Sign in with
Sign up | Sign in
Your question

New build - System Instability

Last response: in Systems
Share
December 27, 2010 6:36:47 PM

Problem
I built my new PC a couple of weeks ago, with mostly new parts. My PC keeps on completely freezing, forcing me to force shutdown and turn it back on again. Sometimes after turning it off it works fine, other times it freezes again at any point during startup (once during BIOS, sometimes during Windows setup, and sometimes minutes after it is fully started). The problem is, this freezing happens randomly. So far it's occurred while the system was idle, while watching a Youtube video, and while opening Google Chrome. It hasn't happened in the 3-4 hours I've spent gaming on it since I got it.

What I've Tried
I've run MemTest 86+ when I initally bought the computer, and it found no problems. I also ran a couple of stress tests (using Prime something...) and the PC never crashed. When I use Speedfan, my System is normally 29C, my CPU is 36C, and the "Core" is 35C. The ONLY thing it has a problem with is something called "AUX" which changes from anywhere between 40C and 130C. I read on the internet that it shouldn't matter, but... that's what it is.

I'm really, really confused on what could be the issue, or how to single out the part. The only thing I can possible think of is the motherboard, and that's only because I'm not sure how to test a MOBO specifically for stability.

Parts
OS: Windows 7 64-bit

Video Card: GIGABYTE Ultra Durable VGA Series Radeon HD 6850 1GB

Motherboard: ASRock Extreme3 870

RAM: G.Skill 4gb 240-Pin DDR3 1600 (Four 2gb sticks for 8gb total)

PSU: CORSAIR 750W (I've had this part for over 6 months now, so this wasn't a "new" part for this build)

HDD: WD Caviar Black HDD (I've had this part for over 6 months now, so this wasn't a "new" part for this build)


Shouldn't matter, but to be comprehensive...
-DVD Burner
-Cooler Master CM690 Case
a b B Homebuilt system
December 27, 2010 7:02:25 PM

Relax your memory timings to auto, set your mem voltage to whatever is recommended...drop to a pair or even a single stick for a while, until you figure what might be going on... (Memtest working 'weeks/months ago' is no assurance of issues or lack thereof today)

Reseat all connections, including main power inputs to MB, video card, SATA inputs, etc....

food for thought....just because your power supply was good a few months ago, does not mean it is good as of when your problems started.....
m
0
l
December 28, 2010 8:32:15 PM

mdd1963 said:
Relax your memory timings to auto, set your mem voltage to whatever is recommended...drop to a pair or even a single stick for a while, until you figure what might be going on... (Memtest working 'weeks/months ago' is no assurance of issues or lack thereof today)

Reseat all connections, including main power inputs to MB, video card, SATA inputs, etc....

food for thought....just because your power supply was good a few months ago, does not mean it is good as of when your problems started.....

Thanks for the advice.

All of my memory timings and voltage are set to auto in my BIOS. I also checked to see what G.Skill recommended for this particular RAM set (for timings and voltage) and that is what my MOBO automatically set it for.

I ran Memtest last Sunday, so it would only have been a few days after running it that I started to use my computer and started noticing these problems. I understand that problems can happen at any time, but a successful test merely days before this started happening leads me to believe it's another issue.

As you suggested, I reseated everything yesterday and made sure every connection was in correctly. However, about 10 minutes ago my PC froze again (it froze while sitting idle about an hour after a 3 hour Bad Company 2 gaming session where I had no problems...).

Anyone have any more ideas?
m
0
l
Related resources
December 29, 2010 9:00:03 PM

Just as a bit of an update on what I've tried.

I ran MemTest 86+ v3.4 for 10 hours last night. It completed 4 complete passes through and reported no errors.

This morning I had my PC freeze again, randomly, while I was watching TV (not using my PC). After I restarted I shut off all of the programs that open on Startup and started playing The Witcher. About an hour into that, the PC froze again.

Finally, event viewer is showing no errors (that I can see) that pertain to applications or system errors.

Does anyone know what is the likely hardware problem, given these random problems...?
m
0
l
a b B Homebuilt system
December 29, 2010 9:49:04 PM

Even if memtest is fine one stick of RAM might be damaged. Also get the newest drivers for the video card. Run Driver Sweeper first to get rid of all ATI drivers after you uninstall them. Check the video card temperatures shown by PCWizard first.
http://www.softpedia.com/get/System/System-Info/PC-Wiza...
m
0
l
December 29, 2010 11:12:54 PM

Hmm, interesting. It was my understanding that MemTest would report any errors for ANY of the sticks installed in my PC. Are you saying that I need to run the test with only one stick installed at a time?

PCWizard reports that my video card idles at about 40C right now. All of the other temparatures seem good, except for "AUXTIN" which is 70C. This goes along with Speedfan which was also reporting very high AUX temperatures. Is this a problem, or just a faulty sensor?

Finally, I've reinstalled ATI drivers, using Driver sweeper, since this started happening and it didn't seem to solve any problems, as it is still freezing at random times.
m
0
l
a b B Homebuilt system
December 29, 2010 11:34:33 PM

Try the PC with one stick at a time and see if it still freezes.
m
0
l
December 29, 2010 11:47:15 PM

Ok, will do.

This couldn't be a problem with my timings at all could it? I've tried them at Auto, and I've also tried them manually set. The voltage has always been set to auto (which is set at 1.54v).

The timings I've used are 9-9-9-24 when I set my MOBO to DDR3 1600, and I've also used 7-7-7-20 when I set my MOBO to DDR3 1333. I will admit that I know next to nothing about memory timings, but these were ones that I've seen posted by G.Skill and owners of the memory. However, the PC has frozen at least once while using each of these memory timing settings.

Anyways, I'll give it a try with one stick at a time and see what happens. I figured I had ruled out bad RAM due to all the memory tests I've done (MemTest and Prime95), but I'm willing to try anything!
m
0
l
January 6, 2011 10:00:33 PM

I would like to report that I tried 1 stick at a time, each in a different slot, and I still encountered freezing. Also, I have completely reinstalled Windows 7 and am still having these issues.

If anyone has any more ideas, I have a (more active) post here: http://www.tomshardware.com/forum/14325-63-build-system...

Thanks a lot!
m
0
l
January 25, 2011 11:35:41 AM

Mosox - just a disclaimer, I have not tried what you suggested yet. If you're still lurking around these forums, please let me know if replacing the MOBO negates the point of doing a latency check.

Now then...

I contacted ASRock customer support about three weeks ago now. They said that the problems I described sounded like a MOBO problem. I contacted NewEgg and they immediately issued an RMA.

After two weeks without a computer, my new MOBO arrived Friday and I installed everything yesterday after I got some new thermal paste for my CPU (yes, I cleaned the old stuff of first, and yes, the temps after I applied new thermal paste seem good, so I don't believe it is overheating). I booted up into Windows 7 and spent about 2 hours backing up some files on my PC so that I could reinstall Windows 7 (due to the new MOBO).

Immediately after I installed Windows and logged in for the first time, the PC froze yet again. It was running Windows 7 Basic theme, meaning there shouldn't have been any Aero or other intense GUI stuff going on. After a few more freezes and a few hours of installing updates and other essentials, I went to bed running MemTest86+. When I woke up it had completed 8 full passes and was halfway through another pass when I shut it off. No errors reported.

So - have I narrowed it down to either the PSU or the Video Card at this point? I was really hoping it was neither because I don't really have any other PSU I can try, and the only other video card I have is an old HD4850. I would like to point out, however, that this freezing happened more often than not while idle, NOT while playing games. I'm no tech genius, but to me this means that it likely is NOT the video card because it doesn't happen more frequently during stressful activities...

I need guidance!

(Also, I was annoyed by Toms Hardware constantly sending me messages to select a Best Answer, so I figured I'd try here again)
m
0
l
a b B Homebuilt system
January 25, 2011 2:01:00 PM

Run the latency checker. You can also try you other video card.
m
0
l
January 25, 2011 7:36:28 PM

Alright then! I am running the latency checker with a "test interval" of 1000. The "current latency" section varies between 100 and 180, generally averaging ~140. I had no spikes at all in the 10 minutes that I ran the latency checker (the bars were always green) and the "Absolute maximum" was 280.

The text at the bottom says that my machine "should be able to handle real-time streaming of audio and/or video data without drop-outs."
m
0
l
a b B Homebuilt system
January 25, 2011 7:55:51 PM

OK. Now try the other video card.
m
0
l
February 11, 2011 11:31:27 AM

So, time for another update I guess...

After replacing my MOBO didn't fix the problem, I swapped my brother's RAM with mine and that didn't change a thing. His computer still runs great, and my computer still randomly freezes.

After trying the RAM, I went back to my old HD4850 video card, but that froze after a day or so too.

So... am I down to the PSU being broken? I checked the voltages of my PSU during startup, and it seemed like all the voltages were spot on (at least during the time I was testing it) - it seems that software voltage test was way out of line.

*sigh* I don't suppose anyone else has any more ideas, eh?
m
0
l
February 23, 2011 7:38:19 PM

Well, I tested Windows 7 on my other known-good HDD and got the same results. After 8 days of no problems, my PC froze yesterday. Then it froze again today after I got home from work.

Still trying to get my hands on a PSU, but that's proving to be very difficult.

Hmm....
m
0
l
a b B Homebuilt system
February 24, 2011 2:46:22 AM

Must be a driver/device conflict. you need to reinstall the OS, the drivers, and then try some other applications. What's in your programs list? Might be the antivirus or something.
m
0
l
February 26, 2011 5:14:10 PM

Well mosox, I've reinstalled Windows 7 *literally* 7 times in the past two months. I always use the latest drivers everytime I reinstall. Plus my PC has frozen without an A/V before. I normally use MSE, which hasn't given me problems before. In fact, one time my PC froze on first boot from Windows 7 after I had installed it, and that was during a state in which it had only the Windows drivers installed.

As an update, I replaced my PSU with a known good one from a friend, and my PC still froze after a couple of days.

Unless I'm missing something, the ONLY thing I have yet to replace is my CPU... that is, if the problem is even hardware related.

I'm really hoping someone can shed some light on this issue :( . My next step, unless someone gives me a better idea, is to reinstall Windows, this time using Windows XP 32 bit (instead of what I've been using every time, Windows 7 64-bit).
m
0
l
a b B Homebuilt system
February 26, 2011 7:58:31 PM

Try relaxing the timings of the RAM. Also if you keep on installing the same apps you don't know what it is. Try some other video card driver and then install only the minimum of apps and see how it goes. Make restore points after each installation.
m
0
l
February 26, 2011 8:10:31 PM

Can you give me a bit of guidance on how to relax the timings on my RAM? I've tried two settings with Memory timings: auto, and the ones that were suggested by GSkill. Both result in the same problems. I don't really know how to "relax" my timings.

Also, is there anyway my CPU is causing these random issues? Like I said, it's the only part I have yet to physically change out... mainly because it's the most difficult to change.

Thanks.
m
0
l
a b B Homebuilt system
February 26, 2011 9:32:30 PM

Change manually the voltage to 1.5V maybe the auto function isn't working properly.

Then if not working try 1333 8-8-8-24 and then 9-9-9-24.
m
0
l
February 27, 2011 1:09:35 PM

Voltage when set to "auto" is already set to 1.54V, which is the closest to 1.5V that it lets me get.

So far the timings I've tried are 1600 9-9-9-24, 1333 9-9-9-24, and 1333 7-7-7-20. Is the only timing left to try 1333 8-8-8-24?

Edit: There is also a voltage setting for 1.477. Would that be better than 1.54V?
m
0
l
March 2, 2011 9:31:44 AM

I really don't think it's a software issue. In fact, I just installed Windows XP 32-bit (instead of Windows 7 64-bit which I have been using), with entirely different drivers, and it still froze a couple of days later. I only installed a few programs, and it never crashed while running those. Besides Windows XP, I had RIFT, Xfire, League of Legends, FixBCD, and Google Chrome installed... that's it. And in the end, it crashed while opening up Google Chrome, one of the easiest operations I ask it to do :p .

So to recap: I've installed two different OS's, tried another PSU, swapped ALL of my RAM with my brothers, got an RMA on my MOBO, tried another graphics card (it was also ATI though, but with a different driver), and tried two hard drives.

At this point, the only hardware I can possibly think that it could be would be CPU, or maybe even still the MOBO.

However, I will start to try the RAM voltages now, although like I said, I set my brother's RAM up in the exact same way (he has the same RAM and MOBO), and his is still fine. It will probably take me a week or so before it freezes using different RAM timings and voltages, but if anyone has any more suggestions, I'm still listening..!

Thank you so much for your work with this issue so far mosox. Seems you're the only one who is still bothering with this unsolvable problem! I really appreciate it.
m
0
l
April 2, 2011 4:36:17 PM

Alright, so I have received my 3rd motherboard (2nd warranty replacement), and after a week of magic, the PC once again shattered my hopes and dreams and froze on me. Same problem that I've been having since December.

I did recently download an updated NIC driver today (after the freezing), as suggested by some other people on another forum, from http://www.realtek.com/downloads/downloadsView.aspx?Lan.... Everything seemed to install fine, and the driver is about a year newer than the one ASRock provides. However, I don't have much hope that this will fix the issue, because my brother is using all drivers from ASRock and has none of these issues.

So, I literally have no semblance of an idea of what this problem is. Is it my CPU? It's the only part I haven't swapped out... I'm pretty sure I've checked all of the possible timings and voltages, but I don't have a comprehensive list of all the ones I should've been trying. Like I keep saying though, my brother has all of his set to Auto and has no issues (once again, he has the exact same PC as me, just a different HDD).

This is probably the most annoying problem I've ever dealt with.

Edit: As an update, I ran Prime95 for 1 hour and there were no crashes. CPU temp maxed at 58C.
Hmm....

Edit 2: Nope, new network drivers definitely didn't fix the problem...
m
0
l
April 2, 2011 8:20:08 PM

I did a "Burn-in" Stability Test using a program called SiSoftware Sandra that I saw recommended on another forum. Here are the results, if they mean anything to anyone:
Quote:
Board Temperature: 30.00°C (30.00°C : 30.00°C - 30.00°C)
CPU 1 Temperature: 41.38°C (41.38°C : 41.38°C - 41.38°C)
CPU 2/Aux Temperature: 55.50°C (55.50°C : 55.50°C - 55.50°C)
CPU 1 Fan: 29348rpm (29348rpm : 29348rpm - 29348rpm)
CPU 1 DC Line: 1.35V (1.35V : 1.35V - 1.35V)
CPU 2/Aux DC Line: 3.00V (3.00V : 3.00V - 3.00V)
+3.3V DC Line: 3.40V (3.40V : 3.40V - 3.40V)
+5V DC Line: 2.12V (2.12V : 2.12V - 2.12V)
+12V DC Line: 13.18V (13.18V : 13.18V - 13.18V)
Standby DC Line: 3.49V (3.49V : 3.49V - 3.49V)
Battery DC Line: 3.40V (3.40V : 3.40V - 3.40V)
CPU 1 Core Power: 185.32W (185.32W : 185.32W - 185.32W)
CPU 1 Cooling System Thermal R:0.06°C/W (0.06°C/W : 0.06°C/W - 0.06°C/W)
GPU 1 Temperature: 41.00°C (41.00°C : 41.00°C - 41.00°C)
GPU 1 Fan: 3382rpm (3382rpm : 3382rpm - 3382rpm)


Quote:
Board Temperature: 30.00°C (30.00°C : 30.00°C - 30.00°C)
CPU 1 Temperature: 35.50°C (38.44°C : 35.50°C - 41.38°C)
CPU 2/Aux Temperature: 99.50°C (77.50°C : 55.50°C - 99.50°C)
CPU 1 Fan: 28723rpm (29035rpm : 28723rpm - 29348rpm)
CPU 1 DC Line: 1.15V (1.25V : 1.15V - 1.35V)
CPU 2/Aux DC Line: 3.00V (3.00V : 3.00V - 3.00V)
+3.3V DC Line: 3.41V (3.40V : 3.40V - 3.41V)
+5V DC Line: 0.81V (1.47V : 0.81V - 2.12V)
+12V DC Line: 13.18V (13.18V : 13.18V - 13.18V)
Standby DC Line: 3.49V (3.49V : 3.49V - 3.49V)
Battery DC Line: 3.41V (3.40V : 3.40V - 3.41V)
CPU 1 Core Power: 134.63W (159.97W : 134.63W - 185.32W)
CPU 1 Cooling System Thermal R:0.04°C/W (0.05°C/W : 0.04°C/W - 0.06°C/W)
GPU 1 Temperature: 41.00°C (41.00°C : 41.00°C - 41.00°C)
GPU 1 Fan: 1196rpm (2289rpm : 1196rpm - 3382rpm)


There were no errors (besides one saying I had no DVD in my DVD drive, which was true, and another which said my two hard drives weren't empty, which is also true!).

I did bold the part that concerns me, which is that AUX temperature again. IT seems very hot, showing at 99.5C? SpeedFan also indicates this temperature fluxuates often, going from 40C - 100+C.
m
0
l
April 17, 2011 1:39:45 PM

Yet another update, although I think everyone is out of ideas.

I swapped my CPU with my brother's two weeks ago, and I've had three random freezes during that time, while he has continued to have no problems.
m
0
l
May 23, 2011 4:51:56 PM

I'm nervous to say this officially, but I think I may have solved my problem, as I've been four weeks without my PC freezing.

Know what I did? I replaced the SATA cable going from my main HDD...

See, every time I shipped back my MOBO, they made me keep the same cables, so I never had an extra cable around to test. I finally decided to swap cables with a new one, and it seems to be working fine! I can't really explain it, but even when I was running off of a secondary HDD, I still had my first one plugged in using the same cable, so my freezing could have followed with the HDD.

I'm not entirely sure the issue has been completely resolved: since the freezing is entirely random, it's very likely that it could come back. All I know is that I've been longer than I ever have between freezes, so I am ready to put this 5-month long problem behind me...

We'll see what happens in the coming weeks!
m
0
l
!