Sign in with
Sign up | Sign in
Your question

BSOD and other problems.

Last response: in Graphics & Displays
Share
February 28, 2010 2:21:04 PM

Hey Everyone,



I keep getting random crashes on my computer and i am not entirely sure how to figure out what is causing them. Here are my specs



intel E8400

Asus motherboard p35( i forget the exact model im at work)

EVGA 8800 GTS 512

4 gig Gskill ram ddr2 1066 (i think thats the speed)

Seasonic 550w psu

Windows 7 (but problems have been happening even when i had XP)



A variety of different problems have been happening with my computer. The first of which is i keep getting BSOD when i am browsing IE. I will be looking at a website or using Itunes and i get a BSOD all of a sudden. No reason.



The second problem i keep having is i will be playing a game and the game will freeze and the sound will stutter. Most of the time the freeze will last 5 seconds and go back to normal. Sometimes it will unfreeze, but my FPS goes down to a crawl. And rarely when the game unfreezes it will come back very pixelatted and i will have to restart my comptuer for it to go back to normal.



Another thing that happens is lastnight i was listening to music and browsing the itnernet and all of a sudden my screen went place and the computer froze. I had to restart to get it back to normal.



My video card is not Overheating, I already checked this. The PSU should be able to handle my system without a problem. Not sure what else to check here to see what the problem is. I am leaning towards the video card.

More about : bsod problems

a b U Graphics card
February 28, 2010 2:51:36 PM

I would recommend you go ahead and get on BIOS look for DRAM settings, Configuration, latencies, and voltage.

Please manually set up your RAM, if you do not know how to do this then google is your friend.
You need to change voltage to manufacturer settings.
Check latencies to make sure they are at stock settings
make sure clock speed is set at 1066 mhz.

You need to find the precise ram model to find the voltage/latency information as even 1-2 digits in latency can result in crashes.

Let me know how it goes.
February 28, 2010 4:40:44 PM

i will try this when i get home tonight. thank you
Related resources
March 5, 2010 1:39:23 PM

So i checked my memory setting and they are set to factory requirements. I tested my system useing Memtest and it has passed with no errors for 20 passes. I did use Windows memory diag and that did find an error and stop the test every time. My comptuer is now getting Bsods more frequently, only when im using IE. It is also turning to a black screen more and more often.

I would hate to start buying replacement parts and finding they arent what i need. Not sure where to go from here to diagnos this problem.
a b U Graphics card
March 5, 2010 3:05:10 PM

mennovh said:
So i checked my memory setting and they are set to factory requirements. I tested my system useing Memtest and it has passed with no errors for 20 passes. I did use Windows memory diag and that did find an error and stop the test every time. My comptuer is now getting Bsods more frequently, only when im using IE. It is also turning to a black screen more and more often.

I would hate to start buying replacement parts and finding they arent what i need. Not sure where to go from here to diagnos this problem.


Run Prime95, tell me if you get a calculation or "checksum" error. If you do, the ram is most likely not stable. Double check latencies, and voltages.

Also have you over clocked the processor?

How old is the 8800 graphics card?
March 5, 2010 4:31:21 PM

AsAnAtheist said:
Run Prime95, tell me if you get a calculation or "checksum" error. If you do, the ram is most likely not stable. Double check latencies, and voltages.

Also have you over clocked the processor?

How old is the 8800 graphics card?


I will run prime 95 tonight when I get home. The processor is not OC'd.

The graphics card is about 2 years old now. Latencies are what the specs say and in monitoring the voltages are where they should be as well.
a b U Graphics card
March 5, 2010 5:36:34 PM

Quote:
must be temp issues with the GPU. checked temps??


Yeah check out the temperatures. You can use software like HWMonitor by CPUID to check it out.
March 6, 2010 2:30:35 PM

Ok so i have been runnign prime 95 all morning and i keep getting this error.

FATAL ERROR: Rounding was 0.5, expected less than 0.4
Hardware failure detected, consult stress.txt file.

First time it was on core 2 and the second time it was on core 1. My temps on my CPU at full load are 51c. Which is acceptable. I will test my GPU later at stress levels to see what the temps are on those. My GPU at idle is at
70c.

I already cleaned out my case of dust and air flow is good.


The PSU is working fine as there is no dip in voltage when running stress tests.

This is frustrating as i dont have parts to swap in and out to test if things are working or not.

I would love to just go and upgrade, but i like to upgrade every other generation. Will post temps on gpu later if i have the time today.
a b U Graphics card
March 6, 2010 3:20:47 PM

mennovh said:
Ok so i have been runnign prime 95 all morning and i keep getting this error.

FATAL ERROR: Rounding was 0.5, expected less than 0.4
Hardware failure detected, consult stress.txt file.

First time it was on core 2 and the second time it was on core 1. My temps on my CPU at full load are 51c. Which is acceptable. I will test my GPU later at stress levels to see what the temps are on those. My GPU at idle is at
70c.

I already cleaned out my case of dust and air flow is good.


The PSU is working fine as there is no dip in voltage when running stress tests.

This is frustrating as i dont have parts to swap in and out to test if things are working or not.

I would love to just go and upgrade, but i like to upgrade every other generation. Will post temps on gpu later if i have the time today.


70c idle on GPU? Dude that is horrible. Up fan speed it to like 100% and check again.

Also that is definitely a RAM error or a processor instability error (or motherboard frequency fail too but your not over clocking)
FATAL ERROR: Rounding was 0.5, expected less than 0.4
Hardware failure detected, consult stress.txt file.

I doubt it's a processor instability because the processor is not over clocked. So that leaves RAM.
And yes I was precisely looking for to mention that only one of the processors fail first, because it often matters which core accessed the moment the ram became unstable. Thus only one core shuts down.

I would like you to go into BIOS again and change the Latency's voltage/speed to what I am about to give you:

Voltage 1.9v-2.0v (try lower first, if error persists go to 2.0v)
Speed (Frequency): 800 mhz
Latencies: CAS(CL)=5 tRCD=5 tRP=5 tRAS=15

If this fails. Your going to need to test one stick at a time. It may be 1 bad module.
March 6, 2010 4:29:09 PM

Ok i set the GPU fan to 100% and now its idleing around 49c, but its really loud now.

I manually set the dram voltage to 1.9v and im running prime 95 now.

Seems since i manually set the voltage on my ram the cpu temp has crept up to load around 53-55c now on full load.

Gonna run out for a while and let prime 95 run and see if more errors pop up. Last time it took about a hour for an error to show.

I will update with results later.
Thank you for the help, appreciate it!
a b U Graphics card
March 6, 2010 6:39:09 PM

mennovh said:
Ok i set the GPU fan to 100% and now its idleing around 49c, but its really loud now.

I manually set the dram voltage to 1.9v and im running prime 95 now.

Seems since i manually set the voltage on my ram the cpu temp has crept up to load around 53-55c now on full load.

Gonna run out for a while and let prime 95 run and see if more errors pop up. Last time it took about a hour for an error to show.

I will update with results later.
Thank you for the help, appreciate it!


The GPU may be loud but 70c idle is really bad even on the 8800 GTS. So endure the loudness or consider buying an after market cooler. This will void your warranty, but I suspect you probably dont have any since it's a rather old card.

Let know how it goes.
March 28, 2010 12:22:15 PM

Ok so here is an update. I have taken out a module of ram one at a time and used my computer over the last week as i normally do. I have gotten a BSOD every single time when i have swapped out a stick of ram. So either all the ram is bad or it is another problem. Which is very well may be. I hate to start buying spare parts to only find i wasted my money.

Not really sure where to go from here. I also tested the ram stick by stick in Windows memory diagnostic and every stick failed the testing. The testing would freeze at 22% each time.

This is very frustrating and time consuming. Weird thing is i have never had an error in Memtest 86. Any Suggestions would be great at this point.
!