Repeated intense crashes during gameplay are a common issue.
Repeated intense crashes during gameplay are a common issue.
Good evening y'all... it's been a hell of a day troubleshooting, and with nothing but failure, I've come with my tail between my legs here. Starting with the specs: 5800X3D with an Arctic Liquid Freezer 240 (solid temps) Sapphire Nitro+ 7900 XTX (solid temps, latest driver, all three cables are dedicated separately from psu) Gigabyte X570 Aorus Elite Wifi (latest UEFI BIOS, SAM enabled) G.SKILL Ripjaws V 2x16 GB 3600 MHz 16-19-19-39 (XMP on) Silicon Power P34A80 1 TB PCIe 3 boot drive/general use (good self test) Crucial P5 2 TB PCIe 3 game library/extra storage (good self test) be quiet! Straight Power 11 1200w 80+ plat Win 11 pro (latest update, repair installed today too) 2 1440p panels, keychron keeb, razer basilisk mouse, fifine mic, speakers, headphones, fractal meshify c case with plenty of airflow This thing is an absolute beast, getting fantastic framerates in games... for about 1-15 minutes, whereupon it'll inevitably hard crash with black screens and loud stuttery audio into a restart. Games include Baldur's Gate 3, Halo Infinite, Splitgate, Fortnite (just to see if it was Steam specific), Deep Rock Galactic, Hogwarts Legacy, etc, with Discord open, maybe a chrome tab or two. Nothing is overclocked, unless you count PBO (is that even a thing with the X3D?), XMP, and default GPU boost. SFC, memtest (100% coverage of all 32gigs), full OCCT suite, FurMark with Prime95, Video Card Stability Test, and 3DMark CPU profile all pass with flying colors. The only benchmark I've run that errors out is 3DMark TimeSpy... which exhibits the exact same hard crash behavior. All drivers on the system are as up to date as possible. Even disabled virtualization as a hail mary. CPU was replaced today from a 5800X, RAM was replaced last week from a similar kit with looser timings(that did help a little, hard crashes no longer deep fry my screen while my pc dies, and they gave my girlfriend's system a stroke when swapped in), video card was replaced a couple months ago from a 3090 (drivers were DDU'ed)... but none of that seems to have fixed the stability of this system. If anything, it's consistently bad at this point. tldr: I'm losing my mind trying to find the ghost in my machine. Any ideas, guys? edit: WhoCrashed doesn't pull up any error logs for the crashes
Start by turning off XMP for any RAM you tested. Adjust it manually to the maximum speed or, if that fails, set it lower. Consider re-seating the GPU. Are you referring to temperature or performance? In Furmark’s case, the score seems linked to the GPU. Try using a different PCIe slot—even though another slot only has 8 lanes, this issue occurred with someone else. It might be because they connected a PCIe connector into a SATA port instead of a proper power hole for their SSD. That change resolved it since SATA power is significantly less (around 54W max).
System failures cause data loss... frequent crashes lead to file damage... the repair tool SFC aims to recover the lost files.
I discovered the problem after purchasing a new motherboard and it wasn’t functioning properly. The GPU driver corruption was the cause. I booted from safe mode and everything worked fine now. Yes, I booted from safe mode during the card swap last time. It’s possible the changes didn’t play well in that state. Perhaps a strange event occurred when I upgraded to Windows 11? Or maybe the old faulty RAM caused issues before. Who knows what happened.