I will print the rules and letters this weekend to see how it works. Has anyone made any videos of how it works?

Not to raise a "dead" thread, but with my state on stay-at-home orders for the fifth week running, I finally decided to give the @Crabbok AI system a try. I've missed this game!

Like many have commented, I did make a lot of small minor revisions and adjustments to the rules, based both on need and preference. For example, I didn't have an opportunity to actually print out the AI cards, so I used a random number generator to determine what ship activated and what command card was then assigned to it (for example, between 1-6 for activating a ship, between 1-12 for choosing a card). On the whole, I found the system to be fun and surprising. It certainly doesn't replace playing against another person, but for those times when that isn't possible, it's not bad!

Some additional thoughts:

1. Thematically, I think the random card selection works better on ships with higher command values, as it somewhat replicates the experience of having the wrong command dial at the wrong time. My AI fleet was mostly made up of corvettes (four out of six ships), and it just didn't seem right for them to be "choosing" the "wrong" command so frequently, being that a ship with a "1" command value can always choose the exact command it needs in that moment.
2. Finding the balance between adhering to the AI system and overriding it when it resulted in an action that no human opponent would ever do was tricky. But then, this system isn't supposed to be as good as a human. Still, sometimes I did override the AI to take an action that made more sense.
3. I mostly ignored the "one defense token" rule and instead took whatever actions would best protect the defender.
4. In my matchup, the AI system was routed pretty easily. I'd be interested to hear if anyone has yet lost to the AI. The AI opponent might deserve an even bigger handicap. A lot of synergies that would work off of each other with a human opponent don't seem to work as well with the AI opponent.
5. Finally, the game was fun! Thank you for creating this system, Crabbok! I think modifying it with some house rules might be the way to go for each person, as in my experience, it still did require some intervention on the part of the player (i.e., Me)

Here's a gratuitous photo of my match-up: The First Order takes on the lowly Resistance above Bespin! (You'll notice the Rebel small ships are kind of all over the place.)

Rmcarrier1,

My experiences are very similar to yours. I do not have anyone in my area who plays Armada, but I love the game, so I started to use @Crabbok AI system.

I found it a good starting place, but like you, made my own house rules.

I use a 12 sided dice to determine which card to use. This has the same affect as your random number generator.

I use the AI cards for target priority, special rules, and move priority. When it comes to command dials, it depends on the ship. If I don't care, I use the AI card. But if a ship is better always doing something, then I make a special rule for that ship. For example, VSDs should use Navigate and Interdictors should mostly use Engineering.

The defense token use was an issue. My AI ships committed suicide too easily. So borrowed a method from DnD called Skill check.

Basically, I would roll for each appropriate Defense dial, except for Contain. My rule is that for Green defense dials, there should be a 50% chance of using it. Orange should have a 25% chance. This allows for the effect of the AI "choosing" not to use a defense token like a human would. For a 12 sided dice, a green token would be used for a roll greater than 6 and a Orange for a roll greater than 9.

Regarding Contain, I always invoke it when it can cancel a Critical result. (until it is gone.)

I do have an exception to this, which is my self preservation rule. If the incoming damage is greater than half the remaining HULL points on the defending AI ship, then that ship must use a defense token(s) to cancel the most dice.

This has created some interesting battles. I mostly win, but there have been a few setup that were stacked against me that resulted in "me" loosing.

I have found that certain ships need special rules to make sure card combos go off when they should and the ships behave more "intelligently". But I try to keep that to a minimum.

I also have not used this much with Rebel AI. Imperial ships are more forgiving for bad AI "choices" and they are easier to handicap.

Thanks for the feedback! I'll probably be working on an update to this system soon, as I've just updated my Legion AI also, and included a really nice little "Elastic Clause" option - I'll summarize.

3 Times per game - you get to Veto the AI and do what you think makes sense. This should be when the AI does something that is generally just a VERY bad choice that no reasonable opponent would do.

I limit it to 3 times, but you are certainly welcome to adjust that number - but the idea here is that if you do it too much... well then you're essentially just playing a normal game against yourself where you have to think about every decision - and hey that's fine and good... but this AI system is supposed to represent a very specific type of opponent - a random one, who makes mistakes.

I was able to try this system out last night and had a blast. I thought I would share my thoughts:

1) the random ship deployment method resulted in a cluster of ships together. After they started heading towards my fleet no amount of maneuvering could prevent collision. Next time I will rule that if two ships would be placed side by side, allow an additional 4-5 inches in between. But I do love this method of deployment in general, it made it exciting.

2) I ruled that if the AI has a lineup of ships headed the same direction, and the one in back would overlap those in front from any maneuver available, it simply got skipped and I would select a new ship to activate and go back to it later. In my game an ISD ran into debris, and then got rammed from behind all in one turn.

3) Squadron movement priority. I think under current rulings if the human player attacks AI ships with a fleet of bombers the AI squadrons don't really care. I might change the priority to be "closest bomber squadron".

4) Someone mentioned this before on defense tokens, I might prioritize preventing the most damage as #1 (or preventing death as #1). If unsure or a tie, randomize.

Excited to play again! Where I moved to in Idaho there are not a whole lot of players available.

On 10/12/2020 at 9:07 AM, GalenErso said:

1) the random ship deployment method resulted in a cluster of ships together. After they started heading towards my fleet no amount of maneuvering could prevent collision . Next time I will rule that if two ships would be placed side by side, allow an additional 4-5 inches in between. But I do love this method of deployment in general, it made it exciting.

This was an issue in my solo match as well. See the photo I attached in my post from April 16, and you'll note that the Rebel Smalls are kind of all over the place; this was mostly the byproduct of trying to avoid collisions with each other.

8 hours ago, Rmcarrier1 said:

This was an issue in my solo match as well. See the photo I attached in my post from April 16, and you'll note that the Rebel Smalls are kind of all over the place; this was mostly the byproduct of trying to avoid collisions with each other.

If you find yourself running into that a lot, I found (back when I was playing a lot of 8x CR90B swarms, lol) that a staggered deployment, like this:

... helps keep me from tripping over myself quite so much, while still keeping a good balance between movement flexibility and concentration of forces.