Tech Support banner

Status
Not open for further replies.
1 - 16 of 16 Posts

·
Registered
Joined
·
7 Posts
Discussion Starter #1 (Edited)
Hello everyone. A few months ago, my old motherboard died, and I wound up getting 4 new parts since I was planning on upgrading anyway. Since then, I have been experiencing two issues related to system stability (I am not sure if they are related or not), and I have not been able to successfully troubleshoot either of them. This will be a long-winded post because I have gathered a lot of data and observations and tried many things to troubleshoot these problems. If you would rather not read a short novel, I would suggest leaving immediately. If you wish to stay and help, thank you very much. Here is a list of my current system specs:

NEW PARTS:
CPU : Intel(R) Core(TM) i5 CPU 750 @ 2.67GHz (Old CPU was a Q6600 quad core)
Motherboard: Asus P7P55D EVO (Old MB was an Asus P5K Deluxe)
RAM: 3x 2GB OCZ DDR3 Memory @ 1333MHz (I am aware this is more RAM than XP can use, I will go into that later in this post)
GPU: NVIDIA GeForce GTX 275 Lightning (1792GB version) - Upgraded from 8800 GTX

OLD PARTS
HD1: 74GB Western Digital Raptor
HD2: 500GB Western Digital Caviar
Optical Drive: Memorex DVD+-RW, CD-RW
Sound Card: Creative Sound Blaster X-Fi Gamer
800W PSU (I can't seem to find the make and model of it)

THE PROBLEMS:
Problem number 1: Random computer freezes. As stated, these freezes happen seemingly completely at random, whether I am gaming, listening to music, browsing the web, or even when the computer is idle (sometimes I will come back to the computer after being gone for a few hours and it will be locked up). Everything freezes for about 5 seconds (mouse/keyboard input, and the screen itself), then it unlocks for another 5 seconds, and then it freezes again permanently until I physically hit the reset button on the computer case. The sound loops continuously while it is frozen. This may happen 2-3 times a night, or 2-3 times a week. It is totally random.

Problem number 2: Somewhat less random computer restarts. This seems to happen only when I am playing videogames on the computer. I will be in the middle of a game, and then the computer will restart as if someone turned the power off and then on again; black screen followed by re-boot. Adding to my puzzlement is the conditions in which the crashes occur. There are some parts of certain games where this crash will happen almost all of the time. However, it will happen completely randomly in other parts of the games as well. Normally, when confronted with a problem like this (where it crashes at a specific point in the game), I would assume it was the game's fault. However, the crashes also happen in a multitude of games at random times, so instead I must suspect that my computer is at fault.

TROUBLESHOOTING:
Here is a list of everything that I have done to try to diagnose and resolve the issue(s).
1.) Check for heat. Heat does not appear to be the problem. I have a massive heat sink with a 120mm fan attached to my CPU that keeps temperatures well under 40C even when it is under heavy load. My GPU temperatures (under load) also remain at around 40C. There is good airflow in my case, and my motherboard sensor remains at even lower temperatures. I do not believe heat to be the cause of any problems.

2.) Uninstall/reinstall latest drivers. I downloaded the latest drivers for my current video and sound cards. I uninstalled my drivers, re-booted the computer, installed the up-to-date drivers, and rebooted again. This did not resolve the issue

3.) Memory diagnostics. I let Windows Memory Diagnostic run overnight and it discovered no problems. I have also tried every combination of my three sticks of RAM (including running them one at a time), and found that the problems persisted no matter which stick(s) I was using.

THINGS I HAVE NOT TRIED AND WOULD LIKE ADVICE ON:
1.) BIOS update. I am currently using the following BIOS for my Asus P7P55D EVO motherboard: BIOS : American Megatrends Inc. 0209 (07/28/2009). The newest version is 1207. I have heard that updating the BIOS could be potentially harmful - is this true?

2.) Re-installing Operating System. Would it be possible that the problems might go away if I were to re-install Windows XP Pro, or upgrade to Windows 7?

3.) Getting a new PSU. Even though mine has more power than what is needed for my system, I have heard that older PSUs can lose their stability over time. However, considering a good PSU upgrade would cost a considerable amount of money, I would only do that as a last resort unless someone with a lot of experience is confident that this is my problem.

Thanks so much for reading (if you haven't dozed off by now). *Any* advice to help me with this situation would be greatly appreciated, as I have been pulling my hair out for months trying to figure this out.
 

·
TSF - Enthusiast
Joined
·
7,670 Posts
Re: Extremely difficult issue(s) to troubleshoot

Your symptoms point strongly toward your noname power supply failing.
BIOS: If a BIOS upgrade addresses an issue that you are having, then by all means upgrade it. A failed upgrade usually means replacing the board.
Reinstall OS: Its possible but I highly doubt that it would make a difference.

a good PSU upgrade would cost a considerable amount of money
Hmmm . . .
Intel i5 750 - $200
Asus P7P55D EVO - $200
3x 2GB OCZ DDR3 Memory - $160
GeForce GTX 275 Lightning - $300

Corsair CMPSU-850TX $130

It's still one of the least expensive components.
 

·
Registered
Joined
·
7 Posts
Discussion Starter #3
Re: Extremely difficult issue(s) to troubleshoot

Drat, I was hoping you wouldn't say that. Thanks for the advice.
 

·
TSF - Enthusiast
Joined
·
7,670 Posts
Re: Extremely difficult issue(s) to troubleshoot

Check your voltages in BIOS and/or with accurate monitoring software. They should stay within +/- 5% of nominal. Voltages even moderately out of spec will cause a lot of grief, including damage to other components.
 

·
Registered
Joined
·
7 Posts
Discussion Starter #5
Re: Extremely difficult issue(s) to troubleshoot

Check your voltages in BIOS and/or with accurate monitoring software. They should stay within +/- 5% of nominal. Voltages even moderately out of spec will cause a lot of grief, including damage to other components.
I was not aware of that rule of thumb. I just downloaded PC Wizard, and found the following results for my voltages:

Voltage CPU: 0.90V (I am not sure what the optimal voltage is supposed to be here)
+3.3V Voltage: 3.44V
+5V Voltage: 6.12V
+12V Voltage: 11.62V

So, my 3.3V voltage is off by +4.2%,my 5V voltage is off by a whopping +22.4%, and my 12V voltage is off by -3.2%. Would I be correct in assuming that the error margin of my 5V voltage means I need a new PSU?
 

·
TSF - Enthusiast
Joined
·
7,670 Posts
Re: Extremely difficult issue(s) to troubleshoot

6 volts on the 5 volt line can permanently damage other components. Your GPU requires a solid, stable 12 volt source. You can start gettting artifacts around 11.5 volts. Confirm the voltages in BIOS or borrow a digital multimeter.

By the way, +/- 5% has always been a quasi standard tolerance for voltages. Personally, I don't like seeing more than a 2% variation, but I can be anal.
 

·
Registered
Joined
·
7 Posts
Discussion Starter #8 (Edited)
Re: Extremely difficult issue(s) to troubleshoot

Thanks for the info. I do not have access to a digital multimeter, but I checked the voltages in my BIOS and got much different results.

According to my BIOS, I have the following voltages:
CPU - 1.15V (as opposed to PC Wizard's 0.90V reading)
3.3V - 3.44V (the same as PC Wizard's result)
5V - 5.20V (*much* lower than PC Wizard's result, and within the +/-5% range)
12V - 12.32V (higher than PC Wizard's 11.62 result, but still easily within 5%.)

Edit: After restarting my computer, PCWizard is still reporting about the same values as the first time (0.94V CPU, 3.3V @ 3.46V, 5V @ 6.12V, and 12V @ 11.62V). Would the voltages change because there is increased load when Windows is running, or should I assume that the software results are inaccurate?
 

·
Enthusiast
Joined
·
16,762 Posts
Re: Extremely difficult issue(s) to troubleshoot

The BIOS is more trustworthy than pc wizard however your symptoms do point towards a crappy power supply.

As said by Tyree there will be a label on the psu indicating make model wattage and amperage.

Freezing, lockups and random restarts are associated with:- faulty or crap power supply, faulty ram or over heating.

download memtest 86, burn it to disk and boot from that disk, run memtest for several passes with only one stick of ram installed for every test (test each stick about 7 times). Memtest runs contunuosly until you tell it to stop. check your heatsink and see if it is completly locked in and does not move and also blow out any dust with compressed air.

if the heatsink does move, you need to take it out clean the heatsink and cpu with 90% isopropyl alcohol and reapply some fresh paste, then make sure the heatsink is attached properly.

if memtest brings up any errors whilst testing a stick of ram then that ram is faulty and needs replaced.

find the info on your psu and we can tell you if it is the likely cause. Wattage isn't the only important factor when using a psu. Amperage is more important and if your psu is crap brand then it may not have sufficient amperage to keep your system stable.

A crap psu can seem fine for ages then you will see lots of different errors and problems happening. A crap psu can also cause other components to be damaged.
 

·
Registered
Joined
·
7 Posts
Discussion Starter #10
Re: Extremely difficult issue(s) to troubleshoot

After some poking around, I found the label. Here is what it says:

Zephyr 750W (not 800 as I thought, but still more than enough for my current parts).

12V 1, 12V 2, 12V 3, 12V 4 = 19.0A
3.3V = 24.0A
5V = 30.0A
-12V = 0.5A
+5Vsb = 3.0A
 

·
Registered
Joined
·
25,116 Posts
Re: Extremely difficult issue(s) to troubleshoot

A) after you run memtest on each stick of ram (solo stick testing only) use only the memory slot closest to the cpu socket >>>>use only bootable memtest / NOT windows memory tester
http://www.memtest.org/#downiso

B) then if no memory errors are found, I would update your bios ........use ONLY the bootable flashing methods / NEVER flash your bios from within a windows environment (use bootable floppy or USB stick; review mobo manual)

C) I would then run the free program OCCT (its a system stress test) monitor the readouts from within OCCT for cpu voltage / 12volt fluctuations / heat fluctuations >>>run that for about one hour with close observations of the voltages, temps etc, does the system crash during OCCT run?
http://www.ocbase.com/perestroika_en/index.php?Download

D) if the actions above have not revealed a hardware problem; then chasing the problem as a driver issue is worth a shot. Perform a win XP "repair" install. The repair install will remove ALL windows sytem drivers without messing with your programs or data; upon completion you will re-boot the system then install ALL drivers again, install drivers in this order:

http://www.michaelstevenstech.com/XPrepairinstall.htm#RI

1) motherboard chipset drivers

2) lan or modem drivers

3) sound drivers

4) windows updates

5) netframework update / direct X update

6) any other missing drivers

7) always save video card drivers for last

8) dont get drivers from Microsoft update unless its a last resort!


post back with your observations



***** question


after you replaced your motherboard; did you perform a win XP repair install ? did you install the new motherboard chipset drivers?
 

·
Registered
Joined
·
7 Posts
Discussion Starter #12
Re: Extremely difficult issue(s) to troubleshoot

I did not do a repair install, but I did install the motherboard and chipset drivers. I'm going to be fairly busy with schoolwork for the next few days, but I will go through your troubleshooting guide step by step as soon as I have the time. Thanks for all of the help.
 

·
Registered
Joined
·
7 Posts
Discussion Starter #13
Re: Extremely difficult issue(s) to troubleshoot

Followup: A new PSU fixed everything. Thanks again for all of the input.
 

·
Registered
Joined
·
14 Posts
Re: Extremely difficult issue(s) to troubleshoot

if the heatsink does move, you need to take it out clean the heatsink and cpu with 90% isopropyl alcohol and reapply some fresh paste, then make sure the heatsink is attached properly.
I would like to clarify something here. You mean wash the entire CPU Core (ie:core2duo processor itself?) with 90% Isopropyl alcohol? Sorry, I'm a newbie in tech stuff. Thank you for your time. Best Wishes to everyone.
 

·
TSF Team Emeritus
Joined
·
6,147 Posts
Re: Extremely difficult issue(s) to troubleshoot

Leave the CPU in the socket, but clean the top (the bit that contacts the heatsink/fan) and the plate on the heatsink with the Isopropyl. Make sure you use good quality thermal paste (Artic Silver is good) when re-applying the heatsink.
 
1 - 16 of 16 Posts
Status
Not open for further replies.
Top