Solved Getting BSOD with these overclocking settings

June 21, 2012 at 11:01:00
Specs: Windows 7, E8200/2x2 GB Kingston 800 mhz
Hello,
I am trying to use my CPU as 3.2 GHz and I can use it with these settings below. However, the PC sometimes gets BSOD especially after playing a game for a long time (error code: 0x000000124). How can I fix that?

Here are the pictures of BIOS settings:

http://oi50.tinypic.com/ie1n34.jpg
http://oi48.tinypic.com/v3ggn6.jpg

My PC's specs:

Motherboard: Asus P5KC
CPU: E8200
RAM: 2x2 4GB Kingston 1.8V (both same)
GPU: ATI HD5770
PSU: OCZ ZS 650W 80+
OS: Windows 7 Ultimate x64


See More: Getting BSOD with these overclocking settings

Report •


✔ Best Answer
June 22, 2012 at 07:45:22
Test it for 24 hours to be sure, if it doesn't give any errors, the instability issue is solved.


#1
June 21, 2012 at 12:15:32
Try updating the P35 chipset drivers: http://downloadcenter.intel.com/Def...

Report •

#2
June 21, 2012 at 12:20:17
Try stressing the CPU with prime95- javascript:void(fileWindow('84aa92208050bf61b75f20c454fd87a7'));.

If it still BSODs, then the CPU is unstable, try turning up the CPU voltage up A LITTLE BIT, not alot, maximum operating voltage is 1.3625V, your current is 1.3125V and 2.66GHz to 3.2GHz is a big jump in frequency, try turning it down to 3GHz.


Report •

#3
June 21, 2012 at 13:51:56
I couldn't find P35 chipset driver anywhere.
I did prime35 test and, worker #1 didn't give any errors for 1 hour but worker #2 gave this:
"FATAL ERROR: Rounding was 0.5, expected less than 0.4
Hardware failure detected, consult stress.txt file
Torture Test completed 5 tests in 8 minutes - 1 errors, 0 warnings
Worker stopped" and it didn't BSOD but after the last BSOD, i turned up the CPU voltage from 1.3000 to 1.3125V as right now.


I did prime95 test again and worker #2 didn't complete any tests and gave the same errors above.


Report •

Related Solutions

#4
June 21, 2012 at 15:26:49
Understanding All Voltage Configurations from the Motherboard
http://www.hardwaresecrets.com/prin...

You know the real meaning of peace only if you have been through the war.


Report •

#5
June 21, 2012 at 20:13:04
If another small bump to the Vcore does not do it, you can try a very small bump to the chipset voltage (northbridge) or back down on the OC for now.
Make sure that you constantly monitor your temperature during the process of establishing stability, especially after you raise any of the voltages. Check your temperature at idle, normal use, during stress testing, and during long gaming sessions. The stress test temperatures are normally going to be higher than normal operating temperature so if they are at least on the safe side, that is fine. If you do game for hours at a time, this temperature is going to be important to you. If it is high, you are going to have to manage that or back down on your OC until you do since this can cause your BSOD's and long term damage to your components. What you need to do will depend on which temps are high and what cooling components you currently have.

You have to be a little bit crazy to keep you from going insane.


Report •

#6
June 21, 2012 at 20:14:36
Also:
Run Memtest to make sure that your issues are not related to your memory since that is definitely a possibility.

You have to be a little bit crazy to keep you from going insane.


Report •

#7
June 22, 2012 at 04:24:43
I turned down the CPU core speed from 3.2 to 3.0, didn't change the Vcore and did prime95. The test didn't give any errors for 2 hours.
Is the problem possible to be solved?

Report •

#8
June 22, 2012 at 07:45:22
✔ Best Answer
Test it for 24 hours to be sure, if it doesn't give any errors, the instability issue is solved.

Report •

#9
June 22, 2012 at 10:04:08
thanks for all help.

Report •

#10
June 22, 2012 at 10:08:30
Your welcome Fraude.

Report •

#11
June 22, 2012 at 11:29:11
I disagree with backing off on the overclock. In fact, you should have no problem running that CPU at 3.4GHz (8.5 x 400MHz). You should be bumping up the CPU voltage as the others have said, not lowering the multiplier. You stated the voltage is at 1.3125V, Intel states the max safe voltage is 1.3625V, so you have a LOT of room to play with.

http://ark.intel.com/products/33909...

Did you google the error message? It seems to be related to the chipset, so possibly bumping up th northbridge voltage will do the trick.


Report •

#12
June 22, 2012 at 11:41:56
To Riider, i think he was worried about burning out the CPU.

Report •

#13
June 22, 2012 at 11:44:14
You can't burn out a CPU by overclocking unless you overvolt the hell out of it.

Report •

#14
June 22, 2012 at 11:50:42
Or you don't have an adequate cooling system.

Report •

#15
June 22, 2012 at 13:17:21
Well, my case is Xclio godspeed one advanced and my CPU temperature is about 40C when it is idle, while doing prime95, it reaches 67C max and while playing a game that you need good system specs to play, it is around 55C. Are these temperatures normal?

@Riider
Yes, I did google the error code and it sounded like a vcore problem to me. After that, I started bumping the vcore with ~1.2 and every time when I bumped the vcore, it took longer to BSOD. finally it didn't BSOD for three days till today. Therefore, I think it is related to vcore but I am new with overclocking you guys know the best.


Report •

#16
June 22, 2012 at 13:18:50
Thats a good temp, especially when overclocked, nothing to worry about.

Report •

#17
June 22, 2012 at 13:23:30
So, Should I turn up the FSB to 400 and keep bumping the Vcore bit by bit until it is stable?

Report •

#18
June 22, 2012 at 13:32:32
You could try putting it back to 3.2 GHz and putting the Vcore voltage higher until its stable BUT remember no higher than 1.3625V.

Report •

#19
June 22, 2012 at 13:34:19
"Should I turn up the FSB to 400"

It's not the FSB, it's the CPU frequency. The FSB would be 1600MHz. I figured you just lowered the multiplier from 8x to 7.5x? Why would you mess with the optimal CPU/RAM frequency settings? Leave it at 400MHz.

BTW, you can NOT overheat a CPU to death. All modern CPUs have built-in thermal protection circuitry. If the CPU reaches a certain "trigger temp", it will throttle back to try to lower the temp. If it continues to get hot, it will simply shutdown. But it will NOT fry.


Report •

#20
June 22, 2012 at 13:36:39
To Riider, you could still shorten the life of the CPU.

Report •

#21
June 22, 2012 at 13:41:45
When I said FSB, I meant FSB Frequency as it is seen in the picture I have shared at fist. When I make it 400, my CPU core speed gets 3.2. Now it is 375 and core speed is 3.0.

Report •

#22
June 22, 2012 at 13:43:31
Isn't the FSB frequency 1333MHz stock.

Report •

#23
June 22, 2012 at 13:49:15
yes it is. Before overclocked, the FSB frequency was 333 which means 1333MHz I think.

Report •

#24
June 22, 2012 at 14:37:45
"To Riider, you could still shorten the life of the CPU"

How long have you been doing this? Even if it does "shorten the life of the CPU" (which I don't believe it does), the CPU would be outdated well before it dies from a shortened life. I have a 15 yr old Pentium 166MMX (2.5 x 66MHz) that's still alive & kicking at 250MHz (2.5 x 100MHz). It rarely gets used anymore but it still works. I've overclocked every single CPU I've owned since the mid 90's & not a single one of then died an early death. And some of them I pushed pretty hard. As far as I know, they're all still alive.

Follow the White Rabbit: Alive And Kicking


Report •

#25
June 22, 2012 at 14:50:05
Did you have crappy cooling?

Report •

#26
June 22, 2012 at 15:20:49
Stock heatsink/fan, one rear exhaust fan.

Report •

#27
June 22, 2012 at 15:30:25
Did they get hot?

Report •

#28
June 22, 2012 at 19:56:17
I am running an E8200 at 3.2GHz on a Gigabyte EP43-UD3L for a couple of years now that is 100% stable. If I remember correctly I only bumped the Vcore up maybe 2 small steps, the Northbridge one small step, the memory on the tightest timings at 1.9V (was listed at 1.8V at loose timings and 2.0V at tighter ones). When the ambient temps are lower, it idles in the med to upper 30's C, when it is warm out, it idles in the low 40's C. It has never gone above the mid 50's C at any time that I know of. I am running a stock Intel heat sink that was from a Core 2 Quad (may have more heat sink mass than the Core 2 Duo ones, but I am not sure) with an aftermarket thermal compound.
It may take a bit of experimenting, but it will be worth it.

Unfortunately, it works so well, I cannot justify upgrading at this time to an Ivy Bridge system. I am Sooo itching to build another system..... Oh well, maybe my daughter will get that new job and need that serious Photoshop/Autocad machine (I already have some of the hardware spec'd out if needed) instead of a new laptop.

You have to be a little bit crazy to keep you from going insane.


Report •

#29
June 23, 2012 at 07:56:30
"Did they get hot?"

Hotter than what? But to answer your question, NO, they did not get "hot". The temps when overclocked may have been a few degrees higher than the non-overclocked temps, but nothing to worry about. My results are similar to Fingers in the above response.

Cooling isn't rocket science. You have a small enclosed box with several components inside that throw off heat. The box (computer case) is only about 1.5 cu.ft. in size. As long as the main heat producers (CPU, motherboard chipset, & GPU) have an adequate heatsink/fan installed (& installed properly), they will run at a reasonable temp, but only if the heat from inside the case is removed & replaced with fresh air. Cooling fans are rated by CFM (cu.ft per minute). Here's a popular 120MM fan that's rated at 70 CFM:

http://www.newegg.com/Product/Produ...

Do the math. 70 CFM / 60 sec per minute = 1.17 cu.ft. per second

That means that every 1.3 seconds (roughly 45 times per minute) all the air inside the case is completely removed & refreshed. And that is just with ONE fan. If the case has a top mounted power supply, it's cooling fan acts as a case exhaust fan too, so the refresh time is even shorter. There's really no need for more case cooling fans than that. More fans doesn't necessarily mean better cooling, but it does mean higher cost & more fan noise. And a high priced aftermarket CPU cooler isn't absolutely necessary either. The stock units are quite good these days. I did buy a Corsair cooler for my last build, but that was only because it was on sale ($10 after rebate) plus the CPU I bought was OEM (no HSF included). It's on sale today for $15:

http://www.newegg.com/Product/Produ...


Report •

#30
June 23, 2012 at 11:25:28
Thanks for the info Riider.

I am a hardware guy not a software guy.


Report •

#31
June 23, 2012 at 13:58:21
One last thing - for load testing, try OCCT 4.2.0:

http://www.ocbase.com/index.php/33-...


Report •

Ask Question