Following the 2021 Supreme Computer Cup I was inspired to have a go at developing a basic AI for FAF. As I had zero experience of coding in Lua, I thought at the same time I would write down things as I learnt them to have both a guide for other people interested in getting started in developing an AI, and a dev log for the creation of my AI.

See below for a link to the current AI development guide and devlog:
https://docs.google.com/document/d/1uQlEoN-kti7G2MnhwD60uaaHwVevPNYH/edit?usp=sharing&ouid=100973959280546778272&rtpof=true&sd=true

You can also access the mod on Github:
https://github.com/maudlin27/M27AI

I've copied the current overview from the guide below:

This document is intended to do 3 things at once (meaning it does none of them very well!):

• Act as a guide to people new to FAF AI development, including collating any other useful resources out there on AI development
• Be a sort of dev log of my thought process and development steps for creating my first FAF AI, with extracts of code thrown in
• Form my main ‘todo’ list for further developments of my AI
  If you see any errors, or have any comments or suggestions (in particular any AI developers who read this since you’ll probably know better/easier ways of doing things than the methods I’ve chosen) let me know and I can update the guide where relevant.

If you’re here for useful resources and/or details of how to write AI and don’t care about the dev log aspect then just skip to section 2) Setup and background and section 5) Index of useful functions, and/or browse the index for points that may be of interest.

Latest changes (v65 summary)

• Refinements to the logic for experimentals to dodge shots (and more generally) so they're hopefully less likely to turn around in a full circle doing no damage while being attacked
• 7 tweaks to T3 arti targeting logic, none of which are major but which for example now assign a value to heavily shielded units (previously none was given), firing infront of mobile units, reducing the value of targets that we have fired a huge number of shots at without killing, and factoring in how long it will take for the T3 arti to turn towards the target
• 15 other misc changes or bugfixes, a number of which relate to decisions on when units shoudl retreat (e.g. MML running from ravagers, skirmishers running from PD, GC/Monkey/Ythotha being more likely to run from a fatboy based on how close it is)

Areas of focus by version (history):
v1-3: Release
v4 - Intel and threat detection
v5 - Use of ACU and hotfix
v7 - Engineers and building upgrades
v9 - Air
v10 - Anti-Turtle
v11-v12 - Pathfinding and misc changes
v13 - Anti-navy
v14 - ACU and mobile shields
v15-v18 - CPU optimisation and hotfixes
v19 - Rally points and air domination strategy
v20 - Late game
v21-23 - Eco management
v24 - Bombers
v25 - Misc
v26 - Nukes
v27 - T3 Arti
v28 - Game options/compatibility
v29 - Optimisation and TMD
v30 - Czar, Soulripper and Defences
v31 - Bombers (again) and TML
v32-33 - Plateaus
v34-35 - Turtling
v36 - Skirmishers
v37-38 - Game enders
v39 - Quantum optics
v40 - Deceiver
v41 - Hives and shield disruptor
v42 - Misc and multi-TML defence
v43 - Misc and UEF experimentals
v44-46 - Teamwork
v47 - Optimisation
v48 - Personality
v49 - Hotfix
v50-51 - Navy
v52 - Desync hotfix
v53 - Shield boats
v54 - Temporary eco strategy
v55 - Experimental defence
v56 - Stealth boat
v57 - Pathfinding (again!)
v58 - Misc (RC tweaks)
v59 - Gunships
v60 - Naval enhancements
v61 - Stats
v62 - Land spam
v63 - T3 arti prioritisation
v64 - Mass fabs
v65-v66 - Skirmisher and T3 arti tweaks

Having only started following the FAF forums and news the start of this year I'd not come across the PC election before, but one thing that's surprised me is just how long the process has gone on for. To summarise the timings for the PC election that's almost now finished:

• 6 May 21 - election announced, candidates requested
• 30 May - Signups closed (includes 1 week extension to accommodate further debate following disqualification of certain candidates/clarification of conditions to apply)
• 6 June - voting begins
• 6 July - voting ends

So essentially, 3 months taken up with the PC election.

Regardless of whoever wins the current election, I'd suggest shortening this period significantly, as 1/4 of the PC's time will potentially be in 'election mode'.

I'd therefore suggest allowing 3 weeks for candidates, and then 3 weeks for voting, to halve the time involved overall, for next year's election, while still giving enough of a period to allow most people who might be on holiday at some point in that time to be able to stand and/or vote, without having the election go on for so long.

I'd also be curious if anyone has stats on the timeframe of when votes were cast, as I suspect they will be weighted significantly towards the start of the voting period (which would give further support for there not being a need for a whole month to vote). If this isn't the case and votes tend to be evenly spread over the time period (or even weighted more towards the end of the voting period) then it may still be worth retaining 4 weeks for voting.

The Spawn ACU at start positions mod allows you to spawn ACUs under your control at specified start positions at the start of the game.

For example, say you're playing a 4v1 on Setons but don't want the disadvantage of only having 1 ACU to start with. This mod allows you to have an ACU at each of the standard start points, under you control.

To use it, enable the mod and then select "Spawn ACU" as the player/AI in each of the slots where you want to spawn an ACU, and specify the faction of the ACU to spawn.

At the start of the game the Spawn ACU AI will then be defeated and their ACUs will be given to a human player on the same team.

Let me know if you encounter any issues using this.

v47+ note - the page numbers for sections 4+ will be slightly different in the actual version (since I'm no longer showing the section 3 topics below I wont be updating the below listings often)

1. Overview 35
   1.1) AI goals 35
   1.2) Background 36
   1.3) Approach 36
2. Setup and background 37
   2.1) Lua editors and IDEs 37
   2.1.1) Intellij IDEA 37
   2.1.2) Visual Studio (VS) 38
   2.2) Other guides which include a section on AI: 39
   2.2.1) General information: 39
   2.2.2) How AI in FAF works 39
   2.2.3) AI markers and how AI uses them: 40
   2.2.4) Hooking (e.g. replacing/appending existing game code) 40
   2.2.5) Creating custom scripts for Sorian AI: 40
   2.2.6) Pathfinding 40
   2.2.7) How platoons work 40
   2.2.8) AI Research (non-FAF related) 41
   2.3) Lua coding 41
   2.3.1) How to code in Lua 41
   2.3.2) Stand alone Lua code testing 41
   2.3.3) Lua optimisation 41
   2.3.4) Common causes of errors for non-Lua coders 41
   2.4) Debugging in FAF 42
   2.4.1) Console 42
   2.4.2) Cheats 42
   2.4.3) Log window and error messages 42
   2.4.4) Debug window and shortcuts 43
   2.4.5) Platoon names for debugging 43
   2.4.6) Function hooks and identifying infinite loops 43
   2.4.7) Destroyed objects 44
   2.4.8) Fast speed 44
   2.4.9) In-game tools and information 44
   2.4.10) Other shortcuts and game startup options 46
   2.4.11) Suggested shortcut 47
   2.4.12) Using FAF Develop 48
   2.4.13) Error handling 49
   2.4.14) Optimisation – benchmarks and monitoring 50
   2.4.15) Debugging via replay 51
   2.5) AI development discord 52
   2.6) Core FAF Lua files 53
   2.6.1) Accessing core Lua code 53
   2.6.2) Other built in functions 53
   2.6.3) Unit blueprints 53
   2.7) Testing AI against other AI 54
   2.7.1) Simple offline testing 54
   2.7.2) Mass offline AI testing 55
   2.8) Getting started – using a simple AI template as a starting point 56
   2.9) Downloading and setting up MicroAI 56
   2.10) Giving the AI mod your own name 56
   2.11) General - Hooks, and getting the code to recognise if it’s running your AI 56
   2.11.1) Hooking – Unit callbacks/events – applied example 57
   2.12) Github 61
   2.12.1) Github advantages and Github desktop 61
   2.12.2) Linking Github with a local repo via creating your own init file 61
   2.12.3) Pro Git ebook 61
   2.12.4) Git commit messages 62
   2.12.5) Own notes – contributing to FAF Git 62
   2.13) Uploading your AI to FAF 62
   2.13.1) Pre-upload checklist 62
   2.13.2) FAF rules on uploading mods 62
   2.13.3) Uploading to FAF 63
   2.14) Other AI documentation 63
3. AI changes 63
   3.1) Build order - Changing the initial build order 63
   3.1.1) Adding/removing buildings 63
   3.1.2) Change build order based on map resource positions 64
   3.1.3) Build order conditions 64
   3.2) Build order – ACU builds without moving at start 65
   3.2.1) BaseTemplate 65
   3.3) Build factory and power to benefit from adjacency 66
   3.3.1) Record all mexes on the map, and reference this information 66
   3.3.2) Determine com start position 67
   3.3.3) Identify mexes near to start 68
   3.3.4) Hook onto the core code relating to building a structure at a location 68
   3.3.5) Call build order from custom aiBrain function, to edit how things are built 69
   3.3.6) Adjust build template based on mex location - conditions 70
   3.3.7) Adjust build template based on mex location – build location adjustment 70
   3.3.8) Build initial PGens by existing factory 71
   3.3.9) Adjacency bonus where ACU is away from start location 73
   3.3.10) Referencing unit blueprints for information such as unit size 75
   3.4) Platoons and sending engineers to reclaim 77
   3.4.1) How platoons work 77
   3.4.2) Send the first engineer on reclaim duty 82
   3.4.3) Determine where to send the engineer to reclaim 83
   3.4.4) Identifying reclaim mass value 84
   3.4.5) Find out if a location is accessible by ground: 84
   3.4.6) Mapping reclaim locations and assigning preferred location to an engineer 86
   3.4.7) Don’t send early reclaiming engi if there’s minimal starting reclaim 86
   3.4.8) Issue new reclaim command if engineer becomes idle 87
   3.4.9) Send more engineers to reclaim later (as lower priority) based on the reclaim currently on the map 87
   3.4.10) Send 5 engineers at the start to reclaim and build mexes 88
   3.4.11) Reclaiming trees when no main mass left 89
   3.5) Building and assisting hydro 89
   3.5.1) Check if hydro near start area and have engineer build hydro 89
   3.5.2) Have ACU assist engineer with hydro 90
   3.5.3) Get ACU to continue build order after hydro complete 90
   3.5.4) Subsequent build order changes 92
   3.6) Attack force composition changes and check of AI performance 92
   3.6.1) Build Moar Tanks 92
   3.7) Pre-v1: AI Progress check 92
   3.7.1) Land factory buildings and other build order refinements 93
   3.8) Raiding AI 95
   3.8.1) Raiding behaviour - overview 95
   3.8.2) Queue up move commands for mexes 95
   3.8.3) Identify mexes to target 96
   3.8.4) Locating mexes near to an existing pathing route 98
   3.8.5) Determine when raiders need new targets 99
   3.8.6) Handling complications (stuck units, move orders built on, platoon units dying) 99
   3.8.7) Change factory rally point so units are assigned orders sooner 101
   3.8.8) Making use of platoon names for debugging 103
   3.8.9) Ensure 1 land scout included in each platoon 104
   3.8.10) Use microbots before switching to tanks 104
   3.8.11) Have scouts stay with the platoon 106
   3.8.12) AI scouting, threat detection, and identifying nearby enemy units 107
   3.8.13) Attack engineers en-route: Overview 110
   3.8.14) Tracking when have reached movement points 110
   3.8.15) Determining the platoon minimum range 110
   3.8.16) Identifying nearby enemy units and assessing their threat 112
   3.8.17) Actions based on enemy threat 114
   3.8.18) Avoid attacking enemy mexes for small platoons 115
   3.8.19) Other issues – cheating, walls and min range 116
   3.8.20) Larger raiding platoons 117
   3.9) Large attack wave 117
   3.9.1) Checking when to form a large attack wave 117
   3.9.2) Generalising raider AI 118
   3.9.3) Creating separate platoon AI for the large attack force 118
   3.9.4) Run nearer to base when outnumbered 119
   3.9.5) Recalling nearby raiders to join attack force 119
   3.9.6) Updating threat assessment for ACU health and upgrades 121
   3.9.7) Don’t chase small numbers of faster units 122
   3.9.8) Switch to hunt and kill enemies once reached destination 124
   3.10) Pre-v1: AI Progress check 124
   (3.11 onwards - left out due to post size limitations)
4. Future possible changes 512
5. Index of useful functions 592
   5.1) Map, game setup and general aiBrain information 594
   5.1.1) aiBrain - general 594
   5.1.2) AI Brain – Economy 595
   5.1.3) Map Markers and start positions 596
   5.1.4) Map size and playable area 597
   5.1.5) Terrain info 597
   5.2) Getting a list of units by category, location, or other method 598
   5.2.1) All units that meet a particular category 598
   5.2.2) Units at/near a location 599
   5.3) Unit information 600
   5.3.1) Units on radar 600
   5.3.2) Blueprints and unit categories 600
   5.3.3) Unit upgrades (enhancements) 602
   5.3.4) See 5.4.10) Other unit commands 602
   5.3.5) Location and pathfinding 602
   5.3.6) Unit’s ai Brain owner 602
   5.3.7) Unit’s current commands and whether it’s alive 603
   5.3.8) Unit bones 603
   5.3.9) Shields 603
   5.3.10) Units with missiles/anti-nukes 603
   5.3.11) Unit resource usage 604
   5.4) Unit and platoon commands 604
   5.4.1) Building 604
   5.4.2) Assisting 605
   5.4.3) Pausing 605
   5.4.4) Reclaim 605
   5.4.5) Factory specific commands 605
   5.4.6) Movement and attack-move 606
   5.4.7) Transport 606
   5.4.8) Combat related 607
   5.4.9) Stopping 607
   5.4.10) Other unit commands 607
   5.4.11) Upgrades and enhancements 607
   5.4.12) Current commands and targets 608
   5.4.13) Unit status 608
   5.4.14) Change a unit’s name 610
   5.5) Platoon additional functions 610
   5.5.1) Units platoon handle 610
   5.5.2) Get AIBrain 610
   5.5.3) Get units in a platoon 610
   5.5.4) Disband 610
   5.5.5) Platoon exists 610
   5.5.6) Platoon position 610
   5.5.7) Get all platoons 611
   5.5.8) Platoon ai plan name 611
   5.5.9) Add units to a platoon 611
   5.5.10) Army pool and removing units from a platoon 611
   5.5.11) Create a platoon 611
   5.5.12) Change or reset an AI plan 612
   5.5.13) Platoon movement formation 612
   5.6) Weapon information 612
   5.6.1) Get a unit’s weapon 612
   5.6.2) Get information on that weapon 612
   5.7) Time and thread related 612
   5.7.1) Game time 612
   5.7.2) System time 613
   5.7.3) Wait/delay the code 613
   5.7.4) Fork thread (run independently) 613
   5.8) Events 614
   5.8.1) Creating custom events 616
   5.9) Changing the game itself 616
   5.9.1) Spawning units 616
   5.10) Other 616
   5.10.1) Visual changes – Draw Circles and Lines 616
   5.10.2) On Intel Change 617
   5.10.3) ScenarioInfo.Options 617

@cpt_cnullrufs sports are banning players that are representing Russia from competing. Big difference to banning/discriminating against someone just because they live in Russia which is what you seem to be suggesting.

Aeon gun's increased range is massive - you can reliably kite enemy guncoms with it, and they're forced to either hope you slip up with the kiting (letting them get within range), have a load more nearby units to charge at you, or get enough T2 PD up to stop you. Oh and a cost decrease to gun upgrade is a buff to the gun upgrade.

Aurora I don't like but it does at least give a clear difference to the other factions who all have a very similar main tank.

Mercy - discussed a load in a separate threat - I'd prefer the oppsosite of your suggestion - instead of making them survive against MAA, make them die to MAA. IMO if the enemy has MAA by their com it should counter mercies. Making Mercies able to tank T1 MAA shots for a while would make them insane.

GC - It gets omni and the claws, which are really powerful against some matchups. As an extreme (unlikely) scenario - I can beat other land experimentals with certain SACU mass equivalent combinations. Try that against a GC and it'll easily crush them due to its claws [Edit: Forgot that the sandbox scenario I was recalling was UEF Shield SACUs with Percies underneath them and for some reason thought the claws got the SACUs - instead it got the shielded percies]. It also easily beats a monkeylord, and its 'crush when walking' is more meaningful due to its huge HP.

Czar - compare this to soulripper, there's just no contest. It's incredibly versatile - you can charge at the enemy com under shields with SAMs nearby and kill it (try that with a soulripper and you've just given them a big mass donation). You can kite enemy asfs with it; you can use it in a more attritional style thanks to its shield; you can do surprise attacks by storing air units inside it, etc.

AI biweekly/monthly tourney: 20km Combined arms
The theme of this map was 20km maps that allow use of land, navy and air. The maplist chosen was:

• Crossfire Canal - FAF version
• Norfair
• The Ditch - FAF version
• Setons Clutch - FAF version
• The Bermuda Locket - FAF version

Setup
Setup is similar to the last tourney - 12 AIs are split into 4 groups. The weakest 4 are eliminated in the first round, with the winners from one group fighting the runners up from the other in the quarter-finals in 3 matches.
As there were 5 maps, maps were chosen randomly for the group and quarters, and then for the semis and finals each map was played.
Starting positions are the default first and second positions set by the map (e.g. on Setons this means a 'mid vs mid' battle)

Changes to AI
5 of the featured AI have released new versions since the last tourney: M27, Swarm Terror, SCTA Arm, Sorian Edit, SCTA Core

The results
Group stages

Quarter finals

Semi finals

Finals

Total wins by AI

Detailed results
The full results can be seen in the 'DetailedResults' worksheet of the following file:
https://docs.google.com/spreadsheets/d/1i2ObsZ49S9gUTgWw0E68qZ8VrOkaP_Jk/edit?usp=sharing&ouid=100973959280546778272&rtpof=true&sd=true

Other Observations

• While some AI attempted to build navy, in all cases it proved ineffective with the mass invested in navy far exceeding the mass killed. Neither of the finalists built a single naval unit the entire tourney.
• The finals were closer than the scoreline suggests, with DilliDalli dominating land on most of the maps and winning map control (e.g. holding 5 out of the 6 islands at one point on Crossfire Canals, and effectively raiding and even claiming mexes behind M27's main base on Setons). The finals ended up being an air vs land contest on most maps.
• SCTA Arm managed to have 2 T3 tanks solo a GC (GC's shot was blocked by cliffs) but it wasn't enough to let it claim a win
• Heroic unit of the tourney: DilliDalli's lone striker vs M27 on Setons which managed 19 kills and 1,720 mass kills raiding mexes
• Notable changes in fortunes: Swarm managed a turnaround in speed, going from the slowest AI to claim wins in the last tourney to getting the second fastest win (11m21, beaten only by Dilli who got a win in 10m48)

Veterancy adds greater depth and makes games more interesting. There have been plenty of entertaining moments both in games and watching casts where for example an ACU is about to die and gets a sudden rank of veterancy allowing it to just escape a kill. It also gives a spotlight to units that have a massive effect for their cost (e.g. units that reach 5 stars).

However Ive no issue with allowing a unit's mass kills and veterancy progress being visible to all players (if it was possible)

Fame/recognition - e.g. news post recognising the winner; part of the client that highlights recent winners of tournaments; etc.

V10 - Implemented anti-turtle logic:

• Use of T2+T3 MMLs and mobile arti to break firebases
• Ecoing (not very well, but better than before)
• Reduced turtle feeding

Misc changes:

• Large bomber attack for if we end up with lots of spare bombers
• T3 air factories can be built/upgraded
• Support factories will be built/upgraded instead of HQs
• Higher tier power is built sooner
• Engineers can request guards
• Various bug fixes (see section 3.30.2 of the guide)
• Updated the guide with details on how to run in FAF Develop offline

My AI also has a new icon

For anyone that's interested in the background to the last 5 versions/2 months - back at v5 I was crushed by gunner1069 (I usually get them to play a match every couple of weeks against the latest AI version) with a simple turtle strategy on Theta Passage - they didnt bother getting any mexes and instead just rushed T2 PD with their ACU and my AI had no answer despite having 4x the mass income. My primary focus in all the versions since then has been coming up with the logic that will give my AI a chance in that scenario.

Unfortunately, a relatively simple plan for a human (dont send units single-file into PD, upgrade mexes and then attack with MMLs) required a bunch of different bits of logic:

• Checking for PD and running away from it (so the enemy isn't fed kills/mass), without running away when we're already in range of it and able to beat it (I had some frustrating to watch games where my AI would charge in at a firebase, overwhelm the shields, be about to destroy everything, before deciding to run and turning a winning position into a loss)
• Realising there's PD when it kills our land scouts without being picked up by their radar
• Making use of air scouts to reveal 'hidden' PD to avoid suiciding units and the ACU into the PD
• Quickly expanding to uncontested mexes (including on enemy side of the map)
• Focusing on maintaining map control and ecoing if faced with a turtle rather than trying to break them immediately, which in turn requires various logic for identifying if we're against a turtle, upgrading mexes, building T2+ buildings, upgrading factories, making use of support factories, etc. (I've still not got basic things done like capping mexes with mass storage)
• Using indirect fire units to attack PD/long range enemy units while both protecting them from enemy mobile attacks, and not suiciding our own units trying to protect them

These changes also mean my AI will be worse in other situations - e.g. against a non-turtle it might choose to eco when it would be able to finish the game by pressing the attack, and it's also still poor at the mid-T2+ stage so extending the game via ecoing means it's more likely to snatch defeat from the jaws of victory. Long term though it needed to happen as a 'T1 or die' strategy only has a decent chance on 5v5 maps.

@stlng At some point! Each update lately I think just after uploading 'right, now I can take a break and focus on writing a new approach to handling land units', then I see a replay where M27 does some stupid or incorrect things and figure I'll just put through one more update to fix it!

Latest changes (v66 summary)

• Refinements to the logic for experimentals to dodge shots (and more generally) so they're hopefully less likely to turn around in a full circle doing no damage while being attacked
• 7 tweaks to T3 arti targeting logic, none of which are major but which for example now assign a value to heavily shielded units (previously none was given), firing infront of mobile units, reducing the value of targets that we have fired a huge number of shots at without killing, and factoring in how long it will take for the T3 arti to turn towards the target
• 15 other misc changes or bugfixes, a number of which relate to decisions on when units should retreat (e.g. MML running from ravagers, skirmishers running from PD, GC/Monkey/Ythotha being more likely to run from a fatboy based on how close it is)

Acknowledgements:

• Radde - several teamgame replays on both land and naval maps
• Jip and Magge for highlighting a missing units error message that appeared a lot, and also Jip for helping me figure out how to determine which direction a T3 artillery is aiming at.

@magge The OP sounds like it's an issue joining games in the first place, whereas those errors would be things that occur during an actual game (that has started). I've not heard of M27 causing any issues with joining games so think it's unlikely, although one thing to try to make sure @zorg-kirill would be to disable all sim and ui mods and see if you still get the crash (and if not then enable them one by one).

In terms of the error/warning messages highlighted unfortunately M27 will generate a lot of warnings and errors in the logs. For example the one about no valid units is something that happens regularly that I've not got around to properly investigating (I'm fairly confident it's a false error the vast majority of the time, but figuring out how it arises and how to not have it show the message when it's not an issue without slowing things down is harder). They're also things that would affect how M27 plays in the game (rather than something that would cause a crash).

If the game is crashing (i.e. during the game) with M27 enabled though I can look into further as that's something I've seen happen before

My thoughts on the latest changes and upcoming changes:

• Jamming for T3 UEF Air - Offers chance for interesting tactics (previously I had to create 30 t1 air scouts for a similar effect which meant I've hardly ever done it), and happy the strat jamming is off by default (since at the lower ratings I play at it feels like a great way of making the enemy team aware you have a strat sooner than they otherwise would have)
• Sparky - ok with the stealth generator, but building all T1 factories feels too much
• Loyalist billy nuke change - I'd rather billy nuke was buffed rather than nerfing the loyalist's ability

Overall with the changes I feel bad for the non-UEF factions.

In terms of upcoming changes:

• Nuke missile vs silo costs - sounds good
• T3 arti E nerf - The main theme with E c osts is that air costs far more than ground, which is consistent with the T2 arti Mass to E cost. in my mind it feels like T3 arti should therefore not cost as much E (per mass cost) as an air experimental so I worry the E nerf will be too great
• Reworking UEF bubble shield - depends on how it turns out (could be good, as it could use a boost to usability)
• GC claws - sounds good, GC feels too OP vs T3 armies at the moment
• Loyalist buff - sounds good, it could use a small buff
• Kennel tweaks - depends on how far they go, I wouldnt want to see them reach mass per BP parity with a hive
• Soul ripper tweaks - not sure how the value over time will be reached - presumably by giving it a decent amount of HP regen which would fit with the Cybran theme, it could use some help, and I like mechanics that favour value over time (its one of hte reasons I really like the shield addition to the Czar) so sounds good
• Reducing crash damage of air experimentals - while a small reduction sounds ok it'd be nice for them to still be impactful
• Bulwark change - no view without seeing the detail (could be good or bad)
• Mini SACU rework - no view without seeing the detail (could be good or bad)
• Nerfing HARMs - if its going to be nerfed it'd be nice if ground firing it by battleships could be removed; would also be nice if T1-T2 torp launchers could get a small buff as I worry this could just lead to no torp launchers being usable
• Buffing higher tech engineers - sounds likely to be good and indirectly help performance by reducing the incentive to build T1 engis at T3 stage of the game
• Adjusting T2 static artillery - I hope this means a small buff rather than a nerf as they're a fairly niche unit
• Billy nuke adjustments - hopefully will be a small buff as this upgrade is hardly ever seen

Latest changes (v65 summary)
Smaller update - I'm experimenting with smaller ad hoc updates with reduced testing for fixing issues highlighted in replays (since I'm not planning any big future changes to M27 for the time being as I'm experimenting with a separate AI to try out alternative land based logic)

11 bugfixes, including:

• Incorrectly calculating Seraphim SACU range as their TML range (leading to strange actions from land units)
• Skirmisher platoons (e.g. sniperbots) should be less likely to suicide (by being more likely to attack move and less likely to try and move close to a target)
• Experimental shots such as ythotha shots will now be considered for dodging

10 Misc changes, including:

• Adjustments to skirmisher logic, including making them run further from T2 arti, and splitting up skirmishers that are far apart into separate platoons
• Increased the value of SMD and Nukes as targets
• Made larger units more likely to try and dodge shots (e.g. summits should try and dodge Aeon battleships, Galactic (colossii?) should try and dodge Ythothas)

Acknowledgements

• Radde for providing an entertaining team game replay against M27 (that also highlighted some of the bugs fixed in this release).

That means players who start off playing just one map (like astro) and end up overrated and then want to try and play different games have to force a teammate to suffer playing with someone overrated, in addition to playing far more games, to have their rating adjust to something more sensible, with it being a miserable experience for both players (since the overrated player is likely to just get abuse thrown at them by the teammate who expects them to play as well as their rank would indicate).

If a custom 1v1 is played and both players are trying to win, why shouldn't the rank adjust? E.g. if 1v1s are for practice, then that sounds like a valid reason to be rated and a decent guide to the player's skill. I also don't understand what makes a 1v1 dual gap game so much worse for judging ability to play than a 6v6 dual gap gap - I'd have thought 1v1 would be a more accurate measure of your skill since in a 6v6 you can play poorly yet still win if your teammates do well, and have a lot less to worry about than 1v1 where you have to do all the roles yourself.

The only issue I see is malicious rating manipulation, but there's already a rule banning that

Smaller update fixing a few bugs and adding in mass fabs:

• Unlocked ability to build T2 mass fabs (I've not checked how it handles reaching the unit cap though, here's hoping it doesn't ctrl-K them!)

6 bugfixes, including

• Bug resulting from the latest FAF patch that caused the Ythotha death ball/energy signature to be controlled by the AI (at the same time I've improved M27's response speed to the death ball from before)
• Shield boats should no longer remain at the naval factory if they had lost shield at any point in the past
• The Bjarg map should now be playable by M27 in the bottom-right start position (previously it would break)
  14 Misc changes, including:
• MML should be far less likely to walk into range of T2 PD
• Added new 'experimental prioritisation' builder logic for late-game (where land factories are less likely to build units to conserve mass for experimentals)
• Slight improvements to M27's response when an enemy tries to get a firebase near M27's base
• Slight reduction in the bomber defence range to make them less likely to suicide into enemy air

Note that updates may be smaller and/or less frequent for a bit as I mull over creating an entirely new AI, and/or testing out changes in a basic alternative AI that might then be incorporated into M27 at some future point.

Acknowledgements

• Jip – Highlighting error on the Bjarg map where if M27 started in the bottom right position it wouldn’t work
• Azraeelian Angel - A number of team game replays
• Radde - Several teamgame replays

Seems like a lot of effort for something that I expect would seem counter-intuitive to a number of people (myself included). Even if in real life nuclear missiles would explode in the air, I'm used in games to seeing any kind of missile explode on impact, and to impact the ground/units (not just in FAF but RTS games in general).

Therefore I worry a significant amount of time could be spent on something that ends up making the experience worse for more people than it improves it for (assuming that from a balance perspective it's done in such a way that there's no change in terms of the gameplay effect).

Generally I like the approach to the changes - I'd much rather small to moderate changes to buff/nerf something to see how it goes rather than overcompensating and ending up either ruining a unit or needing a new balance patch to nerf an overpowered unit.

The only two that surprised me were the soulripper nerf (it makes sense given the changes to other air experimentals but would ben ice if it could have something to compensate it, e.g. a small mass cost reduction, since it's by far the weakest of the three), and the nuke silo change (more for the reasoning, as while seeing 1 early nuke is very common I've rarely seen multiple nukes and if I have it's usually been to that player's detriment - i.e. I'd see the issue more that an early nuke is very powerful in teamgames, especially a heavily hive assisted one that can be both built and loaded quickly, rather than that building a second nuke is really powerful. However, I'm not sure what alternative option would achieve that better - construction build time increase? increasing energy cost of the missiles?)

In terms of each specifically:

• Asylum - fine, it was the best mobile shield
• Titan - fine (they're really good). Feel bad for the loyalist as it feels weaker than the titan
• Sniper bots - Hopefully it's not an overnerf, but most of my experience using them is with my AI since I struggle with the attention/apm to use them as a player so Im not well placed to comment anyway
• Ahwassa - fine, it's really strong if you have air control
• Czar - fine - another strong experimental
• Soulripper - as above feels like it could use something to compensate
• Valiant - good, since it's the only destroyer that is bad vs subs
• Salem - good
• Sparky - good, it could use a small buff and fits well with its theme
• Kennels - good, they could use a buff and this still gives them a clear differentiation to hives
• Nukes - as above, overall ok but I dont see a fast second nuke as that much of a problem
• TML change - nice QoL change

Latest changes (v63 summary)
Various changes based on replays received, including:

• Less likely to overflow mass at T3 stage, with a significantly faster scaling up of build power (6 changes)
• T3 fixed arti prioritisation coupled with target variation (to avoid firing at the same heavily shielded position forever) (9 changes)
• Air less suicidal and also better at defending high value friendly air units (18 changes)
• Chokepoint maps - M27 will now abandon a chokepoint and revert to default logic rather than staying in turtling mode if it's chokepoint is overrun and it somehow survives, along with several other small adjustments (3 changes)
• SAM creep logic - Basic code to have M27 try and protect mexes with SAMs if it's behind on air
• 14 bugfixes and 19 misc changes

Acknowledgements

• Radde - various replays, highlighting M27’s poor ability to cope with lots of mass, flaws with the turtle logic, and tendencies to suicide its air (including air experimentals).
• Azraeelian Angel – Loads of team game replays (so many I had to defer analysing some of them for the next release!), highlighting amongst other issues the need for better T3 arti building logic
• Spikey Noob – A number of 1v1s at increasing difficulties on badlands highlighting the ease with which M27's ACU can be killed with T1 land
• Jip – 2v2 replay against M27
• Relent0r – Flagging a bug in M27 that causes it to break default AI from building on mexes that are covered by reclaim.