Help, I have to underclock my computer to keep it from crashing

[Gethseme]

Alright, I've been having this problem for almost awhile now, and it's getting to the point where my
computer is lagging when playing even older, non-intensive games, like Fable Anniversary and Darkest
Dungeon.

First it started where I tried to play World of Warcraft and my computer would randomly crash while
playing. No BSOD, no error messages, just would go to an empty blank screen for a couple seconds,
then the green light would go off and the computer would completely power down.

I did some research, and everything I read said it was either an overheating or PSU problem. This
is an older (2011 model) Sony Vaio VPCF136FM Laptop, so I know the PSU is actually the AC adapter.

I started out by trying adding a laptop fan (the one that it sits on and blows air up into the
laptop), opening up my laptop and using compressed air to clean the internal fans and such, as well
as applying some heat paste to the heat sink just in case. I also ran HWMonitor and ThrottleStop to
make sure it wasn't an overheating issue, and my comp runs around 79-81 degrees C at all times,
spiking up to about 95-97 during intensive gameplay (not sure how high this is, but I assume it's
not too bad.)

These didn't fix the issue. So I tried downloading MSI Afterburner as a last resort and lowered my
Core Clock (MHz) from 560 to 415 and Memory Clock (MHz) from 800 to 610, and lo and behold, it no
longer crashes, and now has brought on a new problem, which is FPS issues, and playback issues from
running my computer (graphics card I think?) at a lower setting that it's supposed to be ran at.

When I bought this laptop, I could run World of Warcraft on Meduim-High settings, League of Legends
on max settings, and Tera on Medium settings. Now I have to run everything on minimum settings, and
gameplay is deteriorating still. I used to only have to lower the Core Clock to 510 and Memory
Clock to around 740 to prevent crashing, and now I have it dialed down to what I wrote up there.

Anything higher by a few points causes my computer to crash when playing a game.

Anyways, then I bought a new AC adapter, which came in today. Plugged it in, and it still crashes
if I set the settings back to normal.

So now I'm looking around for answers, and the only other thing I can think of is that it's a
graphics card issue. I know that replaceable laptop video cards were around back when I bought this
laptop, so I'm hoping I can do so. If anyone knows if that's the issue, and either whether this
particular laptop has an interchangeable video card or what I need to look for to see for myself
whether I can replace the video card, any help is greatly appreciated.

 

[Supahos]

It is a heat issue those temps are at the max range of not baking the CPU it is throttling to keep from dying. If you just added more paste to the heatsink you may have done more harm than good. Did you clean the old stuff off? Look up how much to use?

 

[Gethseme]

lol, of course I cleaned off the old first. I used a pea-sized drop, like it said on the tube. I bought it a couple years back, so I guess it could be bad, so I'll go out and buy more tomorrow. BTW, the CPU isn't being throttled, only the GPU. Also, the computer crashes IMMEDIATELY, within maybe 30 seconds to a minute of running an intensive game, while the temps are still below 100.

If it is a heat issue and the paste doesn't fix it, should I next replace the heat sink? It has a laptop fan underneath it (which is also raised to allow PLENTY of airflow), and a full-sized standup fan on it at all times and it seems to not really help any.

Also, another question. Why are the core temps roughly the same whether I underclock it or not? I noticed the core temp (before crashing) is the same (around 96ish C) when the computer turns off when not underclocked, yet when it is dialed down, it's still running around 96C and I can play all I want, just with framerate issues and poor graphic rendering. Not saying you don't know what you're talking about, it just doesn't make sense to me unless my sensors on my CPU and GPU are faulty and aren't reporting the correct temperatures up to the time the computer crashes.

 

[Gethseme]

Update: I recleaned the CPU and GPU, making 100% sure there are no scratches on it (from cleaning the old paste off) and made sure there's no paste on it. Now my comp, with the new paste, runs around 68-69 C when it's stressed, not going any higher than that, which is a vast improvement.

The problem is that I STILL am crashing when I turn the GPU back to full power. No, I'm not overclocking it, just hitting the "reset" button on MSI Afterburner to put the GPU and Mem clock back to default settings. When I do so and go into a game, it makes a sorta high-pitch whistle sound as the computer turns off, almost like what would happen if I unplugged it. And I tried this when I had just turned my comp on for the first time in 8 hours. Checked temp before turning on the game (Skyrim on High settings) and temp said 38 CPU, 34 GPU. After game loaded, about 5 seconds went by before the crash, then when I turned comp back on and got to Windows to check the temp, temp said 44 CPU, 42 GPU. So I'm pretty sure the problem now is definitely not overheating.

I used WhoCrashed, but it's not making a crash dump, so I'm out of ideas, unless the built-in graphics card is somehow bad.

Anyone who can help, I'd appreciate it.

 

[Gethseme]

No one can help?

 

[hang-the-9]

Issue is almost certainly a bad video card. For a Sony Vaio model, it's 95% chance or more that it's not a removable video card. You would be able to see if it was soon as you opened it to check the heatsinks.