On The Subject of Secret Bosses

[MAIR]

So, I'm someone who likes to min-max in games, and love to be rewarded with a boss that really tests my skills and stats. most of these tend to be accessible either in new game+, or in the endgame. My current idea is to have two or three optional bosses throughout the game, which are exceptionally strong for the points in which they appear. My biggest concern for this is that it could really mess up game balance. Say, someone wants to fight these bosses, and grinds to do so. This ends up producing a severely over-leveled party, since the player would likely want to grind to stand a chance against this boss. What would you say the best option for this would be? Should I make the bosses new game+ only, where the player is likely already over-leveled to begin with, or should I drop the idea altogether?

 

[Andar]

Option 3: make them challenging by requiring other tactics, not by pushing numbers.


I don't know how you plan to balance your game, but a lot of game developers (even professional ones in AAA-companies) confuse difficulty with higher numbers.


Giving higher numbers requires the player to grind to get similiar higher numbers before the usual attack-spamming works again, but that doesn't make the fight itself more difficult for the player - attack spamming remains attack-spamming.


But if you use a way of balancing where the player needs to develop a strategy on which skills to use, then you can make the bosses more difficult by requiring a more complex tactic instead of higher numbers with attacks.

 

[XxXhelazz]

Make a variable in a field. That variable will be the level of the party. When the variable is more than "X" then you can fight them. That what I did xD

 

[MAIR]

@Andar I did have some ideas for boss mechanics.


1. A boss that forces the player to go on the defensive for a few turns, before leaving them self open.


2. A boss taking a party member away, or completely nullifying an actor's attacks or skills. Unless the player does something about it, of course.


Take Persona 4's secret bosses for example. A secret boss appears in the last dungeon you finished, but you probably shouldn't fight it until you finish the next one. Or, do what I did and grind until you stomp it, and every other enemy that stands in your way. I just want to prevent the player being a bit too over-leveled. 

 

[Liak]

I agree with Andar. Bosses that are only challenging in terms of raw numbers and can thus be overcome by extensive grinding aren't that great. While I do enjoy some amount of grinding every now and then (because of the mindlessness, it's a peaceful activity, much like puzzling or watching TV), I wouldn't really think of it as great game design when the creator forces me to grind a lot at a very specific point in the game when I'd much rather progress. There's a reason why this kind of boss belongs in the postgame where you give players who just can't get themselves to stop playing SOMETHING to work towards.

 

[VicWhite]

I agree with @Andar on this subject. Not only numbers.


IE: Special equipment, and a quest to find that equipment before, could be great. For example a spiritual sword able to wound a ghost. That ghost can be a challenging foe but impossible to beat without that key item.


Secret bosses add that special thing to a game, great idea to put them.

 

[Wavelength]

Actual difficulty over higher numbers is good, but if the game allows you to level up and get stronger with fights, then it's a nice thing to have a few optional super-bosses that the player needs to be level 1200 to feasibly take on.  I like this kind of thing because if I find the combat system fun, I'll endlessly "grind" battles in the postgame with the express intention of getting more powerful (and being able to take on the superbosses if there are any), but with the underlying reason that I just enjoy running around and fighting stuff.


Don't save it for a New Game +, and don't drop the idea altogether.  Just put these superbosses in your postgame!  Make them available only after the player has completed the main plot (or when they reach the game's final save point, if you don't allow the player to save and run around the world they just saved after they defeat the final boss).  That way, balance will never be thrown out the window if the player grinds enough to beat a superboss.

 

[Dr. Delibird]

Yeah like others have already sad, if you design around the knowledge of intending to implement a super-boss of sorts then balancing shouldn't be a problem.


A pretty good example of a post game content type of boss is the raid bosses in borderlands 2, more specifically before the level cap increases as the cap increases cause a bunch of imbalance. They all had mechanics that most regular bosses didn't and they had different quirks that made them unique both from a design perspective and from a difficuilty perspective. The main mechanic that was added in these raid bosses where frequent use of nova blasts which are largely un-avoidable, for better or worse this made sure the player could not just stay just outside of range and pick off the bosses slowly. However these raid bosses where not designed around the level cap increase and now at max level (not including the OP levels as that is an optional difficuilty spike) they are poorly balanced ("bullet sponges", novas kill you if you are not at 1/2 of max health + 2, your damage has not scaled nearly as much as theirs).

So basically as long as you design the super/secret bosses to work with your games balancing in mind they should not be any more difficuilt to balance than a regular boss.

 

[kurt91]

I just finished playing "Hyperdimension Neptunia Re;Birth 2", and it did something interesting with the ultimate boss.


The battle system is a typical turn-based affair, but on your turns, you can choose between "Rush" attacks that fill your EXE Drive meter faster (a shared Limit Break meter, essentially), "Power" attacks that do the most damage, and "Break" attacks that deal medium damage but work the best at breaking an enemy's defense. You can attack multiple times during your turn, but too many actions will actually delay your next turn.


The ultimate secret boss, Delphinus, is strong enough that a full team that's hit the level cap will only survive two hits, and you cannot heal fast enough to keep up, even with the best healing items and spells in the game. Hitting the level cap is only going to get you possibly a single mistake on your part, IF you're lucky. The key isn't to force your way through it, but to prove that you've mastered the game's battle system. A couple of your characters have Limit Breaks that can delay the target's turn, and one character in your team can slightly delay a target through one of her Break attacks. If you've figured out exactly how much each action will cost you in terms of turn order, you can actually manage to constantly delay the boss so that it never actually gets a turn. However, a single mis-count on your part, and it will blow you away in a single shot. You need to know EXACTLY which attacks to use to balance delaying the target, filling your EXE Drive meter, and when to stop your turn early to ensure that your next turn isn't pushed back through too many actions.


Is your battle system complex enough to do something similar, where any dimwit can brute-force their way through the main game with only mediocre strategy at best, and the secret bosses require full knowledge of what techniques and strategies do what in order to win?

 

[Matseb2611]

Suggestions:


1) Put them late in the game, where level grinding no longer plays that much difference, and where players have already gotten their characters built to the standard they're happy with.


2) Don't have the traditional level up system at all. That way, you're encouraging players to use tactics and planned preparation rather than grinding up to higher levels.


3) Make the boss adapt to the player's level. So if the player reaches it at lvl 30, they'd fight version 1 of the boss, if they reach it at lvl 40, they'd fight version 2 of the boss with stats that have been ramped up accordingly, and so on.


4) Make the boss' attacks utilise both %ge HP damage and flat damage in the damage formula.

 

[bgillisp]

, and grinds to do so. This ends up producing a severely over-leveled party, since the player would likely want to grind to stand a chance against this boss.

I've never understood why designers are so worried about this. If a player is going to grind to beat a super boss, they are likely to grind ANYWAYS and they will still mess up your balance. Remember, there are players who grinded to level 99 in Kingdom Hearts on the first dungeon just to do it. I say don't worry if they are overleveled, in fact, maybe being overleveled for a few areas is the reward for beating the super boss.

3) Make the boss adapt to the player's level. So if the player reaches it at lvl 30, they'd fight version 1 of the boss, if they reach it at lvl 40, they'd fight version 2 of the boss with stats that have been ramped up accordingly, and so on.

No offense, but terrible idea. If the player can't handle the level 40 version (or the powered up version), they just irreversibly made it so that they can never beat that boss, especially if they are at the level cap. That's why I'm against level scaling of any sort, because you cannot undo leveling up to try another approach to it, but almost all other decisions in games are reversible (to a degree). Plus, I feel it punishes the player for the time they spent grinding, which, should  still be *a* viable tactic when possible (just maybe not the ideal tactic).

 

[Vox Novus]

As Andar was saying, I prefer bosses that have a particular strategy you need to overcome by figuring out their actions and then what you should respond to it. You could make the secret boss represent whether the player has mastered the available options to them at the time of finding the secret boss.


As far as grinding goes, you aren't going to stop players who already grind through areas (as bgillisp said) unless you have some sort of check in place that limits growth of the player at that point in the game.


Leaving the player the option to come back is nice when they feel they are ready, lets them move on with the game while still gaining growth for their characters without normal grinding (still won't stop grinding though).

 

[Matseb2611]

No offense, but terrible idea. If the player can't handle the level 40 version (or the powered up version), they just irreversibly made it so that they can never beat that boss, especially if they are at the level cap. That's why I'm against level scaling of any sort, because you cannot undo leveling up to try another approach to it, but almost all other decisions in games are reversible (to a degree). Plus, I feel it punishes the player for the time they spent grinding, which, should  still be *a* viable tactic when possible (just maybe not the ideal tactic).

The scaling would obviously have to take the difficulty into account. So the version 2 of the boss still provides the same level of challenge for a lvl 40 character as version 1 did for level 30. It's not like it would be significantly harder suddenly. Also this was a rough example. You could make the boss have many more versions and each one only slightly scaled up, so it feels less granular.

 

[bgillisp]

The issue though is sometimes that challenge turns into reading the developer's mind. Well, what if the player never figures out what cute trick the developer had, at all challenge levels? If you take away grinding, then they have 0 ways to beat the boss. That's my issue.


So on that note (and back on subject), I'd say that if you are going to make a secret boss rely on strategy, please make sure it is a strategy that the player has been learning throughout the game and has a reasonable shot of knowing/understanding. Don't make it they have to use an obscure skill they've ignored since level 5 because it's been useless all game and NOW its suddenly the only way to beat that boss. No one is going to figure that out short of either dumb luck or a guide. In fact, those are so detested, they have a tvtropes name, guide-dang-its.

 

[Vox Novus]

The issue though is sometimes that challenge turns into reading the developer's mind. Well, what if the player never figures out what cute trick the developer had, at all challenge levels? If you take away grinding, then they have 0 ways to beat the boss. That's my issue.

I'd say that's not the biggest deal though in a completely optional secret boss. A standard storyline boss sure, but an optional boss I see far less of an issue with. Not saying that the fight should have such an obscure strategy to figure out though. 

So on that note (and back on subject), I'd say that if you are going to make a secret boss rely on strategy, please make sure it is a strategy that the player has been learning throughout the game and has a reasonable shot of knowing/understanding.

Definitely an excellent point, like don't throw a skill early on to break special barriers on an enemy and never have the player use it only to throw it in later. Also no Yunalesca moments like in final fantasy X where you need to have a specific status ailment on you to even possibly survive the fight when the norm is to remove status ailments.


I would say that a strategic boss fight doesn't only have to be about the skills available to the player. It can be about how the enemy uses their skills to. I'll use something I'm doing in my current project as an example; a boss and his goons use a skill that steals money from the party then later in the fight after receiving some x% damage the boss will use an attack that's damage is equal to the amount of money stolen in the fight so far. In this scenario it is a lot less about the players available skills rather its more about their interpretation of what the Boss is doing and then making sure to take out the goons quickly before they can steal. Of course damage can be avoided by starting with little to no money in the fight but that to is part of adjusting strategy for this particular fight.

 

[Matseb2611]

Well that's true, no-one enjoys a strategy so obscure that it doesn't make logical sense. As @Vox Novus said, it's all about communicating the right information to the player. So whatever the strategy the boss uses, the player needs to understand what it is and how they can overcome it. However, this is not the point here though. It seems rather irrelevant to my idea of having the boss "level up" parallel to the player.


In my opinion, grinding is such an outdated and faulty way of overcoming a challenge. It provides nothing for the player aside from wasting their time. If they grind in order to beat the boss, then they've not really grasped the essence of the game mechanics and simply ramped up their stats in order to beat the boss. Probably would be easier to just provide them with a stat-boosting cheat. And then, what of the games that don't use the traditional levelling up systems? Should these games not have any challenging bosses at all because there's no grinding option in them?


Also, we're talking about an optional super boss. It's not too much to expect for the player to be on top of their game for these.

 

[bgillisp]

Maybe it is outdated, but I still feel you don't want to get rid of the last resort way to beat something. Just for those players who are struggling. Otherwise, it turns into the idea of just telling players to get good, which does not go over well nowadays.


Basically, I just don't like the idea of a boss having one way and only one way to beat it, which is what it can turn into if you level scale it.

 

[Leysos]

Maybe it is outdated, but I still feel you don't want to get rid of the last resort way to beat something. Just for those players who are struggling. Otherwise, it turns into the idea of just telling players to get good, which does not go over well nowadays.


Basically, I just don't like the idea of a boss having one way and only one way to beat it, which is what it can turn into if you level scale it.

The whole point would be rewarding players who have grasped the battle system. Those who have gotten good. Grinding a boss away defeats the whole point, and devaluing the achievement of beating it by making its victory rewards accessible to all players is a much worse crime in my eyes. Like @Vox Novus said, this is a secret boss we're dealing with. Gating story content behind difficult bosses is kinda scummy to me (war flashbacks of that dragon boss on the airship from FF10 come to mind), but holds shouldn't be barred against an optional boss designed to be difficult.


That's just my addition to the discussion. I have read up and do agree that the strategy should be learned over the game or at least be made very clear (it's why I have so many qualms about the final level of Valkyria Chronicles), but it's the mastery of said strategy that's being tested with this kind of boss.

 

[Wavelength]

I agree with @bgillisp that if you have leveling/grinding/growth-of-stats in your game, then that should be a valid way to make any battle, including superbosses, easier.  It is a core concept that you have taught to your player the whole game long and breaking it can feel very frustrating.  It doesn't mean that the battle needs to become easy-peasy with grinding (if the superboss is now doing 20% of your HP with every hit instead of 40%, you still need to figure out the strategy to beat them, but you have somewhat more wiggle room now), and it also doesn't mean that superbosses should only be piles-of-stats without interesting strategies.


But ideally speaking, the strategies to take it down should be diverse and flexible enough that anyone who is very good at battle strategy in general (and appropriately leveled) should be able to beat it.  A boss with unique mechanics is cool; a boss that can only be beaten by the player using a specific series of actions is nothing other than a cheap gimmick.

 

[Kes]

I personally loathe bosses which can only be beaten one way, the way the dev has decided it should be.  I don't always think the same way as the dev does, and I would like the possibility of a boss to be beaten by a variety of strategies - though obviously within the framework that I've been playing the whole game.  I am also uneasy about the whole notion that only a select few who think exactly like the dev should be allowed to win, and that anyone else should, by definition, not get a look in.