Guild Wars 2

Today I installed a geforce gtx 550 ti and sense then when I play for about 20 minutes my comp completely shuts off. I’ve been scouring the web and have come up with the usual culprits such as power supply not efficient enough and overheating. I have been monitoring my temps and while in game shortly before it shuts off it hits about 66 degrees celsius, which doesn’t seem high enough to cause shut down. ( however, my case does feel hotter than normal right after shut down)
I popped open my comp and I have a 500watt power supply which I believe to be strong enough to run the card…but i suppose there is trouble with it trying to run everything along with the upgraded card?
I do have the most recent Nvidia driver….seems like that update is giving people problems.
I played about 30 minutes of Diablo 3 without shut down.
Some Specs:
quad core 64bit processor 3.60ghz
8 gigs of ram
MOBO model number: M68MT-S2

Is the vid card hitting 66 degrees or the CPU?

GPU’s can usually handle upwards of 80-90, but CPUs often top out in the 60s.

500W should be enough for what you’re running. Typically this wouldn’t just flat shut off your PC though. It typically causes legitimate crashes or system resets.

It’s possible that your previous vid card bottlenecked your CPU to a point that it never worked hard enough to overheat, but now that you don’t have that bottleneck, the CPU is able to work at 100% or something along those lines. Just a theory.

66 is the GPU temp. I’ll try and see what my CPU temp is getting to.
Interesting theory..haven’t thought about that before, and not too sure how to go about fixing that..
Fans are working properly, even have my GPU fan going faster than normal.

Thanks Gaile, it might just be my power supply…just wishing it wasn’t hah.

CPU seems to heat up a good bit while playing. Been playing for only abour 3 minutes, and with the game tabbed out in the background its at like 46 degrees celsius. I can only imagine that after about 20 minutes of gameplay it heats up to probably around 60.

Shakawkaw.4053:

Thanks Gaile, it might just be my power supply…just wishing it wasn’t hah.

Well, i assumed that 500W would be enough, but I’m not an expert on the subject, so that would be the next logical step. I encourage you to buy one from a local place, such as Best Buy, etc.

PSU’s seem to have a surprisingly high failure rate, and you don’t want to have to wait for an RMA if it happens to be bad when it arrives.

How hot is your power supply getting. You can try to clean the thing out. If you have a air compressor that works the best if you use it carefully. I take my comp outside and blow the thing out every few months.

skaaz.4281:

How hot is your power supply getting. You can try to clean the thing out. If you have a air compressor that works the best if you use it carefully. I take my comp outside and blow the thing out every few months.

Yeah I dusted it all out when I put in the new graphics card. I dunno about my PS temp, but if im running GW2 the CPU gets up past 45c, and I’m assuming it gets hot enough to cause shut down, seeing as how my GPU only gets to about 66c

The CPU and GPU temps you listed are completely acceptable.

before replacing the PSU, you should check if the Cpu Cooler fan is working fine and doesnt not have dirt or hair that make it work slower, then check also the PSU fan. and sniff the surrounding area for any burned smell, overheated dust has distinctive scent) old and dirty fans are among the first causes of such problems.
if the shut down reason is the CPU you should feel a the performance going down for few seconds before the shut down, in many cases it comes with a blue screen.
in case of PSU the shut down is always imediate.
since you say you can play for 20 minutes or so… it means there is a overheating problem somewhere. and the defense mechanisms shut your system down.
old PSU can also be a reason. as they drop 10-15% in performance every year, if the PSU was 500w (400watts really) brand new. it may be delivering properly only 300-320w after 3 years. happens quicker and more frequently with Generic made PSU’s.

Simon De Borovsk.7460:

before replacing the PSU, you should check if the Cpu Cooler fan is working fine and doesnt not have dirt or hair that make it work slower, then check also the PSU fan. and sniff the surrounding area for any burned smell, overheated dust has distinctive scent) old and dirty fans are among the first causes of such problems.
if the shut down reason is the CPU you should feel a the performance going down for few seconds before the shut down, in many cases it comes with a blue screen.
in case of PSU the shut down is always imediate.
since you say you can play for 20 minutes or so… it means there is a overheating problem somewhere. and the defense mechanisms shut your system down.
old PSU can also be a reason. as they drop 10-15% in performance every year, if the PSU was 500w (400watts really) brand new. it may be delivering properly only 300-320w after 3 years. happens quicker and more frequently with Generic made PSU’s.

Yeah, the first couple of times its happened i noticed a burning smell, like really hot so I freaked out and left it off for a while. I can pop it open and clean them again, or possibly look into buying new fans?
The shut down IS immediate, and the case does feel hot afterward, more so where my mobo is located.
I haven’t had my comp for 3 years, and don’t have it on ALL the time, so I thought my PS was ok for this new card, but it seems more and more like I just gotta get a better one.

one last thing, you can also download a program called OCCT from http://www.ocbase.com/
and stress test your system independently of any game. test the Cpu Gpu and PSU.
this program is used by overclockers to monitor system stability.
a stable system is a system that passes all the tests without shutting down.

if its the PSU fan that is failing, dont bother on fixing it, its either quite old PSU, hence its real performance is much lower than the Nominal (500w). in such case replace it with a new one, and from a known manufacturer (Corsair, Seasonic, TT, Antec etc) and get one with a 80 Bronze label or up (Silver, Gold, Platinum, those mean the efficiency of the PSU is higher).

I might look at replacing the heat sink on my CPU as well as a precaution. I really hope no damage is being done to my mobo.

Shakawkaw.4053:

I might look at replacing the heat sink on my CPU as well as a precaution. I really hope no damage is being done to my mobo.

It’s not your CPU. You said it was at 45C. That’s completely normal.

Hickeroar.9734:

Shakawkaw.4053:

I might look at replacing the heat sink on my CPU as well as a precaution. I really hope no damage is being done to my mobo.

It’s not your CPU. You said it was at 45C. That’s completely normal.

Yeah I’m just paranoid at this point, because of the heat/burning smell I get after it shuts off.
I’ll be looking into new PSs after I check out my fans and stuff

Overheating CPU’s have a protection commonly known as Throttling. i am not going to explain that here, but as example: when my CPU is reaching 105.1C (yes) a process of frequency scaling starts, which slows down the cpu until it reaches acceptable temperature. Every Cpu generation and brand has its own set temp limit point.

45C In your case… is almost cold.

Many Video Cards have a fairly hefty Amperage requirement on the 12V Rail of your powersupply. A bit of Google-Fu shows that a 550ti needs 24amps minimum from the 12V output.

For buying a new power supply… wattage alone doesn’t tell you everything. You need to look at the output ratings, which look like “+3.3V@25A, +5V@25A, +12V@46A, -12V@0.8A, +5VSB@3.0A”, that example lists a 46amp output for the 12 Volt Rail.

However, there are other possible causes to your problem. Most motherboards BIOS will shut down the system if it is going beyond the heat-threshold that it is set to. It is a possibility that your BIOS is improperly set to a low temperature.

I have had one instance of a system shutting down randomly while underload, but there was no fix for it other than to replace the motherboard. A capacitor had gone bad and I’m terrible at soldering.

Waiiiiiiiiit. Do something before you replace your PS.
Remove your Memory Dimm’s and put them again. Make sure that all the dimm’s fits perfectly.

I had a pretty flaky system, which would lock up and also reboot on its own. I ended doing two things which seems to have improved stability:

1) Updated the firmware on my SSD card
2) Increased the PCH voltage in the BIOS to ~1.15V.

Particularly the latter seems to have made a big difference (this is apparently a common issue with mobos with recent Intel chipsets). Of course YMMV.

Welcome to the club of the people whith random sutdown with NO
Inadequate Power Supply and NO Overheating.
Actaully there are a lot of people with that problem, but their messages get lost with people having different crash problems.

Symptoms: after a random amount of time, the screen goes black with “no signal”. However the sound may continue to play.

However the temperatures are OK (passed 3DMark benchmark making the GPU and CPU go to the highest temperature with no crash)
Same for power supply (the benshmark make the computer use a lot more power than GW2, and I have 850W from a good brand, don’t tell me that’s not enough)

Reformating does not change anything, using different version of drivers does not change anything, modifiying graphical options does not change anything (well, maybe it changes the time between crashes, but difficult to be sure)

Underclocking the card seem to increase the time between the crashes a lot and even “solve” the problem for many people with this specific problem (If you call that a solution)

(If you want to try this solution: Nvidia Inspector
http://downloads.guru3d.com/NVIDIA-Inspector-1.9.6.6-download-2612.html)

Good luck

@Kethryes

“Symptoms: after a random amount of time, the screen goes black with “no signal”. However the sound may continue to play.”

This is a slightly different symptom to the OP whos computer shuts down completely. Yours on the other hand just appears to lose video. You mention that GPU temps are fine, I just wanted to confirm that you are certain as this does look like possibly an overheating GPU.

You mention underclocking helps, this could be because underclocking is leaving it running at a cooler temp.

Also you are correct in that 850w would be more than enough for most systems but that doesn’t mean that the PSU is not faulty or necessarily up to the job (see Owyn’s post above), I’m guessing your video card has a connector straight into the PSU? Have you a way of testing a different PSU? Perhaps you can take one out a friends machine just to test for a bit and see if your system becomes more stable?

Also, is it possible you are suffering from this issue?

https://forum-en.gw2archive.eu/forum/support/tech/Black-Screen-Issue-Nvidia

@Laurs: I think the OP might have a similar issue to me because it looks like the whole coputer locked down, when actually it is “just” the video (well the keyboard does not respond either so it is a bit more than the video, but sound still works, for a while at least)
Of course it is true it could be a different problem.

For the temperature, I played Skyrim with “ultra” graphics, for hours and it made my GPU reach 75 degrees, without a crash for more than 200h of playing. With GW2 I never go higher than 70, and I can have the lockup within seconds of starting the game sometime, my graphic card could not have overheated in such little time (unless GW2 is making my GPU fan stop, which is absurd)
And again if it was realy related to my hardware, GPU, power supply or anything else, I should have the problem when doing benchmark, memtest, or playing Skyrim…

And for your last remark it is not that. I tried modifying every parameters in the Nvidia panel including this one, but that did not solve the problem.
Undercloacking did, but I have no idea why…

Edit: Actually this thread just came back to the top.
https://forum-en.gw2archive.eu/forum/support/tech/Crash-Black-Screen-Issue-NVIDIA
But it is the exact issue I was talking about: lots of people with relatively similar symptoms, that actually come from very different problems…

Just to be clear, the game cannot make your computer shut down, software can not cause this. Drivers can cause blue screens but nothing software based will make your computer turn off. (Apart from windows shutdown, before someone tries to be a smarty pants :P)

If it is just turning off like that, this is generally due to protection mechanisms kicking in.

The most likely one as others have said is power supply, if that gets overloaded or too hot it will power itself down to prevent itself from exploding, which would have done even more damage.

It is possible that other components like the graphics card or sound card “might” have something in place, but it’s unlikely. If these are the culprit, it’s more likely with heat expansion, it is causing a short which the power supply then responds with immediate shutdown.

Hi. Had this same problem when I installed my 550 Ti a few weeks ago. My power supply is 500W as well – I thought it might be insufficient even though it meets requirements for this particular card.

However, going into BIOS (Press DELETE when your computer first starts up to get into this menu) and tinkering with fan settings on CPU (setting them higher than they were), as well as cleaning the inside of the computer case of dust using compressed air and a vacuum cleaner (VERY carefully) solved the problem for me and I haven’t had a shut down since. Make sure you get all the dust and lint out of the heatsink (metal grill with fan) over the processor – it really gets stuck in there good!

Also helps to take out your memory modules one by one and blow the dust off them and out of the slots, swap them around a bit too for extra luck

Also, make sure the sides of your PC where the air is getting pumped out are in an open space, not close to a wall or desk as this will also result in overheating. Good luck and hope this helps!

The reason underclocking helps, is because it lowers temps on your card, which lowers temps on everything else. And people stating “oh 45-50 isnt that high” actually it is. And Regardless of people going hotter, it’s still hot. Certain motherboards, and certain components just do not like the heat, and some can take it quite well.

It’s also possible that your Bios has an ‘auto off’ protection for a certain heat range of your motherboard and cpu, and is powering down to protect itself. It definitely could simply be a voltage issue, but that’s an expensive guess. No question that the new power cards are in fact power hungry, and its doubtful a 500 watt will do the trick.

But in addition, even with it getting all the juice it wants, it sounds like you might need to look at maybe more case fans, or higher powered ones, if you don’t want to force downclocking to keep the temps down for your current setup. I’d check Newegg for a good price on a power supply…I know my Radeon 6870 hasn’t been happy with mine…