Jump to content
Rydygier

HETMAN - Artificial Leader

Recommended Posts

Well, you can artifically "feed" him with intel about you by cyclical usage of reveal command, where info goes to any unit under HAL's control, eg to Leader unit itself. However, unless in A3 it is different, as was in A2, this works only, if unit, that gets info about you is not further from you than your view distance setting. In A2 (in A3 too, I suppose) units instantaneously forget about any unit, that is further. I do not know any workaround for that except setting bigger view distance in video options. At least you can try to move his objectives, secondary objectives and/or idle decoy at your position, but it is not the same, still means, that he will send some units towards you from time to time.

EDIT: Oh, there is some other way. I'll update this post later with some code snippet.

So, here is:

if not (player in ((group leaderHQ) getVariable ["RydHQ_KnEnemies",[]])) then 
{
leaderHQ reveal [player,2];
(group leaderHQ) setVariable ["RydHQ_KnEnemies",((group leaderHQ) getVariable ["RydHQ_KnEnemies",[]]) + [player]];
(group leaderHQ) setVariable ["RydHQ_KnEnemiesG",((group leaderHQ) getVariable ["RydHQ_KnEnemiesG",[]]) + [group player]];
};

This will reveal you to the Leader, and add you and your group to the Leader's variables, that contain enemies considered as known, unless player already is known enemy. These arrays are every cycle refreshed, so that code must be repeated often. Eg inside repeteable trigger's act field - this way you may control, when to activate this via trigger condition. This way should be immune to the viewdistance limitation problem.

Edited by Rydygier

Share this post


Link to post
Share on other sites

First of all..thanks for your time and effort for the code but..it's like a hack :)

You see..the MOST impressive with the HETMAN is..all things ARE working almost like IRL without hack-ish approaches from the mission maker..

Even as-is..when the HETMAN get fresh intel i.e from aerial units..hunt will keep going!!

I was thinking on a way he can set up chock-points and far ambushes when hunted group whereabouts are not known.

But i realize..this is getting out of scope multiplying the complexity

Even without it..HETMAN IS shining ;)

Share this post


Link to post
Share on other sites
I was thinking on a way he can set up chock-points and far ambushes when hunted group whereabouts are not known.

But i realize..this is getting out of scope multiplying the complexity

A little - yes. I mean, Hetman has own goals to achieve, engaging enemies is only a step towards these objectives (removing treatening obstacles). He is not designed as "indefatigable pursuer" kind of AI. If enemy is fleeing - is no more significant treat, so no need to spent resources and time to hunt him down, until he return (would be spotted again in the HAL's "region of interest"). So little "hacking" is necessary to persuade HAL to such behavior. Hacking BTW is here anyway partially justrified, as viewDistance "event horizon" beyond which all is forgotten just like that is pure artificial thing, so omitting it is rather way to restore natural state, than violation of it.

Behavior, you described, is more probable for occupant vs partisans situation, where stabilization forces (garrisons) are supposed to keep their area safe and eliminate any local defiance.

Of course, I can add such feature as option, so chosen units will be considered as known regardless of circumstancies, but this would have basically same effect, only achieved in "official" way, not "hack-ish". Mostly formality.

Prediciting, where hunted unit would be after chosen time, finding suitable place for ambush at probable route and performing effective round-up hunt is another thing however, matter for separate script. Interesting matter, I would say. :)

Edited by Rydygier

Share this post


Link to post
Share on other sites

Please forgive me if this has already been asked but does RYD_HAL conflict with either Artificial Intelligence Support Systems or Fire Fight Improvement System

Share this post


Link to post
Share on other sites

Hi Cadmium. HAL and AISS should work together. Make sure you set all your Artillery in AISS to 0. HAL's Arty is much more sophisticated. ;) IF you prefer AISS. There is a way to take out the artillery on HAL in its configs. Its in this thread or HAC's thread on arma2 of how to do that. When I have more time I'll put a tutorial of how to disable either or. But so far the last I tested it. It worked ok. Some features of AISS may be effected, but will work as far as I've seen thus far. ;)

Share this post


Link to post
Share on other sites
Hi Cadmium. HAL and AISS should work together. Make sure you set all your Artillery in AISS to 0. HAL's Arty is much more sophisticated. ;) IF you prefer AISS. There is a way to take out the artillery on HAL in its configs. Its in this thread or HAC's thread on arma2 of how to do that. When I have more time I'll put a tutorial of how to disable either or. But so far the last I tested it. It worked ok. Some features of AISS may be effected, but will work as far as I've seen thus far. ;)

Thank you very much for your prompt reply Mikey74. I just read the basic instructions and got RYD_HAL up and running and it's awesome. :)

I think I'm going to take out AISS for a while and let the very competent battlefield commander in RYD_HAL take it over for a while. It was very effective in the last scenario I played even with AISS and RYD_HAL competing.

If I do "set artillery to 0" in AISS will I have to go into the PBO of AISS for that?

This promises to be much much better than even Armed Assault for generating great missions with a minimum of my own input.

Edited by Cadmium77

Share this post


Link to post
Share on other sites

In your user configs.

Min_Radius = 70; //// Minimum radius you want the rounds to fall from target position. Remember if its out too far they will hit friendly units if within this radius.
Max_Rnd_Radius = 70; ////// Random number for how far out Shells will far from Target position.
Art_Salvo = 4; ////// Minimum rounds you want to see falling. Remember this is per unit in a given group. so 8 x Art pieces in group. example 8 x 4 = 32 minimum.
Art_rdm = 8; ///// Randomizing how many rounds from 1 to 8 + Art_salvo. 
West_Rnd_chance = 0; ///// Blufor: .3 meaning 30%. Always use a . (point) before your number unless you want 100% then 1 so go from .0 to 1.
East_Rnd_chance = 0; ///// Opfor: .3 meaning 30%. Always use a . (point) before your number unless you want 100% then 1 so go from .0 to 1.
Res_Rnd_chance = 0; ///// Resistance: .3 meaning 30%. Always use a . (point) before your number unless you want 100% then 1 so go from .0 to 1.
Marker_pos = 1; ///// 1 = true and 0 = false. If true Marker will pop up where Artillery is targeted. 
AISS_Respawns = 4; ///// times each group will respawn.
AISS_Resdelay = 7; //// time of delay in seconds before each respawn.

Default looks like this:

Min_Radius = 35; //// Minimum radius you want the rounds to fall from target position. Remember if its out too far they will hit friendly units if within this radius.
Max_Rnd_Radius = 70; ////// Random number for how far out Shells will far from Target position.
Art_Salvo = 4; ////// Minimum rounds you want to see falling. Remember this is per unit in a given group. so 8 x Art pieces in group. example 8 x 4 = 32 minimum.
Art_rdm = 8; ///// Randomizing how many rounds from 1 to 8 + Art_salvo. 
West_Rnd_chance = .3; ///// Blufor: .3 meaning 30%. Always use a . (point) before your number unless you want 100% then 1 so go from .0 to 1.
East_Rnd_chance = .3; ///// Opfor: .3 meaning 30%. Always use a . (point) before your number unless you want 100% then 1 so go from .0 to 1.
Res_Rnd_chance = .3; ///// Resistance: .3 meaning 30%. Always use a . (point) before your number unless you want 100% then 1 so go from .0 to 1.
Marker_pos = 1; ///// 1 = true and 0 = false. If true Marker will pop up where Artillery is targeted. 
AISS_Respawns = 4; ///// times each group will respawn.
AISS_Resdelay = 7; //// time of delay in seconds before each respawn.

Share this post


Link to post
Share on other sites
In your user configs.

Min_Radius = 70; //// Minimum radius you want the rounds to fall from target position. Remember if its out too far they will hit friendly units if within this radius.
Max_Rnd_Radius = 70; ////// Random number for how far out Shells will far from Target position.
Art_Salvo = 4; ////// Minimum rounds you want to see falling. Remember this is per unit in a given group. so 8 x Art pieces in group. example 8 x 4 = 32 minimum.
Art_rdm = 8; ///// Randomizing how many rounds from 1 to 8 + Art_salvo. 
West_Rnd_chance = 0; ///// Blufor: .3 meaning 30%. Always use a . (point) before your number unless you want 100% then 1 so go from .0 to 1.
East_Rnd_chance = 0; ///// Opfor: .3 meaning 30%. Always use a . (point) before your number unless you want 100% then 1 so go from .0 to 1.
Res_Rnd_chance = 0; ///// Resistance: .3 meaning 30%. Always use a . (point) before your number unless you want 100% then 1 so go from .0 to 1.
Marker_pos = 1; ///// 1 = true and 0 = false. If true Marker will pop up where Artillery is targeted. 
AISS_Respawns = 4; ///// times each group will respawn.
AISS_Resdelay = 7; //// time of delay in seconds before each respawn.

Thanks a lot. So each of those lines should be set to 0 then?

Edited by Cadmium77

Share this post


Link to post
Share on other sites

West_Rnd_chance = 0;

East_Rnd_chance = 0;

Res_Rnd_chance = 0;

these ;)

Share this post


Link to post
Share on other sites
West_Rnd_chance = 0;

East_Rnd_chance = 0;

Res_Rnd_chance = 0;

these ;)

thanks again. I'm just reading the HAL 1.1 manual here; This is amazing. I foresee vast battles upon the domain of Altis of heretofore inconceivable complexity. :)

Share this post


Link to post
Share on other sites

As you can read here, is planned very soon release of additional goodies for IF conversion mod, although all designed for A2 version. Still, amongst them, there is dedicated version of HAC (A2 Hetman) with all IF+DLC+IFR content classes hardcoded/included under library init variable. That HAC version IMHO may work also under A3 through AiA, same as whole IFA3. Still, if someone want to try HAL or default HAC, here is, not tested, compilation of all these IF classes in form of usual RHQ set.

Share this post


Link to post
Share on other sites
Maybe Add a userconfig to it. BTW I've found some incompatibilites with HAL and AISS. I think I get and ammo error when the Artillery starts firing off. I have it set to reload ammo after every salvo.

Will try to fix this on my end if possible. Going to put an AISS variable to set to tru or false. Also my CAS part does use waypoints. I'll look for a work around. Other than that They both do seem to work pretty good together. Again RYD awesome mod! Been looking foward to the release. :) Thanks man.

Aha now I see why I have to set the ammo to zero in AISS

And Rydygier, thanks so much for all the incredible work you've done on this mod. I'm just getting into installing mods for Arma 3 and yours is amazing. I don't even want to fight. I just want to observe the battle logic unfold around me. These AI commanders are incredibly crafty and cautious, doing the kinds of things I never imagined in a game. You should be hired and paid big bucks to go in and solve the AI problems for the Total War series. I haven't seen such great wargaming thinking since I read Dunnigan's works.

---------- Post added at 11:23 AM ---------- Previous post was at 10:38 AM ----------

Sabotage, bombs, guerilla tactics - This idea isn't forgotten. I have on my HDD something ready in 60% that is not HAC/HAL mode, but entirely separate project concentrating on constant, ambient (spawned) asymetric AI (terrorists/freedom fighters/partisans) vs AI (stabilisation forces/occupant) warfare. And yes, it needs micromanaging AI. A lots. There is implemented whole dynamic ambush mechanics. From predicting, where is a place worthy of set up an ambush, through sending ambushers there, setting a bomb, hiding, waiting for convoy, touching off the bomb and attack, to the dynamic smart withdraw.

There are ready dynamic patrols, sweep the town operations, partisan hit and run tactics (where "run" part is working really)... Still much to do.

So, why it is only 60% done? I do not know. Just after implementing these ambushes about early June somehow lost interest (temporarily, I hope), and took other projects, that keep me busy till now. So I guess, when I'll return to that, I must rewrite all under A3. When? Do not know.

Why separate, not inside Hetman? Too different approach. Hetman mechanics in its specificity will limit too many features, I want there.

I do not pretend to be any kind of tactics/strategy specialist. I probably am awared about part of the basics, but who knows, always can learn something new, useful for Hetman. So - decide on your own, what and how verbose do you want to share with me/us. I'll be grateful for anything, that possibly may improve HAL. Of course not all, what is reasonable in real life, is possible to reasonable implementation...

---------- Post added at 16:50 ---------- Previous post was at 16:36 ----------

Yeah, it is rather deeply modified/expanded Arma, on visual level somewhere between A1 and A2 (for

). It is partially parallel development, but VBS has features, that armaverse never saw, from the other hand, as for engine, it is one step behind Arma (models one or two steps, but much, much more of them, than in Armas). Of course, it is not a game, not very entertaining, so visuals are secondary. As for implemented AI - well. Unless they added something like that lately, situation there was similar, as is in Arma. You want to have advanced AI tactics - you must script them. There may be scripting/other tools, that make it easier, than in Arma, still... Of course some third party companies sometimes developing something like that for VBS2 as external plug-in or something like that.

Also they are working on VBS3, I see... Interesting, what new there. VBS is great, but not so much for gamers, rather for professionalists for training purposes. I guess, AI has there lesser meaning, it is more about trainees working together/alone in the network.

Has anything come of this? Would I ever love to get my hands on this.

Edited by Cadmium77

Share this post


Link to post
Share on other sites

Hi Cadmium77, I'm glad you like it. Yep, also I like often just to see what is happening around. But anytime I'm tried just to play with Hetman, it ends as "testing/bughunting" session - instead of enjoying I'm all the time thinking, if what is happening is proper, what may be improved, what was unexpected or wrong, why it was that way etc... Kind of curse. :)

What's Dunnigan's works BTW?

Has anything come of this? Would I ever love to get my hands on this.

Someday... But you know, I'm single guy and day has only 24 hours... :)

Share this post


Link to post
Share on other sites
Hi Cadmium77, I'm glad you like it. Yep, also I like often just to see what is happening around. But anytime I'm tried just to play with Hetman, it ends as "testing/bughunting" session - instead of enjoying I'm all the time thinking, if what is happening is proper, what may be improved, what was unexpected or wrong, why it was that way etc... Kind of curse. :)

What's Dunnigan's works BTW?

Someday... But you know, I'm single guy and day has only 24 hours... :)

http://en.wikipedia.org/wiki/Jim_Dunnigan

Jim Dunnigan took over wargaming from it's primitve and rather puerile state from Avalon Games and turned it into SPI Games which were incredible, not just for their games but for the intense mathematical analysis behind them. Dunnigan himself was a mathematician. He also created Strategy and Tactics Magazine and ended up going off to work on wargames for the Pentagon.

In 1979, he wrote The Complete Wargames Handbook, and in 1980 How to Make War.

In addition to writing, Dunnigan is a principal in StrategyWorld.Com and the chief editor of StrategyPage.Com. Podcasts of his commentaries on history, military affairs, and the contemporary world are regularly posted on StrategyPage.Com and as at Instapundit.com

Dunnigan regularly lectures at military and academic institutions, such as the Chief of Naval Operations Strategic Studies Group, in Newport, Rhode Island.[3]

http://en.wikipedia.org/wiki/Strategy_%26_Tactics

S_and_T_68.jpg

The greatest pre computer wargame of all time imo was Dunnigan's Firefight which contained an amazing analysis of modern warfare in a ten page booklet. It was the work of a really inspired master. His thesis was that the Battle of Chinese Farm in the Yom Kippur war of 1973 meant that the battle tank was obsolete thanks to the first time use on a large scale in combat of TOW rockets

http://www.youtube.com/watch?v=9vlOzJhujjk#t=81

51EYCoKoMOL._SY300_.jpg

I owned this game and used to play it with a reserve army officer back in the mid 80's. An encounter that would take 30 min to play on a computer would require 8 hours of intense work and calculation of all the units and probabilities of hits etc with paper tables and cardboard counters on a hexagon map.

http://boardgamegeek.com/boardgame/8833/firefight

I used to dream of the impossible day when we could simulate warfare in 3d in computers. It's a miracle come true that we have these amazing machines and software.

Edited by Cadmium77

Share this post


Link to post
Share on other sites

Wow. Thanks a lot. This definitelly looks like something, I must to study despite language barrier.

Share this post


Link to post
Share on other sites

The knowledge on these forums! I love it. Thanks for sharing.

Share this post


Link to post
Share on other sites

Wow! Talk about having flashbacks! I would love to be able to get my hands on the Firefight game again.

I used to dream of the impossible day when we could simulate warfare in 3d in computers. It's a miracle come true that we have these amazing machines and software.

Me too! Although I must admit I'm a little dismayed it's taken *this* long to finally get to the point where we have things like Hetman, ALiVE, etc. I remember coding my first "AI" in the form of a simple, 2D strategy in space game, on the Apple IIe computers we had back in the 80s and thinking that real S&T wargames in 3D were only years away... not decades! ;)

Thanks for the memories!

Share this post


Link to post
Share on other sites
Wow! Talk about having flashbacks! I would love to be able to get my hands on the Firefight game again.

Me too! Although I must admit I'm a little dismayed it's taken *this* long to finally get to the point where we have things like Hetman, ALiVE, etc. I remember coding my first "AI" in the form of a simple, 2D strategy in space game, on the Apple IIe computers we had back in the 80s and thinking that real S&T wargames in 3D were only years away... not decades! ;)

Thanks for the memories!

If you go back and look at the links in my post you'll see it is for sale and trust me that 10 page booklet is well worth the cost of the game even if you never intend on playing it.

http://boardgamegeek.com/boardgame/8833/firefight

$17.99 brand new!

Also I sent this privately to Hetman but I see no harm in sharing it with you if you're interested in this;

http://www.google.ca/url?sa=t&rct=j&q=&esrc=s&source=web&cd=1&cad=rja&ved=0CC4QFjAA&url=http%3A%2F%2Floakes.game-host.org%2Flimeyyankgames%2Fgames%2Fresour~1%2Fcomple~1.pdf&ei=2irFUoyqKY67oQSE0IHwAw&usg=AFQjCNEeCVinscnz9vsx5rnzaj63cV8ptw&sig2=S9oWChrF_A5fQNWiyCyG1g&bvm=bv.58187178,d.cGU

And this is off topic but a friend just sent me this and this is very cool. May come in handy;

https://www.networkautomation.com/automate/public-beta/

AND the closest thing to Firefight in the computer world;

http://www.battlefront.com/index.php?option=com_content&task=blogcategory&id=114&Itemid=163

Created by Major Holridge of the US Army. Comes with a Military Reference library of manuals and books of recent military doctrines..

Edited by Cadmium77

Share this post


Link to post
Share on other sites

Reminds me of some of the old Avalon Hill games I used to play, before computers were in everyone's lives.

Ever heard of "The Book of Mars"? It was intended for paper and pencil RPG or perhaps lead figure combat but I cherish my copy. It was the first system I saw comprehensive enough to pit cavemen with sticks and stones against interstellar marines... and everything in between.

Sorry for OT :)

Share this post


Link to post
Share on other sites

Yeah. Boardgames. Main hobby of my childhood, when I could only dream about C-64 or PC (and I dreamed, oh yes). I think, my road to Hetman began then. I had big problem though - nearly no one else interested in range.

Share this post


Link to post
Share on other sites

I just created my biggest battle yet and added secondary objectives. I also got into heavy mobile artillery, 4 artillery vehicles per side. I've never seen so much smoke in a ARMA game presumably laid down by artillery for the most part. I never thought I'd see that outside of Steel Beasts Pro PE.. I've also never seen such a complex battle. This mod just keep getting better and better the more you use its toolkit. Secondary objectives really works well.

I do have one question though; when I'm playing on the side of LeaderHQ everything works well and the mod loads into the game right away and the AI commander starts giving orders.

But if I attempt to play on the side of the LeaderHQB or LeaderHQC or whatever then nothing happens, HAL doesn't announce himself and that side doesn't jump into action...they (the LeadershipHQB side) just sit there while they are quite active when I play the LeadershipHQ side

Is this something I"m doing wrong or is it meant to function this way?

Also, is there any way you can reset the trees so that armored vehicles can knock them over easier?

Edited by Cadmium77

Share this post


Link to post
Share on other sites
But if I attempt to play on the side of the LeaderHQB or LeaderHQC or whatever then nothing happens,

I noticed the same thing, and it was something that happened in HAC as well, for certain missions. My guesses are:

- You're playing as a SpecFor on LeaderHQB (They rarely get any missions right away)

- You forgot or mistyped RydHQB_Obj1 through 4.

- Not sure how this works in HAL, but you can try a shorter resettime for A and B, RydHQ_ResetTime = 60; I'm confused about the "cycles" , as read in the manual.

But its strange like you said that while playing on Leader-A , Leader-B will activate just fine, according to Debugs. It was similar in my case I will try to get a repro mission.

Another issue I want to discuss is the size of MASSIVE HAL battles. In one particular mission I had about 600 units under 2 Big Bosses. As the big bosses activated, I got froze on the "Big Boss studies the map" stage. I trimmed it down to 300 units and this stage passed. So I'm wondering, does Big Boss get stuck with just too much volume of units? or was there a particular unit (or unit config) that would break Big Boss study -stage? What happens exactly during the Big Boss studies the map stage?

Share this post


Link to post
Share on other sites
But if I attempt to play on the side of the LeaderHQB or LeaderHQC or whatever then nothing happens, HAL doesn't announce himself and that side doesn't jump into action...they (the LeadershipHQB side) just sit there while they are quite active when I play the LeadershipHQ side

Is this something I"m doing wrong or is it meant to function this way?

Is BB demo mission working well? As there you are under B commander. For me - all works fine for A, B, E... HAL is starting, init is going well, tasks for player TL are assigned... I guess, indeed, you did something wrong. So yes, repro mission is needed, I cannot reproduce the issue.

Another issue I want to discuss is the size of MASSIVE HAL battles. In one particular mission I had about 600 units under 2 Big Bosses. As the big bosses activated, I got froze on the "Big Boss studies the map" stage.

You mean during loadingscreen?

600 units is no big deal for BB, but quite a lot for CPU. Perhaps processor is choking? Personally never got any such problems too. Let's explain BB activities and cycles concept then...

1. Cycles. Cycle is "turn", where each Leader repeats his activities. There is stage of gathering data about current situation, stage of issuing orders basing on that analyze and stage of sleep (orders are conducted by groups indepedently).

If there is more than one Leader active, each subsequent is initially activated with 5 second delay after previous one. Then, after one-time initial steps, each at beginning of its each cycle will wait, untill all other leaders are in sleep stage, and only then will act. So at the time always no more, than one leader is issuing orders. It is done to avoid too many heavy script threads processed same time. There is however init variable to disable this waiting for all others sleeping, then each Leader will no longer pay attention, what else Leaders do, just will go in own rythm. Length of sleep stage depends on Leader's personality, amount of groups under his control and custom config. So order of executing cycles by all Leaders is not so easy to predict, but in general the further letter, the longer awaiting for first cycle, also very active Leader, that has not much groups under control can perform even 2-3 cycles, before slower opponents will get their turn. In debug mode each cycle is marked by globalchat message and RPT entry (morale, losses, personality).

2. Reset. It is independent stuff. It happens once per some time. Each reset Leader will consider again sending recon, will reconsider defensive positions, sometimes garrison stances, idle missions. Will also check then, if any objective was lost, and, if such option is active, will try to combine weakened, combat ineffective infantry groups, if near to each other.

3. BB initialization/cycle

Is done for A and then for B side, but some initial stuff is done only once, for A side, and omitted for BigBoss of B side:

- BB (A and B) awaits for all subordinated Leaders until they process to the certain point self-analyze to get data about forces of each. Then all Leaders of that BB side will wait with further action (their first cycles are paused) until BB end own initialization and issue them orders (BB will assign objectives or positions to defend).

- (only A) checks, where is battlefield center and basing on this is dividing whole battlefield to the many square sectors.

- (only A) now begins stage hidden behind loading screen, that allow to reduce vastly time of calculation done during this stage, as simulation is stopped, and all CPU is available for claculations (no simulations means any sleep or waitUntil for some simulation change will freeze script indefinitelly, so BB does not use any script pausing then). It is stage, where for each sector BB conducts topography analyze, saving per sector data as average trees coverage, hills presence, sea surface percentage, urban areas density, average terrain elevation differences (terrain shape). Same data is used by both BB sides, so it is done only once. Nothing is done with units, thus IMHO number of units has no direct impact on this stage and cannot make BB freeze then. CPU load is another story however.

- "Big Boss %1 is looking for strategic objectives." (A and B)

Knowing the battlefield BB can now to determine important spots, that will assign as objectives to the subordinated Leaders. It is based on locations and/or spots pointed by mission maker. There is constructed an array holding all found on battlefield strategic positions with their value and current status. Too close positions are merged.

All above is done once, at the beginning. Below stages are performed for A and B BB sides each BB cycle (which is completely indpendent from Leader's cycles) or also only once, if marked so.

- "Big Boss %1 is analyzing forces..."

At this stage BB gets complex, statistical data about forces of subordinated Leaders, is learning, what actually he has to his disposal.

- "Big Boss %1 is checking own forces placement..." (performed only once)

BB will think in terms of "flanks" and such, so he must recognize, where is center of his army, is defining main "attack axis" - overall offensive direction, to know, where he has front, where flanks, and where rear. At this stage also are determined previously found objectives, that should be considered as initially taken due to close own/hostile forces placement.

- "Big Boss %1 orients the flanks." (only once)

Now mentioned flanks, rear, front are determined. So each sector of battlefield now is assigned as belonging to left/right/central/rear part of battlefield.

- "Big Boss %1 assigns front sections to divisions." (only once)

Now topography analyze data is used to define dominant topography of each part of battlefield and choose, which Leader with his forces will get which part of battlefield (left flank/right flank/center/reserve). It is semi-randomized, but in general depends on Leader's forces composition. The more percentage of foot infantry, the more probable, that this Leader will be assigned to the part of battlefield, where BB found terrain most friendly for infantry/dangerous for mechanized groups. That means more woods, urban areas, hills... and vice versa - "division" of higher "mechanization" will more likely get more flat and open part of the battlefield. Nothing is sure though. If there is enough divisions, one (at least four subordinated Leader's) or even two (if BB of that side controls all eight Leaders) will be chosen as strategic reserve that will be used to reinforce threatened flanks and as primary garrisons pool (rarely used feature, as requires numerous enough infantry groups, too numerous for usual Arma missions).

- "Big Boss %1 issues orders."

Each BB cycle BB my now reassign flank assignment, if BB lost some division.

Depending on forces balance and losses BB will choose for his army defensive or aggresive stance. If stays offensive, each non-reserve "idle" leader gets own "portion"/chain of objectives to subsequent conquering taken from his part of battlefield, and only if there is no more objectives to take in this part - also from other parts. Now Leader is left with his task until done, then again become "idle"/ready for further instructions, but may be also on-the-fly redirected to support another division, if necessary (there is code for that, but frankly still not tested, I do not saw such situation yet), then put back to primary task - his chain of objectives to take.

- "Big Boss %1 will now take a moment to ash his cigar."

Same, as regular Leader, BB has at the end of cycle his sleeping time. What is not sleeping, is objectives monitor, that will "awake"/force to quicken next cycle BB of that side, if already taken BB objective is lost - it is considered as urgent situation that needs immediate handling.

Well, that's generally all as for BB activity. There are of course another details and secondary stuff.

--------------------------

Also, is there any way you can reset the trees so that armored vehicles can knock them over easier?

I would say, its config stuff, that I do not know. Unless new A3 commands setMass and setCenterOfMass have some meaning here.

Edited by Rydygier

Share this post


Link to post
Share on other sites
Let's explain BB activities and cycles concept then...

This is amazing. It must had taken weeks to think of all of this and implement it.

Top cheers to you.

But now as I understand it better, problems during the load screen only come from topography? Could objects placement, such as buildings, bunkers, fortifications, "Warfare" objects etc. perhaps have an effect on BB study of the map, causing the CPU choke?

I tried for hours to find the exact problem, but obviously repeated freezing of the computer is not something I wanted to go through 100 times.

As I mentioned, when I trimmed down each of my 6 LeaderHQ forces by about 80%, this map-study stage passed easily, making me to believe that something I deleted on the map allowed the CPU to process smoothly.

I will get back to the issue after some further testing.

Share this post


Link to post
Share on other sites

Please sign in to comment

You will be able to leave a comment after signing in



Sign In Now

×