So, After reading through the rather excellent battle reports by @TallGiraffe , I decided to attempt to create a Co-op campaign to try at my FLGS. In this, I attempted to address some core ideas:

1. The Campaign would begin fast, so multiple missions could be played in one night

2. The Campaign would allow players to play whatever they want

3. The Campaign would max out with games around 3 hours (Last Arc, with max players).

4. The Campaign would allow anyone to play, relying on an AI to "play" the enemies.

So, I dug in and attempted to create something that matched as many of these as possible. This is the result:

https://docs.google.com/document/d/1aht89dPpS1FAzViLoaix4-fPDHEAY64Be80OZ-aBeXE/edit?usp=sharing

Note that it is still in an early state, and may not be completely balanced yet. Please do point out any glaring issues or confusing wording so that I can try to fix it. Missions for this format are to follow, but I want to balance the core system a little bit first.

Looks cool. You know Heroes of the Aturi Cluster, right? Maybe you'd want to create a movement pattern closer to that system. This obviously can't include yaw values, but at least where the AI points its ship towards.

19 minutes ago, Stefan said:

Looks cool. You know Heroes of the Aturi Cluster, right? Maybe you'd want to create a movement pattern closer to that system. This obviously can't include yaw values, but at least where the AI points its ship towards.

I tried to with the Targeting and movement system, but that might not be random enough. As is, I wanted it to be more of a strike-type AI rather than totally random so that it doesn't do super dumb stuff (as in HOTAC they have turns to regroup, but with only 6 turns that isn't as possible). Or do you mean as in degrees from the front? Like side arc only, double Arc, Front arc only? Thanks for the feedback though, movement is definitely the thing I want to work on more.

My idea would be that the AI has a more sophisticated way of choosing targets and strategy rather than what amounts to a dice roll. Like:

- Check for range to nearest enemy in best arc. If in arc, resolve Concentrate fire. If not, bank token (Nav if far away/no other priority, Engineer if you don't have one, Squadron if Squadron value >2), or something like that.

- Move: Move towards said enemy if your strongest arc is front, or in roughly 45° angle if it's the side arc. Or something to that effect.

What I'm getting at is to give the AI a set of priorities that's easy to check and leads to useful maneuvers.

51 minutes ago, Stefan said:

My idea would be that the AI has a more sophisticated way of choosing targets and strategy rather than what amounts to a dice roll. Like:

- Check for range to nearest enemy in best arc. If in arc, resolve Concentrate fire. If not, bank token (Nav if far away/no other priority, Engineer if you don't have one, Squadron if Squadron value >2), or something like that.

- Move: Move towards said enemy if your strongest arc is front, or in roughly 45° angle if it's the side arc. Or something to that effect.

What I'm getting at is to give the AI a set of priorities that's easy to check and leads to useful maneuvers.

Ah! Okay, I the document with some changes. First, I made it way more likely to select aggressive if it hasn't taken a damage card, and put in the section that describes command dials being set. Basically, they will do a set order every turn of the game, regardless of circumstance (meaning they do mostly/all navigates). I think the idea of making it do a concentrate has merit, but I don't really see a way to do that without making it seem cheesy on large ships. As for moving, I updated the sections to make it clear that the AI will want the enemy in its best arc when being aggressive. The reason for some of it being random is so that the players can't just game the AI and make it do a certain thing - It always has the chance to do something else randomly. As for targeting, that happens at the beginning of the game, and isn't a specific ship, rather a class of ship. For example, a type of Nebulon has a 33% chance each of targeting an objective ship, the slowest ship, and the highest-health ship. While they may seem disparate, categories 2&3 are likely to be the same in the early missions (like this one!), with the possibility of instead going after the objective ship. Additionally, there will be ship that will target a set thing, without a die roll at the beginning of the mission. I can make all of the ships like that, and have them move towards their preferred target, but roll the most potent attack possible. Thanks again for the feedback! I will probably put the first mission in this once it is finished, just so you can see what some of the enemy ships actually look like.

I tried doing some form of AI system a while back. Haven't done much with it since, but I did do a little write up on the batrep subforum. It's probably several pages back by now. Might have some ideas on there to try.

edit: I cant make a link, but it's on page 6 at the moment towards the bottom titled "Battle test of an AI system" if you want to check it out.

2 hours ago, EagleScoutof007 said:

I tried doing some form of AI system a while back. Haven't done much with it since, but I did do a little write up on the batrep subforum. It's probably several pages back by now. Might have some ideas on there to try.

edit: I cant make a link, but it's on page 6 at the moment towards the bottom titled "Battle test of an AI system" if you want to check it out.

Thanks! I read up on it, and it seems that my system is close to the one suggested in that it still requires input, but is semi-random. I might try sketching out the basic maneuver profiles for a ship and making chart based off of that, but for know I think the more-input system will have to do.

HotAC has some great AI cheats in that really help, such as free Focus tokens. Maybe let the AI simply assume that it always has a Nav token?

My feeling is that it will be worst for the fast ships. A VSDII should not pose much problems (point towards enemy, spend Disposable Capacitors asap, fire, fire, fire).

38 minutes ago, Stefan said:

HotAC has some great AI cheats in that really help, such as free Focus tokens. Maybe let the AI simply assume that it always has a Nav token?

My feeling is that it will be worst for the fast ships. A VSDII should not pose much problems (point towards enemy, spend Disposable Capacitors asap, fire, fire, fire).

Add engine techs to the VSD too! XD

11 hours ago, Stefan said:

HotAC has some great AI cheats in that really help, such as free Focus tokens. Maybe let the AI simply assume that it always has a Nav token?

My feeling is that it will be worst for the fast ships. A VSDII should not pose much problems (point towards enemy, spend Disposable Capacitors asap, fire, fire, fire).

Yeah, I was thinking about this at work today. I think I will say that the AI always has the option of increasing/decreasing speed by one. This makes nav. charts reasonable. Things like Dcaps are easy triggers (can use, do use!). the question then becomes: do I still have a command dial system? and when does it spend a concentrate fire token? I think as of right now, I will focus on pushing out the first mission arc for Imperials then start on HOTAC-style nave charts for the rebel ships. As for the 16 different sectors, I'm thinking each arc is once sector, and each double arc is another sector. then, it determines range to be either close-medium(closing) or medium (fleeing) to long.

New(er) plan: Ships can temporarily increase or decrease their speed every activation. Ships with speed 4 may use a navigate command to move from speed 2 to speed 3. this means those ships will have two options (slow, and fast) at speed 1 or 4. Affected ships: MC30, CR-90, Raider. Ships will set the same dials for the entire game, unless forced to do something else by circumstance. For example, a Hammerhead will always concentrate, and a Quasar will always Squadron. Means that there is one less thing to be determined every round.

Sounds good!

Updated with the first two missions, and you can comment on the document with questions and ideas! Also, a teaser for the third mission because I am too lazy to not CTRL+A from my word document. Project: Make a good AI is cancelled because I don't have enough time right now, and am lazy. Maybe some other time. Link again:

https://docs.google.com/document/d/1aht89dPpS1FAzViLoaix4-fPDHEAY64Be80OZ-aBeXE/edit?usp=sharing