Forum Discussion

MichipiGR's avatar
MichipiGR
Seasoned Novice
4 days ago
Solved

Pc hard crashing ever since the latest update introducing RedSec

Hey everyone,

I’d like to apologize in advance for how long this post is, but I think being thorough might help someone identify what’s actually going on.

Background:
I was away from home between October 25th and November 2nd, so I hadn’t updated Battlefield 6 or my drivers during that time. When I returned on November 2nd, I downloaded the latest update (the one introducing RedSec) and that’s when everything started breaking.

The Issue:
Joined a match of Conquest. Two minutes in, I got a kill and instantly my PC hard-crashed. Screen went black, display signal lost, peripherals powered off as if the system was shut down. Had to force reboot multiple times before my monitor and USB devices came back. Same thing happens even while idle in the main menu.

What I Tried So Far:
Here’s a full list of everything I’ve done to troubleshoot, none of which fixed the problem:

  • Updated NVIDIA drivers, crash persisted
  • Rolled back to the September 30th driver, still crashed
  • Tried launch options: -dx11 and -bEnableGlimpse 0, no effect
  • Undervolted and underclocked GPU (MSI Afterburner), no effect
  • Underclocked RAM in BIOS, no effect
  • Updated BIOS (Gigabyte X870E AORUS ELITE WIFI7 to F8d) and AMD Chipset
  • Disabled AI Memory Boost and removed Gigabyte utilities
  • Checked Windows integrity (sfc /scannow and DISM restorehealth)
  • Disabled background apps and overlays (EA, Discord, Afterburner OSD)
  • Turned off hardware acceleration in other programs
  • Tried both DX11 and DX12 modes
  • Limited power target to 90 percent in NVIDIA Control Panel
  • Disabled Smooth Motion from NVIDIA Control Panel and optimized the game for performance using NVIDIA’s performance slider

Nothing changed, the game still hard-crashes the entire PC, usually within minutes.

System Specs:
CPU: Ryzen 9 9950X (16 core, 4.3 GHz, AM5)
GPU: Gigabyte RTX 4070 SUPER 12 GB (Driver 581.80)
RAM: 64 GB DDR5 (Corsair Vengeance 6600 MHz)
Motherboard: Gigabyte X870E AORUS ELITE WIFI7 (BIOS F8d)
PSU: Be Quiet Pure Power 12 M 850 W 80 Plus Gold
OS: Windows 11 Pro (Build 26200 Version 25H2)
Display: 1080p @ 144 Hz (G-Sync compatible)
SSD: 1 TB NVMe PCIe 5.0 Samsung 990 EVO Plus (BF6 installed here)

I’ve also attached my DxDiag.txt file for full details.

Observations:
Temperatures, voltages, and power draw are all stable.
No issues in other games or benchmarks.
Event Viewer shows KERNELBASE.dll and .NET Runtime errors, no blue screens.

My Question:
At this point, I’m convinced this is a post-update issue related to the RedSec patch or the latest NVIDIA driver branch, because my system is otherwise rock-solid.

Has anyone else with similar specs (especially 40-series GPUs or Ryzen 9 CPUs) found any workaround or seen an official acknowledgment from EA or DICE?

Thanks in advance for reading this. I just want to help narrow it down for everyone having these random hard crashes, since I’ve seen similar reports on the forums and Reddit.

  • OskooI_007​ 

    Update:

    I’ve finally fixed the issue — and it turned out to be a memory slot problem, not a RAM issue like I first thought.

    On this motherboard, the primary memory slot is the second one (A2). For two RAM sticks, the recommended configuration is slots 2 and 4 (A2 and B2). However, during testing, I discovered that the fourth slot (B2) is faulty.

    I stress-tested each RAM stick individually using OCCT (maxed-out stress test for 15 minutes). Both sticks passed when tested alone in the second slot, but when I installed them together in slots 2 and 4, the system failed the stress test and hard-crashed — confirming that the RAM itself wasn’t the problem.

    To verify, I experimented by placing the second RAM stick in different slots. The system worked fine with it in slot 1 (A1) but not in slot 3 (B1). That confirmed it’s a motherboard slot/channel issue.

    Now the system runs perfectly again with 64 GB of RAM. From the BIOS, I simply loaded optimized defaults, disabled XMP/EXPO, and everything has been stable since.

    Hopefully, this helps anyone facing similar random crashes or black screens — sometimes it’s not your GPU or drivers, but a faulty memory slot causing instability.

6 Replies

  • Same thing happening to me. My game ran fine in the beta, and from launch until this update. Since the update my game will either crash back to desktop, hard restart my pc, or it’ll be a direct x “device hung” error. I know 100% it’s not my system. I’m on a 5080, 9800x3d, x870 mobo, 64gb of ram with a huge psu. Literally all I have on my pc is Battlenet for COD and Steam for Battlefield. I’ve changed nothing since the beta except for updating drivers. So I’m pretty sure it may be a bug. 

  • MichipiGR​ sounds like a memory error crash to me, the computer shuts down while crashing.

    How many memory sticks are plugged into the motherboard? Four memory sticks puts more strain on the memory controller than two sticks does.

    You Underclocked RAM in BIOS. Does this mean you completely disabled all EXP0/XMP memory overclocking and ran the memory at stock speed of 4800MHz?

    If not I recommend trying that for stability testing.

  • MichipiGR's avatar
    MichipiGR
    Seasoned Novice
    4 days ago

    First of all thank you for replying and showing interest in this issue.

    I currently have 2 sticks in out of the 4 available slots. 2 32gb with 6600MHz speed which i normally dont overclock since i have not found a need to do so.

    I have both tried disabling XMP and manually setting the memory speed to values like 2000MHz ,something i would expect that would be ok for stability testing but nothing happened.

    I will give it another shot this evening as i am currently heading for work and will let you know what happens after I try to tweak my memory settings when I get a chance.

  • MichipiGR's avatar
    MichipiGR
    Seasoned Novice
    3 days ago

    OskooI_007​ 

    Update:

    I’ve finally fixed the issue — and it turned out to be a memory slot problem, not a RAM issue like I first thought.

    On this motherboard, the primary memory slot is the second one (A2). For two RAM sticks, the recommended configuration is slots 2 and 4 (A2 and B2). However, during testing, I discovered that the fourth slot (B2) is faulty.

    I stress-tested each RAM stick individually using OCCT (maxed-out stress test for 15 minutes). Both sticks passed when tested alone in the second slot, but when I installed them together in slots 2 and 4, the system failed the stress test and hard-crashed — confirming that the RAM itself wasn’t the problem.

    To verify, I experimented by placing the second RAM stick in different slots. The system worked fine with it in slot 1 (A1) but not in slot 3 (B1). That confirmed it’s a motherboard slot/channel issue.

    Now the system runs perfectly again with 64 GB of RAM. From the BIOS, I simply loaded optimized defaults, disabled XMP/EXPO, and everything has been stable since.

    Hopefully, this helps anyone facing similar random crashes or black screens — sometimes it’s not your GPU or drivers, but a faulty memory slot causing instability.

  • OskooI_007's avatar
    OskooI_007
    Legend
    3 days ago

    MichipiGR​ wrote:

    I discovered that the fourth slot (B2) is faulty.

    Interesting, it sounds like the copper pins in that slot might be bent and damaged.

    Is your memory running in dual channel mode? You can verify with HWiNFO System Summary.

    https://www.hwinfo.com/

     

  • MichipiGR's avatar
    MichipiGR
    Seasoned Novice
    3 days ago

    It is indeed running in dual channel mode. To be honest I wasn't expecting it to but hey I'm fine with that. I inspected the copper pins and didn't spot anything odd pins look fine. It is what it is. Hopefully this issue doesn't appear again and if it does I will update this post.

     Anyways, thank you once again for your interest and assistance. 

About Battlefield 6 Technical Issues

Having issues with Battlefield 6? Join here to find help with connectivity, performance issues, crashes and more.3,665 PostsLatest Activity: 3 minutes ago