Guild Wars 2

Been having this issue quite a bit, play a little GW2 for a while, then suddenly PC turns off. Can be for just 20 minutes or for 8 hours. The timing seems odd. Used to think it was caused by Chrome, as I would have a large number of tabs open when it occurred. Past few times that has not been the case.

Specs:
Windows 7 professional x64
AMD FX-6100 3.3ghz (no OC)
GA-970-D3 Mobo (8/31/2011 BIOS version, noted as stable for GW2)
32 GB Ram on sticks and another 32 GB as paging (old habit to set it that way)
AMD Radeon 6850 (1-10-2013 version also noted as stable for GW2)

PC and Ram stress tests ran a week or so ago, 12 hours straight for each one, no issues.

Tried turning off cores on start up, issue still ocurred.

Temps (from speed fan 4.49, all temps in Celsius prior to starting GW2);

Temp 1 – 36
Temp 2 – 37
Temp 3 – 32
HD0 – 33
GPU – 40

Temps after running GW2 for 1 hour

Temp 1 – 37
Temp 2 – 56
Temp 3 – 63
HD0 – 32
GPU – 59

HIgh yes, but I’ve had those temps in the high 60s and low 70s and didn’t have my computer turn off.

Does it completely turn off or do your fans still stay spinning and the monitor gives you a message like no video input but you still have background music from your computer?

I, too, am experiencing the exact same issue – or rather, I was roughly one month ago, at which point the frustration made me take a hiatus from playing. Started playing again two or so days ago, and haven’t experienced it yet, but then I have deliberately avoided the steps that repeatedly caused the system to shut down in the past.

The symptoms are as follows: At unpredictable intervals, and only when playing GW2, the system shuts down completely. Identical to cutting the power by pulling the plug or holding the power button. No fans spinning, no lights on. Once I’ve had the system hang instead, showing a black screen, no audio, system still running, which forced me to reboot – but that may be a different issue altogether.

The most interesting thing I’ve deduced so far is that it only seems to happen when I’m playing my low-level thief. It can happen at any point, whatever I’m doing – buying a new skill, standing around town, running past enemies, you name it. But when I play, say, my warrior or engineer, I can play for hours on end without any issues. On my thief, one hour is the longest I’ve been able to play before the system shuts down. No event is recorded in Windows Event Viewer or in the crash log for GW2.

My computer is a little over one year old now, and I haven’t made any changes hardware- or software-wise between when GW2 worked fine even for my thief, and when the system started crashing irregularly when playing said thief. It has an AMD FX-8120 for a processor (overclocked to 4.1GHz, confirmed stable for the past year as well as after rigorous testing less than two weeks ago), 2×4GB of Corsair DDR3 SDRAM, and 2xRadeon 6950s in Crossfire (with stock settings), as well as a powerful 750W PSU that, for all my testing, works perfectly fine no matter what stress I put on it. I’m experiencing similar temperatures as Ferdoc (perhaps a smidge lower), and other applications do produce higher quicker than GW2. Also similar to Ferdoc, I use a GA-970A-UD3 motherboard with the 10/22/12 BIOS. Other quick info: 64-bit Win7, Eyefinity 3-screen setup (running GW2 as windowed fullscreen on the center monitor only), stability tested using AIDA64 and Prime95 for 12+ hours.

I don’t think there’s anything else I can add, but there’s my addition to the issue. I’ll be keeping an eye on this thread to see if you need anything else from me. Hopefully this issue can be resolved.

I had this issue as well a few months ago.
I resolved this issue by getting a new PSU. Specifically a XFX 750W XXX semi modular PSU.
The problem is that GW2 loves raw power. GW2 will max 3 heavy threads in demanding scenarios in the game such as WvW zergs or Dragon events.
If you can, try to use another PSU.
Check the system admin event viewer to see what the error is and correlate that to the issue.
In the System event log, look for EventID 41 Kernel-Power. It’s description is:

The system has rebooted without cleanly shutting down first. This error could be caused if the system stopped responding, crashed, or lost power unexpectedly.

I doubt it is the PSU – as stated, I did test it both with a multimeter and with a PSU tester I was kindly lent by an acquaintance who used to be a hobbyist in electronic maintenance and repair, and both came out perfectly fine. Again, this issue has not happened with any other application, before or since, and I haven’t made any hardware changes to the system since 13 months ago. If it is a case of GW2 drawing so much from the CPU that it’s destabilizing an otherwise stable system, perhaps this ought to be optimized?

Nevertheless, I’ll see if I can borrow a replacement PSU to temporarily switch to and see if it helps, but I’m not yet ready to spend money on a new good one – got a pretty hefty investment I need to make as it is in less than a week, and two more soon after that.

As an afterthought, if it really is a case of faulty PSU and we accept the theory that GW2 is the only application that reveals said fault, why would this happen only on my thief? I just find that theorem rather far-fetched. Pluralitas non est ponenda sine necessitate.

Food for thought.

Post scriptum: You mention that GW2 draws massive power “in demanding scenarios in the game”. If we can assume picking a new skill and/or standing still in a village without any dynamic events going on does not qualify as a demanding scenario, our two problems, while similar, most likely do not share a cause.

Jigain try resetting your overclocks to stock and see how that goes. I know it sucks but GW2 seems to be finicky with OC’s.
The Thief scenario could be related to the zone or area your toon is in. I find some areas are higher demanding than others.
Do you have MSI afterburner installed? Do you use HWinfo for cpu temps?
Hard shutdowns are generally associated to the PSU. Did you run Memtest to see if you had a bad stick?

EtonMessy – it would turn completely off. As in black screen, black keyboard (I have one that lights up the keys) mouse light off, PC light off, fans off. No power being drawn from the outlet.

Ashley Segovia – The newest piece of hardware I put in was my ram 6 months ago when I bought the PC. No error message, kernel dump only had a critical error about not enough power (at full tilt, my computer uses about 75% of my power supply.)

Note – I got the issue solved last night with help from a friend who is suffering from a similar but far worse scenario (Ashley, he’s got 8-30 FPS in low population areas of 3-10 people. And shut downs, his talks with anet support have yielded nothing.)

Jigain, here’s how I solved my issue. It turns out that while my BIOS is listed as compatible with CPU, it in fact wasn’t. I ended up updating to the F7 version (oldest known stable confirmed by multiple sources rather than just manufacturer).

Since then my comp has been running fine. What was the issue? My 6 cors ran as following at moderate useage.

Core 0 + 1 – 1507.2 MHz
Core 2 + 3 + 4 + 5 – 3.552.8mhz

All temps at 26C

Multiplier Core 0 + 1 – 1.5
Multiplier Core 2 + 3 + 4 +5 – 17.0

Core 0 and 1 were not responding and were calling for more and more power, without actually running. It was a fake kernel power supply issue.

Just for kittens and grins, check your BIOS version against known stables.

Stormcrow.7513:

Jigain try resetting your overclocks to stock and see how that goes. I know it sucks but GW2 seems to be finicky with OC’s.
The Thief scenario could be related to the zone or area your toon is in. I find some areas are higher demanding than others.
Do you have MSI afterburner installed? Do you use HWinfo for cpu temps?
Hard shutdowns are generally associated to the PSU. Did you run Memtest to see if you had a bad stick?

In the weekend, I’ll attempt some more gameplay with my thief – first to confirm whether or not the system shutdown still occurs, second to attempt to crash it again with stock CPU clock.
The zones have been all over the place. Just to name a few: Lion’s Arch, fountain; Queensdale, Beetletun; Quuensdale, Bandithaunt Caverns; Metrica Province, Loch Jezt; Caledon Forest, Ogham Wilds. As you can see, some of them include areas that are indeed very demanding, such as Lion’s Arch, but also places that have no reason being demanding, like Beetletun. Also, which zone I’m in does not correlate to my thief in particular – I have visited all of these areas with at least five other characters with either similar or severely more playtime, all uninterrupted by system shutdown.
I do not have MSI Afterburner installed currently, though I know I did at one point. For temperature and other sensor readings, I use AIDA64 – if you do not know of it, perhaps you have heard of its predecessor, Everest.
I have not run Memtest, but I have run Prime95 for over 12 hours (I was away that day, but I believe I came home after roughly 13 hours and 15 minutes), with no errors recorded whatsoever. Prime95 is a program designed to stress test the CPU primarily, in order to ensure system stability, especially after overclocking. It also catches memory errors, so I’m fairly confident that’s not an issue. I will run AIDA64’s cache and memory test, though, just to be sure.

Ferdoc.6429:

Jigain, here’s how I solved my issue. It turns out that while my BIOS is listed as compatible with CPU, it in fact wasn’t. I ended up updating to the F7 version (oldest known stable confirmed by multiple sources rather than just manufacturer).

Since then my comp has been running fine. What was the issue? My 6 cors ran as following at moderate useage.

Core 0 + 1 – 1507.2 MHz
Core 2 + 3 + 4 + 5 – 3.552.8mhz

All temps at 26C

Multiplier Core 0 + 1 – 1.5
Multiplier Core 2 + 3 + 4 +5 – 17.0

Core 0 and 1 were not responding and were calling for more and more power, without actually running. It was a fake kernel power supply issue.

Just for kittens and grins, check your BIOS version against known stables.

Where can I find which BIOS versions are stable? I don’t mind updating them again, but I am hesitant to downgrading them – the one I have is not even fully half a year old yet. Scratch that – I’m already running F7.

My thanks go out to both of you for your assistance.

I had this problem on my last computer and discovered that it was a bad stick of RAM. This is actually very easy to check – download a program called memtest and run it overnight and see if it is still running and reporting no errors in the morning. If it does error and you have more than one stick, repeat the test with each stick to identify which one it is and replace it.

Of course, a computer can crash like that for a lot of reasons hardware related – I mention this one because it’s an easy test to at least eliminate your memory as the culprit.

Not sure if I have an empty CD lying around, my optical drive has hardly been used since I bought the computer. I’ll look around, and if I have one, I’ll give Memtest a spin. See what I did there? A spin? It’s funny because a CD… ah, forget it.

I have used everest before, great piece of software.
Try to use memtest and see if you might have a bad stick.
The good news is that if it is the RAM your probably going to have an easy RMA or return for it.

I bet everyone is going to love me. Its started happening yet again, temps are running a bit cooler than before (cleaned out the PC), started the black screen with a full off (no fans).

Figured something might be wrong with vid card, updated vid card and still the issue remains. Gonna see if there is anything wrong with any other drivers and hardware.

Yay recurring issues…..

Definitely sounds like a hardware problem to me, here are my suggestions. First, check your vcore in your bios. It should be automatically set reasonably, but sometimes it can be set fairly high for stock settings (I’ve heard of boards setting it at ~1.45 for chips that should only need ~1.3 at stock).

Second, does your motherboard have a VRM heatsink (a heatsink directly to the left of the CPU)? Some 970A-D3s do, some don’t depending on the rev. #, but apparently you can call gigabyte and request a heatsink if yours didn’t come with one. Your problems could be caused by an overheating VRM, the shutdown being a protection feature to prevent your board from catching on fire. Regardless of whether or not it actually turns out to be the problem, getting the heatsink free from gigabyte seems worth it to me (you will need the heatsink if you ever want to seriously overclock on this board).

Third, try disabling turbo core in your bios, and see if that helps things. That “mini-overclock” could be just enough to destabilize your system.

Finally, what are your core temps under load?

Ashley – I’ll do that and let people know how it goes.

basketcase – its not a hardware problem at all. temps are below what other games run at, there is no OCing at all. The heatsinks all work properly.

The blackscreen issue as I understand it is 1 or 2 problems. Power supply is faulty or software conflicts.

It isn’t the power supply. The company that built my computer tested it, I tested it, gave it to some local repair shops to have them check it as well as some friends who repair computers at a major company near my house. Power supply came back 100% fine from all of them. Some even had the same event occur with them. They believe its a software conflict.

So here I’m off to contact Anet support and see what they say.. yay….

Little update on what’s happened since last:

*Crashes still only occur on one character
*Crashes still occur with two other flawless PSUs (both tested automatically and manually)
*Crashes still occur with other memory sticks (brand new, the old ones were good according to Memtest but I figured I might as well upgrade)

@basketcase:
I’m fairly sure (this is off the top of my head as I glanced at the BIOS yesterday but without looking for the voltage in particular) that the CPU voltage on my machine is defaulting to something akin to ~1.4. This seems reasonable as the stock voltage for the FX-8120 is 1.39. It does have a VRM heatsink, not that I think that causes my issue on only one character. Since I was adviced to try removing overclocking earlier in the thread, I did, and haven’t turned it on since – this includes turbo core. Finally, my CPU temp idle is ~25-30, running any game save GW2 ~53-56, running GW2 ~59-63. Celsius, natch. Haven’t actually checked if there’s any difference in temp if I play my thief versus any other character… might be worth looking into.

To be honest, at this point I’ve pretty much given up. It’s not like I can’t play the game, after all, only I’m not permitted to play a thief.

Hope you learn something from support, Ferdoc, that can help not only yourself, but also me and the numerous others I spotted glancing over the Tech Support forum having the exact same issue. I’ll fill in a ticket myself, and keep you updated.

Those temps look high for stock, considering the max temp in the amd spec is 61C. Likely not the issue anyway, I would try and reproduce this out of game by running prime&3dmark on a loop at the same time.

Those were admittedly overclocked temps grabbed out of memory. Not sure what they’re like now that I’m down on stock.

Gods, after running OC’d for so long that sounds like a disease… stock. shudders