Tips for an interesting boss fight?

[Jacie0krece]

Hi, I want to get some tips for overall making boss fights. The attempt I did in my game wasn't too good (and it would probably mildly infuriate some players), so I'm looking for tips here.

 

[Trihan]

There are two main levers you need to align when designing a boss battle:

1. How long should it take to beat?
2. How much of that time should the player spend offsetting the damage they're taking?

It goes without saying that a turn in which you need to heal is a turn in which you're not doing anything that directly furthers your goal of defeating the boss. But at the same time, you want bosses to feel like they're genuine threats who are actually capable of wiping out the party, or you have no real stakes.

One metric that's commonly used to determine #1, especially in a turn-based system, is how many turns your player should take roughly to kill the boss. Once you figure that out, you can stat them accordingly based on the level the player is likely to be when facing the encounter. Once you know how long they're going to take to beat and what their stats are, you can give them a moveset/attack pattern that makes them strong without being overpowered, and will allow the player to beat them in the number of turns you've decided on.

 

[ZombieKidzRule]

You might find a lot of valuable insight from this recent thread as well. All about End Game Bosses, but the responses roamed far and wide in their narratives about bosses in general.

Late-Game Boss Design - Some Food for Thought

Alright, we've gone through the desert/cave/temple. Hold on this looks familiar...a suspiciously boss-looking room. Now top off everyone with healing items in the menu, throw on your buffs, and use your best attacks, its time to tackle... ...wait, its dead already? Huh. Whether its level...

forums.rpgmakerweb.com

I am pretty sure a lot of what was said in that thread applies to your question as well.

Good luck!

 

[Willibab]

What kind of boss?

 

[sunnyFVA]

You'll also want to consider the role of boss fights in your game. It has become commonly-repeated advice in the past few years that bosses ought to be direct evolutions/syntheses of previously-introduced enemy mechanics. While useful for getting started, it can be a limiting perspective. Try thinking about why this boss (or "the boss fight" more generally) belongs in the game and then figure out how best to fulfill that initial goal. Even without getting too philosophical, a thematically distinctive boss can smoothly become the basis for a mechanically distinctive one once you hone in on what is so compelling about them.

 

[bigsmiles]

What is the problem you had which you think would infuriate people?

 

[Jacie0krece]

There are two main levers you need to align when designing a boss battle:

1. How long should it take to beat?
2. How much of that time should the player spend offsetting the damage they're taking?

It goes without saying that a turn in which you need to heal is a turn in which you're not doing anything that directly furthers your goal of defeating the boss. But at the same time, you want bosses to feel like they're genuine threats who are actually capable of wiping out the party, or you have no real stakes.

One metric that's commonly used to determine #1, especially in a turn-based system, is how many turns your player should take roughly to kill the boss. Once you figure that out, you can stat them accordingly based on the level the player is likely to be when facing the encounter. Once you know how long they're going to take to beat and what their stats are, you can give them a moveset/attack pattern that makes them strong without being overpowered, and will allow the player to beat them in the number of turns you've decided on.

That's a good idea. How many turns you think that various kind of bosses should take?

You might find a lot of valuable insight from this recent thread as well. All about End Game Bosses, but the responses roamed far and wide in their narratives about bosses in general.

Late-Game Boss Design - Some Food for Thought

Alright, we've gone through the desert/cave/temple. Hold on this looks familiar...a suspiciously boss-looking room. Now top off everyone with healing items in the menu, throw on your buffs, and use your best attacks, its time to tackle... ...wait, its dead already? Huh. Whether its level...

forums.rpgmakerweb.com

I am pretty sure a lot of what was said in that thread applies to your question as well.

Good luck!

The boss I'm making is the first main boss of the game, but I'll look trough this thread when I have some time.

What kind of boss?

Just a smaller miniboss that's there to show the player a mechanic that the final boss of the location will use.

You'll also want to consider the role of boss fights in your game. It has become commonly-repeated advice in the past few years that bosses ought to be direct evolutions/syntheses of previously-introduced enemy mechanics. While useful for getting started, it can be a limiting perspective. Try thinking about why this boss (or "the boss fight" more generally) belongs in the game and then figure out how best to fulfill that initial goal. Even without getting too philosophical, a thematically distinctive boss can smoothly become the basis for a mechanically distinctive one once you hone in on what is so compelling about them.

Due to the mechanic of "exp notes" that give exp to everyone in the party (and are items that can be bought if in a pinch), defeating enemies isn't neccessary. As I said, the boss I'm currently making "showcases" the mechanic that will be important in the upcoming bossfights a total of 5 times trough the whole game (1 per nation). All bosses (exept this dude, but he has different bossfight mechanics each time) will be pretty distinctive, especially if I'll have a good idea for them.

 

[Willibab]

Just a smaller miniboss that's there to show the player a mechanic that the final boss of the location will use.

Right... Humanoid? Beast? Offensive? Defensive? What mechanic is he showcasing? Its easier for people to help or give their ideas if you elaborate a little more Ofc, if you don't want to for whatever reason then you will prolly only get very generic stuff that may or may not apply to you.

The ''how many turns'' thing can apply to any boss. But it also kind of depends on the type of boss, is it supposed to be an endurance test. A tactical fight? Trying to survive the onslaught?

 

[Jacie0krece]

Yeah, I should probably tell some more about the boss.



First, he's a human (just like most of the enemies/bosses), and he's an more offensive type. The first mechanic that I would showcase would probably be elemental resistances - especially rotating ones with a clear mark of which is currently active (that is a state with the VisuStella State Tooltips plugin active). It's supposed to be more of a test if you are able to deal with that mechanic, as it will be used in a more ruthless fashion by the main boss of the nation.

 

[Trihan]

That's a good idea. How many turns you think that various kind of bosses should take?

As already stated, it largely depends on the kind of boss, but it should generally base itself on how long it usually takes to win a standard battle. If normal battles usually take 2-3 turns, a "normal" benchmark for bosses could be 10-15 turns.

 

[TheoAllen]

For me, "interesting" equals memorable. It does not need to be hard. Just interesting/memorable enough. That means the boss must have a gimmick unique from anything else. If you build a community around your game, your player might occasionally talk and joke about it. If it's just buffed normal battle with more HP and occasional AoE damage, it is not going to be interesting. Your player would just forget about it.

To make it interesting, it could be about the mechanic, the banter with your party member, or the unique graphic design.

So, ask yourself. What is something you want your player to remember about this boss battle?

 

[Tai_MT]

Yeah, I should probably tell some more about the boss.



First, he's a human (just like most of the enemies/bosses), and he's an more offensive type. The first mechanic that I would showcase would probably be elemental resistances - especially rotating ones with a clear mark of which is currently active (that is a state with the VisuStella State Tooltips plugin active). It's supposed to be more of a test if you are able to deal with that mechanic, as it will be used in a more ruthless fashion by the main boss of the nation.

Depending on how you've framed this, you could do some interesting things with it.

Ideas off the top of my head:
1. The Boss is actually "invincible", but he "summons" enemies that are hurt by the specific element you want. When the player defeats those summons, the boss becomes vulnerable for a turn or two before summoning the next set.

2. The Boss actually "punishes" an incorrect usage of the element. If you use the wrong one, he hits you really hard or something.

3. You could make it about the player noticing "subtle differences". You could attempt to "trick" the player with an element that is "similar", but not quite. Maybe it LOOKS like it's a Water Weakness... but it's instead a weakness to Ice. You could engage in some light subterfuge with the player.

---
It's honestly going to largely depend. Personally, I don't think a "weakness change" boss is all that interesting as a "final mechanic" for a Boss to utilize, so it wouldn't be something I'd ever teach my players.

Well, not unless I was going to do something with it nobody ever saw before.

If I wanted to teach "Elemental Weaknesses", I'd just have every single enemy in the Dungeon teach that to the player and let the boss do something more interesting.

Elemental rock/paper/scissors isn't typically all that much fun. Nor is "learning weaknesses". Which is basically why I settled for teaching the player "Archetypes" and then mixing and matching those things in order to keep the player on their toes. Players will naturally experiment with trying to figure out "The Weakness" of any given enemy or even the boss, and don't typically need a lesson on which elements are weak to which ones.

I mean... short of a Pokemon game where you've got like 1000 Pokemon and it's near impossible to remember the typings of them all, and you need like a guide or whatever.

If you really want "interesting boss fights", then you should adhere to a single design principle:

"Every regular fight is a lesson to teach the player. Every Boss fight is the test on those lessons."

Put simply, your Bosses shouldn't be there to "teach" the player anything. They should be there to "test" the player on things they should have already learned. Further than that, each "Test" should be something worth testing them on. That is to say... if the player never uses it again, never faces it again, or rarely does either... Then why is it a test? Why is it a boss gimmick?

Each thing you "test" the player on, should come up again and again and again, and you should iterate on the lessons you taught them. Add more wrenches into the gears. If you can't see a way to "iterate" on the mechanic, then why teach the players the mechanic to begin with? The mechanic will just get stale. Boring.

 

[Jacie0krece]

Depending on how you've framed this, you could do some interesting things with it.

Ideas off the top of my head:
1. The Boss is actually "invincible", but he "summons" enemies that are hurt by the specific element you want. When the player defeats those summons, the boss becomes vulnerable for a turn or two before summoning the next set.

2. The Boss actually "punishes" an incorrect usage of the element. If you use the wrong one, he hits you really hard or something.

3. You could make it about the player noticing "subtle differences". You could attempt to "trick" the player with an element that is "similar", but not quite. Maybe it LOOKS like it's a Water Weakness... but it's instead a weakness to Ice. You could engage in some light subterfuge with the player.

---
It's honestly going to largely depend. Personally, I don't think a "weakness change" boss is all that interesting as a "final mechanic" for a Boss to utilize, so it wouldn't be something I'd ever teach my players.

Well, not unless I was going to do something with it nobody ever saw before.

If I wanted to teach "Elemental Weaknesses", I'd just have every single enemy in the Dungeon teach that to the player and let the boss do something more interesting.

Elemental rock/paper/scissors isn't typically all that much fun. Nor is "learning weaknesses". Which is basically why I settled for teaching the player "Archetypes" and then mixing and matching those things in order to keep the player on their toes. Players will naturally experiment with trying to figure out "The Weakness" of any given enemy or even the boss, and don't typically need a lesson on which elements are weak to which ones.

I mean... short of a Pokemon game where you've got like 1000 Pokemon and it's near impossible to remember the typings of them all, and you need like a guide or whatever.

If you really want "interesting boss fights", then you should adhere to a single design principle:

"Every regular fight is a lesson to teach the player. Every Boss fight is the test on those lessons."

Put simply, your Bosses shouldn't be there to "teach" the player anything. They should be there to "test" the player on things they should have already learned. Further than that, each "Test" should be something worth testing them on. That is to say... if the player never uses it again, never faces it again, or rarely does either... Then why is it a test? Why is it a boss gimmick?

Each thing you "test" the player on, should come up again and again and again, and you should iterate on the lessons you taught them. Add more wrenches into the gears. If you can't see a way to "iterate" on the mechanic, then why teach the players the mechanic to begin with? The mechanic will just get stale. Boring.

That's what I also thought, as I decided to go with this approach:

1. The enemies in the location before have "professions" that change their stats slightly, along with making them more immune to a type of damage, e.g. a miner bandit that has a slight physical damage resistance, or an electrician that has a slight electrical damage resistance.
2. The boss summoning these troops and getting invincible until they're defeated is an interesting idea to make (it'll be tricky for me to program that, but I'm sure I'll manage it).
3. Connecting to pokemon, enemies have "natures", which are like these in base ElementStatusCore, but with an additional effect of lowering/increasing resistance to one type of damage respectively by 5%. Good if you learn it, but if you don't, it won't change too much - bosses are excluded from this mechanic.
4. Throwing wrenches into the gears? Like, what about a tanky general that has an army so big, that defeating the troops she summons is meaningless, and the player should resort to things like freezes and stuns on these, and only directly damage the general (while the game tells them that)?
5. I want to signal major elemental resistances to the player - either as a state that the player can peek into, as a visible thing (e.g. a shield with an appropriate colour), or as a natural association (it's a plumber, so probably takes less water damage), or even as the environment (it's a cold region, so people from here are more impervious to cold).
6. Also, the final boss of the nation I'm currently working on isn't evil for the team - they just want to test them if they're worthy (and strong) enough to take the later parts.

Are these good ideas or I'm again misguided by myself?

 

[Finnuval]

The best advice is already given however I will add one smaller thing to consider too when designing boss fights and that is 'risk vs reward'.
It sounds obvious, and it kinda is, yet I see it being overlooked a little too often.

Hard or interesting bosses tend to also come with better rewards, which can be anything from a good item to a cutscene, but a reward that fits the effort put into the fight often also helps cement said boss as a better experience.

Not the make or break of a boss fight and definetly not the sole thing to make a boss fight good, but at the same time it does help a good boss fight become just that tiny bit better.

That's my two cents anyway xD

 

[Tai_MT]

That's what I also thought, as I decided to go with this approach:

1. The enemies in the location before have "professions" that change their stats slightly, along with making them more immune to a type of damage, e.g. a miner bandit that has a slight physical damage resistance, or an electrician that has a slight electrical damage resistance.

How would the player know these immediately, or remember them? Do you have this for every profession in the game? What about Accountants? Things of that nature? There's a staggering amount of jobs and now I'm worried you're going to be approaching "Guide Dang It!" territory here with how your enemies work.

I assume the Electrician takes less electrical damage from all the insulated gear they wear... but wouldn't that make them slightly resistant to cold as well, since the gear they're wearing is insulated?

Are you going to fight a lot of Electricians in the game?

3. Connecting to pokemon, enemies have "natures", which are like these in base ElementStatusCore, but with an additional effect of lowering/increasing resistance to one type of damage respectively by 5%. Good if you learn it, but if you don't, it won't change too much - bosses are excluded from this mechanic.

Silly question:

If it doesn't matter to the player that much, why is it in the game?

Does the 5% extra damage ever swing combat one way or another? Does it reduce the amount of hits that are required to put the enemy down? I'd wager "no" if you're saying it doesn't matter to the player that much. Which just brings me back to, "then why waste time having the mechanic if it's not important?"

You might want to step away from "rule of cool" and start considering "what do I want my players to do, because this is in the game? Does it make them do that? Is it fun?"

Even in Pokemon, the "Natures" are largely useless to everything except min/max in competitive play. It's not a very good system, even in that game, as it largely contributes less than even IV's and EV's do. Pokemon is a game where the most damage you'll do is based basically on Levels and Type Weaknesses and everything else is "tertiary".

So, I'm not sure what a multiplayer mechanic will do for a singleplayer game.

4. Throwing wrenches into the gears? Like, what about a tanky general that has an army so big, that defeating the troops she summons is meaningless, and the player should resort to things like freezes and stuns on these, and only directly damage the general (while the game tells them that)?

Depends largely how it's framed. If the standard enemies are dangerous enough to need to be stunned and frozen, then it would work. If they're "easily ignored", then most players will focus down the boss anyway. Or, just land AOEs constantly if you have them.

5. I want to signal major elemental resistances to the player - either as a state that the player can peek into, as a visible thing (e.g. a shield with an appropriate colour), or as a natural association (it's a plumber, so probably takes less water damage), or even as the environment (it's a cold region, so people from here are more impervious to cold).

"Natural Associations" are going to be difficult. I'll tell you that up front. "Water puts out Fire". Why doesn't Earth put out fire too? Or wind?

Fish are weak to Electricity. They're not weak to Fire too? Or any weapon damage? If they're only weak to electricity because "they live in water", well, now you have to consider that for every single enemy there is. Every enemy would be weak to Wind because you could "Airbend" the air away from their lungs and suffocate them. The act of suffocation would render them unable to act in combat as well.

"Natural Associations" is something you might want to steer away from if possible. What you want to do is teach the player every weakness and which enemies they should expect that weakness to apply to.

Use Crono Trigger as the example: "Thunder stun all dinosaur, you know?".

Signposting is important. Even I do it quite a lot.

"Wolves are afraid of Fire. If you burn them, they run away!"
Cue Wolves that when you inflict the state of "Burn" on them, they run from combat.

I even advertise my "Revenge" Mechanic that all the bosses in the game have.
"It's dangerous out there. If you run into a very powerful monster, you should beware that they sometimes seek Revenge when you hurt them really bad. If you hit them with a type of damage that they really hate, it hurts them a whole bunch, but then they attack you really hard in return!"

Then, the first boss teaches this mechanic in action. It's a "safe" sort of attack that merely "stalls out" the player, rather than destroys them, so that they can learn.

The mechanic is just "it does 150% more damage if hit with their major weakness" rather than the "100%". It's enough to shave turns off of combat, but it can cost you turns as well if you aren't prepared to deal with it.

6. Also, the final boss of the nation I'm currently working on isn't evil for the team - they just want to test them if they're worthy (and strong) enough to take the later parts.

Are these good ideas or I'm again misguided by myself?

Is the Elemental Weakness thing used throughout the entire game? How so? Will it always be required to win combat? If not, then why? Can the weaknesses be bypassed?

Here's my experience with Final Fantasy 13:

"Hey, here's a Break Gauge. You need to fill this to do any sort of appreciable damage to the enemy and get anywhere. So, you should learn all the weaknesses".

Me: "Or, just get massively overpowered for this stage of the game, avoiding all unnecessary combat since I can't make my Crystarium any more powerful right now, and use Summons frequently to avoid having to engage in the Break Gauge mechanic which makes combat take forever".

You need to be very careful when you implement systems like that. Systems that are very "all or nothing".

If I must engage in the mechanic throughout the entire game, then you need to make sure that the mechanic is fun for the entire game. If it grows stale, or boring, or gets time consuming... Then the mechanic actively hurts your game.

If the mechanic is optional, then you need to decide if that's a good thing to be optional, and whether or not the mechanic should exist at all, if the players never need touch it.

You'll want to Test Play your game and look for, "Can combat be won by just mashing attack or using the most powerful skill available?". If yes, then you might want to consider doing more with the combat system in order for players to "engage" with the parts you want them to engage with.

As a player, if I have to expend MP every single round of combat to win, then there are going to be points where I run out of MP. Or, need to carry a ton of Consumables to restore that MP. There are "knock on" effects to everything you do as a dev.

For anything you implement, you need to answer some basic questions every single time.

What behavior does this mechanic make the player engage in?
Is that desirable or the projected behavior?
Is that behavior "fun"?

You will want to run Test Plays of your game frequently. With strangers if possible (friends and family will tell you things are good, even if they aren't). You will want to watch them play and note their behaviors. Add nothing as you watch. If possible, just record their gameplay and not be in the room with them, saying anything about the game. You want to see what they do, and figure out why they're doing it. You don't want to "taint" the tests by offering your input or yourself as a distraction.

Does that make sense?

 

[Grunwave]

This is one of those times where the theme/art/narrative can affect the given mechanic.

If you want to offer up the theme/art/narrative of a particular boss, I will give you some specific ideas for mechanics.

 

[FarOutFighter]

I am not the best at boss battles, either, but what I have learned so far (some of it has already been said) is to make it more difficult, obviously, but also unique, in some sort of way. Maybe they specialize in an element or state, or their skillset has a unique mechanic.