Been getting frame drops in increasing frequency the past month and change. Going from 140/150 FPS down to 70-90 FPS regularly during combat. Updated GPU driver, ran SFC which did find things to be repaired, used the OW performance optimization guide found here, Launcher Repair found nothing to repair.
I have a Razer mouse, but don’t have chroma or synapse installed.
Additionally, it feels like my FPS are quite a bit lower even when they peak than they should be given my set-up.
I had previously manually increased the virtual memory allotment via the performance guide in this forum earlier today, so I’ve just now reverted that back to default settings in case that was added on to the issue. I could certainly get some new physical ram if that’s still an/the issue.
I disabled the XBox Gaming Bar stuff via the Game Bar setting in Windows, yet it still appears as an issue in my DXDiag results?
I had previously used DDU when I swapped from an NVIDIA GPU to my current AMD. I just installed and ran it again, rebooted, ran once more, rebooted, and installed the latest AMD drivers/software. But I see this event called out once again.
It is weird that it would keep showing up after you disabled it. I also see the GPU drivers are still reporting a crash. Are you seeing these frame drops after playing for a bit? Do you think this shut down of the GPU could be related to overheating?
Beyond that, you can open a ticket with the staff so they can collect more detailed files to help troubleshoot. Due to COVID-19, ticket times may be a bit a long. While you wait, you could try some benchmark utilities to make sure everything is working as intended on your system.
Wattman’s been replaced with the “Tuning” Section as far as I can tell, which allows overclocking the gpu / vram on a per app basis, so I’ve now set it to Overclock the GPU for Overwatch. That’s my interpretation of how to implement that successful comment with the current software. Unless, it’s really recommended to roll back the AMD software to whatever version has Wattman.
I just turned off Fast Start Up.
I only had 2 highlights in my highlight folder, but I moved them out anyways.
My Heaven results on High are as such:
FPS:
192.9
Score:
4860
Min FPS:
31.3
Max FPS:
340.9
I started running Prime95 over night but forgot to disable the sleep/hibernate timer, so I’m gonna continue to run that throughout today. No fails so far, but I do see it’s recommended to let it run for a good 24 continuous hours.
I threw memtest on a usb drive and will run that tomorrow and won’t bother testing the game until after that’s completed.
update: I got the following error via Worker #1 in Prime95:
FATAL ERROR: Rounding was 0.499921289, expected less than 0.4
Hardware failure detected, consult stress.txt file.
Torture Test completed 340 tests in 22 hours, 36 minutes - 1 errors, 0 warnings.
The stress.txt file just looks generic and doesn’t pinpoint the exact problem. I opened HWMonitor again, and while the other 3 workers are going, my temperatures appear to still be within range. Which it’s sounding like this is pointing towards a memory issue?
So I’m thinking I’ll just let Prime95 run for a few more hours and then I’ll run memtest unless I get a different error.
update #2: ran memtest, it found a couple errors. I rebooted into windows right after, and then turned off computer for the day. Was going to isolate the ram sticks for another test but low and behold, now my computer won’t POST thanks to the stress tests.