P150HMx - Black screen freeze/crash

Hi guys,

It seems to me I tried quite a lot of things and this forum might be my last chance of actually getting any help. Sorry if this post becomes too long but I'm trying to just be specific as to the state of the PC with the problem. The original specs of the system are at the bottom of the message.

I've got a problem with my laptop which I bought around 2,5 years ago from Schenker Notebooks in Germany - it's an XMG P501 Pro. Since then it's gone through quite a lot: I've spilt tea on it once which resulted in me getting a new keyboard for it - at the same time I decided to swap the DVD-ROM for an SSD drive; then, around last November, the GPU failed me so I've sent it back to the people I'd bought it from to fix it. They swapped the Radeon card for an nVidia GTX 580m which I was quite happy with. I also decided at that time to format the drive and get a clean install of Windows 8 Pro.

I think it was just a bit later that I've started noticing that my PC would, every now and then, freeze with a black screen; the sound would either loop or not play anything or there would be a constant, high-pitched beep playing from the speakers/headphones. The fans would quite often start rotating a bit faster and I think it eventually just turns itself off - most often I do it with pressing the power button for around 5 sec. This started becoming more often around 2 months ago when it'd happen usually at least once a day but mostly when playing some games like Team Fortress 2 or Killing Floor.

I thought of using the warranty again but it'd been expired already and I didn't want to have to pay 200-300 EUR for having it sent to Germany and looked at. I tried using some software that checks drives and RAM for errors but couldn't find anything. I then looked at the Windows Action Centre and saw that there's "Problem with nVidia Graphics Driver" so I did a fresh install of the latest nVidia drivers using some programs to remove everything that was linked to the old ones and am now running the 337.88 version. The problem doesn't occur every day but it persists and also happens sometimes when I'm not gaming.

Is there anything left I could try and do about this apart from replacing the GPU again? Spending a couple of hundred quid on it doesn't look nice when I take into account the fact I'll most likely be unable to sell it for a similar price later on (checked on ebay and there's hardly any such laptops there). I'm also saving up to get a desktop around February so I'd like this laptop to actually last until then.

Here's the original specs written in German (although it shouldn't be a problem to people speaking English):

M4S XMG P501 PRO Gaming Notebook 39,6cm (15.6")
. 39,6cm (15.6") Full-HD (1920*1080) Glare
. AMD Radeon HD 6990M 2048 MB GDDR5
. Intel Core i7-2760QM - 2,40 - 3,50GHz 6MB
. 16GB (4x4096) SO-DIMM DDR3 RAM 1333MHz (nur mit Quadcore-CPU)
. 500GB SATA-II 7200U/Min Seagate Momentus 7200.4 (ST9500420AS)
. DVD SATA Multinorm Brenner
. Intel Centrino Advanced-N 6230 (inkl. Bluetooth)
. Basis-Garantie: 24M. Pickup&Return | 6M. Sofort-Reparatur -DE
. ohne Betriebssystem
. Microsoft Office 2010 Starter (nur bei Bestellung und
Vorinstallation von Windows 7)
. Keyboard U.S. International

 

Could be in result to overheating. Do you run any programs like HWMonitor to see what your temps are.
You can also open up the bottom of the computer to clean out any dust that is in the vents and fans.
You can reapply thermal compound to the CPU and GPU.

 

@Hutsady

Thx for your reply. I've been cleaning the inside of the laptop from dust every 3-4 months at least so I don't think that should be the problem.

I'll start using HWMonitor during gaming times to see if the temperatures are getting too high. Currently, the CPU cores are at ~44 C and the GPU is at 59 C - I think that's normal, isn't it?

Is it worth it to reapply the thermal pastes to both components if the temperatures are fine even during gaming? Do you have any other idea as to what else I might do?

Edit: one of my friends told me he'd had a similar problem in the past and solved it by updating the BIOS. Could that help? I couldn't find any downloadable BIOS for my laptop on the net...

 

what programs did you use to check for errors on the ram and hard drive? mem test and hd tune in sig below.

also can you remember how you cleaned the nvidia drivers and which program you used and also was it in safe mode.

those temps look ok but the gpu is up by a couple of C but that could well be down to ambient room temperature. normal idle is 50-55c but this can vary slightly depending on the various gpu's there are.
a repaste may well bring it down a few more degrees.

the main question is what sort of temperatures are you getting while gaming.

theres lots of free software for monitoring in my signature below. hw monitor is a good one but also try core temps for cpu and nvidia inspector for gpu.

also just incase you didnt do a safemode clean you could try that for the nvidia drivers crap left behind as its amazing how much junk is found using various cleaners. walkthrough in sig below as well.

 

You temps idle seem OK given its summer time and ambient temps are higher. What are they under load?

Also MrDJ suggestions for memetest and HD tune are good next steps.

 

Thanks for your responses chaps. Here's a breakdown of what I've done so far:

- I used HWMonitor to check the temps: CPU min.: 43-44 C, max.: 69-70 C. Today while playing Killing Floor (no freeze today) it was around 59-62 C, depending on the core. As for GPU, min.: 64 C, max.: 93 C. Today while playing Killing Floor it was around 73-65 C afair. I play it on high details so when there's a lot going on the screen I can actually hear the fans working harder but I guess that's natural.
- I'm based in the UK atm, so the ambient temperatures aren't too high - around 17-19 C today.
- I'm fairly sure I used Memtest to check the drives but didn't seem to find anything that'd strike me as out-of-the-ordinary (though I'm no expert on this); I'm happy to perform the test again if you need me to paste any results here.
- To remove old drivers I used something called Display Driver Uninstaller and I remember doing that in safe mode.

Should I use Memtest then? And should I use something other than HWMonitor for the temps? It does show the temperature of each of the CPU cores. I'll also monitor the GPU temperatures straight after the crash (when it happens next time) and let you know what they were.
Is it a good idea to get some thermal paste already or could there be another problem causing this behaviour? Also, I'm assuming no messing with BIOS is going to help, is it?

 

Seeing 93C during gaming makes me really suggest redoing the thermal compound as your next step.
Its not a dangerous but definitely could be lower. Mid 80's C is ideal.

 

@Hutsady, could you maybe suggest a particular brand and type of thermal paste I should use for this? And a tutorial with photos? I've never done this particular bit before...

 

I personally always use IC Diamond Thermal Compound. They also have instructions here Application

 

Gelid GC Extreme is another good compound. It can be applied in the same manner as the IC Diamond compound Hutsady recommended.

 

I've used IC diamond 90% of the time, and even now, but I was wondering if Liquid Pro or Liquid Ultra was a better candidate, in which it is better for overclocking.

 

In a lot of testing, the Liquid Pro/Ultra has bested the IC Diamond and Gelid GC Extreme compounds by 1 - 3 degrees Celsius. It's an improvement, but not a drastic one. Still, the Liquid Pro/Ultra compound is conductive (if I remember correctly), so more care needs to be exercised in application.

 

Ok, I finally got to doing this!

I used some "Arctic Silver ArctiClean - 60ml Thermal Material Remover and Surface Cleaner" to get rid of the old paste and then put "IC Diamond 7 Carat" generously in its place - the amount was at least a pea-sized drop.

The temperature of the GPU after starting the laptop was 55 C, once I started up some programs, it rose to 59 C but it looks like it's still a bit lower than before.

I'll bump this thread if there's still some crashes/freezes. For now thanks for your help chaps

 

Ehhh, unfortunately I'm returning to bump the thread - I just had another crash. Anyone have any more ideas as to what I should try out next?

 

How are your temps now that the IC Diamond has been applied for a while.
You can also install stress test programs like Furmark and Prime95 to see if you get the sudden crashes with those. If they do then you can be more confident its thermal releated.

 

Well, the default temp. seems to be 58-62 C as far as I can see, so it's a bit lower than it was before applying the paste.
The last 2 crashes I had (1 yesterday and then the one about which I wrote in my previous post) happened not during playing anything GPU consuming but rather some times afterwards, i.e. my gf would play Witcher 1 (pretty much all of the graphics settings maxed out) for 2-3h, then I sit to my laptop, run Firefox and the usual stuff, then maybe a game like The Black Mirror (it should run without problems on basically anything) and then after, say, 30 minutes I get a crash.

Now, I'm not 100% sure about this but I think I ran HWMonitor after the previous crash and the temperature was around 86 C but I didn't notice the fan to be running any faster than usually when it's 60 C. I do hear it whilst (properly) gaming, though. Is it possible that the issue could lie there - e.g. fan not working properly?

Also, how would I go about testing my laptop with Furmark? Should I just run it fullscreen as-is and leave it for, say, 2h and check if there's a crash?
I just ran it for 5 min. and the temperature was reaching 97 C and the GPU started throttling but I guess that's to be expected, isn't it?

 

You will want to make sure you have HWMonitor running during your gaming as that is when the highest temps will be. They drop pretty quickly once you shut down the game.

When doing Furmark, again have HWMonitor open like you have been and see what the temps get to. That program will stress the GPU more then any game will so if it shuts down during that it pretty much tells you you're having GPU thermal issues. Also run Prime95 as that will stress the CPU and that could be whats causing your shut downs.

 

It could be the solder going on the card, once cooled down it may be contracting and no longer forming a good connection.

 

Ok, I've used both Furmark and Prime95 for more than 1h each. The GPU temperatures in the former case reached around 97-98 C and the CPU temperatures in the latter -- 86-89 C, depending on the core. I got no crash during any of these tests. The only crash I had was ca. 2h after the first Furmark test during something other than gaming - web surfing, music, etc.

Any other ideas as to what I might check?

 

Those temps seem about right for Furmark. You can try with just one stick of RAM in at a time, but generally RAM issues will first BSOD before shutting down.