk without any reason to fix anything.

Unfortunately you have been measuring using the voltage monitor. It is a monitor for voltage changed. You need a multimeter to see the actual voltages. That monitor is not designed to provide accurate numbers - only to report change.

The multimeter is sold mostly in store that also sell hammers because the meter is that complex and dangerous. Let's get rid of that popular urban myth now. It even sells in K-mart or in Wal-mart for about $18. You must borrow or buy one. There is no other useful solution.

Now, your voltage measurements are only useful when the system is using (multitasking to) every system component. IOW you must play complex graphics (ie a movie), while downloading from the Internet, why playing sound loudly, while reading a CD, while powering a USB device, while searching the hard drive. Only now is the system under any relevant stress.

Use the meter on 20 VDC scale to measure any one orange, red, purple, and yellow wire from the power supply to motherboard. For example, the purple wire will measure about 5 volts. But the number must be reported to three digits. Three digits contain information that only more knowledgeable can interpret. You report the numbers. The next reply will be definitive. There will be no more 'it could be this' or 'might be that' answers.

For example, your power supply could have been defective months ago or even when first installed. Normal is for a computer to boot with a defective power system. But the meter would have found that defect even that long ago. Nobody will provide a useful answer until you collect those voltage numbers only when the system is under that maximum load.

All those other 'it could be' suspects such as a bad power cord? Multimeter numbers will even identify or exonerate those speculations - without doubts.

So, how often is your 120 volts rising to 400 and more volts? Never? Your surge protector does nothing - is nothing more than a power strip - when 120 volts is below 330 volts. Only those with fundamental electrical knowledge or who understood numbers on the box would know that. Another example of so many hyping myths only due to hearsay. Ignore so many silly recommendations that only have you wasting money.

A power system can make any good hardware act defective. Get the meter. Provide those numbers to three digits. Have an answer without doubts. Only then should other suspects be considered.

And finally, the list of suspects for your failure as short. Power system (which is more than just a supply), only a few some motherboard functions (BIOS is clearly not on that list), CPU, sound card, video controller, and drivers for third party hardware. Some of those suspects are quickly eliminate by system log or BSOD messages. Notice so many items not on that list that others wanted to fix. Until you have those three digit voltages, then anything else can act strangely.

Stop trying to fix anything only on wild speculation. Memory, BIOS, surge protector, and so many other speculations were removed from a suspect list at the very start. First learn what is defective. Follow the evidence. Step one starts with a multimeter - and not numbers from a voltage monitor.

I should ask my brother if I can test my GPU in his computer, that would be a better test I think, since a high load crash may occur regardless during a GPU test because of how the ones I've seen work.

Do not swap parts. That is a fundamental No-No in any facility that fixes things right the first time.

If you can create a crash repeatedly, then the next step easily identifies the problem. No, that test does not say the GPU is defective. It says a shorter list of suspects exist.

Only way to have a valid voltage measurement from those six wires is to have the GPU executing stressfully - up to the point that a crash might happen.

If power to a GPU is defective, then numbers from a multimeter will say so loudly and without doubt. Without those numbers, you can be chasing your tail. Do not remove anything. Do not even disconnect one wire. Always collect relevant facts before removing anything. That also means reviewing system (event) logs and Device Manager.

Useful to identifying a defective part before replacing anything is to make the intermittent hard. If you can stress the GPU to always recreate a failure (and do so without Windows), then a hard fact exists. And the number of suspects is down to less than 10 – a serious reduction.

If memory was relevant, then you saw the error message either in the system (event) logs or in a BSOD screen. The numbers in either say everything. No error numbers? Then memory was eliminated from your list of suspects many days earlier.

Most of what you did (ie upgrade BIOS) put your hardware at great risk without any reason to fix anything.

Unfortunately you have been measuring using the voltage monitor. It is a monitor for voltage changed. You need a multimeter to see the actual voltages. That monitor is not designed to provide accurate numbers - only to report change.

The multimeter is sold mostly in store that also sell hammers because the meter is that complex and dangerous. Let's get rid of that popular urban myth now. It even sells in K-mart or in Wal-mart for about $18. You must borrow or buy one. There is no other useful solution.

Now, your voltage measurements are only useful when the system is using (multitasking to) every system component. IOW you must play complex graphics (ie a movie), while downloading from the Internet, why playing sound loudly, while reading a CD, while powering a USB device, while searching the hard drive. Only now is the system under any relevant stress.

Use the meter on 20 VDC scale to measure any one orange, red, purple, and yellow wire from the power supply to motherboard. For example, the purple wire will measure about 5 volts. But the number must be reported to three digits. Three digits contain information that only more knowledgeable can interpret. You report the numbers. The next reply will be definitive. There will be no more 'it could be this' or 'might be that' answers.

For example, your power supply could have been defective months ago or even when first installed. Normal is for a computer to boot with a defective power system. But the meter would have found that defect even that long ago. Nobody will provide a useful answer until you collect those voltage numbers only when the system is under that maximum load.

All those other 'it could be' suspects such as a bad power cord? Multimeter numbers will even identify or exonerate those speculations - without doubts.

So, how often is your 120 volts rising to 400 and more volts? Never? Your surge protector does nothing - is nothing more than a power strip - when 120 volts is below 330 volts. Only those with fundamental electrical knowledge or who understood numbers on the box would know that. Another example of so many hyping myths only due to hearsay. Ignore so many silly recommendations that only have you wasting money.

A power system can make any good hardware act defective. Get the meter. Provide those numbers to three digits. Have an answer without doubts. Only then should other suspects be considered.

And finally, the list of suspects for your failure as short. Power system (which is more than just a supply), only a few some motherboard functions (BIOS is clearly not on that list), CPU, sound card, video controller, and drivers for third party hardware. Some of those suspects are quickly eliminate by system log or BSOD messages. Notice so many items not on that list that others wanted to fix. Until you have those three digit voltages, then anything else can act strangely.

Stop trying to fix anything only on wild speculation. Memory, BIOS, surge protector, and so many other speculations were removed from a suspect list at the very start. First learn what is defective. Follow the evidence. Step one starts with a multimeter - and not numbers from a voltage monitor.

I've had other computers have a similar crash issue due to Physical Memory issues, so testing the Memory wasn't just wild speculation.

Also, neither is the Surge Protector. A ten year old Surge Protector could easily have a million age-related issues that would cause problems.

I had every intention of getting a multimeter today and testing it. No, I didn't say it, but it was part of my plan, I'd been looking into how to get it to work and how to test with it for the past few days, ever since the idea that the Power Supply might be the issue came into my head. I wasn't going to wildly replace it.
Besides, what's so wrong with making sure that the rest of my system is in tip top shape? If nothing else, it confirms that it's not whatever is going on AND something else? It never hurts to update drivers, it never hurts to check your RAM and what not, that stuff doesn't hurt.

Yeah, a BIOS update has the POTENTIAL to screw up your system, but I've never seen or heard of a BIOS update actually breaking a computer in my life. Granted I don't go around on the Internet and look for problems others have, but I've talked to many people long before I posted this.

You think I just ignored this issue for months on end? No. It's gotten slowly worse to the point that I had no where else to go. I've talked to about 40 people, I even sent the damn PC into one of those repair places and they basically said nothing was wrong (Which I know isn't true), and when I brought it back 2 hours later they went "Okay get a refund because I have no idea what's wrong."

I'm not making random checks out of my head, I'm literally testing EVERY part. At this point, the PSU is about the only remaining thing outside of outside sources, like a Surge Protector screwing up. You act like a Surge Protector that's 10 years old CAN'T screw up. I've already said on this topic a few posts back that the very same Surge Protector has problems on one outlet where it flips out, how is it so hard that the problem could have spread to other outlets, causing a power distribution error and thus resulting in lack of required power for the system, causing it to crash? No, normally the surge protector does nothing, HOWEVER, faulty wiring doesn't magically seclude itself to the system. A 10 year old SP can easily have its own faulty wiring. You act like that's impossible, but I know from experience with other electronics that faulty wiring can occur anywhere, and can result in a problem at any point. So the Surge Protector being the problem isn't so hard to fathom in my opinion.

Relax, Polantaris, he's just doing it by the book. Some of us have grown beyond that.

Having the bios set the GPU to PCI probably caused a mess for you (pci is substantially slower than PCI-e), which is why I asked you to run RO and see if you had any problems.

Aye. I played RO for about an hour and a half with no issues. I'm going to move on to something that was more frequently causing crashes, to further test it. I remember having issues that Fallout 3/NV would crash my computer in 10-20 minutes, so I'm gonna play that for a bit and see how it goes.

A ten year old Surge Protector could easily have a million age-related issues that would cause problems. ... It never hurts to update drivers, it never hurts to check your RAM and what not, that stuff doesn't hurt.

Yeah, a BIOS update has the POTENTIAL to screw up your system, but I've never seen or heard of a BIOS update actually breaking a computer in my life.

I have seen motherboards destroyed because a BIOS installation failed. Some have backdoors to fix the damage. Others do not. But nothing in a BIOS is relevant once the machine boots. The OS often replaces BIOS code with its own. Just another reason why BIOS is irrelevant once a system boots. Did the BIOS that was working fine so long ago suddenly change? Of course not.

There are no age relevant issues with surge protectors. Surges occur maybe once every seven years. A failed protector does nothing even on 400+ volt transients. Or the protector continues to perform like it always did - just like a power strip that is not a protector. It has zero relationship to do with your symptoms. Later, if curious why, we can go into hard technical details. But today, that is not relevant.

Either semiconductors work just as good as they did when they were new. Or they are failing. But making changes adds new sources of failure. A long list that even includes static electricity. Many semiconductors are so robust at to withstand 2000 and 15,000 volt without damage. But once removed, even 50 volts can do damage. Just another example of why changing things can only complicate a problem by adding variation and change.

An overwhelming majority do not know how electricity works. Just like an overwhelming majority knew Saddam had WMDs. Only person who is relevant in you inquiry is the one who also says why and says so with numbers.

Most of your tests (from what I read) told you almost nothing about the quality of each part. And you are worrying about a power supply when you should be concerned with the entire power *system*. For example, did you know of a power controller decides when the CPU can execute? When the system powers on and powers off? Power system is much more than just a supply. How many also knew that?

Your surge protector, if failed, has converted itself to nothing more than a power strip. And it does absolutely nothing when 120 v power is less than 330 volts. You surge protector is completely irrelevant. A guy who designed surge protection systems also warns you - that power strip protector can sometimes make computer damage easier. And again, that is also not relevant to your symptoms.

If you think the surge protector is causing problems, then plug an incandescent lamp into that power strip. Does the light bulb change intensity when computer is on or off? If yes, then power strip (or household wiring) has a defect. If not, then incoming power is ideal - is not suspect.

Provided was a short suspect list. Everything that can cause your symptoms. Furthermore, memory test is not performed properly if hunting for an intermittent memory problem. We can discuss that later when it comes time to review that other suspect. Until the power system is defined by numbers as "without any doubt good" or "without any doubt bad", then other tests can only provide vague or misleading results.

Again, I am only telling you why you have been so confused. And why 40 people provided insufficient assistance. The concept is called 'follow the evidence'.

In so many posts, I did not see anything from the system (event) logs. Were they reviewed and posted? That is a very first fact collected when suffering intermittent computer problems.

Appreciate what a major consumer magazine discovered. They created minor computer problems. Most repair shops never fixed the problem. A most common solution instead replaced the power supply - that was perfectly fine. To pass the A+ Certified Computer Tech exams, one need not have any electrical knowledge. Well, my computer hardware design experience is numerous decades. I am pointing you in a complete different direction because I have done this stuff even generations ago on hardware that must never fail.

Numbers from the meter (only when the system is fully loaded - multitasking to every drive and port) will result in an answer that says "without doubt". Accomplishment will be measured by the number of items on that list that are "without doubt" good or bad. Currently your list is empty. Even memory is not yet on that list.

Westom you are full of sh!t. Multimeters are not the end-all, be-all of hardware diagnostics. All types of multimeters have inherent problems that decrease the accuracy of gauging the voltage issues in computers. They frequently suffer from graduation issues or from fluctuation readings (or lack thereof if one is using digital). There's an obvious reason why the pc repair industry doesn't rely on these things. They at most give a generous indication of voltage problems, something one could already deduce from understanding the hardware from the get-go.

PS F*ck you for trying to make the rest of us look bad.

PPS Moderm memtests and stresstests w/ logs are damn accurate. LOL @ they don't show intermittent problems. That's a new one. Oh son, yes they do.

My PC crashed playing Fallout, so it's definitely not fixed, however it did take a lot longer than normal to crash (I was playing for a good 30-40 minutes). I wonder if the setting was part of the issue. Regardless, I'm going to test the PSU with a multimeter regardless once I can get a hand on one (I know my father has one, but I have no idea where he keeps it and he's not around right now so I can't do it just yet).

what did event viewer have to say about it? Did you check that?

I'm guessing a kernel power report event 41.

But try to post the details.

This could very well be a GPU problem with drivers. Some NVidia drivers with windows 7 were making people think they had a PSU problem, and all they had to do was remove their drivers and install standard ones. Sometimes updating drivers makes things worse.

If I remember correctly, when I originally started having this issue I looked at the Event Viewer and it said something about Windows Kernel Power.

I'm about to multimeter test the PSU. I started using this guide: http://pcsupport.about.com/od/toolsofthe...imeter.htm
Step 5 says to short out Pins 15 and 16 on the motherboard? What is the deal with this? Why would I do that? I've already disconnected the entire PSU from the system, but when I got to that step I was a tad confused as to why I would do that, so I stopped. Anyone give me insight on that?

Edit: Nevermind, I just reread it, and it said to short out Pints 15 and 16 on the motherboard connector, not the motherboard itself. That makes sense, I guess it's to trick the PSU into thinking the system is on? That's what it seems like when I read to Step 7.

DO NOT PIN SHORT. Leave that to the crazies.

Sometimes you short the pins to make a PSU work without a mobo. Sometimes you do it to jumpstart a failed device. You don't have that problem. It's a risky way to tell if the PSU is just plain useless (such as it powering a computer but not powering it enough for load).

It seems I cannot get the PSU to start without the mobo. I've tried shorting Pins 15 and 16 as suggested in the guide to get it started, however the PSU doesn't start no matter what I try. I took a good look at the Motherboard connector and noticed that Pin 8 is completely missing. Knowing this, it might just be a good idea to replace it anyway. I don't know why a pin is missing altogether, however it cannot be a good thing. It's not like the pin broke either, the entire wire is missing, it's like the PSU was made without it. The pin chart that goes with that guide says:
8 PWR_ON Gray Power Good
I'm unsure what that means, since it does Power on, but it missing it cannot be the best of things.

Not having an 8 pin is okay so long as you don't overclock the hell out of your rig.

Well since I can't test it manually, I'm going to try out one of those PSU Testers. The ones I can get from stores around here aren't too expensive ($10-$20), so it shouldn't be an issue, and I'll probably grab a Surge Protector while I'm at it because even if the one I have now isn't the problem, 10 years old is basically asking for something to happen eventually. Since I already have the entire PC apart, I'm not going to put it back together to do ATITools testing until after I've PSU Tested.

I've tried shorting Pins 15 and 16 as suggested in the guide to get it started, however the PSU doesn't start no matter what I try. I took a good look at the Motherboard connector and noticed that Pin 8 is completely missing.

Do not do that procedure. Defined in my post is what you do. As with any good diagnostic procedure, you disconnect nothing and remove nothing.

Setup to access (multitask to every peripheral (disk drive, sound card, CD Rom, Internet, USB, etc). Then measure any one orange, red, purple, and yellow wire that goest from the power supply to motherboard. The entire procedure should take a minute. And again, disconnect nothing. Everytime you make a change, you add variables and potentially more confusion.

The procedure as was posted here is the best one – the most informative (in part because reason why is not yet apparent).

I hadn't gotten a chance to test fully, however my friend noted that he had a 700W PSU that was unused, still sealed. I asked to borrow it, and I installed it and what not. After the install, I wiped my computer too because of various unrelated reasons, and ever since I've had no issues at all. I ran several different programs that would crash in the past, Fallout 3/NV, RO, and a PS2 emulator, all for over 2 hours each, with absolutely no problems. It seems that the PSU was indeed the issue.

They also have increased performance since the change. For example, with the same settings, one area in the game I tested on the PS2 emulator would drop down to about 45 FPS, but now it's only 55, and Fallout has had an increased performance as well.

Another note is that when I installed the PSU but before the Wipe, I checked out the HWMonitor App again just out of curiosity. Instead of being .10-.15 under the regular voltages, the voltages were .10 above, and COMPLETELY stable. Even in idle on my old PSU, it would fluctuate often, but when I first turned on HWM after installing the new PSU, I thought the program crashed for a few seconds until I looked at the temperatures to see them moving around, but the Voltages are nearly unchanging (Once in a while the VCORE and CPU would move about 0.04, then return to the normal amount).

Seeing these changes, it seems like the PSU was indeed the issue, and since I've run into no issues either. Hopefully it will stay this way.

I'd like to thank everyone for their input and assistance on this issue.

yay! another southperry success story! welcome back to gaming =p

Another note is that when I installed the PSU but before the Wipe, I checked out the HWMonitor App again just out of curiosity. Instead of being .10-.15 under the regular voltages, the voltages were .10 above, and COMPLETELY stable.

Assuming voltages created your problem, well, the voltage monitor did not report accurate numbers. Did not report the problem exists. Another reason why one uses a multimeter. To make it useful, calibrate a monitor with the meter. Then monitor threshold levels can be set to actually report a voltage problem.

Defective hardware is often apparent in voltages even months or a year before any system crash results. That is what a monitor is supposed to avert. But it cannot if you do not first calibrate the monitor.