Skip to Content
 

Player or AI, a new cardgame concept?

7 replies [Last post]
Fri, 10/18/2019 - 08:17
X3M
X3M's picture
Offline
Joined: 10/28/2013

I often have only like 30 minutes free time.
So I tought of designing a game in that direction.

And seeing that no one else has the time in those 30 minutes. I need to design an AI.

AI is often boring or super complex. And follows rules that a player doesn't want to follow.
The capability to make decisions is very important for players.

An AI might simply be a random opponent that the player needs to fight.
And the advantage for the player should be just that?

How about having a card game, where each unit is designed from several cards and dice.

The player can hold the options open.
The AI is forced to use immediately.

The player can hold resources for a next round.
The player can increase the ammount of any design.
The AI is forced to spend on the last design immediately.

The player decides what it targets.
The AI is decided by the player, abuse is of course allowed.
Suggestions are welcome on how to decide on targets with a game that develops between 1 to 6+ targets.

No real resource managment, no possibility to destroy this resource managment. The first round gives "1" to each player. The second round "2" etc. Not sure on how much yet.

The goal would be, destroying the opponents base.
But this would be a stack of health. That's it. Lifepoints if you will. I don't see the point in having to defend several different instalations. Since the game is 1D. Also, the table should contain less cards. Thinking about 13 or more cards. No more than 18.

On the other and, defences and walls should be possible.
Unsure about having a binary movement system, or a digital. Aka, movement; yes/no, movement; #

Top
Fri, 10/18/2019 - 08:44
#1
wob
Offline
Joined: 06/09/2017
you have given the human a

you have given the human a fair few advantages there, the ability to store resources etc. as the ai has less control i would give it extra resources, or rather make things cheaper and easier for the ai. if it costs you 3 gold 2 cool down to use your cannon it should cost the ai 1 gold and no cool down (or what ever is specific to your game)
its also a good idea to limit the decisions the player makes for an ai. the ai could "attack the nearest troop" or "attack the troop with the highest hp" or each troop could have a target number and the ai "attacks troops with the highest target number first".
an overly complex way to do it (that would add variety) would be to have a way to randomise ai troop priority at the start of the game, for instance having a token for each troop type in a bag, and the ai target s artillery until there is none in range/in play then it targets, light infantry, tanks etc.
i have seen a few games that have booklets and tables dedicated to ai decisions. a sort of "if A then B. if C then D unless E then F" you can use these types of ai to make amazingly complex and realistic games that some people love. personally i am not a fan of spreadsheet: the game.

Top
Fri, 10/18/2019 - 11:50
#2
let-off studios
let-off studios's picture
Offline
Joined: 02/07/2011
Asymmetric Abilities

wob wrote:
you have given the human a fair few advantages there, the ability to store resources etc. as the ai has less control i would give it extra resources, or rather make things cheaper and easier for the ai.
Making AI units stronger and/or more effective at certain tasks (resource cost for certain tasks, as detailed above, is a good example) is what I'd recommend you experiment with to start. The player doesn't have to go through complex steps to make things happen for the AI.

Additionally, if this formula works well, you can likely tweak numbers to provide different challenge levels and even customized scenarios.

Top
Fri, 10/18/2019 - 15:12
#3
FrankM
Offline
Joined: 01/27/2017
Asymmetry rules

One quick way to give the AI a bit of an advantage is to use your original "must play" mechanic, but the AI can play all of its playable cards even if it runs out of the main resource.

Just to explain what I mean, perhaps the deck is a mix of unit cards, fog-of-war ("take that") cards, and initiative point cards. The AI plays its highest-initiative-cost maneuver first, then the highest-initiative-cost maneuver on its remaining cards, etc. until it runs out of initiative points for the turn. Then it still uses each remaining card's lowest-initiative-cost maneuver "for free."

Just have some pre-determined way to break ties, such as order the cards were drawn.

This will ensure there are a small number of worrisome "attacks" each turn, and a fringe of lower-power happenings to keep the player from concentrating on a single kind of defense.

Top
Sat, 10/19/2019 - 20:43
#4
X3M
X3M's picture
Offline
Joined: 10/28/2013
Some more concept

To keep the AI intimidating.
The AI will receive more resources.

To keep the AI simple in attacks.
An attack will occur by rolling a die. Yes/no

An explosive round (no body card) will always occur.

There will be 6 tiers of armor.
1 to 6.

There are several AI possible now.

One is by a die roll. And focusing all fire power on that type of targets.
Two, would be to roll for a front line target.
Three, would be trying to, is the weapons focussed on the best targets.

Priority would be great for most of these AI. But this would be used for the weapon tiers against the armor tiers:
Tier 1; 1,2,3,4,5,6
Tier 2; 2,3,1or4,5,6 (one more die roll)
Tier 3; 3,4,2,5,6,1
Tier 4; 4,5,3,6,2,1
Tier 5; 5,6,4,3,2,1
Tier 6; 6,5,4,3,2,1

***

With 3x6 cards. I allow for a 3x3 design set up.
Having a front line, where range 0 to 3 are of use.
Having a middle line, where range 1 to 4 are of use. But range 0 is allowed for just in case the front line dies.
Having a last line, where range 2 to 5 are of use. Also, range 0 and 1 can be used, just in case.

The base card is the 4th row if you will. Only expensive weapons can reach it.

Designs cannot move at all in this set up. So I need to tackle that first.
I don't want to move the cards on the table.
But movement should be in the game somehow.

Henceforth that one question I asked in another topic.

Top
Sat, 10/19/2019 - 20:59
#5
X3M
X3M's picture
Offline
Joined: 10/28/2013
Designing process

A design consists of 1 or 2 cards.
A body card and/or a weapon card.

Only a body card will become a wall.
Only a weapon card will become an 1 round explosive, doing 3? times the damage.
A body combined with a weapon will form a normal unit.

The player draws a card and chooses to play or hold it for later.
The AI always holds it until it can form a normal unit. Thus no walls, nor explosives.

A new design may be placed only once every round.
A failure may or must be discarded.
The cards remain in the players hand.

***

Body card
Contains the following stats:
-Armor tier
Is fixed
-Speed???
1D-1, unless it is going to be a wall.
-Health factor
2D-2, if 0 is the result, the design is forced to use a weapon card only. The body card will remain in the players hand.
-Cost calculation
Health*Armor*(2*Speed+6)*6

Weapon card
Contains the following stats:
-Weapon tier
Is fixed
-Range???
1D-1
-Multiplier
2D, where lowest gets subtracted from the highest.
If 0, then the design is forced to become a wall instead.
-Accuracy
1D
-Cost calculation
Multiplier*Weapon*(3*Range+6)*Accuracy

Rolled results are translated into proper dice values and placed on the corresponding cards.

***

Are those 2 formula's to complex for players?

Top
Sun, 10/20/2019 - 06:08
#6
X3M
X3M's picture
Offline
Joined: 10/28/2013
got a bit of time for this

FrankM wrote:
One quick way to give the AI a bit of an advantage is to use your original "must play" mechanic, but the AI can play all of its playable cards even if it runs out of the main resource.
It turns out, this is the only way. And to begin with, it will be very random, and stronger than that of the player. The player might have to sacrifice some life points too.

FrankM wrote:

This will ensure there are a small number of worrisome "attacks" each turn, and a fringe of lower-power happenings to keep the player from concentrating on a single kind of defense.

The game contains a RPS mechanism. Focusing on walls only for too long is a bad idea. Focusing on 1 type of design is also a bad idea. Eventually, the entire army will be a mix. Perhaps a certain weapon or body is left out. Due to low RPS mechanisms. But these can still be used as great replacements.

let-off studios wrote:

Additionally, if this formula works well, you can likely tweak numbers to provide different challenge levels and even customized scenarios.

Different difficulties are possible. Simply changing the rate of resources that the AI gets.
I need to play the game to figure out where the average difficulty will be.

wob wrote:
i have seen a few games that have booklets and tables dedicated to ai decisions. a sort of "if A then B. if C then D unless E then F" you can use these types of ai to make amazingly complex and realistic games that some people love. personally i am not a fan of spreadsheet: the game.

I used to make these for the board game, several years ago. Have to say, I dislike this these days. Way to much work. Especially if balance needs to be tweaked.
I want a game that will be playable, where ever you are. Only a deck of cards, with a box of dice and perhaps chips to track health and numbers.

As for prioritising targets. I have thought of 3 ways. And actually expanded on some. Not only that, but to determine, which AI will occur that round, a die is rolled:
1. A die roll determines which body tier is targeted. By all the cards that the AI owns. If this body tier isn't present. The priority is determined by consulting the AI card.

2. Every weapon tier tries to target the most optimal body tier. For this, the weapon cards can also be consulted.

3. A die roll for which line will be targeted. If the player misses a line, the base can be targeted. There is a max of 3 lines. Of course the die roll will occur as soon as 1 line is added to the game. Once the line is determined. There might be several different targets in that line. For this, a die roll is once again administered.
For 2 possibilities: One; 123, Two; 456
For 3 possibilities: One; 12, Two; 34, Three; 56
Easy and simple. :)

4. The AI receives extra resources.

5. The AI, if possible, places a wall. Prioritising the highest tier of body.

6. The AI, if possible, attacks with only a weapon card, thus explosive. Prioritising the lowest tier of weapon.

Top
Tue, 10/22/2019 - 11:54
#7
X3M
X3M's picture
Offline
Joined: 10/28/2013
How could range and speed work?

I can't solve this.

Top
Syndicate content


forum | by Dr. Radut