Crashing and Connection Checks
A reversible troubleshooting route for launch, loading, server-join and in-match crashes, with an evidence-first escalation note.
Troubleshoot a crash by identifying when it happens, checking the newest official patch notes and applying one reversible local change at a time. Keep causes marked pending when the official sources do not identify them.
Quick facts
- First question
- At what exact stage does the failure happen?
- Safe method
- One reversible change at a time
- Evidence rule
- Unconfirmed causes stay pending
- AMD note
- Official workaround may reduce crashes; results vary by system
Do this in order
Record the failure boundary
Write down whether the crash happens at launch, during loading, when joining a server or during a match, plus the current build.
Success signal: You have a repeatable trigger and a timestamp instead of a generic “the game crashes” report.
Check official update notes
Search the latest official patch and news entries for a matching stability, map, audio, UI or server issue.
Success signal: You can tell whether the issue is already listed, fixed or still unconfirmed by the publisher.
Failure check: If the official notes do not mention it, keep the cause marked pending rather than assigning blame to one setting.
Try one documented AMD check
If you use an AMD GPU, test one official temporary workaround at a time: Patch 1.3 notes say disabling gore may reduce crash frequency, while the launch guidance also lists the documented launch options, High graphics and disabling the AMD in-game overlay. Results vary by system.
Success signal: You know whether one reversible change altered the same crash trigger.
Failure check: If the crash continues, restore the prior setting, keep the evidence and do not present the workaround as a universal fix.
Apply reversible local checks
Verify files through the platform launcher, close overlays and return to a known-good display or graphics setting one change at a time.
Success signal: The same trigger no longer reproduces or the changed variable is isolated.
Escalate with evidence
Keep the build, hardware, reproduction steps and official patch references together before asking the community or support for help.
Success signal: Another person can reproduce or evaluate the issue from your note without guessing the context.
Separate a crash from a server or platform mismatch
A missing server, failed join and client crash can look similar from the player’s point of view. Record the stage, server platform label and visible error before changing settings so the next check has a defined target.
What not to promise
The official patch notes can confirm listed fixes, but they do not prove that every unlisted crash has one cause. Avoid claiming that one driver, overlay or graphics option fixes all systems without a reproducible evidence set.
AMD-specific temporary branch
The official launch guidance lists two temporary Steam launch-option combinations: -dx12 -USEALLAVAILABLECORES -malloc=system, or -dx11 -USEALLAVAILABLECORES -malloc=system. It also lists High in-game graphics and disabling the AMD Software: Adrenalin Edition in-game overlay. Patch 1.3 adds that disabling gore may reduce AMD crash frequency. Test one change at a time, keep the change reversible and expect results to vary.
- Use the branch only when the machine has an AMD GPU and the failure is a crash or stability problem.
- If it does not help, restore the previous value before testing the next change.
- Escalate with build, GPU, driver, trigger and the exact change that was tested.