Skip to Content
 

Testing chat GPT for mechanics searching

6 replies [Last post]
larienna
larienna's picture
Offline
Joined: 07/28/2008

I wanted to see again the potential of chat GPT. This time, instead of generating content like in the previous post, I wanted to see if it could help me do mechanics searching for dead or locked ideas.

I un-dusted the idea of making a star wars battle of hoth game. One of the reason is because it a known battle, it could be easier to find games that could work. I would say the results are interesting, there is a few potential games to take a look into. But so far, asking users on forums seems to give better answers. Further refining could probably be done.

Here is the chat transcription:

***

I want to create a board game that tries to reproduce the battle on Hoth in star wars empire strikes back where the rebels try to defend their base until they can escape from the imperial invasion. I need board game mechanics that could be used to create such a game. Suggest me 10 board games containing interesting mechanics that I could use to create my game idea. The theme of those board games does need to be sci-fi.

Quote:

Here are 10 board games with interesting mechanics that you could adapt for your Hoth-inspired battle game:

Memoir '44
Mechanic: Area control, card-driven commands
This World War II-themed game uses a modular board, where players move units using a card system that limits what actions they can take in each turn. It would be a great fit for recreating strategic troop movement and planning under pressure.

Star Wars: Rebellion
Mechanic: Asymmetrical gameplay, hidden objectives
This game involves two very different roles, the Empire and the Rebel Alliance, with hidden rebel bases and multiple victory paths. Asymmetry could mirror the uneven balance of power during the Hoth battle, with one side trying to escape while the other seeks to crush them.

XCOM: The Board Game
Mechanic: Real-time decision-making, app integration
It uses a real-time, app-driven element where players must quickly decide how to allocate resources and respond to threats. A timed mechanism would add urgency to the rebel’s need to escape while facing overwhelming odds.

Battlestar Galactica: The Board Game
Mechanic: Hidden traitor, resource management
This game’s tension comes from players having hidden roles, with some working against the team. You could adapt a mechanic where some rebel players secretly work for the Empire, adding suspense and mistrust during key decisions.

Pandemic
Mechanic: Cooperative play, escalating threats
Pandemic is cooperative, where players must work together to prevent global outbreaks. You could incorporate a system where the rebels must manage escalating threats (e.g., AT-ATs closing in) while coordinating defenses and preparing for evacuation.

Risk: Star Wars Edition
Mechanic: Simplified battle system
A quicker, streamlined battle system that resolves conflicts rapidly could help replicate the large-scale battles without bogging down the game. You could adapt how forces clash in key locations like the snowy plains of Hoth and the shield generator area.

Scythe
Mechanic: Engine-building, area control
In Scythe, players develop their economy and army to expand and control territory. Rebels could focus on building defenses and resources, while the Empire steadily invades, pushing closer to the rebel base as time goes on.

Arkham Horror: The Card Game
Mechanic: Scenario-based play, deck-building
It’s a narrative-driven game where players modify their decks between scenarios. You could create a campaign where rebel players evolve their strategy and forces over multiple stages of the battle, each scenario representing a new phase of the invasion.

Eldritch Horror
Mechanic: Global threat management, action selection
It has a strong sense of urgency with multiple events happening simultaneously. The Empire’s forces could be represented by a constant, global threat that spreads across the board, forcing rebels to divide their actions between defending, retreating, and preparing the evacuation.

Gloomhaven
Mechanic: Tactical combat, card-based actions
Gloomhaven uses tactical, turn-based combat driven by a card system where players choose from a limited set of actions each turn. This could replicate rebel players managing limited resources and positioning in a desperate fight for survival.

Each of these games offers a different mechanic that could inspire aspects of your Empire Strikes Back battle on Hoth, from asymmetric roles to time-sensitive cooperative gameplay and escalating threats. Let me know if you'd like more detail on how to adapt these mechanics!

The time frame will be limited to a single battle so I am looking more for games where the rebels are under siege. Resource management and tactical combat is fine. Having something linear in the game to keep track of imperial progress is important. Some games I have in my mind with similar mechanism is: Siege of valeria, Castle assault, Under falling skies and the state of siege system. Recommend me 10 other potential games.

Quote:

Here are 10 additional games that could inspire your board game’s siege and defense mechanics:

Stronghold
Mechanic: Asymmetrical siege warfare, resource management
One player controls invaders while the other defends the castle, using resources to build defenses and fortify key points. The siege track could mirror the imperial advance on the rebels' base.

This War of Mine
Mechanic: Survival, resource scarcity
In this survival game, players must manage dwindling resources while under constant threat. Though it’s not about a traditional siege, the sense of limited resources and time pressure could reflect the rebels' desperate defense and preparations for escape.

Dawn of the Zeds
Mechanic: Siege defense, cooperative play
Players defend a town from waves of zombies using a combination of resource management and tactical combat. This siege system could be adapted to reflect waves of imperial attacks, with a track representing the growing imperial threat.

Defenders of the Realm
Mechanic: Cooperative, escalating threats
Heroes work together to protect a realm from four advancing enemy forces. The advancing armies’ progress can be mirrored by the Empire's invasion, with players needing to hold them off while managing their resources.

Helionox: Chronicles
Mechanic: Deck-building, linear progression of threats
Helionox uses a system where players manage resources while dealing with crises that escalate over time. A similar mechanic could simulate escalating imperial pressure with players trying to balance defense and preparation for escape.

Atlantis Rising
Mechanic: Cooperative defense, escalating destruction
Players work together to save their civilization as Atlantis sinks. The idea of using a central catastrophe mechanic (in this case, sinking) could be reworked into the advancing imperial forces, giving the rebels limited time before they are overwhelmed.

Ghost Stories
Mechanic: Siege defense, cooperative play
This game is about defending a village from an ever-growing horde of ghosts. Players must balance their actions between defending locations and addressing threats. This could be mirrored with rebels defending their base from waves of Imperial forces.

Space Hulk
Mechanic: Tactical combat, asymmetrical forces
Space Hulk pits heavily outnumbered Space Marines against hordes of Genestealers in tight corridors. The claustrophobic, tactical combat might inspire the intense fighting inside the rebel base as the Empire breaches defenses.

Hoplomachus: The Lost Cities
Mechanic: Tactical combat, arena siege
A gladiator-style game where players defend against waves of enemies in an arena setting. Though it’s not a traditional siege, the tactical combat and waves of enemies could inspire how rebel forces defend against the Empire's elite troops.

One Deck Dungeon
Mechanic: Dice-rolling, linear progress
In this dungeon crawler, players manage resources (dice) to overcome obstacles and enemies. The system of managing limited actions to defeat progressively harder threats could translate well to a scenario where the rebels face mounting imperial pressure.

Each of these games brings mechanics that suit siege-style scenarios with linear progress of enemy forces. You can use this inspiration to design how the Empire's relentless advance pressures the rebels into critical decision-making, balancing defense with the goal of escaping.

questccg
questccg's picture
Offline
Joined: 04/16/2011
I usually choose my mechanics first and don't look for...

I don't search for Mechanics. I usually KNOW what mechanics a game will HAVE and OFTEN when I start designing I ADD other "mechanics" to improve the overall game. I don't need ChatGPT to suggest games to consult the mechanics because honestly from what I see... It doesn't demonstrate how to INTEGRATE all these mechanics.

Basically it's as if you asked ChatGPT to give you a description of a car and it replies: "A car has 4 wheels, a motor, lights, etc." Instead of saying it's "A Ferrari with 19 inch aluminum mags, with an engine with 12 cylinders, etc."

There is no way all the mechanics that it suggested (the 10 of them) fit into ONE COHESIVE design. It's a pretty lame response. But I guess that's the more creative part of making games. Sometimes all you need is ONE (1) MECHANIC and you can derive the entire game from it.

I think much differently nowadays. What I keep to myself is not the designs that I am working on ... But more what I think would make something different and that the end result would be very unique and enticing to players. That being said, I am only beginning that journey ... As I had no way to take notes to this effect. Now that I have a way to taking said notes, maybe I can work on more details and see what I can flesh out (Better).

But the truth of the matter is while ChatGPT can spit out a bunch of games with different "mechanics", it doesn't understand how to LINK any of the ones it said with each other.

The TRUE test is YOU (@larienna): How do you think ChatGPT did in its response???

Does any of that make any sense.

I'll do one more example. Someone asked how to design a GMail Login Form. When asked how to do it, all ChatGPT spit out what the "screen" and Front-End code required not the logic required to IMPLEMENT the "screen". So yeah it spit out some HTML to how the FORM looks and that's it. NO PHP code for the back-end into HOW(?) a programmer would code the "Back-end" of the implementation. That was plain dumb... But it is what it is TBH.

I get the same impression here too.

What do you think @larienna???

larienna
larienna's picture
Offline
Joined: 07/28/2008
Mechanic searching in board

Mechanic searching in board game design is mandatory to make my design progress. All projects stall because I do not have the right mechanics. Therefore, I need to search, ask on BGG or stumble on the mechanic before progressing, this is very annoying. Searching is far from being easy, asking on BGG was the only good solution so far. Remember that I am better at making variants.

I do not expect the mechanics to fit together as is. I intend to modify them but at least it sparks an idea of solution. As for the quality of the results, There is a few games that could be worth looking into, like max 5/20. Many games like Arkham Horror, and Battlestar Galactica, are completely off. I am surprised that siege of Valeria did was not in the results since it's a game I discovered recently that matched the theme pretty well.

I still think this method to be more constructive and less painful to use than previous method of creating mechanics database or having to rely on the memory of the community. Even if the results are not precise, it could give interesting recommendations.

So far, it seems much more useful for brainstorming thematic content generation like shown in a previous thread. I intend to use it to search new titles that could be worth playing instead of using BGG's search tool for example. I will probably also test ChatGPT, by asking questions I know an answer and see if it gets out in the results.


Speaking of mechanic searching, while archiving my files, I stumbled on my Starcraft Lite variant which sparked new ideas handling the combat in my version of the game.

As a reminder, the objective was to create an interesting system to allow Rock- Paper-Scissor interaction between units while allowing designing 32 unique units. SC Lite, used dice to determine which unit got activated, those units could target another unit. The rest of the units are sum up in splash damage. One of the problem of this system is that it requires moving units off board to keep track of unit status.

I put some thoughts into it, and I am thinking of using cards as attacks that can boosted if a unit is present. The units will be the target only. Splash damage could build up from the card, or from surviving units. You play 3 cards/rounds, then the sum of splash damage is applied and the battle is over. This make the system AI friendly (draw the next card). It allow unit synergies on the card, and there is no need to move around unit tokens. Also the quantity of units only affects splash damage. This is important as the number of units per battles will be low (2-4). You also want to make it possible to win an attack by not sending more units than the opponent. Cards can be upgraded with tech.

Now I don't want and hard RPS, where specific units can only kill other units. This makes the outcome too much predictable and can lead to dead end situations.

I only hope that this system can support a lot of unit variety. Resolving the combat system would be nice, as its what is locking the game right now. The rest is pretty similar to the original game.

questccg
questccg's picture
Offline
Joined: 04/16/2011
I use this BGG page and then add my own twist...

https://boardgamegeek.com/browse/boardgamemechanic

That page has just under 200 Mechanics. You can click on a Mechanic and they will give you games using that Mechanic and then you can watch a Dice Tower review of that game to see how the Mechanic fits in with the rest of the game...

Anyhow ... I guess I design differently. I usually know what my PRIMARY Game Mechanic is going to be and then maybe I'll look into the BGG page and see what else I might be able to add, etc.

Like for "Battle Botz" I knew the Primary Mechanic is "Dice Pooling". And then going with the design I realized that there was a STRONG "Push-Your-Luck" Mechanic that is probably next. And then of course there is "Deck Construction" which is an off-line activity Players do to construct their deck of cards prior to playing head-to-head, and of course Resource Management because each "Bot" has STATs which are variable and a part of the "Dice Pooling", etc.

So most of the time it's not Mechanics which block me. It's DESIGNING a BUNCH of "content" like figuring out a bunch of "Botz" and their STATs for each card. I mean I can't playtest Vaporware and therefore need to flesh out ACTUAL cards and see if the interplay is GOOD or NOT(!?)

That's where I BLOCK: designing enough "content" to prove that the design has merit and is FUN to play and is worthwhile fleshing out more "content" as this is the MAJOR blocker.

Completing the DESIGN and writing the RULES are the two (2) hardest aspects for me anyhow and that's where I get blocked between a COOL IDEA and a FINISHED PROTOTYPE. We all know that the IDEAS sound GREAT ... But sometimes when you put them together ... They just don't WORK as INTENDED or there are factors which prevent the IDEA to be designed AS-IS, etc, etc.

In any event I have an issue between the IDEA and PROTOTYPE and usually it means that the PROTOTYPE may have various versions (I think "Monster Keep" is on version 25 ATM...) And it's got me reworking the combat mechanics ... For example.

Any tips for me??? Is there something I am missing?? Perhaps I am not using the proper method to flesh out the design or prototype correctly?

larienna
larienna's picture
Offline
Joined: 07/28/2008
Designing content is always

Designing content is always problematic. You can't design rules because you do not have the robots and you cannot design the robots because you do not have rules.

Thematic content is now easier with ChatGPT. Just let it generate list of ideas, and chose from the list.

For the stats and mechanics, I think the solution is to use the most simple designs( basic stats, no special power) and use a limited amount of quantity. Then playtest. As you play, ability ideas will come, and new robots could be created. Some will be modified. Your possibility space will grow over time

This is how I intend to proceed with my starcraft game idea. Design 4-8 units, no special, just the cards and unit stats. Playtest, get ability ideas, try them out, add more units, etc. Some rules might be adjusted in the process.

When design from variant, it`s easier because you already have base you can use. When designing from scratch, this base does not exist unless you are designing strongly inspired from another game.

Again, in with my starcraft idea, I have starcraft the board game as a base that I could use. I can draw special abilities from this game, or even it`s successor Forbidden Star.

questccg
questccg's picture
Offline
Joined: 04/16/2011
I agree with you again 100%

larienna wrote:
Designing content is always problematic... For the stats and mechanics, I think the solution is to use the most simple designs (basic stats, no special powers) and use a limited amount of quantity. Then playtest. As you play, ability ideas will come, and new robots could be created. Some will be modified. Your possibility space will grow over time...

That makes a lot of sense. Just the process is LONG. It's not like I am going to complete the design in a couple of hours. Clearly this is a lengthy process and it's going to require a lot of patience and effort. That's why MOST of the time, I don't get to a FULL PROTOTYPE until after some time, fudging around with design until I get some solid ideas that ARE "working".

Like "Monster Keep" is an example where the "core" of the game remains the same but the Combat Mechanic has been revised MANY, MANY, MANY times. I'm on version #25 (or there about) and TBH it's still not 100%. I sacrificed some of the combat mechanics from "Monster Keep" and placed them into my Marvel 2024 Tournament which is going to be Next Month (most probably).

So I re-use and borrow Mechanics from my games and introduce them into other games and so forth. The Marvel 2024 is "House Rules" which means that the concept is my own but from the adaptation of "Pars En Mission" from Loblaws and we'll see how the tournament fares...

But yeah I borrowed some Combat Ideas from "Monster Keep" into the Tournament.

Stuff like that happens too! Cheers.

larienna
larienna's picture
Offline
Joined: 07/28/2008
You can try to limit the

You can try to limit the amount of iteration tests, by giving yourself objectives, taking notes during the test, and do a lot of changes for the next iteration. This way, each iteration you will be progressing a lot.

There is still a risk that you never find your sweet spot, normally you use the funnel approach, do a lot of changes early, and little changes in the end.

When the mechanics are solid, Automated testing is also a possibility. You are a programmer, you could program some simulations of the game like I did for my stock market game. This will give you solid data to make design decisions.

Syndicate content


forum | by Dr. Radut