Guild Wars 2

So stable that underclocking made it better? You’ve just listed every situation during which this game would have the highest framerate and highest load on the GPU, then claimed that running your hardware below spec stopped the crashes.
Well that can’t be right since I’m supposed to be an elitist nerdling that doesn’t run my systems every day and hasn’t seen signs of failing hardware. No, you’re system is most definitely fine.
(Hint: the slowdowns were/are the GPU driver automatically underclocking the card to the next lowest perf level (likely ~400MHz) when it detects some instability)

Really, every situation? Walking around a dead server in the map by myself, adjusting my inventory in a field with a rabbit to keep me company, puts more load on me than pvp, group events full to the brim, world events, and the odd champion train i once jumped on in queens? Can I have some of what you’re smoking?

Have you completely missed the fact that this game defers to the CPU over the GPU to run this game? Last chance to jump on the clue train buddy: having to underclock my card – which was running absolutely fine at its factory settings before encountering GW2- to fix a problem in a game that confesses it relies much more on CPU juice in the first place (As opposed to never even encountering a problem that needed to be solved by underclocking in every other game mentioned above and every other game I’ve played over the last 6 months that relies far more on the GPU) doesn’t point to failing hardware. It points to a problem with how the game utilizes my hardware AND the hardware of every other person who has been able to solve this problem by underclocking. I mean, really, you have to be missing a lobe to think it’s more likely that the fact that underclocking JUST IN GUILD WARS works for so many people here points to everyone having failing hardware, and not to the one variable that remains constant on everyone’s system – GW2.

The issue I have with people like you is that you come out and try to stroke your own ego by trying to stir up a fear party that someone’s hardware is failing out FIRST, before even trying to suggest any other far likelier options. No buddy, I’m not going to appease you by wasting 12 hours running a mem test when my system works fine in every other regard and with every other game and every other thing I do in it – as good as the first day I built it. The only reason I can see that your type of people makes this suggestion first and demands it to be the most likely solution is because you love to stroke your ego, insult the intelligence of others who know how their system runs, and knows when something in their system is going wrong, and because your type loves to come back at whatever sap does attempt to appease you with a full night of your diagnostics that result in a bill of clean health with “Sometimes mem test fails.”

You were wrong. Get over it.

Hey, was having the same issue as you for a while. You’ve got better specs on me in everything, but we share the same graphics card. My crashes started out as a black screen after about 3-4 hours of playing, sometimes just rebooting my system for me, sometimes just hanging on the Black screen for 3-5 secs then coming back to my desktop with GW still running minimized.

Gradually problem got worse over time – crashing after 30-60 min in world, 5-10 in every dungeon. Event viewer/bug checks always pointed towards nvidia driver when it pointed anywhere at all – most of the time it didn’t tell me much of anything.

From what I know this kind of black screen locking up without a BSOD (esp given that at times I could return to GW in the bar after a long pause and stutter) is your card performing a memory dump.

I began by rolling back my driver to 314 – didn’t help. I rolled my driver back to 306 – made the whole system much more unstable. I reinstalled windows and tried 314 again – same problem.

In the end what I had to do was install the 320.49 driver (last driver having any changes with the 660 in mind). After that world seemed more stable, but I always still crashed in dungeon after 5-10 min. Finally, i bit the bullet and under clocked with Evga Precision (MSI afterburner and several other utilities can also do it). The underclocking will not harm your system, it made GW run better in all aspects for me, and you can easily revert to default clocking in the program outside of GW while playing something else. It’s all managed in the utility.

Underclock yourself by 100 GPU and 10 Mem. Let the program run in the background while you play.

Also worth looking into is if your Intel Raid driver is up to date – for my gf, this seems to have been her problem with crashes. They released a new driver last month. intel link

Finally, google the program “Who crashed” – it is a much better program for investigating driver related crashing and is much more of a help than event viewer.

deltaconnected.4859:

How did you assert that the system really is stable if you don’t wanna run some basic stability checks like memtest, prime95, or Heaven? “Take my word for it” isn’t a valid answer here given any electrical component can degrade over time. From what I remember, WoW, D3, and DotA2 don’t even come close to stressing a system like GW2 does.

Milkduds.7109:

He isn’t getting stop errors in the sense of his sytem turning blue with a vague code, most of the time the system doesn’t even generate a legitimate bugcheck or stop code. Just lines of zeros. It’s clearly a video issue.

Legitimate bugcheck? Lines of zeros? Do explain.

IMO, this sounds like a dying 660. Pull the card and try it in your own system. Given the frequency of the crashes, it shouldn’t take long to rule this out.

Wow, what a self serving donkey show of a reply. I am the boyfriend that HAD the problems she posted. The system was completely stable before, had no problems before GW2 and outside GW2, and the frequency of crashing was not related to my system being stressed. I had no degradation in my overall system stability – the crashes only intensified while playing GW2. Rolling back to 314 drivers helped a little. Rolling back to stock windows update drivers made it worse.

What I did was do a clean wipe of my drivers with Display Driver Uninstaller. I then installed the nvid 320.49 drivers, the last driver created with specific changes relating to the 660. Then used Evga Precision to underclock gpu by 100 mhz and memory by 10. Also went into nvid control panel, tweaked settings (single display optimization, favor performance in power management mode).

Since then, not one crash, and in general the whole game is more snappy and runs much more smoothly. Finally got to finish a dungeon instead of crashing 5-10 minutes in (please note i had no problems with pvp or world events before this, my crashes were concentrated to dungeons and when I’d walk around by myself in the world, once in a story quest, and once in Lions Arch). I also never have the random fps drops anymore that would occasionally happen (Not just in heavily populated areas but even more often in the world walking around by myself).

Clearly, there is some issue out there that far too many people here experience to just first write them off as having dying systems. To be honest, that attitude comes off as incredibly elitist and dismissive, and pretty ignorant because no matter what you may think, this game doesn’t require a nasa supercomputer to run. If it did, the crashes would be related to relative stress at the time of crash, not all over the place as people report.

In any case, I’d chalk this up to the architecture of the coding – there seems to be a weird disconnect in how the game handles distributing stress to the cpu and gpu, then the resulting back and forth feedback between the CPU and GPU. Much more likely than some oligophrenic theory like a massive wave of people’s hardware failing coincidentally upon installing this game.