• Welcome to Overclockers Forums! Join us to reply in threads, receive reduced ads, and to customize your site experience!

Random PC Freezes

Overclockers is supported by our readers. When you click a link to make a purchase, we may earn a commission. Learn More.
As my friend RGone said, please disable C1/E, C6, Cool N Quiet, APM, TurboCore and in windows performance manager itself set to "performance" mode and check back. All the memory issues i have seen always ended in blue. Maybe try to render with only one stick of RAM to see if the problem is still plresent.
 
All the memory issues i have seen always ended in blue.
Not 100% accurate tung, when I was dealing with trying to run my 8350 @ 1866 or above I had hard locks and freezes. The IMC on my chip needs 1.35 Cpu NB voltage to run 1866-2000 Mhz stable, which will in turn raise the Socket and Package temps. It seems that running at 1600 Mhz on the memory will be a solution until a better heatsink has arrived.
 
Not 100% accurate tung, when I was dealing with trying to run my 8350 @ 1866 or above I had hard locks and freezes. The IMC on my chip needs 1.35 Cpu NB voltage to run 1866-2000 Mhz stable, which will in turn raise the Socket and Package temps. It seems that running at 1600 Mhz on the memory will be a solution until a better heatsink has arrived.

I was thinking more in the lines of the sticks themselfs. I do not have experience with IMC issues
 
I was thinking more in the lines of the sticks themselfs.
Then yes most problems with the sticks themselves result in a BSOD. That said I have and have seen that some of these 8350's need more Cpu Nb voltage then others to run 1866 and above. My thinking is that the issue lies with that. The problem is raising the Cpu Nb voltage will increase the heat and there is no headroom in temperatures. Therefore, lowering the memory speed to 1600 may resolve the issue. Until the new heatsink arrives.
 
The problem is raising the Cpu Nb voltage will increase the heat and there is no headroom in temperatures.

U can say that again. Exactly what I experienced with the "exercise in futility" rig I wrote about in the FX Info thread. Initially the ram was Auto set to DDR3-1600 and all seemed okay. Then the TW3AK3R in me got lose and upped the cpu speed to its rated DDR3-1866 and raised the CPU-NB to match the HT Link Speed and the 'wheels' fell off. Hehehe. I had to decide on using my DDR3-1866 or lowering the CPU-NB speed so I did not have to add more CPU-NB Voltage. I dropped the CPU-NB speed and fixed the flats and off and running.

I went to the Maya site but to tell exactly what is recommended as hardware depends geatly on the Year of the software, so I came up a little short on knowing what hardware is said to be good. I know they said NO ATI video cards but he seems using Nvidia so no harm there.

5 Mins of P95 Blend and that quick rise those temps is surely a bad thing. I would run Maya using 'only' one monitor and have HWMonitor was open and logging in the background and then post the capture made after logging for 10mins of heavy Maya use.

Anytime that specialty software gives issue the issue is only found by diligent testing. I expect taking care of the heat and then adjusting CPU-NB voltage up may help the current problem BUT what is still tripping me up is that this 'issue' has been in the system most likely since the rig was built and Maya installed. But surely the heat must be gotten in check. That spiking heat can go unnoticed in some software. They poll to slowly.
RGone...
 
I would run Maya using 'only' one monitor and have HWMonitor was open and logging in the background and then post the capture made after logging for 10mins of heavy Maya use.

Thanks. It will also crash if I am just using an empty maya scene (pretty much zero performance hit) with a video on the second monitor. I can use Maya all day without a video playing and It won't crash. I remember it also crashed a lot playing that Watchdogs game that came out earlier this year (or other Ubisoft games for that matter). I can do really intensive rendering or play intensive games. Although it will crash if I have dual monitors on with a video on the second monitor and doing something on the other. Other times I can be doing some web browsing and it'll freeze up but that is far less likely.

Also I remember when I upgraded my other install of Windows 8 to 8.1 (and stupidly didn't make a restore point) it would freeze up constantly with any form of media playing so I had to reinstall Windows 8 which is the install I am using now. I also had Windows 7 on this system before that, which also crashed but IIRC I would get blue screens, although I can remember what they said.

Anyway, as I said, it seems to be really inconsistent; I can do some really processor heavy things and it won't crash but other times I'll be doing something low performance, It will give up on me. Thanks.
 
Are you using and AMD GFX card. I do remember reading they sometimes had problems with 2D video setting voltage too low.
 
You have a better cpu cooler coming...

...the new cooler is surely truthfully needed since a 'real' load on the system causes too much heat. So get the new cooler and install accurately and then run P95 Blend mode according to the outline and images below which describe what it is we really need. Okay I have already shown what 4 images we need in post #37 > Link Post #37. Those 4 captures with new cooler and settngs in bios as suggested and Windows Power Management set to "performance", should let us all see how the system is responding in a more overall way.

Because I was a mechanic for 29 years, car references come naturally to me even if some dislike such reference. Getting the cooling in check so we know the system 'can' withstand a load without going into over-heat mode is simillar to taking a car to the shop for an A/C problem, but having to jump start the thing everytime to do any A/C testing. Heck the car may not have enough electrical current to even pull the A/C clutch in and turn the compressor. Fix the jump-start issue first. Hehehe. Get the cool in place and then that is 'removed' from the list just as using and Nvidia video card is suggested at Maya site.
RGone...ster.
 
OK. The new Noctua cooler came today; I am really impressed with the quality. I'm getting about 25 degrees idle temperature which is awesome. Some images below.

new-cooler.jpg
new cooler.PNG

Bad news though, I am still getting crashing. I took a photo of the screen at the point where the computer froze up whilst rendering video and watching youtube at the same time, so you guys can see the readings at the point of the crash.

new-cooler-readings.jpg
 
mynameisollie,

Please create a Signature with your system info in it. You have a gold star below your user name now so you have access to this feature. As threads get long like this people trying to help you don't want to have to go back and glean all the info about your hardware that you may have included here and there. It's a common courtesy on the forum.
 
I think you should try what I suggested earlier now that you have better cooling. Set the CPU_NB voltage to 1.25 and ram voltage to 1.55v and see how it goes.
Also there have been issues lately with ONline video and Adobe HW acceleration, FF has been having a lot of issue as well. Until we get this straightened out try doing one thing at a time. First off run the P95 test and post some pics as RGone suggested. This will get us started.
 
OK. These are the results from CPUz & HWMon after 20mins Prime95 blend. The CPU_NB voltage was set to 1.25 and the RAM 1.55. I also disabled C1/E, Cool N Quiet, APM and TurboCore. HPC was enabled and windows was set to performance.

cpu1.PNG
cpu2.PNG
cpu3.PNG
cpu4.PNG
hwmon.PNG
 
That looks good the temps are in line now. If you intend on going any faster you'll want to get some active cooling on the VRM area and behind the mobo on the CPU socket.

I would suggest letting P95 run for a full 2 hrs so we know it's "stable" at these settings. This will eliminate that in the problem solving process.
 
Back