Graphics Cards

[Jabroni]

I just started on a complete new build.

PC components as follows:

Case: NZXT H510 Flow; PSU: NZXT C850; MoBo: AMD MSI MPG B550 Gaming Plus; CPU: AMD Ryzen 9 5900X 12/24; CPU Cooler: Cooler Master Hyper 212 w/ Push and Pull fans; DRAM: Corsair Vengeance LPX 32GB; GPU: XFX Speedster SWFT 309 AMD Radeon RX 6700 XT; SSD Samsung 970 EVO Plus NVME M.2

Started out with downloading CoD MW2 (2022) (after drivers obviously) and played easily 10 hours no issues whatsoever. hopped off for about 3 hours hopped back on started Cod again and in multiplayer online game game froze and then forced reboot. Problem continued for about 2-3 days until I had time to sit down and really troubleshoot. was always randomly in the first game I played. never on start up or anytime else.

I tried playing lower demanding games, such as Insurgency Sandstorm, and no issues whatsoever. then i tried to download Red Dead Redemption, and I got Blue Screen two times in a row.

After this i did multiple stress tests and everything seemed to be running very much OK by the written temps. I used OCCT to run stress tests, and AMD Adrenalin as well. OCCT I ran for 2 hours with zero signs of over heating.

After that just in case, and because it was a brand new PC with nothing but drivers and games on it yet decided to do a full factory reset on the Windows System settings and start fresh, just in case the drivers i downloaded were not operating correctly.

With Windows 10 reinstalled for the second time i jotted down all the Drivers i downloaded to keep better track of my steps...

MSI Mobo in box DVD Drivers as follows:

AMD Chipset Drivers

Realtek PCIE Network Drivers

Realtek HD universal Drivers (that was it on the DVD)

AMD Radeon Adrenalin 22.11.1 (restarted computer)

*AMD Software Installer*... on their suggestion:

AMD Processor Power Management Support 7.0.4.10

AMD PCI Device Driver 1.0.0.89

AMD GPIO Driver 2.2.0.130

AMD GPIO Driver (for Promontory) 3.0.0.0

AMD SMBus Driver 5.12.0.38 (Restarted Computer)

Installed Ryzen Master, Samsung Magician, and then continued down the road to Call of Duty on Battle.net.

first game in same exact problem ruling out software issues entirely.

Back to Overheating or malfunction with hardware. I tried to run cod again after messing with settings, running as admin, graphics settings in windows and in cod, etc. started to get a notification from one of the drivers I installed and I managed to screen shot it before it disappeared, and I also managed to back out of game before it froze and rebooted again: 1. GPU Temp Too High 2. Latency 3. Packet Loss.

Now that my long story is over, this is where i am at. I have my case side panel removed, running all case fans on full blast through BIOS, and have tuned the GPU fans full blast and surprise surprise, it worked for like 2 games instead of malfunctioning on the first. My question to everyone on here is as follows:

How the heck do i lower the temps during in game stresses without voiding the warranty with a liquid cooler?

Or does this sound like a malfunction with the GPU from factory, maybe the thermal pads or thermal paste isn't properly applied? I dont know, but to be overheating with full speed fans?

Also my case has (2) 140mm NZXT F140Q with 101 CFM as inlet fans and (1) 140mm F140Q as upper exhaust (only one spot for upper exhaust which I considered to be the issue) and (1) NZXT F120P with 73CFM. Also my CPU Cooler has (2) of the identical NZXT 120mm fans with push and Pull. And yes every single fan is aimed correctly.

PLEASE ANY SUGGESTIONS!!!

[elstaci]

You never mentioned the GPU temperatures when it crashed. It maximum operating temperature (Hot Spot) is 110c.

Are all three GPU fans running at maximum speeds when it overheats?

Did you run OCCT PSU Test by any chance?

What were the PSU outputs during the test (3.3/5.0/12.0 Vdc)?

Many Users under volt the GPU card so that it will run cooler. But someone else will need to help you with that.

It may not be your GPU card but some other hardware like RAM or PSU that is misbehaving unless you see the GPU's Hot Spot go above 110c and freeze or shut down.

[Jabroni]

I did run PSU test with zero issues.

I ran OCCT labeled the GPU hotspot at 93 degrees C, and yes all three fans were running at full speed

[elstaci]

93c is still within its operating temperature of 110c.

Have you tried changing the Game playing settings in either Radeon Settings or the games itself to see if it stops crashing since it only occurs when you playing games.

What were your CPU temps when the computer froze or crashed?

[Jabroni]

Running PSU test currently what are the figures you need to know

[Jabroni]

During test my RAM isnt reaching over 20% of useage actually only using 6.6 GB of utilization, unsure if that is normal

[Jabroni]

CPU was at 62 degrees. I have not touched Adrenalin settings other than to turn my fan speed higher which seemed to delay the inevitable. the hotspot temp is low you are correct but seconds before crash it jumps from 75 degrees straight to 90-93 then crashed. The PSU ratings i am not entirely sure of anymore i do not exactly remember them, but because the problem persists i should be able to run them again.

Also good to note i did windows memory diagnostics about 15 times, and haven't had much of an implication that the RAM is damaged.

[elstaci]

Thanks for the update. If your GPU is crashing below 110c Hot Spot I would open a XFX Support ticket and see what they say. Unless it is another issue that is causing the crashes.

That GPU card is engineered to run normally up to 110c before it starts to throttle. Even though, I have heard from other Users here on AMD Forums on other GPU cards that it also would crash below 110c. But I don't consider that to be normal. That is why I suggest you open a XFX Support ticket to see what they say.

I do remember another User with similar issues was resolved by changing settings in Radeon Settings.

You do have a high quality Gold rated PSU. All PSU outputs should be within -/+ 5% tolerances. The 12.0 Vdc output shouldn't go below 11.4 Vdc under heavy loads or stress as an example.

At least you eliminated defective RAM. I just mentioned that because bad RAM can cause those types of issue.

NOTE: Since the GPU card passes all the OCCT Tests it seems to indicate a Settings issue. Either with Radeon settings or the games itself.

But some one other than me can help you with that since I am not a gamer.

[Jabroni]

I really appreciate you help and suggestions i am currently running stress test and with fans maxed hotspot is at 90 degrees pretty steady. i will look up any Radeon settings i may be overlooking and file the XFX support ticket. Thank you for your time.

[elstaci]

If XFX says anything useful will you post their response back?

Another User can help from this point on if it isn't a hardware issue.

Thanks for the update again.