Asus G73SW Notebook BSODs

[johnm123]

Thanks for the pointers and explanation for setting up prime95, Dave.

I'll get back to run the seagate tool, then memtest again, then prime95 with the settings you suggested.

For how long should I run prime95, 4 hours?

Also, can all of these stress tests cause any damage to the hardware (given the temperature stays below the 77°C you mentioned)?

Edit: Ran all seagate tests available, all passed successfully.

 

[Dave76]

You're welcome.

Glad to hear the HDDs have past the tests.

Hard drive tests check for errors with low stress, and the memory test is fairly benign.
The CPU test Prime95 runs the CPU at maximum frequency which will use maximum voltage and cause high heat, damage can occur to the CPU if the heat is too high. This is why it is very important to monitor the temps with programs like RealTemp and CoreTemp.

Intel's maximum CPU core temp is 77degC, many overclockers go above this, I like to stay below it.
During the Prime95 test the CPU temp on your laptop should be around 60degC.

The test length is an ongoing controversy, hard core OC'ers say minimum 12-24 hours. Some OC'ers and Enthusiasts say 8-12 hours. Then there are some that say 1-4 hours are adequate.
It depends on your usage, self built desktop intended for heavy usage like gaming/folding would need the maximum, self built desktop OC'ed for 'normal' usage somewhere in the middle. Any OEM desktop or laptop would require the minimum as these can't be overclocked and are testing for component stability.
This is a disclaimer for anyone that may disagree with my recommended testing period.

Long story short, 1-2 hours will be enough for your purpose.

 

[johnm123]

Thanks Dave, sounds good.

Wondering, you say the CPU temp on my laptop during testing should be 60°C? I just did a small test and within minutes I was at, believe it or not, 77°C. Fortunately it didn't go higher but I'm not sure if 1-2hrs of testing are going to be too much for the hardware given that it reaches the max temp within a matter of seconds?

 

[Dave76]

I wouldn't run it that hot.

That is indicating a cooling problem, I searched for reviews of your laptop max stress temps and found 58-60°C and they mentioned that this laptop runs cool, several times.

At that temp the CPU will throttle down which if not causing the crash may be contributing to it.

Since it is new I would recommend returning it for repair or replacement.
It shouldn't be filled with dust yet, so that means loose heat sink or faulty fans.

If you want, open it up and see if there is anything obviously loose like heatsink and fans.

Let us know what you decide to do.

 

[johnm123]

Thanks for your reply, Dave.

I just looked as well and while I found several reviews stating ~60°C under max load was normal, Anandtech reported a max temp of 75° so I'm not sure if the manufacturer will take back that model if I just say "those reviews say it should be cooler than it is so please replace my version"? I never had to replace a laptop that's why I'm not sure if a reason like this would qualify.

Now when I tested again, I hear that on the left side of the laptop the fan is actually louder than on the right side. Also, on the left fan outlet (where the air comes out), a LOT more air gets pushed through than on the right side. I guess that definitely points to a fan problem then which could be causing those high temps? My idle temp is between 40-45°C as well which appears to be quite high given the 30°C other customers report.

 

[Dave76]

Your Intel Core i7 2630QM spec sheet does say max core temp (Tjunction) is 100°C, this is the max temp before the CPU throttles down to save itself. Doesn't mean you should run near that temp.
Strange that some sites are say max temps they see are 60°C and others saying 75°C.
I would check with manufacturer for the max temps.
If you are getting idle temps of 40-45°C that confirms a cooling problem.

Check your manual if the right fan is suppose to be exhaust or intake. I would think it's exhaust, which means that needs to be looked at.

 

[johnm123]

The design itself already suggests that the right fan should be exhaust as well but I looked through the manual and couldn't find anything. It just covers basic stuff like what kind of fn keys are there, status lights, setting up power options in windows etc. but it doesn't talk about the interiors. Here is a link if you want to take a look at it (after specifying the OS, scroll to the bottom for "manual"): Asus Support

Already contacted the manufacturer and will see what they have to say on the issue. They'll probably get back to me sometime around next week since we have the eastern weekend now.

Thanks for your help so far, I'll keep this thread updated.

 

[Dave76]

Let us know the progress.

What is your 'Power Option' setting?

Have looked at the manual, the OEM manual is very basic as most are.

 

[johnm123]

Here are a few news:

1. Completely reinstalled windows yesterday, it really seems that the bulk of the bluescreens came from daemontools (buffer overflow) and zonealarm (wouldn't allow some service to start).

After I got rid of them, I could only partly reproduce bluescreens when using verifier.exe on all selected drivers. (when split into test groups, the BSODs vanished and even with all drivers selected BSODs happened only about 5 times every 7 restarts).

This is fine with me since I just want to work and play on this laptop so it seems those BSODs are gone now (it's not like I'm running verifier.exe every day).

2. After having tested 3d mark vantage (a graphics benchmark), the right fan is blowing out air as well! I still haven't found any hardware sheet that shows where each component is and what it does exactly, but it seems that when stress testing the CPU the left fan goes up and the same happens on the right side when stressing the graphics card.

However, when I let these tests run for like 30 minutes, I saw max temps of 80°C which is probably still higher than it's supposed to be. I've seen my PC's graphics card freeze at around 78°C already so this is definitely not good (even though my laptop didn't freeze/restart).

Dave, where would I find my "power option" setting?

Thanks,
John M.

 

[Dave76]

Good news the crashes have stopped.

I still think they are/were being caused by the temps.
Check with the seller or Asus about fixing this issue, it might just be that the CPU heatsink needs the thermal paste re-applied or the heatsink is loose.

Go to the Control Panel, select 'Power Options', should be set to 'Balanced'.
This will allow the EIST, which will let the CPU slow down when the processing speed isn't required.

 

[johnm123]

Thanks for the suggestions Dave, but today I've sent my laptop back since the seller agreed as well that those temperatures would be too high.

I'm not 100% sure if they'll send me a replacement straight away or take a few weeks to look at the issue and then decide if they give me a replacement or send back the repaired one. Fingers crossed that whatever comes back will work without any issues.

 

[Dave76]

This will be the best choice, even if it means not having your new laptop for a while, you'll be glad you did after you get it back.

Let us know when you get it and how it's running.

 

[johnm123]

Ok, so this was really just a big waste of time. Got it back today, they said they couldn't find any problems and didn't repair anything which I find really stupid since the retail store told me as well that I should send it in for repair since those temps would be too high. Just ran some tests again with prime95 etc. and temperatures are as before.

Well, I guess this is the end of the story and I have to hope that my CPU and GPU both live as long as possible

Edit: To add a little bit to this, I just see they ran Furmark and put the results.htm on my desktop, in which I can see that all furmark tests passed. Wow, this is how technicians fix hardware errors?

 

[Dave76]

Sorry to hear that, they probably sent it to a third party to be checked, I hope Asus wouldn't do that kind of warranty work.

I would be sending an email to Asus about this issue.

Let us know how it is going.

 

[johnm123]

Unfortunately, they did send it back to the manufacturer, Asus. I read a thing or two about their support and well, I'm not too surprised. Not sure if it makes sense to email them nevertheless because they'd usually either say "sorry, we couldn't find anything" or "we can't solve this via email, please send your laptop in for repair" just to get the same result.

Just hypothetically, what would happen to the notebook if either CPU or GPU would overheat? What kind of damage would be caused?

 

[Dave76]

If the CPU gets too hot it can throttle down and/or crash.
The GPU will usually just stop working.

Worst case they will burn out, become non-usable, and possibly take out the motherboard.

You might consider returning the laptop for replacement, maybe another model, or refund.

 

[johnm123]

Well, if something like that happened (burn out, become unusable, needs a new motherboard), that would fall under a guarantee replacement, wouldn't it?

 

[Dave76]

Yes, that should fall under the warranty.

How long is your warranty?
Take that and add one day and that's when it might fail.

 

[johnm123]

The warranty is valid for 24 months so the notebook should be safe until march 2014. By then the same model should cost just a couple hundred dollars and who knows what else is going to be available.

I guess this case can be closed then, I thank you for your help and all the others that analyzed the BSOD's and put their knowledge and intelligence together to figure these things out!

Kind Regards,
John M.

 

[Dave76]

Good luck and let us know if you have any other issues.