Frequent computer freezes even after reinstall!

Page 2 - Seeking answers? Join the Tom's Hardware community: where nearly two million members share solutions and discuss the latest tech.

PresidentDylan5

Honorable
Apr 30, 2013
127
0
10,680
I am being driven insane by this. After a lot of troubleshooting that just led me in false directions, I still haven't got anywhere. My computer has completely random freezes. It ranges from time to time how often it is. Just now it lasts a few minutes so I can't do a resource scan.
I have completely reinstalled the operating system.
Resource scan showed everything fine.
I think it is hardware related, perhaps my AMD CPU. I am having some heating problems (massive lag spikes) which I pretty much got rid of with a hair dryer (was getting freezes as well) but was still planning on redoing paste and stuff soon. Maybe it's not a coincidence that these two issues arose at once? The lag spikes appeared before the freezes, which would make sense. However, HW monitor (in the brief time before a freeze just now) said my proccesor was at 30C!
A friend told me that a common windows problem is Ram, and another (who basically copied my build:) has the same ram and motherboard. He was getting blue screens and stuff which he managed to fix through voltage settings.
I was playing Titanfall beta all afternoon today with no problems whatsoever but (this could be coincidence) when my friend called me on Skype IT CAME BACK. I reinstalled OS for no reason. It was the refresh feature on windows 8 by the way.

Someone please help me!

Fx 8350
GTX 770
ASRock 970 extreme 4
(Ram has really strange name but it is corsair veangance)
SSD vertex 3
Caviar blue 500gb
Black 1tb
Zalman z9 plus
 
Solution
Since it failed, return the RAM settings to normal. As nostall suggested, if you haven't already, reset everything to default. This should be under Exit>Restore Default Settings.

Then reboot. If it crashes, change the Command Rate to 2T (I didn't see you post that you tried this).

I agree with nostall as well about the PSU, it's just I'd rather try the little things before making you spend money or have a non-functioning computer while you're waiting for the replacement to come.

Swapping out PSUs should take about 15 minutes if you're semi-experienced with it, could take up to 45 if you're really incompetent (it's just remembering where the cables used to be then plugging them back in).

Did you run the two tests I suggested? They're...

PresidentDylan5

Honorable
Apr 30, 2013
127
0
10,680
[sigh again] More freezes this morning with same ram settings. I have just used cotton buds to get more dust from the CPU cooler and used the hairdryer on the PSU through the vents on the bottom of the case. I couldn't get to it otherwise as I would have to take it out of the case. I might resit CPU today just to get a look at it, despite thermal paste not arriving yet.
 

gumbykid

Honorable
Jan 15, 2014
505
0
11,160
Since it failed, return the RAM settings to normal. As nostall suggested, if you haven't already, reset everything to default. This should be under Exit>Restore Default Settings.

Then reboot. If it crashes, change the Command Rate to 2T (I didn't see you post that you tried this).

I agree with nostall as well about the PSU, it's just I'd rather try the little things before making you spend money or have a non-functioning computer while you're waiting for the replacement to come.

Swapping out PSUs should take about 15 minutes if you're semi-experienced with it, could take up to 45 if you're really incompetent (it's just remembering where the cables used to be then plugging them back in).

Did you run the two tests I suggested? They're important. If you run one then your computer shuts off, crashes, or freezes soon after then there is an obvious issue with that component. If it does crash I would repeat it once or twice more just to be sure it wasn't your normal freeze.
 
Solution

PresidentDylan5

Honorable
Apr 30, 2013
127
0
10,680
Ok I will return RAM settings to normal. Will resetting to normal reset my boot drive? What does the command rate mean and where would that be?
Oh it seems that the cpu test is an overnight thing as well. Will probably run that tonight then but not sure my dad will be all too happy about power consumption! :)
 

gumbykid

Honorable
Jan 15, 2014
505
0
11,160
If you had to change your boot drive in your current configuration, then yes you would have to change it again. Command Rate is under the RAM Timing Control. It may also say (CR) or have an N instead of a T.

Command Rate is basically how many times your RAM has to review the information before it sends it back out. 1T is faster, 2T is more compatible.

And I don't want you running the test overnight. I want you to run it while you can monitor the temps with a program like HWmonitor. If there is a serious issue it will crash fairly quickly due to heat or from just being faulty. It will only take hours to crash if there is a slight voltage issue, which is more common for Overclockers.
 

PresidentDylan5

Honorable
Apr 30, 2013
127
0
10,680
Well I have had the test (prime95) running for about half an hour now and no freeze, and they fans seem to be running relatively quiet which is slightly confusing since it's meant to be a torture test... However, it was the 3rd option which said lots of RAM as it was checked by default and I didn't know what they meant! :) The cpu temperature is about 55C and it usually goes a bit above that on demanding games.

So far it said this:
[Feb 22 17:30] Worker starting
[Feb 22 17:30] Setting affinity to run worker on logical CPU #1
[Feb 22 17:30] Beginning a continuous self-test to check your computer.
[Feb 22 17:30] Please read stress.txt. Choose Test/Stop to end this test.
[Feb 22 17:30] Test 1, 6500 Lucas-Lehmer iterations of M12451841 using AMD K10 type-2 FFT length 640K, Pass1=640, Pass2=1K.
[Feb 22 17:36] Test 2, 6500 Lucas-Lehmer iterations of M12451839 using AMD K10 FFT length 640K, Pass1=640, Pass2=1K.
[Feb 22 17:41] Test 3, 6500 Lucas-Lehmer iterations of M12196481 using AMD K10 type-2 FFT length 640K, Pass1=640, Pass2=1K.
[Feb 22 17:47] Self-test 640K passed!
[Feb 22 17:47] Test 1, 800000 Lucas-Lehmer iterations of M172031 using AMD K10 type-1 FFT length 8K, Pass1=32, Pass2=256.
[Feb 22 17:50] Test 2, 800000 Lucas-Lehmer iterations of M163839 using AMD K10 type-1 FFT length 8K, Pass1=32, Pass2=256.
[Feb 22 17:53] Test 3, 800000 Lucas-Lehmer iterations of M159745 using AMD K10 type-1 FFT length 8K, Pass1=32, Pass2=256.
[Feb 22 17:56] Test 4, 800000 Lucas-Lehmer iterations of M157695 using AMD K10 type-1 FFT length 8K, Pass1=32, Pass2=256.
[Feb 22 17:58] Test 5, 800000 Lucas-Lehmer iterations of M155649 using AMD K10 type-1 FFT length 8K, Pass1=32, Pass2=256.
[Feb 22 18:00] Test 6, 800000 Lucas-Lehmer iterations of M153599 using AMD K10 type-1 FFT length 8K, Pass1=32, Pass2=256.



PS: Can I play games while this is running? :D
 

gumbykid

Honorable
Jan 15, 2014
505
0
11,160
The Blend test factors RAM into the test, making it less CPU intensive. You can play games but they will be slow and can mess with the results (e.g, the PC crashes b/c of a graphics issue and you now think it's a CPU one).

Try Furmark. This seems more like a PSU issue.

Do you live close to a BestBuy?
 

PresidentDylan5

Honorable
Apr 30, 2013
127
0
10,680
Should I have done one of the other tests? Should I stick to very undemanding games then?
-Furmark seems to be a gpu test...

-Um, no...

Haven't froze since about 1:30 I have been playing Fallout 3 (not the whole afternoon!)
 

gumbykid

Honorable
Jan 15, 2014
505
0
11,160
Furmark is for GPUs and is a simple test to run so its worth it before buying parts.

I asked because they will let you return a part at no fee, so it would be good to buy a PSU and quickly return it if it doesnt solve the issue.
 

gumbykid

Honorable
Jan 15, 2014
505
0
11,160
Okay, those temps are fine for a GPU.

This is either a CPU or PSU issue. Before you buy any parts make sure to try reseating your CPU when you get the thermal paste.

And just for safe measure, try hooking up your computer to a different monitor with a different cable, just in case there is some type of display issue. Your computer doesn't restart or shutdown, right? Just screen freezes?
 

PresidentDylan5

Honorable
Apr 30, 2013
127
0
10,680
Wow problems with the monitor didn't even occur to me, but really I think it's unlikely as the problem resembles overheating or at least something similar. I played Battlefield which was notorious for overheating my CPU and that freezes after half an hour, then I leave the PC off for 5 minutes, play Fallout for hours with no freezes.
Well I will get back to you once I've done the thermal paste; I don't think there's much worth doing before that. The latest it [should] arrive is Tuesday. Keep checking back and don't disappear! :D
 




Here's a link from Tom's re. cleaning and reapply thermal past:
http://www.tomshardware.com/forum/333968-28-thermal-paste-removal-installation-guide
Gumby's recommendation re. PSU is the next step if the CPU reseating doesn't solve the problem; but before you go to all the trouble of buying-trying-maybe returning a PSU Please do this: Go into your BIOS, look under (probably) Monitor or Hardware Monitor and post the listed voltages for 3.3, 5, and 12V.
Then install HWInfo (whichever version fits your Operating System)
http://www.hwinfo.com/download.php
Post the voltages or a screen shot of the voltages shown on HWInfo.



 

PresidentDylan5

Honorable
Apr 30, 2013
127
0
10,680
Yes a lot of the guides I've looked at, including this one you've sent me, say to use alcohol to remove the paste but I bought that ArtiClean stuff (http://www.amazon.co.uk/Arctic-Silver-ArctiClean-Material-Purifier/dp/B000BKP306). That will work fine surely? Also can I use cotton buds instead of a dry cloth if I don't have one?
I will try the stuff for the PSU sometime tonight as I'll be out all day.
 

gumbykid

Honorable
Jan 15, 2014
505
0
11,160
Yeah the Artic stuff will work, just know for the future that's just overkill when alcohol works fine.

You CAN use cotton swabs, however they are likely to leave behind lint. Many people say coffee filters will do the same cleaning without the lint. You could even use a papertowel, but it has the same issue as the cotton swabs. Q-tips also work. Just clean up after yourself and you'll be fine.
 


+1^
An old, clean t-shirt works well also. The Artic Clean works REallyy well; forgot you had purchased it.
 

PresidentDylan5

Honorable
Apr 30, 2013
127
0
10,680
Um, *sigh*, my temperatures have got worse with the new thermal paste. Before the highest was 68C, possibly 69; now it got to 72C on BF3. It was extremely awkward; I didn't do a perfect blob and it is really hard getting it on properly. I expect you've experienced this: you have to get the hooks over those things (hope you know what I'm talking about) and it results in moving the heatsink around slightly. Really don't want to have to it again.
Oh and it froze.
 

gumbykid

Honorable
Jan 15, 2014
505
0
11,160
You must've applied it incorrectly or the heatsink is slightly out of position. Did you reseat the CPU like I suggested and look for any damage?

As for your temperature issue my only advice is to make sure the heatsink has good contact with the cpu and is centered well. How thick did you apply it? Too much will ruin the purpose.
 

PresidentDylan5

Honorable
Apr 30, 2013
127
0
10,680
I don't think I applied it to thick. I thought too much came out at first so I took a bit off with the cloth, and that meant a bit got smudged around the blob which was still a fair size anyway. And it was awkward getting the heat sink on. I also reseated the cpu and I couldn't see any damage. Do I really have to redo it? It took ages.

I looked at those voltages. In the BIOS, it was +12.091 for 12V; +5.040 for 5V; +3.168 for 3.3V. The main voltage is 1.384-1.4. I couldn't really find the same things on that program other than the main voltage, which is 0.8750 on summary but when I look at the processor in detail it says 1.4250. Also this is probably normal but loads of things were marked 'not present' in standard feature flags.


PS: Could it be possibly possible that the thermal paste takes time to settle? [Say yes] Right now my temperature is 11C. Maybe the BIOS is a demanding application now.
 

gumbykid

Honorable
Jan 15, 2014
505
0
11,160
There's a possibility that you may have to reapply it. It needs to be even and spread out as a thin layer, any bulges, bubbles, or areas with more or less paste can cause poor transfer of heat. Is the heatsink stable? As in it won't move if you gently nudge it. And it can be hard to see, but does the center of the heatsink appear to be on the CPU? Slight adjustments there can cause heat spikes.

As for your voltages, i'll leave it to nostall since I don't understand power supplies all that well.
 

PresidentDylan5

Honorable
Apr 30, 2013
127
0
10,680
The heatsink does move very slightly if I nudge it, yes. Also there do seem to be spikes, as my temperature is around 10-15C but it says the maximum in this time 45C!
Oh my god I have to reapply it...
How do I get it out the tube so it's a blob and not a 'spiked blob' if you know what I mean? Is that important?
 
PresidentDylan, I wouldn't worry to much about the thermal paste, your CPU is within specifications. Here's another link with a picture of how the paste should have looked before mounting the heatsink - at least for their tests. If the paste isn't oozing over the side of the cpu socket it should be o.k. for now.
http://www.hardwaresecrets.com/article/Thermal-Compound-Roundup-February-2012/1490/3
What does concern me is the voltage on the 3.3 rail. If the BIOS is showing 3.168, that is 4% below specification - with the computer NOT doing anything. I would suspect it would get lower as the pc warms up. You never did tell us, or I missed it, what the EXACT model of Corsair PSU is that you have, but it is now time to call Corsair customer service and tell them or Tech support all that you have done to rectify the problem and you suspect the PSU may be a the problem. Be sure to mention testing the RAM also. (It has probably been mentioned, but if you have someone else willing to let you install your RAM in their PC that is a good test - if their working PC fails with your RAM then the RAM is bad; Corsair would probably agree).
See what they have to say (I assume it is still under warranty). +3.3 votage rail used to be one of the primary power sources for the video card, and it is still a major source for that PCI-E slot and video card; if it is going lower, it probably is the cause of your problems.
 

PresidentDylan5

Honorable
Apr 30, 2013
127
0
10,680
Um ok so the voltage gets lower as the load gets higher? And the rail you are concerned about supplies the video card mostly? In that case why is that the gpu test (Furmark) didn't freeze? The exact model is the CX750. It had a five year warranty! Do you think they'd be willing to send a replacement without too much hassle, and would I be without a PC for a few days at least?
Just checked again the 3.3 voltage is still about the same. Both of my readings have been after freezing though; in a period of n freezing I will have another look. Would this confirm your thoughts?
So you think my CPU temperatures are fine? Interestingly I'm not having the lag spikes I was having before but am running at similar temperatures...

PS: my blob didn't seem to be as big as the one on the picture. Guess I might have been too cautious :D
 


Call Corsair. CX units are good, but not as well built as the TX, HX, or AX units.