Jump to content


  • Content Count

  • Joined

  • Last visited

  • Medals

Everything posted by nkenny

  1. I'm building a small project of simplified formations and danger.fsm I'll put them on gitHub of course. -k
  2. I am 100% we are overwriting when changing these values. In fact, I have a lite formation mod I'm just about to release as a showcase. Also, I've used this already in my ork AI mod enhancement. These functions cannot, to my knowledge, be easily called. I wish they could, because they govern some pretty interesting features! That said, some of the values can be changed, see: https://github.com/nk3nny/LambsDanger/blob/master/addons/danger/CfgFSMs.hpp That is the result of some painful experiments. ^^ -k
  3. Short answer. You can't. These FSMs have been turned into engine-level commands. There are two ways to effect these however. Check cfgFSM to see the fsm turne dinto config-- along with a series of interesting functions which are unfortunately inaccessible to seeing or editing. The danger.fsm is fsmDanger config setting, and the formationEntity.fsm is the fsmFormation config setting. It IS possible to change the config settings however. And I have done so for LAMBS danger fsm. In my ORK mod I have also altered fsmFormation-- which is really, really powerful. HOWEVER. The price you pay is loosing all the neat cover-to-cover bounding movement. -k edit: 1. Because of 3 2. Only for the civilian fsms! 3. I believe they are hardcoded
  4. LAMBS Improved Danger.fsm by nkenny / diwako / joko Current version: 2.3.3 I am very happy to announce the release of a new AI improvement mod. At its core it is a considerable expansion of vanilla Danger.fsm. Added are features which extend the enemy response and reaction to fire. The overall goal is to make buildings part of the AI's available terrain. The AI will dynamically enter and clear buildings holding enemy soldiers. The nature of the the danger.fsm does not make these features a replacement for dedicated scripts and clever use of waypoints-- instead the mod should work seamlessly with existing missions. It puts no special demands on mission makers, nor does it add universal game mechanics with strategic implications. With the mod enabled the AI is a significantly more active combatant. The biggest changes is that the AI is now able to view buildings and structures as part of the tactical terrain. The AI will search through buildings, hide and garrison within, as well as provide suppressive fire against suspected enemy positions. In addition AI reactions to fire, even when the origin is unknown is expanded. The AI will also communicate threats to nearby allies. Armoured and tracked vehicles receive new custom behaviour which will see them always try to turn towards known enemies and hammer suspected locations of enemy forces. Features * AI will move inside buildings * Perform suppressive fire (Vehicles will dismantle buildings) * Panic, with special effects * React, even when exact enemy position is unknown * React by hiding in the presence of powerful vehicles when without equipment to deal with it * Check dead bodies, rearms when possible * Tanks will react by rotating towards threats * Improved civilian reactions, hiding in buildings and checking dead bodies * A special Close Quarter Battle mode when DELTA or FILE formation are used (formation can be configured) Some limitations of this implementation are twofold. For one, the Danger.fsm only triggers when the enemy feels threatened. Which makes the actions always purely reactive. This means that you should not expect pro-active searching through buildings, or changes in stance or formation based on terrain. Such things are simply governed elsewhere. Second limitation is myself. I am by education a philosopher, not a coder or computer developer. Please excuse my bumbling attempts. I intend to develop the mod further. In a sense it is already outdated based on my future plans. That said I wanted to release something now in a public sphere to gain more experience, feedback and perhaps garner more interest in FSM level modifications to AI behaviour. I see a lot of opportunities to add or fix annoying aspects of AI behaviour. I write that this release in a sense already outdated, because I intend to redo the current FSM flow structure; which is currently an expanded version of the Bohemia variant, which itself appears to have remained basically untouched since Arma2. Video Here is a (rather clumsily made) demonstrations video. The scenario consists of eight Spetsnaz soldiers assaulting three buildings held by 10 looters. The looters are garrisoned using CBAs taskDefend module. The attackers are given a single MOVE and SEARCH AND DESTROY waypoint. Their formation is set to DELTA/DIAMOND-- which triggers a special CQC mode. The attackers defeat all defenders and suffer zero casualties. No other special mods or scripts are in play. --- STEAM Workshop --- GitHub --- Development Discord
  5. @Kurtosis Default Arma feature in the editor, mate. 😉 -k
  6. @Storm Rider Not without a complete makeover of Arma3's animation and pathfinding system. Especially as the pathfinding ties into formations. Can't do much about the first two, the third, well. I am exploring things. -k
  7. @madrussian We'll be looking into it. -k
  8. That is essentially how LAMBS danger works 🙂 Most if it is built on stuff documented by .kju and others in Arma2 days-- but extrapolated to Arma3. If you are running LAMBS danger you can track these variables by getting <unit> getVariable ["lambs_danger_FSMDangerCauseData",[]]; Which outputs : [_dangerCause, _dangerPos, _dangerUntil, _dangerCausedBy]] --- I have experimented quite a bit with the different types of events, and there are some intricacies to them. I unfortunately do not have time to write it out right now (dinner is calling!). The next evolution of LAMBS danger.fsm is, as I envision it, to read these variables and feed them into an even more nuanced brain than LAMBS danger.fsm currently runs-- while also rewriting (or somehow enhancing) the existing engine based formation.fsm. DCDeadBodyGroup and DCDeadBody absolutely do fire. It is only that their priority is very low and therefore will rarely come to the fore. You can experiment with LAMBS where it is possible to configure the priorities on the fly: // FSM priorities GVAR(fsmPriorities) = [ 3, // DCEnemyDetected 1, // DCFire 6, // DCHit 5, // DCEnemyNear 3, // DCExplosion 1, // DCDeadBodyGroup 1, // DCDeadBody 2, // DCScream 4, // DCCanFire 3 // DCBulletClose ]; // GVAR(fsmPriorities) translates into the variable 'lambs_danger_fsmPriorities'. edit: For your use you could do this: Read "lambs_danger_FSMDangerCauseData", or extend the codebase with an eventhandler. Then continuously set "lambs_danger_forceMove" to true on the unit to override other manoeuvre schemes. 🙂 The latter variable is a first-run attempt at getting a more agile combatant. But like I said, it will require an ever deeper delving into A3 AI methinks.
  9. LAMBS Danger.fsm New Release Version 2.3.3 Changelog Added TaskReset Module Added Hide Area when moduleTaskArtilleryRegister is Synced to Object Added TaskArtillery now uses ETA time for check round distance Added FiredNear EventHandler to TaskCamp to Improve Reaction on Enemy Contact Added Voice callouts to more functions Fixed Issue with Modded Voices Fixed Possible Scripting Errors in TaskArtillery Fixed Scripting Errors in danger.fsm Fixed Possible Scripting Error in assaultBuilding Fixed Issue with taskCamp where units would occasionally get stuck in an animation Fixed Reduced AI leaders reaction to hostile air and tank vehicles Fixed Ability to break out of taskAssault/Retreat Fixed TaskAssault waypoint in retreat mode Fixed Callout ranges and speakers across some functions Fixed Suppression was using bad positions Changed Some small background performance Improvements Changed remove some Duplicated Config Data From the team: diwako / nkenny / joko HOTFIX Following feedback from testing sessions and Discord we've quashed all known bugs and pushed out a hotfix. Available from the usual places. 🙂
  10. The effect of weapon on AI accuracy by nkenny Abstract* Testing the effect of weapon on AI accuracy. Introduction The weapon an AI soldier is equipped with greatly affects killing power. Even when the class of weapon is nominally the same, weapon configuration plays enormous part. This is especially evident when similar weapons from different mods are used together. To explore the effect of weapon on AI killing power I built an experimental suite. Experiment The test was conducted while running only CBA, CUP Weapons, RHS Russian Armed Forces and NIArms. The player profile was newly created and no AI skill tweaks were made. Description I placed a single shooter on flat terrain, back turned towards a four man group of enemies. The four man team is concealed from the shooter (his back is turned). Each soldier is identically equipped with no body armour or helmet or weapon. After five seconds the shooter is given a weapon and given perfect information about the enemy. This results in him turning and engaging the enemy. The targets are spaced at two meter intervals and the test was run on a 100m and 200m range (normal combat engagement ranges). Shooter and targets were forced to remain static and remain in a standing position. My point of comparison was the AKM rifle. Ubiquitous in enemy hands and mod packs alike. I provide data for the modern MX rifle with a HAMR RCO as a frame of reference. The exercise is over once three minutes has passed, the shooter has expended all ammunition, or all targets are dead. Measurements are printed on screen and recorded on the clipboard. I paste that information into a document. Measurements I ran the exercise six times with each weapon. While not enough for statistical evaluation, I nonetheless deemed it sufficient to get a sense of the performance characteristics of each weapon. - Hits, each bullet that impacts a target. - Shots fired, maximum is 6 magazines: 180 rounds. - Accuracy, hits divided by shots fired. - Time, time in minutes and seconds. Record After each round I recorded weapon class name, Accuracy, (hits / shots fired) and time. Each measurement is averaged (in red). Expectations Mod configuration will vary somewhat. With CUP weapons being more accurate than RHS counterparts. Data @ 100 Meters Data @ 200 Meters Findings @100 Meters At close range the vanilla AKM and CUP AKM are closely matched. With a slight edge to the vanilla gun. RHS weapons are considerably less accurate than all other counterparts and require four times the amount of time and rounds fired for similar effect. The numbers for the NIArms AKM belie their effect on target. Toadie's AK tended to fire in bursts, but with a highly accurate first shot. In effect the gun killed with the first shot, but burned off a few rounds into the air-- wasting time and bullets. @200 Meters At mid range the differences between the weapons become evident. CUP weapons are almost twice as effective as their vanilla counterpart. RHS weapons are considerably less accurate than the others. Which is not to say ineffective. Rate of fire is high and spread provides greater coverage-- area of fire-- than the other counterparts. As for NIArms: the rifle comparatively is laser accurate, out shooting all other alternatives. Contrasted to RHS it provides almost eight times more killing potential. It even compares favourably to the vanilla MX with a optic. Conclusion Not all AKMs are created equal. The killing potential of each gun varies wildly from mod to mod. The most deadly close quarters AK is the vanilla, while in most combat ranges, the NIArms gun outperforms all other alternatives. One obvious take away is that there are radical differences of capability with two presumably identical weapon systems. This is one reason why I would advice consistency in mod sets. Another is that this vast difference in capability need not match the players performance, expectations and experience, with the same weapon system. In the case of the AKM, I believe most players find them very similar-- aside from minor differences in sight picture and sound. In fact to test that very thing I used the same suite to record my own times. I won't post the results here, but they were largely similar across mods. The test itself remains neutral to the intended design intention of the style of configuration each mod adopts. The comparable lack of accuracy of RHS weapons need not be read as a flaw. There are arguments both from the perspective of realism and game play to prefer less accurate weapons. - Firstly, the sort of 'white room' engagement against a non-responsive, non-moving target is hardly applicable. Handling performance on sterile ranges fail to convey the stress of combat shooting. Most studies into the effect of combat on marksmanship show a massive decrease in effective accuracy. RHS weapons simulate combat shooting, rather than target plinking. - Secondly, more shots fired and fewer one-shot-kills, is generally good for game play. At 100 meters vanilla, CUP and NIArms equipped soldiers can eliminate an inattentive fire team in less than 10 seconds. While demanding some caution is good, such a quick end also means a quick end to the session for the player in question. - Thirdly, RHS comes with an entire dimension of weapons. Accuracy comes with modern quality. A RHS M4A1 is more accurate than an AKM. So is a an AK74. See references. Many communities depress AI accuracy values. The considerable accuracy of CUP and NIArms weapons need not be a flaw. Follow the links in the references and one can see the extreme degradation of accuracy based suppression levels. The considerable default accuracy of vanilla weapons, the presence of multiple stances, all are suggestive of a very specific type of gameplay. Soldiers should be cautious and reveal their positions only momentary. Suppressive fire is very effective. It is more important to suppress all dangerous enemies than it is to pursue time consuming killing shots! When the AI is accused of laser accuracy, it will in some cases be due to failure to adequately suppress-- through fire superiority-- the enemy position. Finally, I suspect, particularly in regards to the user made modifications there are considerably differences. Test are done in different AI setting ecologies-- tweaking AI skill is easily (if somewhat obtusely) done. If I shall brave a conjecture I expect CUP weapons, being added more along open source model will show greater variance in configuration. NIArms weapons are developed with an eye towards delivering an uncompromising experience for the shooter. The AI comes much later. All of this means that mod makers and mission designers must pay attention to the source of the weapons given to the AI. Future tests 1. The same platform can easily be used to test other weapons or family of weapons 2. In the future I would like to test ranges out to 300 and 400 meters** References - Original thread on nopryl.no, further numbers, weapons and experiments are here - Experimental mission, links to mission for your own testing -k * Dear mods: Feel free to move this thread. While relevant to mission makers and mod developers equally. It seems sufficiently meta to concern general players, but it is not pointed enough to belong to the mod discussion forum. ** My initial run with standing RHS weapons yielded one hit across 180 rounds. In other words. If faced by a lone, standing gunman armed with a RHS AKM. At 300+ meters you might as well remain stationary and standing yourself.
  11. @Devastator_cm Seems like these are pre-defined in ACE3 by classnames, not by config entry. https://github.com/acemod/ACE3/blob/master/addons/medical_treatment/ACE_Medical_Facilities.hpp Best bet may be to set it by the facility variable? Also: https://github.com/acemod/ACE3/blob/master/addons/medical_treatment/functions/fnc_isInMedicalFacility.sqf -k
  12. LAMBS Danger.fsm New Release: Current version 2.3.0 Changelog Added TaskCamp ~ scripted camp behaviour Added TaskArtillery, TaskArtilleryRegister, TaskAssault, TaskCamp, TaskCQB, TaskCreep, TaskGarrison, TaskPatrol, and TaskRush Zeus and Editor modules Added Target Module for Zeus Task modules Added Ability to define Areas in TaskArtilleryRegister, TaskAssault, TaskCQB, TaskCreep, TaskGarrison, TaskPatrol and TaskRush Added Parameter in TaskArtillery to skip adjusting rounds Added Parameter to TaskAssault and TaskCQB to use dynamic (target) module position Added Smarter distance-based sorting to TaskCQB Added Parameter to TaskPatrol to move waypoints randomly after patrol cycle is finished Added Parameter to TaskPatrol for Waypoint Count Added 3DEN Setting on vehicles to mark units as pre-registered artillery Added Share information will share to all friendly sides Added Option to disable enhanced fleeing function Added Improved position analysis to Suppression, Vehicle Suppression, and Vehicle Assaults functions Added APC assault routine for armoured vehicles (keeps infantry mounted) Fixed Issue that TaskX functions get called from the wrong environment Fixed TaskPatrol did not Loop Waypoints Fixed Issue that Zeus Module window did not Open Fixed Share information MP compatible Fixed Enhanced fleeing function is much less intrusive Fixed Issue with taskArtillery when weapon out of ammunition Fixed A race condition in taskArtillery Tweaked HideInside to be more defensive Tweaked Callouts so contact shouts are not always run through formation leaders Tweaked ShareInformation default range Tweaked Civilian priorities in civilian.fsm Changed Moved Automatic Artillery Registration to WP module ZEUS and 3DEN This time our focus has been directed towards enhancing the Zeus and Misison edit user interface. Many of the core component scripts and waypoints have been updated, rewritten or enhanced. Which feeds into the biggest new addition: the new MODULE system. The module system adds an easy way to use and configure the enhanced AI waypoints and scripts available in the mod. The modules integrate seamlessly to both ZEUS and 3DEN. This system supersedes and enhances the original waypoints in many ways. We've also added a new TaskCamp module. This script makes the group appear safe and dismissed and is perfect for camps or having troops just sitting around. Units set to taskCamp will occasionally garrison buildings and man static weapons. If the group is large enough a small patrol will check the Discord Got questions, suggestions or other feedback? Come hang with the cool kids on our Discord. From the team: diwako / joko / nkenny --- STEAM Workshop --- GitHub --- Discord
  13. @mandaloin It may be the reaction state interfering with your commands. You can enable or disable a number of player group features (to create more dependable units) in the CBA mod settings. It is also possible to configure a key to hot-toggle the LAMBS mod for your own group-- so if you need them moving, just toggle it and do your thing 🙂 -k
  14. "Added: Static M2 (low / raised), with optional ballistic shield and collimator sight" What? Hallelujah! Christmas has come early this year. -k
  15. nkenny

    Cold War Rearmed III

    Much excite. What about Global Mobilization as a sort of crossover? -k
  16. nkenny

    The effect of weapon on AI accuracy

    Actually this is precisely one major design flaws of vanilla Arma3. It was initially released with a lack of dimensionality which make it possible to create such asymmetric scenarios. Within the same mod ecology I think the ideal solution would be to set base accuracy to chronological origin of the weapon. It would make perfect sense to have, given a 2020ish timeframe, that poor quality troops carry older AK rifles, whereas high level troops would carry modern and accurate AKs with RIS and polymer bits and all that jazz. CUP and vanilla configs are quite similar. I wonder how much of a difference the dispersion values above actually make. Perhaps I'll make a tiny mod to test the theory. *shrug* -k edit: I too greatly appreciate the efforts of the CUP team.
  17. nkenny

    The effect of weapon on AI accuracy

    @froggyluv As in previous tests my skillsettings are default. Units are set to 0.5 skill. The skill setting system is in my experiences rather opaque and difficult to get a proper read of. The testing suite makes it easy to run multiple tests of it. -k
  18. nkenny

    The effect of weapon on AI accuracy

    Restested AKM rifles by nkenny Introduction This is a repeat of the entry experiment where AKMs from different mods were compared. It utilises the new testing suite developed mid-thread. The report coincides with the latest @CUP release, which seemed as fitting time as any. Experiment 1. This test evaluates three larger mod/DLC packs and the vanilla AKM as control: RHS, Global Mobilisation and CUP. 2. It develops and compares numbers from 25 to 400 meters. 3. Because this is primarily an accuracy test, and the new Magwell technology allows it-- the same magazine is used for all weapons : "30Rnd_762x39_Mag_F" (vanilla 30 round 7.62x39 magazine). I will immediately recognise that this may slew some numbers as weapons should in some case be considered within their mod/conversion/DLC infrastructure, but as the original post suggest. This is not commonly how these mod packages are actually used. 4. As before the only mods run was CBA and the relevant mods in their most recent Development versions. - Arma version 1.96.146114 - CUP Weapons DEV version r1419 - RHS Dev version Build 83 - Global Mobilisation version 1.1 Description and measurements The measurements and platform of testing is identical to earlier tests save that units were set to "AUTO" stance at all ranges. Expectations As per previous tests: RHS weapons to be the least accurate, CUP weapons to be the most accurate with other AKMs falling somewhere in between. I know that the CUP team have recently been reviewing some of their weapona ccuracy settings according to https://dev.cup-arma3.org/. Data As before I will present a summary of my findings. To access the raw data check the references link below. Findings - The vanilla AKM reinforces and corresponds with numbers developed in the previous test. Peak accuracy resides around 100 meters and the weapon reaches maximum engagement range at exactly 250 meters. The maximum range is not borne from the figures listed above but other tests. - The CUP AKM is the weapon that follows the vanilla gun the most closely. Even so it remains considerably more accurate than the vanilla weapon at all ranges. At sub-100 meters it as similar times to kill, but at 200 meters it is siginifcatly more deadly than its vanilla counterpart. It is also far more ammunition efficient at all ranges. The CUP gun is subject to the same maximum range as the vanilla AKM; tested to exactly 250 meters. - The Global Mobilisation AKM, actually a East-German version of the Soviet AKM, is by far the most accurate weapon at all ranges. At 200 meters the GM gun is 62% faster killer than its vanilla counter-part. It is also the only weapon which can engage with any degree of accuracy out to 400 meters. Without mitigating factors such as ping, suppression or skill settings, a GM AKM will kill a four man fire team at 300 meters in around 30 seconds, dealing damage every four shot. - The RHS AKM reaches peak accuracy at 25 meters at which time it is comparable to all other weapons. From there it has a slower time to kill and much increased ammunition expenditure to the point of being in a different category. It is worth explicitly mentioning is that all weapons, aside from the RHS AKM, are configured with a drop in accuracy at closer ranges and a peak accuracy around 100 meters, with a loss of accuracy further out. - Other findings: (a) Though not timed, all modded and DLC weapons appear to have a slower magazine reload time than the vanilla weapon. (b) Only the GM AKM fired true automatic fire at 25M Conclusion The hierarchy of deadliness goes as follows: - GM AKM - CUP AKM - Vanilla AKM - RHS AKM 1. Referencing numbers from earlier weapons test one can see that the Global Mobilisation AKM is simply in a class of its own and rivals most modern 2035 weapons. This could in-universe be explained as (1) the weapon being closer to factory new and (2) the weapon should be compared only to weapons from its own mod set and ammunition. These are in my opinion poor explanations. If used blindly alongside weapons from other mods or vanilla it will in the AIs hands outclass all other takers. From a personal perspective as a mission maker and server host, I would say that without severe reductions to default server accuracy and intelligence settings, the weapon is unusable for most scenarios. 2. The CUP AKM remains deceptively accurate. While its numbers are close to vanilla stats-- it routinely outshoots the vanilla weapon in terms of 'time to kill' and 'rounds fired'. This is especially noticeable at average combat ranges of 100-200 meters. 3. RHS weapons again stick out as being configured in a very different manner than other weapons. I have defended the approach of configuring weapons for the 'combat accuracy' rather than 'bench accuracy' above and will not repeat that here. 4. It is my personal opinion that both CUP and vanilla AKMs suffer from the lack of pluss 250 meter engagement range. Not only from a simulations perspective, queue any reports of modern firefights, but even more accessibly from player vs player warfare in Arma-- were humans will often engage at ranges where chances of hitting are low for reasons of both weight of fire and incompetence. 5. Final words remain that mission makers and server/mod managers should remain aware of which weapons are used in which scenarios. Attempting to use RHS and Global Mobilisation weapons interchangeably in AI hands may result in effects wildly diverging from expectations. Future tests CUP, vanilla and RHS all feature 10-11 inch HK416 carbines. References - Raw Data folder Most pertinent dispersion values for 'single' shot at maximum range. vanilla dispersion = 0.00203; cup dispersion = 0.00125; rhs dispersion = 0.0021; gm dispersion = 0.0014992; It is not entirely clear to what the basis of these values are.
  19. @Melody_Mike I'll answer here, but if it still fails to work please redirect to this thread. 🙂The script works for me on Altis. I have however updated on the gitHub to 0.28b. Just a little bit of cleanup on my end. Script seems to run quite well when execVM'ed. It is worth noting that the script sorts for only internal positions within buildings. The variable you are asking about is defined in line 55. You can edit this line to alter if only indoor sites are selected. Which map are you running this on? It may be one without defined building positions and paths.. Especially older Arma1 content ported into CUP or similar has this problem. Oh, and I'm happy to help. hehe There is no need to unpack the mod. Just check the gitHub. It isn't difficult, but will require some familiarity with arma scripting and modding. The modded version is both more advanced and performance friendly, but if used will require you to comply with the license. -k
  20. @cpt.ghost I have added it to our to-look-at list. Quick fix is to write <group> allowFleeing 0; in the relevant units inits. In fact the function should already have a max distance. But as mentioned, it is difficult -- if not impossible -- to unflee units by script once they've started. -k
  21. @cpt.ghost Glad you like it. I am guessing you are running an op with a considerable number of armoured elements? The AI will respond with much improved sense of self-preservation. 1. The only instance which triggers a change in behaviour to STEALTH mode is when the AI is fleeing. Fleeing is a vanilla behaviour which is enhanced by the mod to have the AI view buildings and bushes as valid places to stop. One way to mitigate this behaviour, if you want better control of the AI, is to set allowFleeing to 0. It is admittedly an imperfect implementation on our end as it can be difficult to get fleeing units to unflee. 2. While I have seen units disembarking early, from time to time, this is much mitigated with the latest version. Units becoming stuck is however new. Is this on some custom terrain? What orders/waypoints are the units under? 3. This I have not seen. I see you are playing with a long list of modded content. Can you verify that this works without the mod loaded? 4. As per (1). Come to think of it, we should add a setting for the Fleeing too. -k
  22. @autigergrad That is a negative. I have thought of ways to implement it, it is fairly straightforward to do it for specific weapons with the eventhandler system, but not come up with the conceptual framework of one which is universal enough to suit our needs. Out of interest. Did you experience a bug of some sort, or do you have something specific in mind in terms of functionality. 🙂 -k
  23. @Melody_Mike https://github.com/nk3nny/taskRush-Creep-Hunt-for-Arma3 TaskCQB from that GitHub, mate. edit: Script is fully standalone. You do not need the mod. Use it as you would taskRush or any other script on that hub. -k
  24. LAMBS Danger.fsm New Release: Current version 2.2.1 Changelog Added AI unit voice callouts Added setting to disable unit gestures Added setting to disable unit Immediate Reactions Added verbosity to debug messages Added side configured colours to 3D debug helpers Added debug messages recorded to RPT Added visual checks to suppression to make function more reliable Added ability to hot-configure fsm lambs danger priorities in variable "lambs_danger_fsmPriorities" - Fixed reverted CfgFSM settings to increase unit cohesion Fixed added more time to check body events Fixed tweaks to unit speed and distance check in CQB to add more fluid combatant Fixed reduced building search range on reaction phase -- makes for more responsive AI Fixed enhanced fleeing state intelligence Fixed prevents squad vehicles from being drawn into movement Fixed refined FSM danger priorities -- increased importance of canShoot Fixed various aspects of taskGarrison and taskPatrol and taskCQB Fixed taskAssault and Retreat will now allow users to move waypoint actively Fixed incapacitated units would sometime speak or give orders Fixed issues with LAMBS Zeus modules deleting units Fixed additional checks when selecting artillery - Changed default suppression fire minimum from 25 to 20 meters Optimize fnc_findClosestTarget (Thanks to Dedmen) Optimize fnc_findBuildings (Thanks to Dedmen) Optimize fnc_findOverWatch (Thanks to Dedmen) Shout out to groups that have played with LAMBS Danger fsm and given feedback -- or simply have had us listen in on their troubles! Pori Brigade, Gruppe W, CNTO, Special Forces Group, and nopryl.no Developments in 2.2.0 A big tactical focus for this release has been to tweak settings pertaining to squad cohesion. This has been combined further refinements to close quarters AI movement skills. Expect to see less scattering of AI squads. Another area of improvement has been debug information. With this release: debug information is stored to RPT (if the setting is enabled), debug descriptions are more verbose, and there are additional colour coded debug helpers to let users understand what the AI is thinking. Finally, the release comes with bugfixes and improvements to waypoints and Zeus modules. These enhancements will set the stage for the next release which will have a significant focus on Zeus and Editor interface. Discord The team has created a public discord were we answer questions and discuss various aspects of developed features. Feel free to join in. 🙂 From the team: diwako / joko / nkenny --- STEAM Workshop --- GitHub --- Discord Edit: Terribly sorry guys. Released a small Hotfix 2.2.1 which fixes some errors relating to scripted waypoints. - nkenny ~ A small glimpse at one of the micro-features in the LAMBS Danger fsm mod
  25. @AirShark Incidentally, that option is ready for the next version🙂 -k