PC Crashing

[Simon Case]

When I play a game on my PC it will crash immediately or up to half an hour later. The screens go black with a No signal message, the sound loops like a robots voice and the fans slow down. I was on teamspeak using my phone testing it out when it crashed but the PC stayed connected and transmitted the occasional popping sound. Has now started crashing all the time, As soon as I log in the screens can just go black.

Specs
Processor - AMD FX8150, H105 cooler
Motherboard - M5A99X EVO
Graphics Card - Powercooler HD6970
HDD - WD 2TB Black
RAM - HyperX Genesis 16Gb
PSU - Corsair TX650
OS - Windows 7 64bit

Have uninstalled and reinstalled AMD Drivers and Audio Drivers
Removed 4.2GHz Overclock
Ran Prime95 and Memtest with no errors
Ran FurMark, Crashed once at 66 degrees then ran fine all the way up to 77 degrees when I stopped it
Thinking it was an overheating issue I took it all apart, cleaned it out and still no changes
Any insight would be lovely, I have been pulling my hair out all week.

 

[Bill_Bright]

and transmitted the occasional popping sound.

Popping sound from where? The speakers or from inside the PC case?

Since sudden crashes can be due to power, I always want to make sure I am supplying good, clean stable power. So if me, I would swap in a spare power supply and see what happens. Hopefully you have one, or another computer or friend you can temporarily borrow one from.

 

[Simon Case]

From the speakers
I will try and get hold of a power supply from a friend

 

[Pill Monster]

Yeah that happens when the gpu gets stuck halfway through a driver reset. If you've had it since new and everything has been fine till now then it's prob either drivers or the card, but being Powercolor my guess would be the card. Does it happen a lot when running flash apps or games,? Or is it all the time?

Btw Furmark is pretty hard on your gpu....


If you have a look in your dump file is there a bunch of 0x116 or 0x117 bugchecks.......?

 

[Bill_Bright]

From the speakers

Good. If from inside the case, that could indicate arcing and potential fires - not good.

 

[Simon Case]

Yeah that happens when the gpu gets stuck halfway through a driver reset. If you've had it since new and everything has been fine till now then it's prob either drivers or the card, but being Powercolor my guess would be the card. Does it happen a lot when running flash apps or games,? Or is it all the time?

Btw Furmark is pretty hard on your gpu....


If you have a look in your dump file is there a bunch of 0x116 or 0x117 bugchecks.......?

Its been fine up until a week ago then it started occasionally while playing Warframe then it started happening all the time, its got to the point where it crashes on the desktop before I have a chance to do anything.

Quick update, it is now crashing on the log in screen if left long enough.

 

[CAPSLOCKSTUCK]

Genuine Windows ?

 

[Simon Case]

Genuine Windows ?

Yes its genuine

 

[Bill_Bright]

Still need to eliminate power as the possible cause because everything else (CPU, RAM, graphics, motherboard) depends on good power run properly.

 

[Steevo]

I would say either power issues, or your card is dying.

Try running the GPU fan at 100% and see if that changes anything, but before that get a multi-meter and test power at the cards power plug, between the yellow and black wires.

 

[OneMoar]

sounds like the board is on its way out
if its been overclocked not surprising the 8150 will draw a metric ton of power overclocked also has a reputation for killing even the best boards
77c is pretty dam hot for a h105

 

[Tatty_Two]

I think definitely a board or PSU issue, as Bill said, the PSU is the easiest to swap out and eliminate, although have a close up visual check of your board and see if you can see any degradation in caps etc.

 

[Pill Monster]

I would say either power issues, or your card is dying.

Try running the GPU fan at 100% and see if that changes anything, but before that get a multi-meter and test power at the cards power plug, between the yellow and black wires.

I agree...... If he can run P95 on a 4.2Ghz Bulldozer without crashing it's prob not the board.

Anyway good luck OP.

 

[xfia]

dont forget to do a good air dusting.. i have had crashing that was from a bit to much dust under my fat air cooler and around my vrm sinks plus i have to take my gpu's completely out every few months or its impossible to get both sides of the heat sinks.

 

[Bill_Bright]

dont forget to do a good air dusting..

He said in his opening post he took it all apart and cleaned it out, but no luck.

 

[Simon Case]

Quick update
Couldn't find a PSU to test but could find a brand new motherboard, swapping them out has had no effect
PC will still randomly black screen however the GPU fan will speed up slightly then go back to idling speeds

 

[Simon Case]

Replaced both the motherboard and the PSU, the issue is still present!!
Does this mean my graphics card is dying?

 

[95Viper]

I was on teamspeak using my phone testing it out when it crashed but the PC stayed connected and transmitted the occasional popping sound. Has now started crashing all the time, As soon as I log in the screens can just go black.

When you used the "phone" on teamspeak... did you or windows show installing drivers or did you install any?

You may want to check your device manager, and check "show hidden devices" in the menu, for any problems.

If you have another GPU... try it.

You may just have to bite the bullet and do a re-install of your OS; or, maybe, try un-installing teamspeak; also, the game may need an update to fix some compatibility problems.
If you have a restore point that is before that crashing, try rolling back to that.

 

[Bronan]

I would test the memory if possible as well.
My current system has been so called able to run memory up to 2133 Mhz but i can promise you if i try that it crashes instant like you posted. When i put them back at 1333 Mhz and kept it at default so no XMP profile it runs.
The shop where i bought the system never got it to run at the higher speeds which where advertised for the memory. Even corsair had send me 2 times new 16 Gb sets which never ran at speed over 1333Mhz.

It kept running fine, if thats not the issue the other possible culpritts has been mentioned.
If it was the gpu it would not even try to boot at all, so i kinda can not give other advice then those already given

 

[Simon Case]

I would test the memory if possible as well.
My current system has been so called able to run memory up to 2133 Mhz but i can promise you if i try that it crashes instant like you posted. When i put them back at 1333 Mhz and kept it at default so no XMP profile it runs.
The shop where i bought the system never got it to run at the higher speeds which where advertised for the memory. Even corsair had send me 2 times new 16 Gb sets which never ran at speed over 1333Mhz.

It kept running fine, if thats not the issue the other possible culpritts has been mentioned.
If it was the gpu it would not even try to boot at all, so i kinda can not give other advice then those already given

Ran multiple Memtests and never got thrown any errors and ram seems to be fine

 

[Simon Case]

Just picked up an Old HD6950 from a friend, going to swap that out and test from there

 

[Bronan]

Its often an nasty long search to find what is causing the problem
In my system i have issues with the nvidia drivers causing to share the gpu with the soundcard on the same irq ... well i can tell you its a tough job to get them no longer to not share that irq. Its a tech story but you have master devices and slave devices normally the motherboard takes care that 2 masters do not share a irq. Simply because thats impossible to work proper. A master and a slave device is however often not really an issue at all. But two master devices will fight for the same irq ( masters are hd controllers, vga card , network card and sound cards and ofcourse the cpu ). I got my problem solved to reflash the bios and reset it to default and reinstall the hardware. The funny but also annoying part is you have to repeat this step till your os forces the devices to jump onto another irq if thats the case.
Lets hope its not the same issue i had.

 

[Atomic77]

I hope its not a lemon PC.

 

[Pill Monster]

Its often an nasty long search to find what is causing the problem
In my system i have issues with the nvidia drivers causing to share the gpu with the soundcard on the same irq ... well i can tell you its a tough job to get them no longer to not share that irq. Its a tech story but you have master devices and slave devices normally the motherboard takes care that 2 masters do not share a irq. Simply because thats impossible to work proper. A master and a slave device is however often not really an issue at all. But two master devices will fight for the same irq ( masters are hd controllers, vga card , network card and sound cards and ofcourse the cpu ). I got my problem solved to reflash the bios and reset it to default and reinstall the hardware. The funny but also annoying part is you have to repeat this step till your os forces the devices to jump onto another irq if thats the case.
Lets hope its not the same issue i had.

Your system must be very old because IRQ assignment since XP is virtualized and handled by the Operating System, not the BIOS.
Even if IRQ assignment ordering is a bios option it's translated by APIC and Windows will assign whatever IRQ address it wants.

You're talking about a specific problem which affected nF2/3 boards with Soundblaster installed. So you have a very old motherboard or are running Windows 98.........


The driver reset causes the black screens, trick is to find out why.
Driver reset is called by the scheduler after IRQ timeout happens 7 times in a row (I think default is 7).
Windows will BSOD if the reset fails, but sometimes the reset gets stuck and hangs causing a soundloop. TDR basically means the GPU is hung..

Drivers can do it (early W10 drivers made blackscreens galore), but mostly it's voltage issues, either through the drivers, card or PSU, vrm's, ripple etc can do it but with a bad PSU normally you'll get at least freezing or something on a system level not just GPU resets.