Win 7 Crashes

I've recently built my myself another PC and I've been having the occasional Win 7(64) crash. I'm vaguely familiar with looking at crash dumps, but need a bit of help rooting out the problem.

When it crashes the system locks up(no BSOD) and there's a distorted audio oddity(not the mobo) which leads me to think it might be a sound card issue. This happens when using a variety of apps, from games to browsing Firefox and listening to Pandora.

 

Some more details would be helpful: motherboard make & model, are you using on-board audio or a separate sound card? if a separate sound card, we need the make and model; have you run any diagnostics on the RAM or hard drive? and what is your hard drive make? was everything OK for while? if so, what did you do recently that may have caused the problem? did you update a driver? install more or different memory? etc....

. . . . and WELCOME TO MAJOR GEEKS! :wave :major

 

Sorry about the lack of detail, I should know better. Thanks for the welcome and help too.

Mobo: Asus P8P67LE B3 revision
Sound: Creative X-Fi Extreme Gamer

I'm using a 120GB Corsair F Series SSD and a 640GB Western Digital HDD. I haven't changed my hardware setup since setup, and I've kept my drivers consistently updated, but wasn't able to correlate the issues to any specific update.

The problem has been happening intermittently.

 

I was hoping someone might have an idea on this. As an update it seems to NOT be sound related(it's happening without now), and seems to be happening more frequently.

My system also seems to NOT be generate crash dumps for these events either.

 

My first thought would be to remove the SoundBlaster audio card (or disable it in Device Mgr) and use the built-in sound on the motherboard with the newest driver available from Asus. If the problem persists, then it is likely the MB. If the problem disappears, then we know it's a problem with the SB device. In that scenario I'd recommend that you completely remove the drivers for the SB card, power down, remove the SB card from the PC, run the PC for a day or two w/o audio or with the on-board audio, then re-install the SB card using the newest drivers from the Creative site (one of the worst sites in the world besides Gateway for finding correct drivers).

Do the crashes occur only when running specific games or programs? Are they random? Have you checked the Event Viewer > System logs or the Event Viewer > Application logs for info? Look for any errors or warnings that occur at approx the same time as the crashes.... Have you done any temp monitoring? I like using HWMonitor; it's free and accurate and will logs Min/Max temps, voltages, fan speeds, etc.

 

OOOOPS - I just read that the problems occur independent of any sound problems you have been having... so, you can probably ingore my suggestions above concerning the SB and on-board audio, and just adhere to the other stuff I mentioned: temps, voltages, event logs, and a full driver update for ALL devices wouldn't hurt either....

 

Thanks, more things I should know better than to forget about.

It DOES appear that heat and the consequent crashes are the symptoms, but what is the cause :confused? I keep my fans and heatsink clean, and there's plenty of room at that back of the chassis. I double checked my BIOS settings and the CPU isn't being OC'd, nor have I played with OCing the GPU. The SS below is when the system is running idle.

http://img.photobucket.com/albums/v321/RainMan5419/Temps_FanSpeed.png

The CPU is an Intel i5 2500k, with the stock heatsink and the GPU is an EVGA 470 GTX as shown.

 

An update: I'm not sure if there's a problem the with HWMonitor software, or a temperature sensor is malfunctioning, but there may be more than one problem here.

The idling temperature is initially displayed as around 60-65C(which is still unreasonably hot for idle) but jumps 10-25C after the application is open for just a few seconds.

 

You could try a different temp monitoring app (check out this free program), but it could very well be a problem with the CPU cooler. Maybe it isn't mounted correctly, not enough or too much thermal paste, slow/failing/dead fan....

 

So the temperature display lag I was seeing was just a function of how it reports. Open Hardware Monitor was giving the same temps, just less erratically.

So, here's where I'm at now. After calling EVGA the GFX card is idling ~10C higher than what it should be, so that will be RMA'd. The thermal paste between the CPU and heatsink was the stock pre-applied stuff so I cleaned that off and applied new paste. While it seems to have helped a bit, I'm still seeing idling temps peaking at 85C.

 

Maybe write down the various temps shown in your monitoring program while at idle, then restart, go to the BIOS and check the hardware monitor there. Then compare the numbers. It's not uncommon for monitoring programs to report erroneous numbers because of the number of MBs they have to add to their programs is huge, and they can't always keep up, and considering you have a fairly new MB, it could be that the temps are being misreported. You might want to check out the ASUS PC Diagnostics V1.12 which is a suite of system tests and diagnostics (DUH rolleyes) for your MB. I'm not sure if ASUS PC Probe II is compatible with your board or not (it works with my Asus MB), but it's another option. You might also consider some general diagnostic that may not be covered my the ASUS suite (like a full HD test; free from your HD maker's site).

GOOD LUCK!

(did you check the Event Viewer for any warnings/errors in the "System" logs that occur at approx the same time as the crashing?)

 

Event Viewer didn't have anything useful to share, and I'm still getting the crashes.

Checking out some of the Asus diagnostics today.

 

I didn't glean anything useful from the Asus apps unfortunately.

When getting a hold of Intel support, they suggested the problem is NOT overheating, due to the fact that the system is locking rather than shutting down or restarting. They did however remark that the idle temps were too high. The suggestion was that the memory was to blame, given that i5 2500 supports 1066/1333, not 1600. I'm tempted to call BS here, but want to be certain before I do.

 

Yeah, their comment about the memory speed causing a thermal problem seems a bit far-fetched. You can try "underclocking" the RAM to 1333 to see if it makes a difference, but I can't imagine RAM causing the CPU to run hot. It doesn't make sense (at least to me). What type of cooler is on the CPU? Is it the stock Intel cooler or something a bit more robust from a 3rd party? Also, I was just reading in another forum that some Asus boards had a BIOS routine that mis-read CPU core temps, so maybe an upgrade to the newest BIOS would help? I'll research this a bit more and post anything I might find....

Did the Asus diagnostics include a full memory diagnostic? If not, I suggest running MemTest86+ or Windows Memory Diagnostic. Have you run the diagnostic from your hard drive maker?

 

The said that while the temp was high, it wasn't what was causing the issue. Their argument was that using RAM speed higher than 1333 was causing stability issues, not heat issues. I'll flash the BIOS to see if that helps with the temp reporting.

 

I have run RAM rated at a higher speed than what the MB/CPU is rated for many, many, many, MANY, MANY, M A N Y times, and have NEVER had a stability problem, but this does not mean that this isn't the cause of the problem (yeah right ). As I suggested earlier, run some RAM and HD diagnostics, try to 'underclock' the RAM to 1333....


Oh yeah - you didn't answer - is your CPU cooler stock Intel or something else?

 

About the RAM, that's what I thought. It is the stock heatsink/fan though.

 

Also, just now I had a crash after lowering the memory speed to 1333, so that rules that out. I'll check the HDD and RAM now.

 

I didn't think the RAM speed had anything to do with it.

I don't mean any disrespect, but are you SURE there is nothing in the Event Viewer in either the System or Application sections that might indicate anything? I know we discussed it before, and it's not unusual for the Event Viewer to have no clues... here's an example: at work, I have a PC that would boot to Win7 desktop w/o any problems at least 15 or 20 times in a row, then it would suddenly crash early in to the boot process, and this was with a brand new hard drive and brand new memory; everything else is pretty freekin' old; the PC is an older HP tower that advertised the "greatness" of the Geforce4 MX440 AGP 8x video card and was sold with WinXP Home (to illustrate the age of this PC: no drivers for Win7 are available for Geforce4 GPUs; the owner of the PC thought that installing Win7 Ultimate was a good idea with 512mb RAM and an Intel P4 2.2ghz single core CPU w/o 64bit support rolleyes); the Event Viewer offered no insight to anything; the PC should never have had Win7 installed on it IMO due to its age; after installing WinXP Home again, everything (so far) is perfect......

....... uhhhhh ..... hmmmm ..... :confused I have no idea why I decided to share this, but I was going somewhere with it....
:-o

OK - my excuse - I've had a beer or four LOL

 

Yah, the RAM was just fine. I ran the Windows memory diagnostic with Extended settings and 3 passes overnight, no problems there.

No disrespect taken, I totally understand it's easy to overlook something. The only Critical events in the event viewer are Event ID 41 Source: Kernel-Power, which is a result of me having to shut the system down after it hang and becomes non-responsive. A review of the others shows no Hardware events and only errors related to unexpected system shutdowns.

 

Hmmmm..... this one is a bit of a stumper.... have you flashed the BIOS to the newest version available? You can use this direct link for the latest BIOS download. If unsure about how to flash the BIOS, read the info in the manual for the MB. I could be wrong, but a quick check at the Asus web site shows that they only have a DOS-based BIOS flash tool, available from this direct link. It's weird that they only have a DOS based update tool since just about every other MB maker has a simple .EXE file that you can run from Windows. There might be a BIOS update tool on the CD that came with your motherboard.... at this point, I really think that a BIOS update is a good idea, mainly because I don't have many ideas or suggestions left. rolleyes

Did you run a hard drive diagnostic? (use the extended/advanced tests; this program is a free download from your hard drive manufacturer's web site)

OH!! Another thought! It could be that your power supply is on the fritz. If you have another one laying around (or can borrow one from a friend), it wouldn't hurt to temporarily swap your unit out just to see if it makes any difference. Just be sure that the temp unit is healthy and strong enough to power your rig.

There is the chance that Intel support was right LOL Have you contacted Asus support?

Do the locks/crashes occur with any sort of regularity? For example: it mainly occurs when watching YouTube videos, or it usually happens when gaming, or it's most likely to lock/crash when simply surfing the web, or there is no rhyme or reason to it: it's completely random and can occur while sitting idle.

 

Ok, I have a few answers for you. I have not had a chance to run a HD diagnostic, but will do so if the problem reoccurs. The PSU I'm using is a Corsair TX750W, which makes me miss my previous modular PSU :cry, but is more than enough to power the GFX Card. I updated the BIOS today to the latest revision, but I won't have time to use the machine enough to test for problems for another 24-48hrs. For some reason, the website displays only the DOS tool, but the manual gives directions to install Asus AI Suite, which lets you update the BIOS from Windows. Here's what I got out of that so far.

http://img.photobucket.com/albums/v321/RainMan5419/AsusSuiteII.png

I did contact Asus, but that was using the high temps that the two monitoring apps reported, and they offered to RMA the board. With the temps that the Asus software is reporting though, heat doesn't seem to be the problem. There's no set app that I'm using when the system becomes non-responsive, and it happens often when I'm gaming. That said, I game quite a bit and it's happened more than once when browsing/listening to music.

 

Maybe it's just me, but I wasn't able to find any Corsair built apps to test my SSD. On the PSU front I downloaded Prime 95 and tested my system for 2hrs watching the 12V and 5V rails. The 12V +/-.2V and the 5V was +/-.1V.

While I wrote it off as an audio problem, I'm thinking that there MUST be something to the fact that when the system hangs it's frequently accompanied by an odd audio distortion. I'm going to pull out the sound card to see if the problem reoccurs.

I'm not ruling out the idea that Intel may be right, I have very little faith in Tier 1 support for most large companies. I also think that if RAM compatibility was the source of the problem it would have caused problems when I first assembled the PC rather than becoming incrementally more common over the last month.

 

As a follow up I really appreciate the help, and I was able to fix the issue. I'm not sure exactly when I made the mistake, but at some point the memory voltage was dropped to the 1.5V default instead of the adjusted 1.65V. Problem solved once I corrected this.