Phantom Battery is officially documented by Hunt the Phantom Battery and gold award Master Gunner. A phantom target invites confirmation bias: many lines can appear to support the point an operator already expects. Label source, time, and confidence for every mark.
2950790, Windows public 1.0, checked 2026-08-09. Exact battery positions, decoys, phase order, and gold criteria are not asserted without current evidence.Build an evidence register
Number each observer/reference message. For every bearing or range, record origin and timestamp. Put direct reports, inferred intersections, impacts, and recon icons in different categories.
Do not erase a line because it disagrees. Re-read its origin and whether it belongs to an earlier battery state.
Reject attractive false intersections
Two rays crossing is useful only if they describe the same target at compatible times. A range circle can eliminate a visually neat but unsupported crossing. Map terrain or target description can narrow candidates but should not override measurements without explanation.
If two candidates survive, seek later intelligence or recon. Document why the selected point wins.
Confirm a battery can move or multiply
Treat a new dispatch as a possible new state. Preserve old positions and label them inactive rather than shifting them silently. If multiple sources are present, make a separate firing card for each confirmed objective.
Do not combine an old range with a new bearing solely to produce one answer.
Convert the confirmed point
Measure from the Nest, calculate with actual shell/powder, choose barrel, and verify bearing/elevation. A perfect target plot can still be defeated by crossed gun state.
Wait for response after impact. If another battery appears, start a new card. If nothing changes, distinguish miss, wrong effect, decoy/candidate error, and unread dispatch.
Use abilities as evidence, not shortcuts
Recon or battlefield-control abilities can clarify ambiguity when available. Plot first, activate for a named purpose, and attach the result to its time. If a reposition changes the Nest, rebuild every gun line.
Record ability cost and effect from the current console; do not rely on Demo unlock tables.
Earn Master Gunner through verification
The official award means gold on all Phantom Battery medals, but the title does not reveal requirements. Capture live medal hints and test accuracy, target discrimination, assistance, resources, shots, and time independently.
A low-shot solution copied from coordinates cannot prove the intended reasoning and may be version-specific. Publish a reproducible evidence chain.
Troubleshoot the hunt
Check unread messages, mixed timestamps, wrong origins, candidate ambiguity, and objective effect. Update for PATCH 1.0 (1558) if a crash causes entry trouble or stale elevation state. Preserve the save.
A useful report includes the evidence register, chosen candidate, card, impact, next message, medals, and public patch, with strong spoilers clearly labeled.
Audit for duplicate and decoy evidence
Before counting reports, ask whether two items originate from the same observer or event. Repeated wording can reinforce attention without providing an independent measurement. Keep suspected decoys or superseded positions visible but visually demoted.
When a candidate is engaged, record whether the mission confirms a battery, merely shows an impact, or stays silent. Silence does not by itself prove a decoy; the shot may have missed or used the wrong effect. Use the next report to choose among those explanations. A reliable walkthrough should show this decision chain and clearly label any coordinates by patch and stage. That approach remains useful even if a later update adjusts a target point, while an unexplained screenshot becomes stale immediately.
When comparing two supposed battery locations, require the same mission stage and public patch. One may represent a later relocated source rather than a contradiction. Use teleprinter wording and recognition response to align stages before merging community evidence. If alignment cannot be proven, keep both as labeled observations and omit a universal coordinate. This confidence boundary is more reliable than pretending uncertainty has disappeared.