Max Caine
msg tools
Hey all, I'm looking into developing a dedicated support class of enemy. By "support", I mean something on the lines of a classic healer/buffer/debuffer/stunner/blocker/etc. which does support effects as its main role rather than as an extra bonus (e.g. from a random serpent mage spell or the Repair ability of the Custodian). The key to developing a support class is making a support behavioral ruleset and if anyone's free I'd like to ask your advice in developing a solid universal ruleset that can be applied to any support class of enemy.

Here are my thoughts. A support class should:

Focus on its support role, but not exclusively. If a Hero starts wailing away in melee, then it should be able to respond to that.
Be able to do something other that support. If there are no valid targets for its support then it should be able to do something else, e.g. attack.
Be able to easily identify which enemies are valid targets for support.
Be able to prioritize which enemies should be given to support

My thought for how this might be realized is this:

SUPPORT (Ability:[List of Keywords])

SUPPORT indicates that the enemy follows the Support ruleset. Ability indicates which ability in the ability box it should use. The List of Keywords indicates which enemies with corresponding keywords may be targetted by the Support Ability.

So SUPPORT (MEDIC: Alien, Soldier, Trederran) would mean the enemy with Support ruleset uses the MEDIC ability, and any enemy with the Alien, Soldier or Trederran keywords can be targeted.

That's fairly clear, don't you think? Anyone with a better idea?

Targeting, I think, is the hardest bit to work out. The Support class should maximise the value of its support abilities, however the ability could be anything. I think that each support ability should have a single Priority requirement. Any enemy that satisfies the Priority requirement will be targeted first. By setting a priority, we can create a subset of valid targets which are modeled specifically on what the ability does. It would look something like:

MEDIC
Priority: Enemy with most wounds
[Main Text]

Or.

MELEE BUFF
Priority: Enemy in melee combat with Hero
[Main Text]

That guarantees there's a subset of appropriate valid targets but that subset may be everyone! More refinement would be useful, but if there's too much refinement then the ruleset gets bogged down with going through lots of steps. So... two more levels of refinement? Is that too many? I think that something like: if more than one target then 1) select from this subset the enemy which the support can affect with the ability (otherwise they're wasting their time). 2) select all targets from the second subset which have attacked a hero (as not targets may have attacked, so that focuses on enemies which either have or are doing something to heroes). But there is the possibility that refining that far might reduce the valid target pool to zero, so is it worth it? Perhaps it's just worthwhile prioritizing a subset of targets which the support can reach, and which are appropriate to the ability being used.

Now, if there are no valid targets for the support (e.g. all other enemies are dead, or it cannot reach a valid target to affect it), then the Support needs another behavior ruleset. The basic targeting ruleset would, I think, be a suitible backup. And the support should be able to the basic ruleset if a Hero tries to melee, otherwise it will be very vulnerable. (Or is that a good thing?)

Well, those are my thoughts on a Support behavioral ruleset. If anyone has any ideas, I'd love to hear from you.
 
 Thumb up
 tip
 Hide
  • [+] Dice rolls
Sid Rain
United States
Columbus
Ohio
flag msg tools
badge
Avatar
mbmbmbmb
I think the Custodians of Targa fit into a "medic" support class and could work as a template (with some alterations to fit the specific Enemies that it heals).


Electro-shock - At the start of this Enemy's Activation, every adjacent Hero takes 3 Hits.

Defense Laser - Ranged Attack: Range - 8 / Shots - 2 / Damage 3

Repair (2) - At the start of this Enemy's Activation, if there are no adjacent Heroes, Heals 2 Wounds from itself and all adjacent Targa Pylon and Robot Enemies.

Repair Protocol - When selecting a target, include any damaged Targa Pylons (will move to get adjacent for Repair, but targets a Random Hero with Defense Laser).


So when he's moving around, he's still attacking the Heroes. If a Hero gets in melee, he'll stop to attack. Even when he's going for Repair, he's still attacking. You could probably tweak the behavior a bit more depending on how you wanted it to act if you want it more/less aggressive.
2 
 Thumb up
 tip
 Hide
  • [+] Dice rolls
David Griffin
United States
Georgia
flag msg tools
designer
Avatar
mbmbmbmbmb
One of the reasons PC parties win is that they usually have a better mix of abilities (and work together better). If you give the enemy a cleric and maybe a mage etc., you might make them better than the players which might have more game effect than you might think.

One of the things I am not fond of in boardgame versions of roleplaying games (though I accept this is inevitable) is that all the enemies work together like a delta force unit. In reality, lots of monsters would attack each other just as easily as the heroes. When I was a DM I could make this happen and the results were often a lot of fun with the players arranging to aim one set of monsters at another.

At the least, with intelligent monsters, they should have to check morale when almost all of them are dead. That's tough to do too though so I don't expect to see it.

Anyway, playtest it to make sure you're not going to murder all the players. Regeneration (e.g. Clerics) is tough to beat.
 
 Thumb up
 tip
 Hide
  • [+] Dice rolls
Freelance Police
United States
Palo Alto
California
flag msg tools
designer
Avatar
mbmbmbmbmb
Have you played Warhammer Quest? I think that's the "best" AI I've seen. Rather than explicit if-then code, a monster will perform every action it can. So a medic might be AID - ATTACK, with the rest of the card filling in any details.

IMO, NPC design means "efficient play", not "mimicking a PC". You want the players immersed in the game, rather than slowing down play.
 
 Thumb up
 tip
 Hide
  • [+] Dice rolls
Max Caine
msg tools
Having mediated on the wisdom of Sam&Max, carbon_dragon and paddirn I've had these thoughts.

SUPPORT ABILITIES.

If an ability is marked (SUPPORT) it must be resolved before targeting and attacking Heroes. Target an enemy according to the targeting requirements listed on the ability (e.g. target a random damaged Targa Pylon). Move the enemy so it's within range and LoS to use the ability and then use the ability. Repeat this for every ability in the abilities box listed as (SUPPORT) reading from top to bottom. After all (SUPPORT) abilities have been used, target and attack a Hero as normal. Any movement used when using (SUPPORT) abilities counts against the total movement of the enemy.

----

From a mechanics aspect, allowing support abilities (such as healing) to be usable on every enemy in combat isn't the best of ideas, as it makes dangerous enemies more OP. (Although that's exactly what a Void Magus does with half his spells, but oh well) and it stretches suspension of belief. A Trederran medic would reasonably be expected to perform triage on a raider or legionnaire, but it would be unreasonable to expect the same medic to heal a void spider! Having the support ability list what targeting requirements are necessary means the ability can be tailored to a specific set of situations, and there are no complicated rules about how to target an enemy. Moving support abilities entirely into the abilities box also means that there's more flexibility in building enemies, as a support character could be built entirely out of support abilities, or feature a mix of attack and support. Any thoughts on this change?
 
 Thumb up
 tip
 Hide
  • [+] Dice rolls
Front Page | Welcome | Contact | Privacy Policy | Terms of Service | Advertise | Support BGG | Feeds RSS
Geekdo, BoardGameGeek, the Geekdo logo, and the BoardGameGeek logo are trademarks of BoardGameGeek, LLC.