Windows 7 BSOD (maybe due to low ram?)

unclezhuge

New Member
Joined
Mar 19, 2013
Hello all:

I have been having a few BSOD problems lately. It always seem to happen when I am playing a game, or sometimes even a few hours after I finished playing a game. I think it might be due to low RAM issues, although I am not entirely sure. So i hoped you guys can helped me figure out what is wrong with my PC.

I attached everything that is required by the sticky thread to the zip file. Do let me know if you need any more information.

Thanks!

View attachment uncle_zhuge bsod.zip
 
Thanks for the reply.

I tried to get the BSOD using the method you recommend (3 times), but my system just hangs. No BSODs occured. However, after I deleted the settings of Driver Verifier and wanted to type this post, my system went BSOD. However, when I checked the minidump files, no new file was created.

Also, I liked to point out that sometimes during my BSOD when the BSOD says "Physical Dump Memory", my system just hangs itself even before it reaches 100%.

Also, during the hanging of my PC, my speakers start making a loud buzzing noise. I don't know whether this is significant or not.

Sorry I can't get more BSODs, I will let my PC run overnight with the Driver Verifier settings and see I can get any more BSODs. Is there anything else that I can do?
 
That's not a driver problem or a low RAM problem. That's a hardware issue, and if it's hanging even during a BSOD dump procedure, then it's a pretty serious issue. I suspect the power supply or the CPU is bad, but there's also the chance the hard drive is.

From the looks of it this is a custom PC, yes? Did anyone else build it for you? If you got this from a shop, you should see about them taking a look at the hardware to see what the problem is. They have a lot more tools available to work on it than we do here on this forum.

If you want, we can run through a couple diagnostic tests:

RAM: Memtest86+ - 7+ passes
CPU: Prime95 - Torture Test; Blend; overnight (9+ hours)
Drives: Seatools - All basic tests aside from the Fix all, Long generic or the advanced ones.

All of these are included in the UBCD if you prefer a Live CD environment (which is the best environment to test hardware on). Note that Prime95 currently does not work on the latest version of UBCD due to a bug. Also, please provide us temps/voltages using HWInfo with Sensors only option checked. Log two 30-minute instances: one for idle, and one for high load. If you can get the system to crash during high load logging, that's even better. Last, make sure that when running Prime95 that you check temps during first 30minutes or so to see if temps are stable. Prime95 runs very hot so be sure your cooling can handle it before letting it run extensively.
 
Btw, you may wanna start backing up all your data as your system is in an unreliable state and may get worse. Do it before the hardware diagnostic tests.
 
Ok thanks. I have backuped most of my data. Also, this is a custom PC. I have took my PC to a shop to have a look at it, but that was like last year October or so. Back then, they just told me that my PC is still usable, and recommend me to just get more RAM.

I have some questions regarding the stuff you need me to do though.

1) When you say high load, do you mean I just run my PC with high RAM programs? (Like gaming?)

Also, one more thing is that in the past, my PC often comes out with the below picture. The problem comes and go, and recently it stopped.

BIOS-CMOS-Checksum-Bad.png
 
Yes, gaming, benchmarks, Prime95, anything that taxes the CPU/RAM heavily.

Hmm, that CMOS thing is a concern. You'll want to replace the CMOS battery, which is the round disc battery located on the motherboard. This will revert settings on your motherboard to defaults, but they should be at default anyways for the most part. Of course, turn the system off when removing/replacing the battery.
 
Ok I will get around replacing the CMOS battery. I should be able to give you the HWinfo tests soon, but the rest will have to wait for a while.

Btw, based on the state of my PC is in right now and the information you know, are there still any possible remedies? (Like just replacing certain parts of the PC). Because I have been considering to get a new PC, but I still want to give this old one a try. Can you tell me what is the best case and worst case that could come out from after I did all the tests?

Also, I managed to get another BSOD when I was trying to type this post.

View attachment 032013-15646-01.dmp
 
The problem is we don't know what hardware is causing it, and replacing hardware will be a shot in the dark until we get a better idea what's causing it, which is what these diagnostics tests will - hopefully - turn up clues on. I personally am eyeballing the power supply, CPU or hard drive as potential suspects, but unless you're willing to replace em all right now, it's best to just go through the tests first and see what they may come up with before we start swapping out hardware.
 
Sorry for the inactivity for the last few days. I managed to get the HWInfo done.

Regarding the seatools, I ran into a problem. The tests from the UBCD only have the following options:
Short test
Long Test
Acoustic test
(I picked both the GUI and Text interface, and both only have these options)

I did all 3. The Seatools put me as passed for both the short test and long test. It didn't really say anything for the acoustic test. I am not so sure whether u want the logs for the seatools or not. But then I am also not so sure where are the logs saved for the seatools file (I am using the UBCD)

I will try the memory test tonight. I will do the prime test last.

View attachment HWINfo.zip
 
CPU and Aux temps get up into the 70's under load. GPU temps get into the 80's under load. Not a real bad situation, but I would investigate the cooling solution inside your case. Make sure you get lot's of airflow to the CPU and GPU coolers.

More free hardware diagnostics here: Hardware Diagnostics
 
I used my PC in a room without an air-conditioner, so sometimes even the room temperature gets rather warm. These few weeks are getting pretty hot in my country (I lived in Singapore)

I always thought my computer gets very cranky due to RAM. There are times where I have played like 5-7 hours of gaming continuously and nothing happens. Then there are times when 5 minutes of gaming will instantly crashed my com.

I am not very good with investing what happens inside my CPU though. :( I don't even know how many fans I actually have inside my CPU. (I think it is 1?) Any suggestion to improve my cooling system then? Or any cooling system that you recommend for my system?

I will try out the link you suggested. :)
 
Ok guys, so sorry for the long absence. I got a problem that I need to settle with my relatives, so I couldn't do the tests as promised.

Right now, I have done the MemTest86+. It crashed when the screen shows 3 passes. It then said 712 errors. Below the text it said "Unexpected Interrupt-Halting". I couldn't do anything to resume the tests. Duration of mem test is about 2 hours and 20 minutes.

I am still waiting for my com to finish the Prime 95 test. It has now passed Self-Test 16k. Almost 2 hours have gone by. I also checked the temperatures, it recorded like 88-95 degrees celcius for the CPU Core #0 to #3. (The previous high load test that I did for gaming only hit a max of 70 degrees celcius). Is this normal? No crash has occurred for the Prime 95 test yet.

Again, so sorry for the lack of reply.

EDIT: I have opened my CPU casing while doing the Prime95 test. Also using my house fan at full blast to hopefully get rid of the heat generated.
 
Last edited:
Ok, tried to finish the Prime 95 test. It crashed after 2 and a half hours. The last message that appeared was this:
180,000 Lucas-Lehmer iterations OFM565247 using Pentium 4 type 1-FFT length 32 k, Pass 1-128, Pass 2=256

What now?
 
Those temps
I am still waiting for my com to finish the Prime 95 test. It has now passed Self-Test 16k. Almost 2 hours have gone by. I also checked the temperatures, it recorded like 88-95 degrees celcius for the CPU Core #0 to #3. (The previous high load test that I did for gaming only hit a max of 70 degrees celcius). Is this normal? No crash has occurred for the Prime 95 test yet.

are far over what you're supposed to be getting, even on Prime95.

The fact that you're living in a place where it tends to get warm also makes me think you're dealing with temperature problems.

Could you take a picture of the inside of your case (try and get the full picture)?

(Extra note: Looking at your dumps, is your processor overclocked? From my point, you've got speeds higher than stock - if so, could you return them to stock levels for now?)

Memtest and dumps:
Memtest errors suggest faulty RAM. The dump files I've gone through so far all reference memory issues (corruption, problems etc.)
Do you have any spare RAM sticks lying around? If so, swap them around and give it a test.
IF this works:
Put in one stick of your original RAM set. Run memtest, and see if that gives any errors. If yes, that stick is bad. If no, that stick's good.
Repeat with the other stick/s - any that show errors should be the ones causing the problems (and if all of them do, I'd just get a whole new set).
 
Those temps

are far over what you're supposed to be getting, even on Prime95.

The fact that you're living in a place where it tends to get warm also makes me think you're dealing with temperature problems.

Could you take a picture of the inside of your case (try and get the full picture)?

(Extra note: Looking at your dumps, is your processor overclocked? From my point, you've got speeds higher than stock - if so, could you return them to stock levels for now?)

Memtest and dumps:
Memtest errors suggest faulty RAM. The dump files I've gone through so far all reference memory issues (corruption, problems etc.)
Do you have any spare RAM sticks lying around? If so, swap them around and give it a test.
IF this works:
Put in one stick of your original RAM set. Run memtest, and see if that gives any errors. If yes, that stick is bad. If no, that stick's good.
Repeat with the other stick/s - any that show errors should be the ones causing the problems (and if all of them do, I'd just get a whole new set).
[/COLOR]
Thanks for the reply.

1) Photos of inside my PC. (The whole album is about 10 MB, sorry forgot to compress. I took about 5 pictures.) I counted two external fans.
Inside my PC - Imgur

2) About overclocking:
I can't really tell whether my processor is overclocked or not. I attached the CPU-Z images that I took during my first post here again, maybe you can see whether is it overlocked?

all slots.jpg

3) Memtest:
I don't have any spare RAM sticks lying unfortunately. Can I just go ahead and do the testing of one memstick at a time? (I got like two mem sticks inside my PC right now, i think.)
 
Last edited:
Alright:

For the inside of your PC:

4 total fans - one for the CPU (boxed), two for the case (default with the case I assume) and one for the Power Supply.

First:
Replace the CPU cooler. It'd cost you about $80 AUD (not sure the conversion to your currency) to get a decent aftermarket fan cooler for your CPU.
If you're unable to get a new cooler for various reasons, have you tried cleaning the heatsink and fan, and reapplying thermal paste? If not, buy a bottle of compressed air from your local electronics shop - remove the fan (should be screws on each of the 4 corners) and blow the nozzle under each of the fan spoke thingies and between the fins of the heatsink. You can also get a tube of thermal paste for about $5 from the electronics shop too.
To reapply thermal paste:
1: Clean the processor and underside of the fan (which should be the two points that touch each other) with a good cleanser (eg: ArctiClean). You can also use a solution with a high-percentage amount of alcohol in it.
2: Apply a few drops on to the processor and the bottom part of the fan, and leave it there for about 5 minutes.
3: Using a lint-free cloth, wipe the surface. Repeat until both are fully clean and shiny.
4: Lock your processor in the socket.
5: Get your tube of thermal paste and apply a dot of it on to the center of the processor (about the size of a baby's fingernail).
6: With the cooler, press it straight down on to the processor.
7: Lock down the cooler (with screws, pins, etc.).
Done.

For the extra case fans on the sides - using compressed air clean those two, and check the dust filters on the out side of the case too.

Not sure about the make and model of the PSU (looks like a Revolution 85+?).
Check the PSU fan for any dust buildup and clean that as well.
Looking at the dust filter right next to the PSU - that also seems dusty, give that a check and clean it if necessary.
Think of any other entry points for air in your case (my guess would be under the drive bays in the front, but there could be more) - check those for dust and clean as necessary.

Seems I made a mistake in the CPU overclocking - doesn't look overclocked to me, and seeing since this is a custom build it probably wasn't overclocked in the first place.

And yes, you can go ahead and do the testing with the mem sticks. I just suggested different sticks of RAM in case it might be your RAM slots that are faulty.
 
Yeah, stock CPU coolers are never that great. There's also the potential that it was not installed correctly, especially dealing with applying the thermal paste. Follow the instructions Hining gave you.

Btw, while Memtest displayed plenty of memory errors, the unexpected interrupt means an internal CPU problem, so the errors may have actually originated from the CPU, not the RAM. Unfortunately, I'm worried that your temperature issue has caused permanent damage to your CPU, and no amount of cooling will fix that. There's the possibility you will need to replace the CPU anyways.
 
Back
Top Bottom