[SOLVED] GPU Usage drops to 0! Help would be kindly appreciated!!

[Light888]

I've been having this problem with my AMD GPU to which I've been trying to find a solution to all over the internet to no avail. I know this is a long read but I would be grateful if you have the time to read this to the end and perhaps assist me!

So right off the bat, I have an old laptop which is 5 years old. Please don't read this and instantly start typing a reply to just upgrade my laptop because I would love to and am saving up to get a good upgrade eventually but I have to make the best out of my current one for now.

SPECS:

System Model: HP ProBook 450 G3

BIOS: N78 Ver. 01.05 (type: UEFI)

Operating System: Windows 10 Pro 64-bit (10.0, Build 19042) (19041.vb_release.191206-1406)

GPU: Hybrid of Intel(R) HD Graphics 520 and AMD Radeon R7 M340

CPU: Intel(R) Core(TM) i7-6500U CPU @ 2.50GHz (4 CPUs), ~2.6GHz

RAM: 8GB

The AMD GPU is the one with the problem. I could upload my DxDiag or give any other information necessary if needed.

Anyways, for the amount of time I've had this laptop I have not run into any issues with it when playing games which my AMD GPU can handle. I play this one game a lot: Rocket League. My GPU is good enough to run the game at 50-ish fps on low settings and has done so since I started playing the game in 2016.

A few weeks ago though someone carelessly handled my laptop and it fell down from a distance big enough to crack the screen. I had it sent to be fixed and took the opportunity to have it serviced as well since I hadn't had the fan cleaned in a while. I had the HDD changed as well and with that came some loss of data unfortunately so I had to reinstall the drivers. The laptop essentially came back "brand new."

So when I ran Rocket League it ran fine at the usual consistent 50ish fps but about 4 mins into a match I had severe fps drops down to 15-20 fps from 55 fps to the point where even the audio was lagging and crackling in my earphones. It was unplayable. Then it returned to normal after about 15-20 secs before it started happening again after a min or 30 secs. After every consecutive match I played though, the fps drops seemed to get more frequent. I tried another game Far Cry 4 and again, normal fps then massive fps drops.

With AMD Radeon Software's metric overlay and MSI Afterburner I noticed that the GPU usage was dropping to 0 and fluctuating back up causing the fps drops as shown in the pics below.

Also, the GPU gets pretty hot. It's always gotten pretty hot though and it's been fine for years. It is overclocked by default I think. I have never tried messing with overclocking or underclocking it as I've never felt the need to since the games ran fine. The fan was actually clogged with dust before the laptop got dropped. Now after being serviced, the fan's clean.

https://imgur.com/vq7iEPn

View: https://imgur.com/vq7iEPn

https://imgur.com/agrm5qZ

View: https://imgur.com/agrm5qZ


The GPU memory clock also becomes static when the fps drops happen if that indicates something too.

https://imgur.com/MHC6a7g

View: https://imgur.com/MHC6a7g


The fan seems to be running fine.https://imgur.com/p3LFN4L

What I've tried so far:

- Uninstall and reinstall drivers manually from the AMD website numerous times using AMD's cleanup utility

- Update windows 10, it's up to date now.

- Update BIOS

- Try older drivers

- Set Unified GPU Usage Monitoring in MSI Afterburner

- Check power options, I play on High Performance anyway

- Run a a memory diagnostics test, there were no memory problems detected

I'm hoping this is just a software related issue cuz I have had stuttering or fps drop problems before but not this bad and I've usually resolved them by just updating my driver. Windows updates have been notorious for me for messing with the drivers. I don't think Windows updates is the culprit this time. Could it be malware? Windows Defender has detected nothing and all the files got removed anyway when I had the laptop serviced and the hard drive changed.

My worst fear is that something could be wrong related with the hardware. Especially when my laptop got dropped, or it could be something related to the heat. The GPU gets pretty hot when it starts working intensively so maybe the GPU usage goes to 0 to cool down? But again, this has never been an issue before for the 5 years I've had this laptop. Not to mention my fan was very dirty before and there were no problems. Now the fan's completely clean. I've read online that the thermal paste on the GPU can dry up cuz of running at high temperatures? I hope that's not the case.

Sorry for such a long post but I wanted to make the situation clear. I hope someone will be able to give me a solution to this problem or any advice as I feel like I've pretty much run out of options from the ones I found after searching on the net.

 

[Ralston18]

Very much doubt that the repair service truly tested the laptop by playing games or otherwise running some demanding tests or apps. Once the system boots and sits for X hours (if that long) the laptop is deemed repaired and returned to customer.

Contact the repair service.

You may or may not have some warranty on the repairs.

Unfortunately, under the circumstances, the repair service may have enough wiggle room to say that whatever they fixed is ok and that there is now another issue. And in all fairness that may indeed be true.

Remember to backup all laptop data if you have not already done so.

Document all calls, emails, chats, etc. Keep all records, invoices, to track the repair history.

Hopefully whatever else has happened is repairable under warranty or at least will not be too expensive to repair.

 

[Light888]

Thank you for the reply! I think that's what I'm going to have to do. I suspect there might be something wrong with the cooling or the thermal compound. I was hoping this was just a software issue rather than a hardware issue but I guess there's no running from it. I'll back everything up before I send it.

 

[Fiorezy]

The GPU is throttling since you mentioned high temperature and massive fps drops after a few minutes, the cause of this is more likely a bad thermal paste application by the repair center, so as stated above, take it back and tell them that you never had this issue before