25 replies to this topic

[Mazzyplz]

like the title says...

i played mwo for a long time.. and i upgraded to this vid card gtx770 about a year ago.

everything works fine. i would record gameplay with geforce experience.
but my vid card started giving me a blue screen of death error. something about nvidia's new drivers messed up my video card completely.

i had that problem for a while until i got rid of geforce experience completely and installed old drivers for my card.
this is the driver number that i installed.

347.88-notebook-win8-win7-64bit-international-whql

so it worked perfectly. seems my pc was saved.
since i could not record footage with geforce experience anymore and fraps kills the Fps...
i switched to Open Broadcast Software. which is apparently a popular program for streaming, and it can also record gameplay.

i recorded about 20 matches with it, and all of a sudden my pc started shutting down at random when playing mwo. i mean the power source would shut down my pc.
i guess to prevent damage.

i had that happen 3-4 times already! at first i thought it was maybe the wall plug or something, but after playing street fighter 4 for days and having no crash i went back and it shut down the pc again during mwo.

i decided i had to try something else and switched from dx11 to dx9 in mwo, as well as use the 32 bit client instead of 64 bit client.

after a couple of games.. mwo crashed during a match into black screen. and the last half a second of sound loops endlessly. luckily this time i could alt tab back into windows and kill the mwo process from task manager.

now i don't know what to do.
at least mwo is not physically destroying my pc and halting the power to the 750w thermaltake power supply!!
but it is still crashing badly. not even an error message

can anybody help me?

-my card is geforce gtx770.
-my power source is thermaltake 750w smart series.
-my drivers are 347.88-notebook-win8-win7-64bit-international-whql
-additional software installed on my pc is mcaffee antivirus and open broadcast software. i don't have geforce experience installed any longer.
-i ran mwo patcher tool and fixed the client's files. it still did not help.


i am afraid to even play mwo now. before i switched to dx9 and 32bit client mwo was literally killing my pc. i had never seen a software forcefully shut down a computer. that's a very scary error that causes damage to the system!!!!

what can i do!?!?!
helpp!!!!

Edited by Mazzyplz, 01 July 2016 - 11:23 AM.

[D V Devnull]

Wait a minute... Are you trying to use a NoteBook driver file with a Desktop Computer? Ya know what, I think we need you to provide a listing of your hardware, and I mean a hell of a lot more than what you've stated. Without a proper listing of System Specs, people just can't help you properly.

~D. V. "Too much confusion in the info you're giving... More details needed!" Devnull

[Mazzyplz]

no i am pretty sure that's just a driver that is for desktop and notebooks and for windows 7 and 8.

it was the only version on the nvidia site.


what details should i add?

my cpu is a amd fx8350 black series.
my c drive is a solid state drive
and my d drive where the main install of mwo is- is a hard disk drive. the kind that spins.
my OS is windows 7 ultimate

i am not sure how these would affect the power supply in any way.
so long as i don't play mwo my pc is safe too. i can play other games. i had batman arkham city installed not too long ago on this pc and i beat the game almost to completion (got the collectables and achievements) without any problem.
i also play street fighter 4 for hours on end. no problem

it's only when i play mwo that my pc gets really loud. more than with any other game. and eventually it crashes. not immediately, but after a couple of games the risk of crash becomes higher.
i don't have any crash report or anything because mwo doesn't give me any error codes.
at least with dx9/32bitclient i can alt-tab out and it tells me the application is not responding. so i kill the process.

[Mazzyplz]

i should also add that mwo started killing my pc before i got these drivers. the drivers i got actually helped for a while. a week or 2.
i had attributed the problem before i got the new drivers to the geforce experience. and it did help to get rid of it. and a reinstall did quadruple the likelihood of crash to black screen.

[Mazzyplz]

regardless of what the problem is, i am 100% sure geforce experience is crappy software that is compounding the problem.
i also heard from the guys on teksyndicate.com (which i trust cause they helped me before, and it's a community that knows their stuff and they were involved with tiger direct at some point) that geforce experience is developed by some shady russians and not nvidia's american or even EU branches

the problems all started when geforce experience updated my driver automatically i'd say 1.5 - 2 months ago
and it was full crashes with that one. even rolling back didn't improve things until i actually got rid of geforce experience completely.
i read somewhere that for the geforce gtx 770 the optimal driver was 347.88

so i downloaded that driver install. it says notebook but that's the only version of 347.88 i found on geforce drivers site.


apparently the good folks at geforce make new drivers that will destroy your old video card and on top of that geforce experience updates it automatically.

at first when i experienced the first shut downs and video card failure i am sure the voltage for the video card was being set to the erroneous ammount by the new driver. so the power supply shut down my pc to preserve my hardware. that is what i can guess.

i was faced by a choice back then. to set the voltage of my card with a software, which seemed like a bad idea and still does. or roll back to the ideal drivers which i was told was 347.88

Edited by Mazzyplz, 16 June 2016 - 03:47 AM.

[Tarl Cabot]

1. Stop playing MWO or other games that put a serious load on your system until this is resolved The game is only the trigger.
2. System gets loud? That sound is due to fans ramping up, we do not know which fans atm.
3. CPU will shutdown the computer if it gets too hot.
4. Can of compressed air and blow out the heatsinks.
5. Check that CPU heatsink/fan assembly is actually seated properly.
6. Software to check temps for both CPU/GPU. Again, fans ramping up means something is getting hot, and not necessarily the GPU.

Edited by Tarl Cabot, 16 June 2016 - 05:05 AM.

[Rushin Roulette]

Mainly what is written here. How does the inside of your computer look like? Is everything dust free (or at least just a light dusting that wont affect anything, or does it look similar to this?

Spoiler

Fans getting louder means that your PC is heating up too much and teh fans are trying to get more air to cool whatever is causing the heat. If that doesnt help, then the PC will auto shut down once a certain temp has been reached.
This could have many reasons. The easiest to rectify and check is to get rid of any excess dust clogging fans and heatsinks (and basically everything else while you are at it).
After that you may want to check if your fans are all working correctly (no blockages or stuck bearings). This really means every fan. Case fans, GPU fan(s), CPU fan (check if it is also fixed to the CPU and not loose at all) and PSU fan.
Once all easy to check hardware related issues are accounted for, then it is time to carry on with the software (you already updated the drivers following the posts above). Check if you have any malware causing your PC to work overtime.
Folloing that... keep us updated and we will see what else can help .

Edited by Rushin Roulette, 16 June 2016 - 05:42 AM.

[Mazzyplz]

no it does not look like that. it's super clean. that computer looks like it's 10 years old.
mine has a very open case and i can see inside it at all times, the ventilation on it is pretty good and it does not get too hot.

no i'm not going to get the latest drivers because i had them last month and it was 100x worse.

even if these are notebook drivers they are doing better than the desktop driver UP TO DATE from last month.
the official release is what caused all my problems to begin with.

i am not going back to it. my video card wasn't free. i'll look for a more up to date driver than the one i have - but i am not updating to the current one.

i was told by several sources that for older cards you shouldn't always update to the new drivers.
there was also several times where nvidia has released drivers that have caused mass BSOD for players around the world, not just me.
i will look into getting a different install even though there was no other installer on nvidia website for that driver release ver.

i only fear that i get a release that is worse and my card gets damaged. or even my hard drive with these random reset/shutdown.
so i'll get the driver following 347.88 release and go from there

again - the card functioned perfectly for about a year before the most up to date version gave me these voltage issues

i could tell immediately that the current drivers were the problem cause it was night and day.

the card has always been super loud only when playing mwo. and my card on my previous computer did the same thing.
i guess it's just the way mwo is optimized.. the loud sound does not worry me much

[D V Devnull]

Mazzy, a few things:

• I'm running the 355.82 Desktop Driver with an NVidia GeForce GT 440 here, no problems of the grade you have mentioned. Try getting a hold of that version in Desktop Driver form, as it may help your GTX 770.
  
• If you're currently trying to run MWO at maxed Graphics settings, stop doing so. Pushing your computer too hard can trip problems in tons of different ways. MWO loves to push things, sadly.
  
• I don't know if you're overclocking or not. But if you are, stop overclocking immediately. You'll only kill your hardware if you try to overclock and you don't fully know what you're doing. Personally, I actively avoid overclocking in order to make sure my hardware can live a crazy long time. MWO can and will actively trigger problems on overclocked systems if they're pushed too far to the edge.
  
• Pardon me yelling in all caps here, but... MCAFEE IS CRAP, KNOWN INFAMOUSLY FOR A WIDE RANGE OF PROBLEMS. GET RID OF IT AND GET A BETTER ANTIVIRUS AND/OR FIREWALL. And before you ask, yes, it has been known to kill computers in a myriad of ways.

~Mr. D. V. "And that's the stack of bullet points for now... with a need to skip emojis." Devnull

[Tarl Cabot]

http://www.cpuid.com.../hwmonitor.html

You need to be monitoring/logging details, because regardless of drivers, if the system fans (and we do not know WHICH fans) are ramping up, you need to know which ones!!! So stop BS around.

F9 in game, what does your FPS max out at and what is your average?

Create a user.cfg in your MWO folder, with the following line command. If it max @ 70FPS and aver around 50, put in 40.

sys_MaxFPS = 40

If it is your GPU fans, this will cap FPS and help to reduce the workload.

Again though, until you are able to first identify WHICH fans, if not all of them, are ramping up, you are simply throwing darts.

Could the driver installation, either current or previous ones be corrupted and doing weird things? /shrugs.

Could the GPU be dying? Maybe. Could the PSU be dying? Maybe. Could the CPU OR GPU heatsink/fans come loose? Maybe.

Right now you were getting BSOD, and now you are getting complete computer shutdown, either due to CPU or PSU. Again, until you are able to identify and narrow down the issue, tis simply a dart game until your system dies on you. And if it actually dies, the problem component may take more components with it, becoming a very expensive lesson.

[Mazzyplz]

i will try the tools and drivers you suggested thank you for your help and taking the time to share your experience!

[Tarl Cabot]

And just to touch on it once more. When was the last time a can of compressed air was taken to the heatsinks, especially the CPU and the PSU? It may be time to redo the thermal paste on the CPU/heatsink.

Edited by Tarl Cabot, 17 June 2016 - 05:03 AM.

[Mazzyplz]

well i tried doing all the software related stuff and it worked for about 24 hours, but now i played for the event and it shut down again. with dx9.

yeah i will have to take my pc somewhere. but i guess i will be uninstalling mwo as well. too bad

this sucks.. i've been playin since dec 2012 and i guess finally i have to leave the game due to these issues..
i might reinstall on my old 2010-2011 pc... which is a piece of crap.

i checked out the rpms on the fan and it went up to 5900 during a mwo match

last can of air was taken to it never because this computer is about 1 year old. i had my old one cleaned but not this one yet.

i might do that sometime soon but i don't have the time to take it myself to the shop. and there's no geek squad or anything like that available to me

[Tarl Cabot]

Quote

i checked out the rpms on the fan and it went up to 5900 during a mwo match

last can of air was taken to it never because this computer is about 1 year old. i had my old one cleaned but not this one yet.

Don't worry, if it is the dust-heatsink issue MWO was only the tip of the iceberg, you will be uninstalling other programs and over time eventually something will give out due to excessive heat through lack of actual cooling. It is not MWO fault that the servicing of the computer hardware has been non-existent. There are tons of youtube videos to walk you through things though.

Do you have a friend that is computer literate or mechanical abilities who can look at it for you? Your heatsinks are like window blinds, or fan coverings, as air is being blown through them, they pick up dust particles that build up over time. Higher humidity adds moisture to that dust, making it stickier.

That at least one of your fans (you did not mention which one though) is ramping up high due to excessive heat that it can not expel, likely due to dust build. And that you have not done anything to it for a year.../sad face. It is not like it has a self-service nanobots to do it for you. This is part of the servicing you should be performing on your system on a regular basis. And with summer upon us, that also means the ambient temps will make the issue even worse.

Are you in a position to replace your system if something fries and takes several components with it? ugghh. never mind.

My nephew was in a similar position. Everything appeared to be clean but it was not. The cpu heatsink, under a large fraking fan, was caked with built up dust. Had to remove the fan/sink then the fan from the sink to wash/dry the sink cause at that moment even 2 cans of air would not have helped much. He had been blowing dust out of the case every few months but he had not really worked on the CPU sink, basically cleaning the edges but never reaching the sink interior. After quizzing and observing his motions that he reenacted when cleaning it out, he knows better now.

Edited by Tarl Cabot, 18 June 2016 - 11:10 PM.

[B L O O D W I T C H]

Mazzyplz, on 15 June 2016 - 10:39 PM, said:

what can i do!?!?!
helpp!!!!

From the informations you provided i would place my bet on a heat issue and/or faulty hardware.



Check the eventlog (system tab) and look if you have an error logged by the time your system shuts down.

We can help you more if you tell us the errorcode or the information of the error log (or logs).


Tho, i'd say either faulty powersupply or heatsink/dispension issues with your cpu/gpu.

[Mycrus]

sounds to me like a defective psu... it can't meet the power requirements when your GPU spins up... street fighter 4 is very low gpu usage... and arkham city is also an older game (my gpu doesn't even go to full clock speed on those games)...

[Tarl Cabot]

Mycrus, on 19 June 2016 - 05:40 AM, said:

sounds to me like a defective psu... it can't meet the power requirements when your GPU spins up... street fighter 4 is very low gpu usage... and arkham city is also an older game (my gpu doesn't even go to full clock speed on those games)...

Thing is, we do not know which fans, if not most of them, were spinning up, he simply said he saw where one spun up to 5600 rpms. That he has not cleaned out the system for over a year is not a good sign either. He could have the PSU replaced and still have problems cause the root cause was never identified, nor has even basis maintenance been performed.

Basically, he could replace an burnt out engine in a vehicle but it still have heat problems cause the radiator and fluid had never been serviced, and the radiator has been clogged up long enough to kill the engine.

His system needs basic maintenance before he starts having components replaced, and he needs to do it before something actually dies, provided something is not already on its way out.

OP, that you do have another computer that you can use to view youtube and look up relevant information, I would suggest getting involved yourself before paying for unnecessary labor and parts elsewhere.

Edit, just to note, I am on the downside of 50, but I have been tinkering with things since I was little, a few times not with the best results though /sorry mom for cross wiring the toaster that one time....

Edited by Tarl Cabot, 19 June 2016 - 10:06 AM.

[Mazzyplz]

ok i gathered some more data and i concluded the cpu is the problem.

the cpu fan was the one that was spinning, i identified the sound and looked carefully at the spin.

i also touched near the cpu and it was Hot. the video card was the complete opposite. it's not so close to the cpu and that part of the computer was cool.

the temperature reached up to 90 degrees when playing mwo... no wonder it shuts down.

temperature is cool where i am it's not summer but winter. i live in south america temp is a cool 16 degrees C - or 65 Fahrenheit and its windy enough. i'd be wearing a jacket most of the time

i got a workaround by setting the client to 32 bit force as admin, killing windows explorer process before a match and setting compatibility mode to win7, the temperature only goes a bit over 60 at some points. but 60 is the max temp recommended for my cpu.

i am going to take the pc to have it cleaned and the thermal paste re applied and the cpu fan to be mounted carefully.
and i am going to get new cables too if i can for whatever i can get. and an extra fan for the case

but even after doing that i am probably not going to go near the 64 bit version of the client

by using the workaround the cpu is a lot cooler, and the fan makes no noise

i wouldn't have known the temps if it wasn't for hwmonitor, though

Edited by Mazzyplz, 21 June 2016 - 08:49 PM.

[Mazzyplz]

http://www.tomshardw...-safe-temp.html

this thread says if i underclock my card it can run safely at temperatures into the 80s.

so i will underclock in the meantime, until i can get a fan. it seems these issues with heat spikes are not that uncommon... i didn't know fx8350 needed this much cooling.. it also seems like it's not the biggest issue ever because i also read in there it can last a long time even with "high heat".. i am going to choose to believe it's not been damaged at all yet cause i am an optimist and i read just enough to trigger my confirmation bias

Edited by Mazzyplz, 21 June 2016 - 10:37 PM.

[Mazzyplz]

okay. i set the voltage a little bit lower and now my temps dont go above 61 on 2 temps and 65 where i presume is near the vidcard.