Repeatable disconnect during map load / matchmaking handoff (BF6, Steam)
SUBJECT: Repeatable disconnect during map load / matchmaking handoff (BF6, Steam) — persists after full clean install, driver reinstall, and clean boot
SYSTEM
- Laptop: Dell Precision 7750 (hostname DELL-7750)
- GPU: NVIDIA Quadro RTX 3000 (mobile, Turing)
- OS: Windows 11, build 26100 (later 26200)
- Game: Battlefield 6, Steam version, currently patched to latest available build
- Anti-cheat: EA Javelin (EAAntiCheat), confirmed loaded via Event Viewer
SYMPTOM
Matchmaking finds a match, "found match" sound plays, player character stands up during deployment, then the game never loads into the map. After a period (~60–200 seconds), the client silently returns to the matchmaking search screen. No crash, no error dialog, no client crash-to-desktop. Other in-game tabs remain usable during this state; a "searching" banner persists at the top.
TIMELINE
- Issue first appeared in this exact form after a full reinstall (previous install had a corrupted/mismatched Steam manifest, since resolved with a full clean 143GB reinstall from Steam).
- Worked correctly on [date ~Aug 11]. Same symptom returned the next day with no known change on our end other than a routine Windows security update, later ruled out (see below).
- Confirmed NOT hardware-specific to this laptop: identical symptom reported by other players on unrelated high-end hardware (e.g. RTX 5080/7800X3D) in EA Forums thread [link if you have it], suggesting a broader client/server handoff or anti-cheat validation issue.
WHAT WE HAVE RULED OUT (each individually tested, not assumed)
1. File integrity — full clean reinstall of the game from Steam (143GB, no manifest tampering). Issue persists.
2. GPU driver — clean removal via DDU (Display Driver Uninstaller) in Safe Mode + reinstall of Dell-certified driver package (ID M6MGP, v32.0.15.9658, released 2026-07-14). Issue persists.
3. Hardware-accelerated GPU scheduling — disabled in Windows Graphics settings. Issue persists.
4. Windows VBS / Hypervisor — confirmed via `bcdedit /enum` that hypervisorlaunchtype=Off and isolatedcontext=No at the boot loader level (not just the Settings UI toggle). Issue persists even with hypervisor confirmed off.
5. Windows security update KB5121003 (Secure Boot certificate rollout, installed the night the issue reappeared) — uninstalled, updates paused. Issue persists.
6. Secure Boot state — confirmed enabled and clean via msinfo32, no warnings.
7. VMware Workstation / Hyper-V / Windows Hypervisor Platform services — stopped and disabled (VMware Authorization Service, DHCP, NAT, USB Arbitration). Issue persists.
8. Bonjour / Winsock namespace provider (mdnsNSP.dll) — fully uninstalled; confirmed clean Winsock catalog via `netsh winsock show catalog` post-removal (no orphaned entries). Issue persists.
9. Clean boot (msconfig, all non-Microsoft services and startup items disabled) — issue persists even with effectively zero third-party software running.
10. Background software (Discord, RGB utilities, VPN/proxy) — none present; ruled out by inspection.
WINDOWS EVENT LOG EVIDENCE
Every disconnect event correlates within ~60–90 seconds with an EAAntiCheat filter driver reload cycle in the System event log:
Source: Microsoft-Windows-FilterManager
"File System Filter 'EAAntiCheat' (Version 10.0, [build timestamp]) unloaded successfully" / "has successfully loaded and registered with Filter Manager"
This reload pattern is consistent across multiple independent disconnect instances (examples: 2026-08-12 13:20, 14:34, 17:31; 2026-08-24 12:40:20).
EARLIER CRASH DUMPS (before the above troubleshooting, different symptom — client crash to desktop, since resolved by clean reinstall but included for completeness)
Two minidumps captured 2026-07-26 showed identical EXCEPTION_ACCESS_VIOLATION (null pointer read) at bf6.exe+0x1832bac, thread 10832, both instances. Root cause at the time was traced to a manually-edited Steam appmanifest.acf pointing at EA App-sourced game files; resolved by full clean Steam reinstall. Mentioning for context only — current issue is a different symptom (silent matchmaking return, not a crash).
REQUEST
Given the breadth of client-side elimination above, we believe this points to a server-side matchmaking/deployment handoff issue or an EAAntiCheat validation timeout, not a local configuration problem. Please advise on:
1. Whether EAAntiCheat server-side logs show any validation failure correlating with the timestamps above.
2. Whether this matches a known open issue (the symptom is also reported by other players on unrelated hardware).
3. Any additional client-side diagnostic EA support can request (e.g. enabling verbose EAAntiCheat logging) that we haven't already covered.
Happy to provide Event Viewer exports (System/Application, filtered to relevant timestamps) and the original crash dumps on request.
