• Hello and welcome to MSFC. We are a small and close knitted community who specialises in modding the game Star Trek Armada 2 and the Fleet Operations modification, however we have an open field for discussing a number of topics including movies, real life events and everything in-between.

    Being such a close community, we do have some restrictions, including all users required to be registered before being able to post as well as all members requiring to have participated in the community for sometime before being able to download our modding files to name the main ones. This is done for both the protection of our members and to encourage new members to get involved with the community. We also require all new registrations to first be authorised by an Administrator and to also have an active and confirmed email account.

    We have a policy of fairness and a non harassment environment, with the staff quick to act on the rare occasion of when this policy is breached. Feel free to register and join our community.

A2 AI Modding

Rifraf

I know just enough to be a danger to myself
Joined
25 Aug 2013
Messages
1,205
Age
51
I'm not sure how many of you actually play A2 anymore, but this should be applicable to FO as well? I started a thread on AI over at FF and FO last year/this year respectively and Freyr and Dominus_Noctis were very helpful in talking about how the AI works and such. Earlier this year I PM'd with Freyr quite a bit and put together a redone AI in hopes of continuing that aspect of any previous work. I thought I'd share the results of my testing if anyone is interested. I did release some new AIP files over at FO and can do so here if anyone would like to try it. Some feed back would be nice.

The values listed are from stock Armada 2, but I did my actual testing on my personal build and continue to refine the values. By watching the minimap in the debugger and lining up freighters with the grid overlay and how they corresponded to the colored squares of the threat, goal, explore, and death maps I was able to determine what appears to be happening.

So, each gameplay map grid is 100x100 units or a 1x1 grid as evidenced by the grid overlay on screen (alt-G) and sensor/weapon ranges are bound to this.

Each strategic map grid is 400x400 units or a 4x4 grid as evidenced by AI-Grid-Size=400 in the RTS-CFG under the strategic AI section.

The strategic-ai-cfg.txt file in the AI folder contains all the strategic parameters:
// How many tiles per threat mapgrid?
#define threat_map_resolution 800

// How many tiles per death mapgrid?
#define death_map_resolution 800

// How many tiles per our empire mapgrid?
#define empire_map_resolution 800

// How many tiles per enemy empire mapgrid?
#define enemy_empire_map_resolution 800

// How many tiles per exploration cell?
#define exploration_map_resolution 800

// How many tiles per goal cell?
#define goal_map_resolution 800

// Iteration for relaxing grid borders
// NOTE: DO NOT CHANGE
//
int relaxation_cycles = 1;

// Relaxation bleedover
// NOTE: DO NOT CHANGE
//
double relaxation_coefficient = 0.5;

Every ship/station with an "attackpower=XX" value in its odf (anything with a weapon should have it) projects an initial threat of 800x800 units or an 8x8 grid as specified by #define threat_map_resolution 800. This is 800x800 units total from its center point not in each direction from its center.

The "int relaxation-cycles=1" setting controls how many strategic map grids to spill the threat over into. So a setting of 1 would mean the threat gets spilled over into 1 strategic grid (400x400 units or 4x4 grid) in each direction from the edge of the initial 8x8 threat grid. 2 would be 2 strategic grids each direction and so on.

The "double relaxation_coefficient = 0.5;" is the percentage of the threat that gets spilled over into other strategic grids. So .5 is 50 percent and 1 is 100 percent. I feel this should be 1 or at least above 90% so that all the threat gets spilled over into the next grid to help the AI better judge where the danger is.

Now the relaxation-cycles and the relaxation-coefficient parameters each appear in the all the AIP's in the AIPs folder so you can further fine tune each AIP separately.

There is also this:
// How much to reduce danger recorded in grid cell each cycle
// NOTE: DO NOT CHANGE
//
double danger_diminishment = 1;

There's no way to test that but I think it means that 100 percent of any threat is reduced each cycle which to me seems pointless. Why reduce the threat completely and so quickly. I've changed mine to .05 or 5% of the threat gets reduced each cycle to keep the threat around a little bit longer. May help the AI play better and not just rush into danger head on so quickly.

When the AI loses a unit it references the death-map-resolution and paints an area of danger from the grid it lost its unit in of 800x800 units. 800 is a bit large to me depending on weapong ranges.

The exploration-map-resolution is the size of the area it considers explored. 800 is a bit large in my opionion unless you have your sensor range at 800.

The empire and enemy-empire-map-resolution is the size of your and your enemies empire. I feel you would want this to be a little bit bigger than your total threat area to emphasize your region of control and then beyond that is open space or the perimeter.

The goal-map-resolution corresponds to the strategic goals. This you can watch in the minimap in the debugger as it's the first mode to choose. You can see the areas the AI chooses to explore and what it assigns to protect and where it marks resources it finds.

I can change the strategic-ai-grid size to any number (I went as low as 100) and the game still played. However, the threat-map-resolution has to be a minimum of 400 or the maps start to crash (in stock A2.) And the threat-map-res and goal-map-res values must be linked somehow because even if they are different values the strategic grids for those are always the same size and it's the threat-map-res setting that determines that size.

Also, the empire-map-res and enemy-empire-map-res values don't seem to be affected by the relaxation-cycles setting or threat spill over. The value set for each (and I think they should both be the same size) as represented in the strategic map view is the size that it appears on screen.

For your star base no matter what the threat map is set at the threat map grid is never centered on your base like one would think it'd be which kind of sucks. So one or two sides of your base might project a threat 4-10 grids out but the other two sides would be left with the remainder of 10-18 grids out which kind of skews the threat locations slightly depending on your threat-map-resolution value.

One of the most important things that needs mentioning (if you're modding stock A2) is that your attackpower = XX value in each ship/station odf (that have weapons) should have a minimum value of 1.0 as Freyr suggested years ago in his AI work. I see FO as of 3.2.6 has this, but not sure of earlier versions. The reasoning is that it's used to calculate how the AI determines attack parameters and anything less than 1 is actually dividing its outcome which is less than optimal. Most people probably already know this, but thought it was worth mentioning.

Regards, Rifraf
 

MrVulcan

Crewman 2nd Class
Joined
17 Aug 2011
Messages
142
Great explanation Rifraf, cheers :)

I've seen a paper by Dr. Ian Davis on path planning in Arnada, which is largely based on quadrees (split the space into halves up to a maximum resolution to map out the empty space). Helps with memory saving, and faster computation, but does introduce some unusual artifacts in that things don't always go in straight lines when possible. This was caused by the quadree divisions not being aligned with stuff on the map.

A wild guess might be that the AI maps (grids) are produced in a similar fashion to pathing quadrees, so stuff on the physical map isn't centered / aligned with the threat maps and so on. If so, there would be no way to fix this problem besides raising the resolutions and increasing the number of spillover tiles.

I like the fact that when it says DO NOT CHANGE, you went ahead and changed it. Heheh

Great job :)
 

Rifraf

I know just enough to be a danger to myself
Joined
25 Aug 2013
Messages
1,205
Age
51
Thanks Mr Vulcan. Freyr may have privately experimented with the settings that say do no change when he was working on the AI, but I'm not aware if he ever shared that with anyone. I thought, what the hey and changed them anyway. :) Sharing PM's with Freyr for a while it seems clear that the programmers didn't have a lot of time to flesh out the finer details to make the AI work better. Freyr really changed that and I was just hoping to improve on it if possible. I'm not sure it it'd be allowed, but I can post my zipped AI files here if anyone wants to try them out?
 

Rifraf

I know just enough to be a danger to myself
Joined
25 Aug 2013
Messages
1,205
Age
51
I'm doing some more testing with the AI and wanted to get some opinions from others. I know, the AI is a bit hopeless, but any tiny bit helps.

So if intrinsicValue determines where to attack and attackPower determines what to attack once there Freyr always advocated making valuable targets like shipyards, mining/orbital/research stations of strategic value have high intrinsicValue and ships used for tactical value have none.

Of course you can make the values whatever you want, but I'm setting up more testing with research/orbital/mining stations with a value od 2 and shipyards with a value of 1.5. No, not much difference I know, but would others just make them all 2? They're all important and would cripple your enemy in meaningful ways.
 

Terra_Inc

MSFC's Cheshire Cat
Staff member
Site Manager
Necromancer/Troll hunter
Kitten Commander
Joined
16 Dec 2009
Messages
3,121
Age
34
I'm doing some more testing with the AI and wanted to get some opinions from others. I know, the AI is a bit hopeless, but any tiny bit helps.

So if intrinsicValue determines where to attack and attackPower determines what to attack once there Freyr always advocated making valuable targets like shipyards, mining/orbital/research stations of strategic value have high intrinsicValue and ships used for tactical value have none.

Of course you can make the values whatever you want, but I'm setting up more testing with research/orbital/mining stations with a value od 2 and shipyards with a value of 1.5. No, not much difference I know, but would others just make them all 2? They're all important and would cripple your enemy in meaningful ways.
I do think mining and orbital stations in particular should attract the AI's attention. They're often farther away from the base and undefended (if the player is not careful), so attacking them could hurt the player quite a bit without doing a risky base assault.
Research stations IMHO should have a higher priority than shipyards, because most players will have one research station but many shipyards. So unless you can spend a lot of time under concentrated fire from the starbase and/or defensive forces to take out all (or the majority) of the shipyards, destroying the research station is much more crippling than going for the shipyards.
 

Rifraf

I know just enough to be a danger to myself
Joined
25 Aug 2013
Messages
1,205
Age
51
I appreciate your opinions Terra. I agree. For testing I think I'll do IV of 2.25 for mining/orbitals, 1.75 for research stations and research pods and 1.25 for shipyards.
 

Rifraf

I know just enough to be a danger to myself
Joined
25 Aug 2013
Messages
1,205
Age
51
I'm starting to get back into modding A2/FO and wanted to continue work on the AI again. Some may say give it up, but I like tinkering with it to see what else can be discovered. Freyr may have squeezed every last ounce out of it, but it keeps me busy and I enjoy it.

I've set up the base game with the 1.1 patch then made all the necessary edits to get a baseline. Then I'm rewriting the build lists so the AI produces ships in a more logical manner, gets better ships faster and moves up the tech sooner and finally changing the AI personality files.

Here's the baseline:

---Testing parameters: stock A2 with 1.1 patch applied---

In game settings at main menu set game difficulty to hard and adjust AI cost handicap and

build handicap to hard
--------------------------------------------------------------------------------------------

**NEED TO COMPLETELY REDO ALL BUILD LISTS TO BUILD SMARTER**--IN PROGRESS


Added to craft.odf and station.odf: sanecrewlossmethod = 1 and shieldprotection = 1

In craft.odf and station.odf changed shieldDelay = 3.0f to 1.0f to assist the AI with
transporting crew right in the middle of battle and getting blown to bits

attackPower and instrinsicValue settings:

attackPower:

assault, battle, battle2, cruise1, cruise2, cruise3, destroy, frigate, saucer, scoutBase, and
special (These base ODF's all have attackPower=1.00)--DONE

ships w/ weapons (combat ships) = 1.00f (kcolony, kconst, kcargo, kfreight = 0.00f)--DONE

stations w/ weapons (bases, turrets, 8472Sentinel, 8472Mine, 8472Mother, 8472HeroMother,
cquantum) = 1.00f--DONE

intrinsicValue:

ships w/ weapons (combat ships) = 0.00f--DONE

bspecial, cspecial, fspecial, kspecial, rspecial = 0.50f (left stock)--DONE

csuper, ksuper, rsuper = 0.50f (left stock)--DONE

const.odf, freight.odf = 1.00f--DONE

cargo.odf = 0.50f (left stock)--DONE

colony.odf = 0.75f (left stock)--DONE

Upgrade Research Center = 2.00f--DONE

Mining Stations and 8427Transmuter = 2.00f--DONE

Orbital Stations = 2.00f--DONE

Advanced Research Stations = 1.75f (left stock)--DONE

Advanced Shipyards and 8472_FluidicGate = 1.50f--DONE

Borg Portal = 1.50f (left stock)--DONE

Basic Research Stations = 1.25f--DONE

Basic Shipyards = 1.00f--DONE

Trading Stations, Borg Recycler and Technology Assimilator = 0.75f--DONE

pod.odf = 1.75f (left stock)--DONE

bsuperbl, fsuperbl, ksuperbl and rsuperbl = 1.50f (left stock)--DONE

sensor stations = 0.50f--DONE

left weapons on kconst, kfreight, ksrepair, kcolony, kcargo, but changed combat=1
and can_sandd=1 all to 0 so they don't try to search and destroy or guard--DONE

changed construction time of 8472_cruise1 and 8472_cruise2 from 80 to 40--DONE

changed health of 8472_mother from 2500 to 3500--DONE

Added photon back to bcruise1 and added phaser back to bcruise2--DONE

Redid build hardpoints of all upgrade stations so they can research all items on separate
pods (cupgrade and rupgrade appear to have invalide hardpoints or many that show in
milkshape, but aren't selectable??)--DONE

--To keep all races the same for baseling testing--
changed retreatStrenght to 0.00 in each races odf file--DONE
Changed crewRetreatRatio to 0.35 in each races odf file--DONE

Added 600 to normalCrew in each races odf file for 1800 total because each race recrews their
starbase to 750 at the start of the game, the minimum normal operating green alert level.--DONE

Build list alterations (Will completely redo build lists to build smarter):
Moved scout to be built after the main shipyard for borg, card, fed, kling, and rom. (In
stock if the scout gets destroyed before the race can build a ship yard then it goes into
it's defensive AIP and defensive buildlist, but is then stuck because it can't build a scout
first before the yard gets built)--DONE

SINGLE_PLAYER_OFFICER_LIMIT = 750 -from 600--DONE

Changed planet constants for colonizing from 1000, 2000, 3000, 5000 to 1000, 1250, 1500, 1750
so we get crew faster and minimize the human colonizing a planet more than once since the AI can't do that --Done

--RTS_CFG changes:--

HARD_DAMAGE = 1.0 -from 2.0--DONE
EASY_MINING_MODIFIER = 1.5 -from 2.0--DONE
HARD_MINING_MODIFIER = 1.0 -from 0.5--DONE
HARD_AI_COST_HANDICAP = 1.0 -from 0.6--DONE
HARD_AI_SPEED_HANDICAP = 1.0 -from 0.7--DONE
HARD_AI_CLOAK_PERCENT = 25 -from 75--DONE

cfg_MIN_MOVE_DISTANCE_FOR_CLOAK = 100 -from 1500 (will test further)--DONE

cfg_MIN_MOVE_WARP_DISTANCE = 1100 (left stock)(will test further)--DONE

SHIELD_DAMAGE_PROTECTION = 1.0 -from 0.8--DONE

cfgTOO-CLOSE_TO_WARP = 100 -from 800 (will test further)--DONE


I know everyone is busy with life and their own works, but wanted to see if anyone would be interested in play testing with me? I'll provide all the necessary edited files. If not no biggie. I know the prime of A2 has passed. You'd be included in any credits should anything come to fruition.
 

Terra_Inc

MSFC's Cheshire Cat
Staff member
Site Manager
Necromancer/Troll hunter
Kitten Commander
Joined
16 Dec 2009
Messages
3,121
Age
34
I know everyone is busy with life and their own works, but wanted to see if anyone would be interested in play testing with me? I'll provide all the necessary edited files. If not no biggie. I know the prime of A2 has passed. You'd be included in any credits should anything come to fruition.

I'm so tempted to play test, but I'm really really busy atm. Perhaps in a week or so? I guess I'll have a bit more time then.
 

Rifraf

I know just enough to be a danger to myself
Joined
25 Aug 2013
Messages
1,205
Age
51
Like I said, no worries if you don't. Just throwing it out there, but do appreciate it. I'll get some files put together and PM you once ready.

Everyone knows now that simply rewriting the stock build lists in a more logical manner will make the AI play better, but I want to see if it can be made a teeny bit smarter and surprise you when you play. Always looking for that one thing to try with the game :)
 

Atlantis

Master Chief Petty Officer of Starfleet
Joined
1 May 2006
Messages
518
Well it all appears well thought-out and all makes sense, so it's bound to be an improvement.
 

Rifraf

I know just enough to be a danger to myself
Joined
25 Aug 2013
Messages
1,205
Age
51
Well I began my testing and got mixed results. Tested on Wide Open with all 6 races. I was Roms. First the Feds then Cardies got knocked back into their starting bases and didn't have enough dilithium to build a construction ship so they just sat there. 8472, Borg and Klinks kept building what they could as dilithium or crew built up and only really concentrated on me so feel that proves "int human_target_raw_bonus = 1500" in the AIP files works because they never finished off the Feds or Cardies.

Having only exhaustable dilithium moons really limits play. All the races were in their "very_aggressive AIP and "endgame" build list which is strange as you'd think if the Feds and Cardies got beat back into their starting base they'd go back into their "defensive" AIP. I think once the AI gets into it's "endgame" AIP it stays there no matter what?

Playing with this: "double resource_distance_multiplier = 1000.0;" in the AIP's and setting it higher should make the AI branch out a bit to build dilithium mines further away where the inexhaustable dilithium moons are.

I started testing with the following settings and remmed out all the int max settings:

int min_escort_force = 1;
int min_perimeter_force = 1;
int min_defense_force = 1;
int min_exploration_force = 1;
int min_attack_force = 2;
int min_resource_force = 1;

I did this with the thought that the AI would choose how many ships to commit to attack, or the perimeter or defense etc. based on other events events. Most people make these settings higher like attack_force 8 to get the AI to attack with 8 ships minimum. In my testing with a value of only 1 or 2 it still attacked in fleets of many ships, defended with many at once etc. It would occasionally send one ship here or there though which I'm not sure why?

I also have turrets set at no intrinsic value yet a lone ship would wander in to attack one instead of a research pod or something of higher value. More testing needed. As it is the AI starts strong, builds up forces quickly however, the end game is weak and it can't steam roll me. I'll have to test different values and move things around. Any ideas or thoughts welcome.
 

Rifraf

I know just enough to be a danger to myself
Joined
25 Aug 2013
Messages
1,205
Age
51
I thought I'd post some more stuff I've been testing in hopes that anyone finds it useful. Where to begin?

Species 8472. Most probably already know this, but as they are implemented in stock A2 the AI doesn't properly assess the mother, heromother or transmuter on the threat map. From the FO guide:
First Mode: Strategic Goals

Higher priority goals are brighter.

  • Purple - Explore
  • Red - Attack Base (Craft that have is_starbase = 1)
  • Yellow - Attack Vessels
  • Green - Defend Base (Craft that have is_starbase = 1)
  • Blue - Resource
Stations all have "is_starbase=1" in their odf or the odf as part of #include. This is how the AI determines which bases to attack and which to defend. However, the mother (other races have a starbase) and the transmuter (other races have a mining station) are actually ships with some station properties and specifically have "is_starbase=0" and "facility=0"

Don't get me wrong they'll still be attacked, but if you use the debugger tool while playing a game you'll see that these two units are not displayed properly on the threat map compared to all the humanoid races and so are not assessed as a threat to the degree that other races stations are.

Next is personality settings. I'm still tinkering with these and the strategic_AI_config and RTS_cfg files. I feel I've made progress, but the one area I'm really trying to improve is keep the AI out of danger, but it essentially had no sense of self-preservation. Oh well, more testing.

What go me more interested in it is getting my A1 disk and seeing how it was set up. The AI in A2 is not very different from A1 I'm using that as inspiration. What I'm focusing on still is how the AI determines a threat and at what range or more to the point how to use it to its advantage.

I've attached a picture of an in-game shot of the debugger to show some of the things you can see and what I'm testing. Image 001 shows the quadtree visible as well. More to come.
 

Attachments

  • Snapshot000.jpg
    Snapshot000.jpg
    279.9 KB · Views: 13
  • Snapshot001.jpg
    Snapshot001.jpg
    295.6 KB · Views: 11

Rifraf

I know just enough to be a danger to myself
Joined
25 Aug 2013
Messages
1,205
Age
51
I had one of the best games last night so far in my testing. All the AI's attacked each other in fleets over and over which is better than how stock plays. I'm trying to figure out a couple mysteries though.

Over at Starbase 34 Freyr still has his AI notes up from his testing years ago and he posited that the AI enters the exploration AIP and instant_action_buildlist either after discovering an enemy base or it's scout getting killed. I'm not totally sure either of those are the answer. Reason I say is because I've been noticing the AI goes into the exploration AIP and requisite built list several times during a game. If the AI already knows where the enemy is at or has scouted pretty much the whole map why go back into that AIP to explore?

Also, occasionally in some games one or two of the AI's will not immediately start scouting the map upon game start. This is usually Klingon or Cardassian though occasionally 8472 as well. I'm not sure why a couple races would not start scouting until later in the game? Interesting behavior.

Obviously the AI goes into it's defensive AIP and defensive buildlist when it's beaten back into its base. This is where I think the following comes into play from the strategic_AI_config file in the AIPs file:
// How many tiles per our empire mapgrid?
#define empire_map_resolution 800

If you have operations (I'm not sure if this is only mining/orbital stations or ships too?) outside of that value then the game considers you not beaten back. I'm experimenting with a larger value to make your home base bigger. At least that's my thinking.

One thing to add though is that the thinking has been when an enemy gets beaten back into it's base it goes into its defensive AIP/buildlist and the attacker goes into its endgame AIP and build list however, I continually see all AI"s in the endgame AIP and buildlist with no one in their defensive AIP and buildlist so something else must be at play that I'm not understanding as of yet?Almost as if things are set on some kind of internal timer or something? Anyway, it's been fun testing and I'll keep posting as I find more.
 

Beta_1701

Cadet Sophomore
Joined
13 Jan 2015
Messages
31
Age
33
Hopefully you don't mind me asking this but I had a more basic question about AI, but I didn't think it deserved a whole threat nor could I find anything in the guides over at FO or on the forums.


Basically I've change my federation tech tree so the shipyards build pod bays which inter build ships. I though it was a good idea as it cuts down on the the need to have loads of shipyards. So I edited the AI, by simply copying the line for each occurance of shipyard and copying just below it.

e.g.

"fyard", 1,0, -1, -1,
"fyard_a", 1,0, -1, -1,
"fyard_b", 1,0, -1, -1,
"fyard_c", 1,0, -1, -1,

But the AI just sits there...I only recently made the lvl 2 yard poded but its never worked. Unless I miss understood what I'm doing? I though a pod-yard worked because unless I'm mistaken fleetops uses them? (although I can't be sure).

Any idea where I'm going wrong?
 

Rifraf

I know just enough to be a danger to myself
Joined
25 Aug 2013
Messages
1,205
Age
51
Is this for stock A2 or FO? The only thing that really stands out is the last number in each sequence. The -1. I don't know if FO uses tech levels, but normally your yards as you have them will build up to tech level 1. After you reach tech level 2 if you've set the -1 means those yards will no longer build.

In the techtree you've set each successive yard requiring the one before?
 

Beta_1701

Cadet Sophomore
Joined
13 Jan 2015
Messages
31
Age
33
Is this for stock A2 or FO? The only thing that really stands out is the last number in each sequence. The -1. I don't know if FO uses tech levels, but normally your yards as you have them will build up to tech level 1. After you reach tech level 2 if you've set the -1 means those yards will no longer build.

In the techtree you've set each successive yard requiring the one before?

This if for your upgrade project mod in fact. Also got to say now If I'd put two and two together before I'd have said your mod is amazing!

Anyway back to the point at hand.

I've tried the techtree like this:

fyard.odf 1 fbase.odf // shipyard
fyard_a.odf 1 fyard.odf //shipyard pods
fyard_b.odf 1 fyard.odf //shipyard pods
fyard_c.odf 1 fyard.odf //shipyard pods
fyard2.odf 2 fyard.odf fresear.odf // advanced shipyard
fyard2_b.odf 1 fyard2.odf //advanced shipyard pods
fyard2_a.odf 1 fyard2.odf //advanced shipyard pods

And as
fyard.odf 1 fbase.odf // shipyard
fyard_a.odf 1 fyard.odf //shipyard pods
fyard_b.odf 1 fyard_a.odf //shipyard pods
fyard_c.odf 1 fyard_b.odf //shipyard pods
fyard2.odf 2 fyard.odf fresear.odf // advanced shipyard
fyard2_b.odf 1 fyard2.odf //advanced shipyard pods
fyard2_a.odf 1 fyard2_b.odf //advanced shipyard pods

In both cases the AI (in standard tech) build the fyard and fyard2 (and all the stations in between) but not the pod. It then stalls and does nothing.

In freetech the AI just sits there...
 

Terra_Inc

MSFC's Cheshire Cat
Staff member
Site Manager
Necromancer/Troll hunter
Kitten Commander
Joined
16 Dec 2009
Messages
3,121
Age
34
It's been a while since I've done anything A2-related - never mind its AI - but it might help to turn on the AI debug mode. It shows how the AI works its way through the build list, and can usually help to find out what's going wrong. You can activate it by pressing Alt-Shift-A, and cycle through the list of players with Ctrl-Up and Ctrl-Down.
 

Beta_1701

Cadet Sophomore
Joined
13 Jan 2015
Messages
31
Age
33
Thanks Terra, that seems to have answer some questions at least...so in AI stalls in freetech as it's asked to build fyard_a and tehcnology not available and the same is the case for standard tech but it manages to build the fyard first before stalling.

I also want to correct the AI tech I quoted before

Hopefully you don't mind me asking this but I
"fyard", 1,0, -1, -1,
"fyard_a", 1,0, -1, -1,
"fyard_b", 1,0, -1, -1,
"fyard_c", 1,0, -1, -1,

It should have been

"fyard", 1, 0, -1, 0, // Basic shipyard
"fyard_a", 1, 0, -1, 0, // Basic shipyard
"fyard_b", 1, 0, -1, 0, // Basic shipyard
"fyard_c", 1, 0, -1, 0, // Basic shipyard
 

Terra_Inc

MSFC's Cheshire Cat
Staff member
Site Manager
Necromancer/Troll hunter
Kitten Commander
Joined
16 Dec 2009
Messages
3,121
Age
34
Interesting. Just to make sure, did you add the pods to all the relevant techtree files?

Another thing you can try is to just remove the dependency from the techtree and see if it builds then. It might help determine what sort of problem we're dealing with here.
 

Beta_1701

Cadet Sophomore
Joined
13 Jan 2015
Messages
31
Age
33
The pods are present in both fulltech and tech1.tt, do they need to be in anothers? Also I've tried removing the techtree dependance and the same thing happens....
 

Terradyhne

Petty Officer 1st Class
Joined
19 Jan 2011
Messages
253
The pods are present in both fulltech and tech1.tt, do they need to be in anothers? Also I've tried removing the techtree dependance and the same thing happens....

Are the pods and the HP's to build them on, defined in the Shipyards odf ?
 

Beta_1701

Cadet Sophomore
Joined
13 Jan 2015
Messages
31
Age
33
Yep, maybe it something in the odf the AI doesn't like?

I'll attached the .odf and you can passs judgement
 

Attachments

  • stations.rar
    8.7 KB · Views: 1

Terradyhne

Petty Officer 1st Class
Joined
19 Jan 2011
Messages
253
Yep, maybe it something in the odf the AI doesn't like?

I'll attached the .odf and you can passs judgement

Did you try it without the builder_ship = 1 option in the Fyard and Fyard2 odf-files ? All other things in the odfs seem to be corret.
 

Beta_1701

Cadet Sophomore
Joined
13 Jan 2015
Messages
31
Age
33
Did you try it without the builder_ship = 1 option in the Fyard and Fyard2 odf-files ? All other things in the odfs seem to be corret.

Sorry to repor that didn't work either, and it made the fyard and fyard2 unable to build anything for me...I'm open to any other ideas...
 

Rifraf

I know just enough to be a danger to myself
Joined
25 Aug 2013
Messages
1,205
Age
51
As I continue to work with the AI and build lists to attempt to get a more challenging, "smarter" AI I've played over 70 games in testing so far. I was wondering if anyone had any interest in playing with my edited files to get opinions?

This is would be for stock A2. The AI files will work for stock A2, modded A2 and/or FO. It's the build lists that would be different depending on what ships you have in your game. Also build lists written for a map with 2 planets and 4 dilithium moons would/should be very different than a map with 3+ planets and more than 4 dilithium moons, etc because of the available resources. Your build list should be written with the available resources in mind.

I continue to test and tidy up the files, but if anyone is interested just let me know. Or maybe I'll just post the files for download?
 

Archonon

Master Chief Petty Officer of Starfleet
Joined
13 Aug 2016
Messages
539
Age
46
Hi Rifraf,

I would be more than happy to help you test your AI, I'm not sure how it will work with my current game.

I have tried to leave stock race tech trees intact, instead I just add new tech that they can build with mostly their own separate units. There are some exceptions of course, fbase for example builds fconst2 which has the new units. But I haven't tested the campaign to make sure that the stock units build without problems (They should...). The only changes I have done to stock units is cosmetic, upgrading stock models for better versions made by the community. Again, I don't know if this would affect your AIPs in any way.
 

Rifraf

I know just enough to be a danger to myself
Joined
25 Aug 2013
Messages
1,205
Age
51
I'm getting back to my stock Armada 2 AI testing, I was on a roll early last year having played over 60 games in testing and I've recently come across some other resources related to the game engine. I found mention of and AIP manual and found it listed on an Auran website, but the link is broken.

I found it listed on a Game Front download site, but it says it's unsafe so didn't try to download it. I'm asking if anyone might happen to have this AIP manual or the Tactics Engine manual. Probably a long shot, but thought I'd check. I'm not sure if it'll be helpful, but could provide some useful info.

Most probably know, but Battlezone, Dark Reign and Civ 3 use versions of the engine and many of the AIP entries appear in the AI code of those games so I've been trying to dig up more info.
 

Rifraf

I know just enough to be a danger to myself
Joined
25 Aug 2013
Messages
1,205
Age
51
Sorry, time expired to edit post. I should specify it's the AIP manual and Tactics Engine manual for Dark Reign.
 

CABAL

<< ■ II ▶ >>
Staff member
Administrator
Star Navigator
Rogue AI technocrat
Joined
15 Aug 2009
Messages
3,511
Age
32
Here you go! Found them buried in the 1.7 fan patch for Dark Reign.
 

Attachments

  • DR_AIP&TE_Manuals.zip
    1.7 MB · Views: 4

Rifraf

I know just enough to be a danger to myself
Joined
25 Aug 2013
Messages
1,205
Age
51
Thank you so much Cabal.
 
Top