Class systems and Design choices

Having the liberty to change classes at will is something amazing but it also kinda kills the narrative purpose and uniqueness of characters. After checking Octopath Traveler and how it handled such system in a way makes it part of exploration and plot as well.
I haven't played Octopath nor have seen many reviews or even watched walkthrough videos so my assumptions may be wrong or is part of the game itself.

• Characters can't change their class
• Characters can change their subclass
• You have to find a shrine in order to unlock each subclass

Octopath has 8 protagonists and each of them has their own story. Because of this model the this system can't be opmized or i can't find a way to opmize it further, consider the following ideas as my own approach to make such system better considering plot and exploration as well.

• Each class has their own "guild"
• In order to unlock a class you must pass in the guild's test
• Some guilds requires that you have a certain character in your party that belongs to the guild's class
• You earn access to new areas in the guild's building and even items and also quests
• Guild quests may be tied with the lore

I also have a couple of ideas myself that i wanna test either in a future project. Those small pieces are personal and not ways to improve such system.

• Subclasses can be changed only in safe zones, cities or in the world map
• Ignoring base parameters when assigning a subclass (only special and extra parameters would be used)
• Leveling other classes gives extra traits (ex: lv50 Warrior grants 500 HP even when it isn't your current subclass)
• Limiting the number of passive skills by equiping them (i like broken combinations but we never know)

What's your take on it?

I think the greatest consideration for governing class change is balance, followed by overall game design philosophy. Class change also calls for menu time, which might make the game that much less fun. Ideally, there is a way for the player to quickly change class on their actors with a minimum of menu grind. Like a dedicated button to open a class change scene.

Your idea for guilds adds more content that requires you as a developer to make that many more events and the player to sit through that much more dialogue. I suggest that class advancement be tied more to gameplay than events.

Unlocking classes over long periods could be problematic. If the player has made considerable progress in one class (or set of classes) and then unlocks some more, the player would be punished for changing into the underleveled class. Personally, I think *all* classes should be unlocked as early as possible. Breath of the Wild approach.

Balancing your game around the player's ability to change classes is problematic. I don't think any version of RPG maker will allow the player to change classes in the middle of battle, so the player will be changing classes outside of battle. Which means the player will need to be able to know in advance what kind of enemies they're facing so they can preemptively adapt.

If none of the classes are unique enough from each other, then too much variety could be a bad thing. Each class should play very differently from each other. If more than one class is just buffing, casting spells, or guarding, then scrap 'em.

Equipping and mixing passive skills from among classes *will* break the game. It will lead to "one build to rule them all." But, it can lead to interesting emergent gameplay. My suggestion is to just stick to the class/subclass as the only class-based options for customization.

There is a reason why many games with class change in its core gameplay have a small number of party members. There is a reason why games with large numbers of party members don't have too many customization options.

I dunno. Class systems are hard on developers. I suggest you keep as much simple and governed by the core RPG maker features as possible.

What you dont like ninjamageawarriors?

I can dodge and barrier and block and resist and melee and ranged (magically or physically) i also s$&t gold.

@Eschaton Can't say that you're wrong but you need to consider that you can only manage the subclassing part, you're not in really deep trouble.
The number of actors in your party and classes are the same so you'll rarely find yourself in a situation where you'll not have the right character in your party assisting you.

The guild approach is something that i find more appealing ratter wander around the world map to find a shrine that unlocks a class option by chance because i have nothing better to do besides grinding. Octopath decided to go with the exploration path alone and i think it's a mistake compared to raise immersion with its own lore.

I'm actually not a huge fan of class systems for several reasons:

• Hard to balance as a developer. Anyone who's ever played FFT knows how utterly broken some of the class combinations are, as well as how useless others are. Unless you pay very close attention to all possible pairings, add severely limiting restrictions on abilities that carry over, or ban ability carryover altogether, you're going to end up with a bunch of bladegrasping Calculators capable of wiping out entire battlefields with Holy spam.
• Grindy. Generally, every time there's class changing, characters restart their new class at the lowest level and have to work on it to get the good stuff. This means a massive amount of time is spent running in circles killing boars in the woods instead of experiencing the story, exploring, etc.
  
• No real character identity. FF5 really had this problem in spades because each character pretty much was the same. FFT tried to get around it with base classes, but all that really did was make some of them mediocre (Mustadio/Rafa/Malach) and while others were game-breakingly overpowered (Orlandeau/Agrias/Reis)

That said, if I were to implement a class change system, it'd just be a role/skillset change with no mix-and-match and no grinding required. For example, Aesica is a level 34 White Mage with Cure, Raise, and Esuna. She class changes into a level 34 Monk, losing access to those spells while instead gaining Burning Fist, Hadouken, and Meditate. However, I'm unlikely to even do that, as I'd rather just make 2 separate characters, one with the White Mage skillset and the other with the Monk skillset.

I like the guild idea. It makes the world lore fit together with the mechanics nicely. There are a million RPGs out there so I definitely encourage devs to explore such ideas as opposed to conforming to all of the "best practices". You can always toss it if it does work well.

As for class change systems, I've never been a fan for most of the ones outside of D&D. Having a level 30 Fighter change to a Black Mage and forget how to swing a sword and instantly become a total weakling (and a magical powerhouse) is just too illogical for me. In D&D, you sacrifice advancement in one class to study in another. You usually lose out on the best stuff from both and end up with weaker abilities from multiple classes which is almost always less optimal.

You guys really need to see how class system works in Octopath Traveler cuz it's not like what you guys are describing at all.
I understand the frustration and that's why i've started this thread, there's so much to learn with this game about mixing class system with in game's lore and also exploration.

@Frogboy Octopath handle classes like in DnD but you loose access to active skills when you change your subclass.
You don't get huge bonuses when assigning a subclass but the extra skills increases your options in battle and this by itself is more than enough IMO.

To be quite honest, if you're going to do a "Class System" like that... I'd rather it just be like Seiken Densetsu 3. You start with a base class, you level that class up, you get to pick which class it becomes after that while retaining most of your same skills (or those skills upgrade with the class) and then you complete a small "trial" type thing to obtain the proper item and level in order to obtain your final class, which does the same as the first. This lets you "change class", but also optimize the characters for your particular play style or what you particularly need.

The only other way I would ever really implement something like that is if the "class change" was always unique to the character involved, and thus, it was an extension of the character's personality. Like say... everyone can be a White Mage... But, the guy who is normally a Thief does the role differently, has access to different skills for that class, or adds his own mechanics to the existing skills that make them work entirely differently.

I wouldn't muck about with subclasses and carrying over passives and other such things. All that invites is "min maxing". And let's face it, if you allow players to do that... there's going to be an "optimal build". One you're going to have to balance around. And if you try to cater gameplay to that optimal build, players who don't find it are going to have a tough time of your game... while if you don't balance around that optimal build, players who find it are going to find your game far too easy and not engaging at all.

Personally, I don't look at a "class change system" as anything more than a simple extension of "combat strategy". You know, akin to equipment. Because, really, that's what your classes are. They're easier to balance for and around if you think of them like pieces of equipment. If you wouldn't make a piece of equipment that was broken, why would you ever allow options in your game for classes that render things broken?

Not enough devs consider that. Your "Classes" are simply a Piece of Equipment like the Plate Mail or the Infinity +1 Sword. Unless you handle it like those things in terms of balancing... you're better off removing classes entirely.

I think @kovak is asking for a discussion around Octopath's class system...relative to their own ideas?

Ability to class change is something I could see only work for games that are very lengthy, since people will spend a long time in the game, they might want to make sure they're playing the thing they like the most. For shorter games (most RPG Maker games), people don't get invested so much and just want to complete it using what they're given.

Okay, so, if your game has class change. How am I going to know I want to use it? Like, I might have a Warrior/Monk character, and it's working out decently for me, but I feel like I could get a better deal. How am I going to know which subclass I might want to change to? How much time/resources is it going to cost to get the benefits from the class change? Is it going to be easy/convenient to switch back if I end up liking it less than my previous subclass? The more obstacles you put in the way, the less people will want to try it out. And the less information I have on what a subclass change will provide for me, the more likely I am to stick to what I currently have unless I get 100% stuck on something (but then I'd consider that poorly thought out subclasses and just quit playing instead).

Most important you have to ask yourself is: what would make a player want to change their subclass? If your game is designed well so that every class+subclass combination is viable for completion, what will drive people to want to change their subclass? And is it going to be worth investing the time in developing it?

@Eschaton The post is a little vague. At first, it sounded like that to me, but then there were lines in there like, "I've never played the game" and "I don't know how it works" and then other lines like, "Here are some ideas I have to improve it". Then, it was followed with the question, "What's your take on it?"

I've never played Octopath Traveler. I can't comment on its system. Same as the OP. I've simply based my reply on the limited information provided on the subject of Classes and Subclasses.

To me, the post was sort of a discussion on how to improve a Class System or make it feel more organic and less like you could game it or have it be overpowered.

That being said, I think Class System only really makes sense from the standpoint of "It continuously improves" and you can't "go backwards" in it. From a narrative standpoint, this makes the most sense. After all, it's how it works in real life. You do not start out as a Garbage Collector for the town you live in and the next day turn into a Police Officer and then the next day turn into an Office Supervisor. You take the skills you learn from "job" and they grow into new jobs as you get better with those skills. Sometimes you would take something outside of your skillset, but the way you would approach that new job (or Class, since the two words are interchangeable in RPG terms) would be in relation to all the job experience you had up to that point. Namely, you'd use skills you already have to put your own spin on the new job, to make it easier for you to grasp.

That's why I think the best way to approach a Class System is simply to have a base class that upgrades as you go along (like Fire Emblem or Seiken Densetsu 3, or Final Fantasy Tactics) while also removing the ability to travel backwards on it.

Class change systems are overrated, if I'm to be asked. They're cool, but "cool" does not always mean "fun" or "balanced." Seiken Densetsu, or hell, even the original Final Fantasy are examples of class change done with economy and balance that's well tied to the story and not in an overly-grindy way. They're flavor. Like @Tai_MT said, they're just another means of altering an actor's stats.

I believe that if a player is going to be given multiple class options, each one should play very differently from one another. Their core gameplay mechanics should be very different from other classes. But RPG Maker, even with Javascript, isn't powerful enough to change its gameplay in very meaningful ways. The best the average RPG Maker dev can do is create differing flavor for the usual "attack," "magic," "buff/debuff" mechanics of the default engine.

Personally, I am not much of fan of the class or subclass system.

My current main project I set up for high degree of personal freedom for the player in regards to their own characters; as such I tried integrating a direct class change mechanic, a direct class change & subclass change mechanic, a fixed class choice with a directed subclass change mechanic, an evolution based class change mechanic (class changed defined by variable evals derived from skill usage), etcetera, etcetera.

I rebuilt the system from the ground up at least half a dozen times, & each time the results ended up the same.

Specifically, you either make your classes defined enough so that they aren't effectively interchangeable; in which case you make some classes that are effectively hobbled by their own constraints, when compared to other classes. So your player has no desire to choose them, & those characters that use those classes, the player is keen to replace as soon as possible.

Or you make your classes flexible enough, that many perhaps even the majority of classes are entirely superfluous. So either changing classes becomes a largely irrelevant mechanic, or one where most of the choices on the menu are irrelevant, save for the few that provide something unique compared to all other classes.

Finally I ended up discarding not only class changes as a mechanic, but classes entirely. In favor of an organic playstyle based evolution of player characters. The player uses a limited character creation suite to determine the sex, name, appearance, starting parameters, starting weapons, starting armor, & starting skills of each of the four characters in their party. They learn new skills each time they level up, but what skills they can learn is determined by what skills they have learned previously. Their parameters evolve over time based on playstyle (weapons used, armor used, skills used, etcetera), in place of an assigned class. & the characters appearance is determined by playstyle as well.

Personally, I rather like your guild tie in idea; however, I also think it's potentially rather limiting. Why can't your Ogre slaying adventurer who joins the fighter's guild, be a sneaky rouge like assassin, who is just in it for the money? Why can't your character your who joins the squeaky clean band of holy knights, be a bloodthirsty "kill them all & let the gods sort them out" berserker, that normally is played as a Dark Knight/Black Knight?

I think when it comes to fantasy in any medium, but especially in RPG games, we too often have a reflexive "this is just the way it is" perception of the tropes. We have to actively remind ourselves to consider, am I using a trope because I think it's the best way to tell this story, or simply because it is what I think is expected because it's a trope?

When I initially shrank my player characters down to a single band of "heroes" in my current main project, my plan was to have each one of them have their own personality, & the evolve within the usual the scope of their archetype within the RPG trinity. But, then I realized I was taking away player choice for no reason other than fulfilling tropes. Why can't the self serving hedonist who only cares about gold, wine, food, & getting laid be a mage instead of the typical rogue archetype? Why can't the taciturn & levelheaded stoic also be the teams sneaky rogue? Why can't the team's cerebral, smartass, sarcastic, bookworm, also be the one that wears plate armor & swings a battle axe? Why can't the teams heroic leader, be the healer in the back ranks, rather than the front line warrior with the McGuffin sword?

So I decided that each of the characters basic personalities would be preset, rather like their hair, eye, & skin color; but that the player would decide what skills & attributes they developed, just as their decisions determined the character's appearance via the kind of armor they wear & color scheme.

Challenge ... accepted.

Interesting. I look forward to the "class gameplay mechanics ideas" thread.

@XIIIthHarbinger The design itself can be improved further, the "guild" is a generic way of thinking about a place that offers training, pretty much like a school or college, so it gonna look way less refined in terms of plot development when you cross classes with actors and their personalities under the idea of "giving soul to the characters".

It doesn't mean that the Paladin that servers under the church and has a huge bloodthrist for slaying infidels cannot still be the Paladin that serves under the church and has a huge bloodthrist for slaying infidels...all he did was to train to get access to new skills of another class. I know that people would argue about it as a reson to enrich my plot but if it's for the gameplay sake i could just ignore this "richness".

It seems that i'll have to face the challenges that i'm taking about to learn and see if it's a mistake or not.

So I did some quick research on Octopath Traveler and came across this
http://www.shacknews.com/article/10...e-bonuses-and-skills-list---octopath-traveler

I then noticed there are some huge differences between stats with each class and subclass. Apparently there are also 4 rarier classes that the player can get later on in the game still at these "shrines" but with the requirement of having to defeat more challenging bosses.

I also read in Octopath Traveler the subclass stats don't affect stats growth and you can't change main classes only subclasses like you said.

I also read that when you unlock passive subclass skills and then swap subclasses, the same passive skills remain unlocked no matter what. Woah! That surprised me.

But with your ideas, I would agree with Eschaton that it'd be best to have "main" classes/subclasses unlocked early on if you go with that approach, since you'll want them to balance out in a way that each has it's own advantage and disadvantage while they still each remain useful in certain areas of the game.

However, with Octopath Traveler it looks as if all the classes are useful but with each main character class some subclasses aren't advised. So some subclasses are not as good when compared with pairing with other subclasses, yet with all 8 actor classes all subclasses with the 8 of them together still have some use, and I think that would always be the case even with a good subclass system, since they can be tricky to balance out.

If you decide to do the shrine thing though, or your expanded guild verison of it, then that would naturally lead you to a game where a lot of locations are unlocked near the beginning.... and I have mixed thoughts on that.

I'm sure it can work somehow, but you'd need to really think everything out ahead of time.

My thoughts with a way around it is to have all these main locations for swapping subclasses unlocked early on in the game with a clear manual on the different advantages per class for the player to get used to the system, and keep the shrine/guild locations short distances from the other on the world map, then have real/extra locations in the game after that where most of the game takes place while still making it easy for the player to change subclasses.

The main problem I have with the "guild" system is, if it's not easy to get to each guild and meet the requirements to change subclasses, then it'll be difficult for the players to switch and change them with ease, so if they make a mistake they'll be greatly punished for it, and I'm against that outcome without a quick solution to fix it....

Could you also elaborate on what a "guild test" would be?

Think about a requisite to join the "guild". They don't need to be overcomplicated nor to have many steps. This is up to the developer, the test could be a puzzle, a payment, mob slaying, finding and then delievering an item, etc.
Also consider that the "guild" is a place for trainning, so it provides the basics only. If the test is too hard then only strong people will be able to face it. The test works as an entrance fee to joing the "guild" and also to unlock a class and maybe some benefits from the said place.

Thinking about benefits you could have access to some stuff if you have a character who already belongs to this "guild" but he wouldn't be able to join the "test" since he has done it before (it could also lead to cheating). I think it falls into the plot part as well if you dive deeper into it by the time.

What do you think of the way The Elder Scrolls does guilds? Skyrim and Oblivion reduced guilds to just more questline with equipment as rewards, but in Morrowind and Daggerfall, guilds were more mechanically entwined with player character development.

In the older games, advancement required the player to have met minimum stat requirements, and Morrowind even had stat requirements for admittance.

But what was always missing was the large impact being in the guild had on character development. Sure, youd get nice equipment and access to spell shops and training. I'd like to see flavorful rewards. Passive skills with a well-described guild-related reason for why your character is stronger, faster, or smarter.

I haven't played the older TES games and since there isn't much to do in those guilds i find them not so close to my idea.
The only guild that i've found interesting when playing SKyrim were the Dawnguard (i know it's DLC related but you had to help to rebuild it not only by recruiting NPCs but by doing so you were also repairing it as well) and the Dark Brotherhood (their questline was supposed to be longer).
You had better immersion with both of them compared to the other guilds.

I'm aware that questlines were better in previous TES games but you had nothing better to do after finishing them, no reasons to stick around for a longer time.

In Elder Scrolls Online you have skills related to guilds but i don't find them as good as they were supposed to be.
https://elderscrollsonline.wiki.fextralife.com/Guild+Skills