Finally fixed my GPU

I disassembled my PC to clear out the dust buildup and reapply some thermal paste.
The GPU is only a year old and hasn't been opened before, but after I opened it up, blew it out, wiped off the old compound, and applied the new stuff, I clearly messed something up. Now the card is overheating like crazy and my games are stuttering constantly.
I’m certain I applied the paste correctly; there are no gaps, and it’s making solid contact with the copper die.
Is there a specific type of thermal paste designed specifically for graphics cards, or will any standard brand work?
More importantly, what steps should I take to get my rig back to the way it was functioning just two days ago?
The GPU is an xfx Radeon R9 270a.

It’s possible you applied a bit too much thermal paste. Usually, I find that if you just put a small amount—roughly the size of a grain of rice—right in the center and then seat the heatsink on top, the pressure will spread it out naturally.
Also, I don't think you should have gone tearing that GPU apart so quickly, especially if it's still covered by a warranty. My own card has been running for over five years now and I haven't felt the need to swap anything out yet since the temperatures have stayed right where they should be.

I took it apart again, applied a thin layer of thermal paste, and tightened the cooler down. Ran a stress test, and it's hitting 72°C after about 5 or 6 minutes. Is that too hot for this piece of junk?

It still needs work in games,
I just finished tweaking it.

Gregory Hayes87 said:I tore it apart again, applied a tiny bit of thermal paste, spread it out with the cooler, and ran a stress test. It’s hitting 72°C after about 5-6 minutes. Is that too high for this little card?

Look... if you're seeing 72°C during a full stress test, it's not a disaster. You'd probably see something closer to 65°C while actually gaming. What are your CPU temps looking like? Make sure you didn't mess up something else. 😬

Gregory Hayes87 said:I took it apart again, applied just a tiny bit of paste, spread it out using the cooler itself, and threw it straight into a stress test; it’s hitting 72°C after about 5 or 6 minutes—is that really too hot for this little card?

Spread it out? What on earth does that even mean? Are you seriously trying to smear the paste manually by pressing the cooler down? That is an absolute disaster waiting to happen. If you go around smearing paste like that, the only thing you can realistically expect is higher temperatures than what is normal.

Just put a single grain of rice right in the center, exactly like Amy Vaughn12 suggested, and then mount the heatsink on top. There, job done.

72°C is perfectly fine during a stress test, so you should look elsewhere for the actual culprit. Check your CPU temperatures; I'd recommend downloading EVGA Precision XTX so you can monitor both your GPU usage and individual CPU core loads. It's possible your CPU is overheating and triggering an automatic downclock or something similar.

Hitting 62 degrees on the CPU; I might need to pull the cooler off.

Stress test results:
CPU 61.5°C
GPU 83°C
Should I take the entire rig into the service center or just the graphics card?
I've accepted that I might be without my $67, but I am officially out of patience.

If you're seeing those kinds of temps during a stress test, I honestly wouldn't sweat it too much—after all, a stress test is designed to push your components right to the absolute edge, isn't it?

Maybe try downloading Core Temp and just leaving it running in the background. Then, hop into whatever game you're playing and check the program periodically to see what the temperatures actually look like under a more realistic load...

Gregory Hayes87 said:during the stress test:
CPU 61.5°C
GPU 83°C
should I haul the entire rig into the repair shop or just the graphics card?
I’ve basically accepted that I’ll be living without my $67, but honestly, my patience has completely run out.

And tell me, do you actually have EVGA Precision XTX or some similar utility installed? What specific card are we even talking about here? Because 83°C isn't necessarily an absolute disaster, and frankly, isn't it quite possible that your fan curve is simply set to cap out at 83?

All I know is that there was no stuttering before, but now it’s hitching constantly and it's driving me crazy.
The GPU is an ASUS R9 270a, as stated above.

I think your CPU should be perfectly fine for a stress test, but that GPU? Yeah, it’s pushing it a bit too hard. Speaking from my own little trial-and-error experience here—and I've definitely been there—you really have to be careful with how much thermal paste you use; just a tiny amount will do. I actually had a similar situation once where I used about a pea-sized amount and hit 83 degrees during a stress test, but after I went back in and applied maybe three times less than that, the temp dropped right down to 73. My advice? Don't try to spread the paste manually—that usually just makes a mess—just put a small dollop right in the center of the chip. Then, when you're mounting the heatsink, give it a gentle, even press with your hand to help it spread out naturally. If you just tighten the screws without that extra bit of pressure, it might not be making enough contact. So, if I were you, I'd pop that graphics card open again, give the old paste a thorough cleaning, apply a bit of the new stuff, and you should be golden. Also, it honestly wouldn't hurt to pick up some high-quality paste—nothing too crazy, but maybe skip the absolute cheapest stuff you find at the drugstore.

I haven't touched a single setting since the day before yesterday.
I played about two hours of Call of Duty:AW and one match of NBA 2K16, and there hasn't been any stuttering yet; the system probably just needed a little time to warm up.
ps: temps remain unchanged and it's getting pretty hot near the tower, so I'm turning off the radiator.

Honestly, I am about two seconds away from chucking this entire rig out the window.
While I'm gaming, my CPU sits at 55°C and the GPU stays between 64-66°C, yet my frame rate just tanks from a steady 60 down to 10 for a few seconds before bouncing back. It is incredibly frustrating.
This is happening specifically while playing NBA 2K16 and Call of Duty:AW.
I’ve even tried cranking the fans to maximum in EVGA Precision XTX, but it does absolutely nothing except make more noise.
Does anyone have any actual ideas? I am losing my mind here; I'm on vacation and instead of actually enjoying myself, I'm stuck staring at a frozen screen.

Gregory Hayes87 said:Honestly, I’m about two seconds away from tossing this whole rig right out my window.
While I'm playing, my CPU stays at 55°C and the GPU sits around 64-66°C, everything looks fine with 60 FPS, but then suddenly it just tanks down to 10 for a few seconds before bouncing back, and it is seriously driving me up the wall.
I'm talking specifically about playing NBA 2K16 and Call of Duty:AW.
I tried cranking the fans up to max using EVGA Precision XTX, but it doesn't change anything except making more noise.
If anyone has any bright ideas, please let me know because I'm losing my mind here; I've got my vacation time all set aside, and instead of relaxing, I'm just sitting here unable to actually play.

To be honest, I don't think your issue has anything to do with temperatures, because those numbers look perfectly fine to me—in fact, I've seen rigs run up to 100 degrees without any stuttering at all, provided the card is powerful enough to handle the settings you're running.

Just lay out the whole mess so we can figure out our next move.

To start, I’d probably just go into the UEFI and disable Intel SpeedStep or whatever other junk is in there depending on your CPU.

Phenom II 965
Gigabyte GA-970A-DS3
8GB 1600MHz
Fortron 550W
ASUS R9 270

It's definitely the DualShock 3, not the DualShock 3 ProP?

When you get into the UEFI, I guess you just hit Ctrl + F1 to unlock those extra settings

If I were using a Dell, I’d probably disable Turbo Boost and then turn off Intel SpeedStep in the UEFI features. For the testing phase, I'd just set all the smart fans to disabled in the PC Health Check and run them at 100% manually.

And tell me, does your CPU usage spike straight to 100% right when those stutters start happening?