RPG Equations

This is a really vast topic and in the end it'll come down to personal taste. There are thousands of RPG games out there and nearly as many implementations of how stats affect outcomes. Even "The" D'n'D rulset (or whatever it's referred to as), which is one of the largest and most robust declarations of all of this stuff, is constantly changing and evolving. Look at the difference between old Sierra Dragonlance games, then Baldur's Gate, then Neverwinter Nights, then Dragon Age: Origins. You can sort of draw a line of spiritual progression through those titles and the rules changed immensely at each turn.

My personal preference is always to have a levelling system that forces compromise. You want to be making choices that craft an identity, and then the way you make your way through the game naturally begins to follow those choices - you look for weapons that especially match that identity - you approach each situation in ways that keep your weaknesses guarded. That's a rewarding RPG for me.

A brutally strong thug is not the same as a sneaky ninja, and those two respective toons will have different advantages and disadvantages against different enemies. It's a great layer of depth to navigate. To me, levelling becomes an utterly pointless exercise when all roads lead to a homogenous superhero that can handle anything.

I don't like to be railroaded either though. There's no good reason a wizard shouldn't be able to learn to use a weapon effectively, unless he's 103 and has the body of Mr Burns. He would compromise some spellcasting effectiveness (by not concentrating all levels in that area), but that's a valid choice.

Of course, this is logistically difficult to program, because you need ensure that all archtypes of character builds are capable of beating the game in solo - this is different to your options when designing the levelling system of a multiplayer RPG, where players can (and should) rely on one-another.

Also.. I've always hated the luck stat. I don't know why, but I always ignore it, never train it, and just generally wish it wasn't there. Maybe it's because I don't feel like luck makes sense as an actual "stat". Not as something you can train and increase, anyway.

The best info you're going to get here is from playing some well known RPGs and observing how it's done. Better yet - observe what you don't like and think of ideas to improve.

I have a huge number of ideas on this topic, but I'm not going to flood this thread with pages and pages of ranting about it. One day I'll eventually get back to working on my own RPG and prototype all the stuff in my head.

This particular subject is just about the most fun aspect of coming up with a game like this though.. you should be scribbling down insane diagrams like a madman and trying different things out.

For me it always comes back to fostering unique identities. You want players to feel like they've created a *unique* hero. Some magic sword you find in a chest should be the most awesome thing ever not because of it's own stats per se, but because of how it's incredibly suited to your levelup choices - you've FINALLY found this amazing piece you were looking for, etc.

Getting it right will only come from a lot of testing. Also, it's more important to make it fun than to make it right. It could take years to find some kind of true stat equilibrium, even a simple small-scope game. It's a worthwhile thing to spend some time on to a point, but don't get totally lost in it.

If you really want to go deep into this, you could add some anonymous analytics data to your release-build game sessions (use something like Flurry). Track things like - if some one is easily winning every encounter all the time no matter what - what weapon do you they have, what's the stat sheet, etc... This is kind of balancing occurs constantly in online games like WoW (and even COD/Battlefield), because the information is freely observable on the online servers. With a single player game, you'd have to track it and datamine it that way.

Two games that come to mind with extensive weapon sets are Diablo II and Demon's Souls (/Dark Souls). Look at Wikis for those pages and then find the weapon listings.

Another strange but potentially useful info source might be if you can find some recap pages from the "Deadliest Warrior" tv series. Although I find that show pretty silly most of the time, they do often go into some really interesting history about the origins of certain weapons, as well as their respective strengths and weaknesses.

Good luck with it. I'd also love to do a dungeon crawler but I have other projects I need to finish first.

 

Yeah I think keeping it simple is good, but when you boil everything down to a core set of stats you want to focus on and build from, there are still design decisions that can be made to create a sense of individuality from levelup choices.

For combat it usually (in terms of common practice) boils down to three primary things:

- strength (hitting power - either universally or only with say blunt weapons, ability to use heavy armour well)

- agility/dexterity (dodging, ranged attacks, sometimes hitting-power with agile weapons like small swords)

- magic/mana/sprit/energy/insert similar word (efficacy with spells, ability to access spells, efficacy with magic items & artifacts, etc)

Having a speciality in one of the above areas makes up the bulk of the most commonly seen character-build archtypes.

I'm always disappointed though when only presented with an option to choose an archtype that automatically levels up. It's always the most fun when I have freedom of choice to mix and match. Even when I make bad decisions and create an ineffective build, that's part of the fun and will simply drive the desire to replay a game (if it's good).

As for equations, I think the first decision to make is whether you want a dice-roll system or a rigid-outcome system.

Take these two games for comparison:

Neverwinter Nights is a classic DnD dice-roll system.
-
Human has an "attack bonus" of 4, Goblin has an "armour class" of 13.
Human attacks Goblin - d20 dice is rolled, lands on 7 - attack is 4+7 (11), does not beat AC of 13, attack misses.
Human attacks again, this time rolls 11, 11+4 (15) beats AC 13, attack hits.
Dwarf has "attack bonus" of 25 (extremely high level character), dwarf attacks Goblin, rolls only a 2, attack hits anyway. There's basically no chance to miss, unless a 1 is rolled, which in the NWN system, means a failure regardless of stats.

the above examples are mirrored in every aspect of Neverwinter Nights, and it's all intricately balanced. e.g. a paralyze spell is cast at an enemy - a number to beat is generated based on various stats of the caster, then the target rolls a d20, adds their willpower bonus, and resists the spell if the roll+bonus is higher.

Then, for comparison, take Demon's Souls:
-
player attacks enemy - if they are blocking, little or no damage, if they are not blocking, apply full damage - no variance. If blocking, apply damage in stead to stamina, and when stamina is eventually depleted, blocking is ignored.

Demon's Souls obviously uses a combat system that's expertly crafted for a real-time action game, but I don't think the ideas and the mechanics on offer should necessarily be completely ignored for a game with a much simpler scope.

I absolutely love the strategic element of being opportunistic with sequences of attacks, pauses, blocks, dodges, etc. It's very satisfying to control a battle strategically rather than crossing fingers on dice rolls.

But again, I really really recommend grabbing a bunch of good RPGs and observing some of these painstakingly crafted and balanced systems. There's a wealth of fantastic work out there to leverage and build upon. I would even pay special attention to strategy guides, where players have actually taken the time to dissect and analyze how the mechanics of a combat engine unfold, and how to take advantage of them.