Skip to content
X4 Modding Wiki

mlog6 fork fixes

The mlog4 fork of DA-Eco is primarily a compatibility fix release. The mod’s gameplay logic is preserved unchanged from DeadAirRT’s v1.13HF1 — what mlog4 restores is 7 major library patches that had gone silent through the 8.x → 9.x upgrade cycle, plus cleanup of ~140 dead patches.

The 7 restored XPath patches

Section titled “The 7 restored XPath patches”

Vanilla 8.x wrapped several code paths in do_else blocks due to the introduction of hasstagedconstruction. DA-Eco’s original XPaths targeted the top-level if clause, which no longer received the actual production logic (that had moved into the do_else).

Result: 7 major mod patches were silently no-op through 8.x and 9.x until mlog6 fixed the XPaths to target the actual code location.

1. Request_Factory ProductionLimit 1→5

Section titled “1. Request_Factory ProductionLimit 1→5”

The signature “bigger factories” patch. Vanilla Request_Factory cue in factionlogic_economy.xml decides how many production modules to plan for a new NPC factory.

  • Vanilla: ProductionLimit=1 — one module per type.
  • DA intended: ProductionLimit=5 — up to 5 identical modules per type.
  • Actually applied 8.x/9.x pre-mlog6: unchanged (patch aimed at wrong branch after do_else wrap).
  • mlog6 fix: re-target XPath to include both branches. Now actually 5.

Impact: the “bigger factories” gameplay that DA is famous for was NOT actually happening in vanilla playthroughs of 8.x/9.x. Post-mlog6, NPC factories grow properly.

2. safepos radius 3km→21km

Section titled “2. safepos radius 3km→21km”

Request_Factory also sets a safe-position radius for new station placement — how much clearance around other stations/objects is required.

  • Vanilla: 3km — very tight, stations crowd.
  • DA intended: 21km — real breathing room.
  • Pre-mlog6: unchanged.
  • mlog6: 21km applied.

Two occurrences — one in Request_Factory, one in Request_Wharf. Both restored.

3. Request_Production_Module WantedProductions 1-2 per tick

Section titled “3. Request_Production_Module WantedProductions 1-2 per tick”

When a faction wants to expand an existing factory, the vanilla logic adds 1 module per tick. DA wanted:

$DALocCount min=1 max=2 — expansion adds 1-2 modules per tick.

Combined with ProductionLimit=5, a fresh factory grows from 1 module to 5 in about 3-4 ticks (instead of vanilla’s 5 ticks).

Pre-mlog6: patch not applied → factions expanded at vanilla speed. Mlog6: patch applied → factions grow faster to full capacity.

4. DAIncreaseStartingStorage actually called

Section titled “4. DAIncreaseStartingStorage actually called”

The storage-sizing subsystem was DA’s centerpiece. Vanilla 8.x wrapped God_DefaultFinaliseFactory in a do_else guarded by staged-construction detection. DA’s XPath targeted the outer if.

Result: DA storage sizing wasn’t happening. Every new factory shipped with vanilla-default storage, matching pre-DA behavior.

Mlog6 fix: XPath re-targeted to include the do_else path. DAIncreaseStartingStorage now runs on every new factory creation.

Effect on a save: new factories going forward get 12h-appropriate storage. Existing pre-mlog6 factories keep their undersized storage until they’re destroyed or re-built.

5. DAImproveStationLayouts actually called

Section titled “5. DAImproveStationLayouts actually called”

Same problem, same fix as #4. DAImproveStationLayouts was the racial-connector subsystem.

Pre-mlog6: all races got vanilla-default connector pool → stations looked homogeneous. Mlog6: racial connectors actually applied → Argon looks like Argon, Split looks like Split, etc.

6. Mining efficiency 90% (silently no-op)

Section titled “6. Mining efficiency 90% (silently no-op)”

DA-Eco’s design intent: NPC mining should be ~10% slower than vanilla to prevent NPCs from over-mining and crashing resource prices. Implemented as a miningefficiency=0.9 XPath patch.

Pre-mlog6: patch silently no-op. NPC miners ran at 100% vanilla efficiency. Mlog6: patch restored. NPC mining now 10% slower — resource prices stabilize.

7. 16 god-quota patches (food/graphene/hullparts/refinedmetals/claytronics/silicon/microlattice)

Section titled “7. 16 god-quota patches (food/graphene/hullparts/refinedmetals/claytronics/silicon/microlattice)”

Individual per-ware quota-scaling patches. DA-Eco wanted higher station density for key intermediate wares (graphene, refined metals, etc.) so trade chains had more nodes.

Pre-mlog6: all 16 patches dead. Vanilla quotas ruled. Mlog6: all 16 restored. Wharves and shipyards spawn with proper module counts.

Cleanup: ~200 error/session → 0

Section titled “Cleanup: ~200 error/session → 0”

Beyond restoring dead patches, mlog6 also removed patches that had been superseded or made obsolete by vanilla changes:

69 jobs.xml patches removed

Section titled “69 jobs.xml patches removed”

Vanilla 9.x natively adopted the traderoutine aiscript that DA had been patching around. The DA Middleman → TradeRoutine transformation patches were superseded — removing them lets vanilla’s own implementation work.

62 region-yield patches removed

Section titled “62 region-yield patches removed”

Vanilla 8.x completely rewrote the resource-yield system. DA’s yield-scaling patches were targeting a schema that no longer existed → errors on every load.

6 dead constructionplans / baskets patches removed

Section titled “6 dead constructionplans / baskets patches removed”

Individual patches that no longer matched the new plan structures.

1 dead Boron cross-faction patch removed

Section titled “1 dead Boron cross-faction patch removed”

Small cleanup.

Total impact: DA-Eco went from ~200 =ERROR= lines per session to 0 errors in 5+ hour multi-mod soak tests on 9.0 release.

The version bump

Section titled “The version bump”

content.xml: dependency version="700"→"900" (mlog006rel, 2026-07-04). Bumps mod’s declared X4 version target from 7.0 to 9.0. Metadata / publication release.

Verification protocol for the fixes

Section titled “Verification protocol for the fixes”

To verify the fixes are actually applied in your save, you can:

  1. Check for ProductionLimit=5 — look for NPC stations with 5 identical production modules of the same type. In the DA Scripts fork’s Production Module Count report, a mature Argon Hull Parts should show 5+ Operational Hull Parts modules on their oldest factory.
  2. Check for 21km safepos — new NPC stations should have visible clearance from each other on the map.
  3. Check racial connectors — visit a large Terran/Split station in-game. Should see struct_ter_* / struct_spl_* racial cross-braces, not homogeneous connectors.
  4. Check script.log — no lines starting with =ERROR= referencing libraries/god.xml, libraries/jobs.xml, etc.
  5. Look for DAIncreaseStartingStorage debug lines — DA-Eco emits [DAECO/STORAGE] log messages when it fires.

Testing baseline

Section titled “Testing baseline”

5+ hour multi-mod soak on X4 9.0 release with:

  • DA-Eco (mlog4 fork)
  • DA-Scripts (mlog4 fork)
  • Apus
  • ETW
  • Galactic Heroes
  • Zero-Transporter

Result: zero =ERROR= from DA-Eco. This is the mlog6 baseline that lets the fork be considered stable for 9.0.

Why these fixes matter

Section titled “Why these fixes matter”

Pre-mlog6, someone running DA-Eco on 9.0 would experience:

  • Vanilla-sized factions (small stations, no “bigger factories” effect).
  • Vanilla-sized station storage (empty silos problem persists).
  • Homogeneous station visuals (no racial connectors).
  • Full-speed NPC mining (resource prices crash).
  • ~200 errors per session cluttering the log.

They would technically be “running DA-Eco” but getting almost none of DeadAirRT’s design. Mlog6 restores the actual gameplay experience the mod was supposed to deliver.

Section titled “Related pages”