Is Windows trying to tell me something?

[aragond]

Pardon the length of this, but do you think my computer is trying to tell me something?

 

And this isn't all of them, just the ones I shot computer

(Pic ID mfwQt.png )

 

Inb4

 

 

 

 

No, seriously, I am actually asking a question.:

 

Does this definitively mean my vidya card is scrood, or could there be another cause?

 

As you can tell, this has been going on for ... well, the whole three years I've had the PC, but it always seems to "get better" all by itself, or with me playing the DVI-plug-swap game. And then for a few more weeks, it'll be fine, no incidents. And then, out of the blue, BAM and the dirt is gone.... and the GPU stability with it.

 

The PC is 3 years old, so I am also wondering is my old computer (refer tagline) going to hold back a 980 replacement card to the point of wasting my time.

 

Questions, question, questions. Anyone got anything resembling answers?

Edited September 23, 2014 by aragond

[uNi]

What card is it? Also make sure the gpu is not overheating and you're using appropriate drivers.

[aragond]

What card is it? Also make sure the gpu is not overheating and you're using appropriate drivers.

 

Per sig': GTX580.

 

GPU is not overheating. I almost never hear the fan. The reason I know this is (a) because Tyler knows this and (b) because as the gpu carked it last night, the fan went ballistic for about two seconds, so (i) I heard what it would sound like, and (ii) I know the fan is capable of hypersonic speeds and isn't broken allowing overheating

 

Nvidia driver ver. 331.82. It was made for the Nvidia GPU so I am assuming that makes it "appropriate".

Yes, 344.11 is out. Please don't make me upgrade. 9/10 times, I get gpu issues when I upgrade. Just like I am rarely able to run 3Dmark without my machine freezing and dying.

[aragond]

And on it goes:

 

[uNi]

Do you get any issues while playing or using any other applications/windows? Other than those reports that is.

[aragond]

Do you get any issues while playing or using any other applications/windows? Other than those reports that is.

 

Yes, but MOSTLY something video-intensive. Like this evening, I started watching an mp4 file and the sound went all digital and **BAM**, the dirt, my screen, and ultimately that run of Windows was gone. Screen blacked out multiple times, but never showed the screen again, it just ran through several iterations of blackness.

 

I say mostly because earlier today it black-screened because in Firefox I opened up my bookmarks dropdown button, and began scrolling through the list of unsorted. So, that is kinda video intensive.

I am pretty sure I have had it black screen in the middle of scrolling (again quickly) through a webpage, but not necessarily when it is loading the page, for example.

I am pretty sure it's also done it in the middle of Excel, again, when I was intensely scrolling.

Finally, after a black-screen of death it may come back to the windows7 load screen, that shimmering set of four colours and freeze with that image, and then I can hear the windows start-up music but I am still staring at a fixed image of windows starting. I can log in, fine, but I cannot see anything. I'm looking at the load screen. Windows thinks it is showing me the desktop.

So, it's a video image freeze in this case, not a O/S freeze.

 

But, I do not recall it ever black-screening when I am typing an email or reading a webpage (not flicking about).

 

And I say "black screen" because I never get a nice neat BSOD with a dump. Whatever happens happens at a more fundamental level than Windows can recover from, implying a h/w fault, implying either my mobo or my gpu are royally, as we say in the business, f**ked.

But, which one?

And how do I test to find out?

[aragond]

Anyone?

[FriendlyPsychopath]

if you don't want to upgrade your nvidia driver

 

try this do a registry clean with CCleaner

try downloading the same version as yours and doing a clean install http://www.nvidia.com/Download/Find.aspx?lang=en-us

try opening your case and see if she is well placed on your mobo

try cleaning the part marked in red with an eraser

 

 

 

 

Try cleaning the cooler.

Try applying a new and fresh thermal paste

[aragond]

 

 

if you don't want to upgrade your nvidia driver

 

Did this just then. GPU took all of ... I guess 20 minutes into its first game after two reboots (just to be sure) before I got my first "driver has stopped working"-error.

 

try this do a registry clean with CCleaner

Yep, done this before and it incinerated my computer's memory of itself. Not good. It really should come with a big red emblazoned warning saying, "This nukes your machine. Are you suuuure you don't just want to reinstall everything?"

 

try downloading the same version as yours and doing a clean install http://www.nvidia.com/Download/Find.aspx?lang=en-us

Done this before, too. Sure, not recently, but it didn't help. I mean, sometimes it does, for a few days, but as you can see from the above charts, it is very infrequent and unpredictable, happening for a week or two and then not at all for a month or two.

 

try opening your case and see if she is well placed on your mobo

 

Is.

 

 

try cleaning the part marked in red with an eraser

 

 

Hmm. Not a fan of pulling it out and putting it back in, but if it really IS likely, I'll try it. But, I have pulled it out before and reseated it, just not the eraser bit.

 

Try cleaning the cooler.

Yep, also tried this before. Same result set, refer above.

GPU temperatures are a stable 44 degrees (celcius) albeit when not playing a video/game, but that's fairly normal having just been doing so and then crashing out. CPU is 35 degrees. Voltages are all stock-standard and unchanged.

(Data gathered using HW monitor, which has never shown anything is amiss.)

 

Try applying a new and fresh thermal paste

Now, this one intrigued me: to where?!
I cannot imagine you mean to the CPU because that would seem crazy. It's a GPU error, not a CPU error which would be spread across the computer's operation like a crazy man.
...
Woul'n'it?


I'll just reiterate that this has been going on for years now. This is not a recent phenomenon. Couple of years ago, I even sent the card back and the computer vendor said they tested it and sent it back to the manufacturer before being caught in a lie about their testing when the mfr sent it back as "no problem".

Oh, and, yes, I have run memtest several times.

I ask because even upgrading the drivers does nothing for me, so I guess I am considering just buying a new gpu and be done with it. If that don't work, I buy a new pc. I can't afford a new pc, but f-ck it, I am just exhausted from this unusability sh!t.

Anyone care to offer some last minute ideas?

 

(PS; LOVE that gtaf is no longer allowing multiquoting of the same person. Stealthy move, admins. Score!)

Edited November 10, 2014 by aragond

[Red Hat]

 

Try applying a new and fresh thermal paste

Now, this one intrigued me: to where?!

I cannot imagine you mean to the CPU because that would seem crazy. It's a GPU error, not a CPU error which would be spread across the computer's operation like a crazy man.

Actually you can change the thermal paste of your GPU.

 

It's a pain in the ass to get the job done right, tough.

 

Edited November 10, 2014 by Kevin3014

[Flachbau]

If you're worried you might not be able to complete the task of applying thermal paste onto your GPU, don't bother

[RyanBurnsRed]

I'm no computer engineer, but I'm almost certain that your video card is dying.

If not the video card, then the power supply.

 

If the power supply isn't outputting enough power for the card then you'll have unexpected shutdowns like that.

Edited November 29, 2014 by RyanBurnsRed