Game shuts down my PC, at my wit's end here...

[A Tree]

I posted about this in the pinned thread but it seems this is a bizarre and unique issue, not connected to the more common crashes.

I have an EVGA nVidia GTX670, Win 7 Pro 64-bit.

I ran the game for about 4 hours straight, getting pretty much everything in world A and got through B-1. Upon returning to the main map, my computer shut down. Ever since this event, I can't seem to run the game for more than 5 minutes without a shutdown. It'll happen in menus, it'll happen in game. It's just an unplayable waste of space at this point.

I've played many older-non intensive games since the incident, along with more recent (but still old) games like Garry's Mod and Brutal Legend, as well as 2015 games Life is Strange and the resource-intensive GTA V...all for extended periods and with no crashes.

It wasn't until I tried to run a Unity game, Dr. Langeskov, The Tiger, and The Terribly Cursed Emerald: A Whirlwind Heist, that the issue cropped up with that one. I run it for a few minutes and my computer shuts down.

I've tried reinstalling the game.
I've updated my graphics card drivers.
I've tried running the diagnostic tool "WhoCrashed?", but no crash dump is being made for either game even after doing all their recommended troubleshoots.
I've inspected my graphics card as little as I could, the fan still functions heh. Not really much I can do here...it's not like I overclock it or anything.
I've googled the hell out of this problem and it seems to just be that obscure. You can imagine the feelings of hopelessness I've got with this problem.

It really doesn't feel like a "aging PSU" issue or a "GPU overheating" issue. This game can shut down my PC in menus of all places and it's capped at 70FPS. The Unity game that has the issue doesn't cap its FPS, but neither does Garry's Mod which works perfectly fine. Either way I'm not in the position to just slap in a new PSU or GPU here and now.

My best guess at this point is that it's a software problem -- like some shared library or resource that Talos Principle and the unity game both rely on got corrupted while I was playing your game. I was wondering if maybe the devs could point me into some useful directions with regards to this? Any libraries that Talos Principle uses and doesn't just install its own instance of?

(one that Unity might also use? heh...)

Thanks...

[AFFM]

This sort of behavior is typical of dying powersupplies or dying graphics cards.... (or faulty graphics card drivers, etc.) If it was a problem with a game, the game would crash... but since it sounds like your whole system is locking up, it's a deeper problem.

Trust me, I've been there many times. I had an issue with my computer locking up that would ONLY happen in Left 4 Dead 2 during a map change. No other game was affected, and I could play the current map as long as I wanted with no issues. Eventually, over the course of many weeks the problem got worse, crashing more often, and with more games.. eventually until the computer was basically unusable.

[A Tree]

Yeah, PSU/GPU are what tend to come up in similar discussions (including one in this very forum)...but I'm just not seeing the logical connections. PSU you would think would be unlikely to be randomly triggered during a couple specific games and GPU...well I just ran FurMark stress tester for 20 minutes, spending most of the time at 83C and 61% fan power which is harder than I've ran it since one fairly simple indie 3D game decided it liked to run at 700fps, heh. (I quickly forced it to stop that nonsense)

Power was more or less a static 113% TDP during the test, whatever significance that holds.

Plus it's been in this broken state for three weeks and hasn't degenerated further. The Unity game was a new acquisition so I'm guessing it would've been the same three weeks ago. Other than that there have been no other signs that this is a general problem not specific to Talos Principle and Unity/that one game.

I mean yeah there's wishful thinking on my part seeing how hardware problems are more of a hassle to deal with, but as far as I'm concerned the notion of a software problem hasn't been ruled out. One thing I noticed while just now doing some cleanup is that the many Visual Studio redistributables that games install are actually shared, there's one possibility for the culprit if one of those is needed by Talos Principle and is corrupted...

By the way, how similar are Serious engine and the one Talos uses? I could try running Serious Sam 3 and see if I get the same issue.

[Master]

Originally posted by A Fancy Fridge Magnet:

This sort of behavior is typical of dying powersupplies or dying graphics cards.... (or faulty graphics card drivers, etc.) If it was a problem with a game, the game would crash... but since it sounds like your whole system is locking up, it's a deeper problem.

This.

Might also be overheating: did you try cleaning your CPU/GPU heat sinks, to check inside the PC to make sure it's all clean?

Originally posted by sarysa:

By the way, how similar are Serious engine and the one Talos uses? I could try running Serious Sam 3 and see if I get the same issue.

If I remember correctly is the same engine.

[AFFM]

Talos is closestly related to Serious Sam 3, engine-wise. Talos has a lot of updates and improvements, though.. for instance Talos supports DirectX11, which I believe is the default API that the game will use.

[A Tree]

Originally posted by I'm not 30 they framed me:

Originally posted by A Fancy Fridge Magnet:

This sort of behavior is typical of dying powersupplies or dying graphics cards.... (or faulty graphics card drivers, etc.) If it was a problem with a game, the game would crash... but since it sounds like your whole system is locking up, it's a deeper problem.

This.

Might also be overheating: did you try cleaning your CPU/GPU heat sinks, to check inside the PC to make sure it's all clean?

They're clean...in fact the GPU heat sink faces the ground which in light of this is a pretty good design choice. (CPU faces the side)

Anyway, it turns out everyone else is right, sadly. :P

I did a better test running the GPU stress tester again along with a CPU stress tester on the other three cores. Running both at the same time crashes my computer fast. Running only one at a time does not with 25+ minutes of running it straight.

This pretty much tells me that it's a PSU problem. I know source games only use one CPU thread but I guess GTA V and Life is Strange do as well, and Talos Principle/Unity don't Guess it's time to order a replacement.

Weird that it suddenly permanently lost voltage or whatever in the middle of a playthrough and stayed that way for 3 weeks, but it's not like I know how a PSU works. :P

(my ignorance is showing...I'd love to hear what an expert on the subject thinks...a website says my maximum load wattage is around 407W and it's a 750W PSU, maybe there's a significance to why it's failing around half wattage levels?)

[Interknet]

Originally posted by sarysa:

Originally posted by I'm not 30 they framed me:

This.

Might also be overheating: did you try cleaning your CPU/GPU heat sinks, to check inside the PC to make sure it's all clean?

They're clean...in fact the GPU heat sink faces the ground which in light of this is a pretty good design choice. (CPU faces the side)

Anyway, it turns out everyone else is right, sadly. :P

I did a better test running the GPU stress tester again along with a CPU stress tester on the other three cores. Running both at the same time crashes my computer fast. Running only one at a time does not with 25+ minutes of running it straight.

This pretty much tells me that it's a PSU problem. I know source games only use one CPU thread but I guess GTA V and Life is Strange do as well, and Talos Principle/Unity don't Guess it's time to order a replacement.

Weird that it suddenly permanently lost voltage or whatever in the middle of a playthrough and stayed that way for 3 weeks, but it's not like I know how a PSU works. :P

(my ignorance is showing...I'd love to hear what an expert on the subject thinks...a website says my maximum load wattage is around 407W and it's a 750W PSU, maybe there's a significance to why it's failing around half wattage levels?)

What exact PSU do you have, and how long have you had it?
Also on average, how long do you run your computer? i.e 6 hours and 5 days a week.

[A Tree]

Originally posted by Interknet:

What exact PSU do you have, and how long have you had it?
Also on average, how long do you run your computer? i.e 6 hours and 5 days a week.

This one: http://www.newegg.com/Product/Product.aspx?Item=N82E16817139010

It's my personal computer, and it's usually awake while I am. I'll only shut it down while sleeping myself or when remote access is unlikely. (shopping, going to the park, etc.)

I'm estimating it's had about 17000 hours of usage out of the 100k MTBF it claims. (though I'm guessing MTBF means total failure. for the most part I'm unhindered and any shutdowns I get are predictable)

My IT professional friend considers my situation to be very weird. Sadly I'm more of the software type. ;)

[AlenL]

MTBF is _not_ what most people think it is. It merely means that if you ran 1000 such PSUs for 1000 hours, you would have found around 10 malfunctioning ones by then (1000 units*1000 hrs/10 failures = 100k hrs MTBF). It is useful for big deployments like datacenters, to be able to calculate expected yearly cost in replacements. But it is not a reliable way to determine how long a single unit "should" last. For that, you'd need expected lifetime. And even then you are subjected to the "infant mortality" area of the "bathtub curve".

http://www.controleng.com/search/search-single-display/learn-or-review-the-difference-between-mtbf-and-lifetime/cc1b974477.html

[Interknet]

Originally posted by sarysa:

Originally posted by Interknet:

What exact PSU do you have, and how long have you had it?
Also on average, how long do you run your computer? i.e 6 hours and 5 days a week.

This one: http://www.newegg.com/Product/Product.aspx?Item=N82E16817139010

It's my personal computer, and it's usually awake while I am. I'll only shut it down while sleeping myself or when remote access is unlikely. (shopping, going to the park, etc.)

I'm estimating it's had about 17000 hours of usage out of the 100k MTBF it claims. (though I'm guessing MTBF means total failure. for the most part I'm unhindered and any shutdowns I get are predictable)

My IT professional friend considers my situation to be very weird. Sadly I'm more of the software type. ;)

It may not be your PSU after all, though bare in mind not all PSUs are going to last as long as intended.

I find it unlikely for the issue to be GPU related.
As far as my knowledge goes, I'd say it was motherboard or CPU related.
Leaning more to motherboard logically, though. Only problem with that is whether it 100% is or not. Nobody wants to waste time pretty much rebuilding a computer just to change a motherboard, especially if it isn't the problem.

You may wish to check out this:
https://community.spiceworks.com/how_to/112266-how-to-efficiently-troubleshoot-power-supply-motherboard-and-cpu-problems