Designing Character Classes: Introduction

Beat, loot, repeat was born from a desier to create a roguelite ARPG focused on class fantasy. Each class would have its own lore, thematic abilities and several playstyles to explore.

This is the first part of a devlog series on class design in the game. Here, I will go over general thoughts and ideas when creating character classes. Following posts will be dedicated to specific classes.

Class fantasy is the reason this game exists

I am a huge fan of class fantasy as a concept. I cannot really figure out why, but having a fantasy attached to whatever I'm playing can make or break a game for me. I'd rather play an interesting class fantasy than a story-wise interesting character. I loved playing Final Fantasy Tactics Advance for the sheer amount of classes available, and for the ways those classes were unlocked. And I really dislike it when the "class" specialization only gives you a numeric benefit.

So I set out to create a system which would support all my dirty class fantasies.

The class/spell system is currently the main cornerstone of the game, with a lot of functionality to power all sorts of spells and abilities which can come to mind. However, thinking about different kinds of RPG and ARPG games, the philosophies between their "spell" and combat design are extremely different. For me, it was a realization I had when already deep into the spell system design: the majority of ARPGs and roguelikes aim for a simpler spell system and focus on making the combat dynamic and fun. What I have is more like Diablo or PoE, which are extremely fun games, but not exactly roguelikes. Unless you play on hardcore...

I want to talk about what the general thoughts and ideas were behind creating currently existing classes and spells.

Right now, there are three classes in the game: the Blademaster,  the Witch and  the Dragonslayer. These are not your classic warrior-mage-priest et cetera class fantasies, so it might be a little harder to immerse into one of those. But I believe that the names, in-game descriptions and presentations can create a desirable class image, After all, you  can somewhat anticipate what every one of those is about, if not necessarily in the playstyle, but rather in theme.

Unholy Trinity+

Tanks, healers and damage dealers are the "holy trinity" of RPGs. There's another trinity prevalent in RPGs(which is often expanded into more than just a trinity.. But as someone who played too much WoW and Diablo, it's a trinity for me): Strength, Intellect, and Agility. While it seems to be a means to customize your character, there's usually a "main" stat out of those three which you benefit from, while the other two's effects are negligible, if exist at all. I tapped into this trinity for my class system: each character has str, int and dex, but only benefits from one of those. It's mostly a way of reinforcing class fantasy; I could functionally replace all three with a single "damage" stat, or give it a thematic name for every single class. I'm thinking about that option, but the "unholy trinity" is often so tied to the class fantasy in the minds of players(I personally cannot make myself play a STR class, ever), that removing it for functional simplicity can take away from player experience.

Class Inception

The first two classes created(the Blademaster and the Witch) were built as a means to expand the capabilites of the spell system. The way their spells were created went somewhat like this: 1) I thought of a spell in a different RPG/ARPG game I like 2) I though "does my spell system support something like that?" 3) I made sure it did, and a similar spell was put in the game. There are no huge spell adoptions which really stand out; "making sure they would work" mostly meant being able to control how and when Effects are applied(on hit, on kill, when using an action, when taking damage etc) and how they are managed. (An Effect is the basis for interaction between spells and units. I guess I will elaborate more on how the whole spell system is organized, but that will be in a different devlog.) 

Since then, both of these classes have been touched up and expanded upon with more Spells and Passives, making sure they are not mismatched messes, but rather - thematically-consistent playable "fantasies".

Class Design Philosophy: Fantasy

Class fantasies are, in my opinion, a key factor in any class-based game. They consist of visuals, lore, playstyles and mechanics.

Since I don't have an artist and all my characters are, unfortunately, just small squares, I don't have to worry about visuals, for now :D If you are an artist and you are interested in this game - please send me a message.. :)

Lore is both the story behind the character or the class, and what it represents. It's a curious differentiation: your character could be a "named" hero who has personal impact on the lore, or a representative of a group of people with a common background. In Beat, loot, repeat you play as a single character with story significance, and the ability to switch between classes and spells is granted through the use of "memories". This idea lets me use both the "person-based" classes and "group-based" classes. You unlock classes as you discover memories. Some of them could be of generic fighters belonging to a prominent group - for example, you assimilate memories of several Dragonslayers/one of the generic Dragonslayers. Others - you could find memories of a specific, prominent and feared Witch, and, essentialy, channel that specific Witch's powers on the battlefield. These are just examples, no actual decisions like this have been made regarding the Dragonslayer or the Witch. While this difference will not have any real gameplay differences, it is a cool flavor and can really augment the class fantasy. Another idea - all classes can be "group-based", while a class advancement system can be implemented with "person-based" classes. For example, you start off as a "Blademaster", and eventually advance to "Blademaster Ren". Doing that gives you an extra skill or modifies existing skills in unique ways. Current systems easily allow for such manner of modifications.

Playstyle is all about how you precieve the class should play. For example, a Warrior class is often expected to have STRONG attacks which deal a lot of damage, and should survive in the heat of battle. Giving a warrior a damage-over-time or debuff oriented build can be an interesting variance for Warrior class fans. But if the core fantasy of strong attacks and tanking capabilities are not covered, it can drive away fans of the class. I am using fairly unorthodox classes, which frees my hands from common playstyle tropes, but poses a challenge of delivering the message about the fantasy to the player in a clear and interesting fashion.

Mechanics are a baby of lore and playstyle. They fuel the playstyle, but should take their root in the lore. For example, the Witches, by their lore, come to the playable world  from the Other Side, which is a volatile space beyond our plane of existance. It is inhabited by vicious lifeforms, and Witches learned to control some of them. As a result, one of the Witch's playstyle is focused on summoning Locusts: tiny insects which gnaw on the enemies, and dissipate back to the Other Side after a short period of time. When creating classes outside of common fantasy trope, I feel like designing lore is key to creating unique and interesting playstyles.

Class Design Philosophy: Builds

Beat, loot, repeat is a roguelite where you loot Spells, Gear and Passives as you go. That means that every run your character and their build sequence is going to be unique. The primary goal is to create several themes for every class, with multiple spells and passives to support each of the theme. The secondary goal is to create systems to control, to an extent, the random nature of roguelite loot, to provide the players with a way to steer their character towards a desired build.

As a result, every class is intended to have 3 "main" playstyles within that flexible paradigm, which can be achieved with a few spells, passives and mods. Ideally, 2-3 spells, 2 passives and 1 mod(spell modifiers found on gear) would form a build, but right now this isn't strict at all.

Due to the random nature of roguelite loot, there can be a lot of inconsistency and imbalance in the possibility of putting together builds during the run.  The fact that, to upgrade spells/passives, you need to find duplicates, complicates this even further. I came up with several ways to control this:

1) Balance the amount of spells which fit each of the playstyles. I decided not to go with this option, since artificially making sure there are equal amounts of spells and passives that compliment each of the builds feels too artificial.

2) Create loot "buckets". Spells and passives could be split into "buckets" based on builds which they compliment best. Given 3 build per class, every class would have 3 loot "buckets". Whenever a player gets loot, it would first randomly choose a "bucket", meaning that whichever of the 3 general builds you favor, there's an equal chance to get the loot for a desired build. This is a consideration, but isn't yet implemented.

3) Transformation. There can be a way to transform spells and passives, with limitations. I mentioned upgrades before: each spell and passive exists in 5 Tiers. To get a Tier 2, you need to combine 3x Tier 1s. To get a Tier 3, you need to combine 3x Tier 2s. And so on. Right now, as a means of controlling the build RNG, you can combine one spell of Tier X with a different spell of Tier X+1, to consume the latter and turn the former into Tier X+1 version. Fore example: you have a Tier 1 Slash and Tier 2 Sheathe. You can combine the two, consuming Tier 2 Sheathe and recieving Tier 2 Slash. This is currently in the game, behind a small unlock cost.

That's it! For now...

For now. The nex devlog is going to focus on the Blademaster, the first of the designed classes. If you're interested, but don't wanna wait - check out the game! The Blademaster, the Witch and the Dragonslayer are all available to play.