Hi there,

Do you guys know of any good articles/books discussing RPG mechanics?
By mechanics I mean experience, levelling, statistics, rewards, ... but most importantly how to keep all of this in balance.
Fights shouldn't be impossible to win, but also not too easy, etc...
So actually a more theoretical approach.
I'm not really finding what I'm looking for, maybe it's not even out there

Tnx in advance!

M.

I don't know of any articles or tutorials about that particular issue. However, balancing is more art than science, and I don't know that there's a cut-and-dried approach other than testing and tweaking.

How you balance things depends on the level of challenge you want the game to have, as well as the play experience you want the player to have.

A fairly standard approach (from my experience playing RPG's) is to seperate the game world into areas, and then scale monster difficulty/experience gained from battles in such a way that the player is clearly expected to move on once a certian level is reached. That is, a player can grind levels for a while, but at a certain player level the experience gained from monsters in that area becomes insignificant compared with the amount required to level up, so the player is encouraged not to stay there and move on. Similarly, if a new area is too difficult, then the player should be able to get some gains from a previous area before levelling there ceases to be effective.

But that's a very broad mechanic. You can also tweak things like encounter rate with enemies (or number/composition of enemies) in an area, and then since you know how much time a player will be required to spend in that area to advance the plot, you can rough out an estimated player character progression over the course of the story.

There's a lot more to balancing a game of course, but this is at least a place to start. It's a broad topic, and there are a lot of ways to do it-- it all depends on what you want your game to be like in the end.

While I don't know about any books, etc. there are some general concepts. I actually lectured on a little of this at GDC Austin last fall. A lot of it can be broken down using game theory.

For example, think rock-paper-scissors. Everything is equal in that there is 1 way to win and 1 way to lose (ignore the tie). If you were to adjust it so that one of those relationships was 80%/20% instead of 100%, the whole system would be out of balance. Not only would no one play the nerfed play, people would play the intended target more often.

The initial matrix:
R 100% -> S
S 100% -> P
P 100% -> R

If you make it:

R 80% -> S
S 100% -> P
P 100% -> R

That means that you also have to add:
S 20% -> R

Therefore:
S 100% -> P
and
S 20% -> R

It just makes sense to play S because you always beat P (as before) and now you also have a 20% chance of beating R as well!

The way to remedy this would be to make sure the entire system is in balance... for example:


R 80% -> S
S 80% -> P
P 80% -> R

Now, everything is equal. There is no "flow" through the system that ends up collecting at one point.

In RPG games, you will find these sorts of balance points all the way through. Remember economics as well... Something can be 5x as strong if you make it 5x as expensive -- either in money or even time to construct.

The key is to remember to deal with the entire system rather than individual relationships.

By mechanics I mean experience, levelling, statistics, rewards, ... but most importantly how to keep all of this in balance.

I have very little experience in the programming of anything but the most basic of games, but I do have more experience on the playing end

Generally there are several ways to deal with experience/levelling. Experience can be given per point of damage (useful with multiplayer) or per kill, as well as from quests and other rewards. The amount of experience earned for a given action tends to either diminish as you level up, or the amount of experience to get to a new level gets higher as you level up.
Statistics tend to be upgraded at level-up, some games give bonuses to statistics whose skills were used during the past level (for example in Oblivion, if you leveled up your Acrobatics Skill, the Agility Stat would get additional points if you chose to upgrade it at level-up)

Balance mainly requires planning ahead and play-testing. I would store all of the values for the skills, experience, etc earned from everything in a separate config file and tweak that until the game seems balanced.

If you are doing a single player game or co-op type game, you can also have the game dynamically scale the difficulty of the bosses. For example, the game can detect if a boss is absolutely crushing the human player, and make the boss's attacks do slightly less damage and make its armor slightly weaker. This can also be applied in reverse; if the boss is getting killed in one hit, the game can make him stronger.

Also, many (single player) RPGs tend to have a difficulty control, so people can choose how tough they want it.

I don't know, maybe a form of PID control can be used even here (PID control is basically an algorithm used in control systems to prevent jittering)

Actually a really good reference book is probably one of the older Dungeons and Dragons rule books and a Dungeon Masters guide. I am sure out right using them would not be proper, but they show all of the rule sets, probability tables and stuff that have been used for years (what like 20 years or more?) in actual play. The play balancing, i.e. enemy difficulty and economy were controlled by the dungeon master, but the general guide lines are laid out.

[quote=IADaveMark]Something can be 5x as strong if you make it 5x as expensive [/quote]

I guess the problem here ist to judge how strong that something really is.
Sure, you can build an extremely complex algorithm to judge how much e.g. an ability that increases a characters speed enhances it´s strength in combat.
Judging how much players will want to have that ability simply because it cuts down time to move from one site to the other is not as easy though.

Say you want to implement that "speed" ability and are going to assign a value to it, say, an experience point cost to level it.
You will have to make decisions, such as "Do I want to give this ability away nearly for free because players desire it so much, or do I assign a heavy cost to it BECAUSE players desire it so much and will spend the XP on it anyway?".

Your post was interesting to read non the less; seems like I have to learn more about game theory.


@Wrathnut:
I don´t know the first edition D&D Books but I have found some (A)D&D publications to be quite good when it comes to such things.
The TSR-wrought "Alternity" roleplaying system, published not long before they were taken over bei WotC, often offers really nice ideas that are clearly structured; such as different FTL drives, Weapons, Ship types, etc. per technological development level.

However one thing that those books, and almost certainly also the old D&D books, don´t offer is perfect balancing for a computer RPG.
Simply because it is not necessary there, nor does it even have to be wanted.
In a pen&paper RPG you have the gamemaster to judge on a case by case basis, in a computer game you can´t do that.
So when TSR published the Player´s Option books which allowed you to practically build your own character class by choosing from a list of abilities, the authors explicitly covered the topic of onesided powergamer charakters and both appealed to the players not to build such characters as well as to the GM to be wary not to accept such characters.
For a computer game on the other hand you need definite rules because no GM is there to judge the situation.

Now what WotC did with D&D 3rd Edition and especially D&D 4th Edition is to mitigate exactly those problems that arose of the necessity of the GM to individually judge a situation.
I have one exact example for this.
The disintegration spell had been with (A)D&D for a long time and so also appeared in D&D 3.0.
The rules for this spell said that it would disintegrate a certain volume of material, be it inanimate or living.
This made this spell into practically the deadliest spell in the game, because you would just assume to disintegrate vital parts of an enemies body, theoretically even being able to disintegrate part of a dragons head and at least causing loads of damage.
But there trouble arose, because normally you could count that spell as a special instantdeath spell, but under some circumstances it would not mean instant death for the victim, because, say, the dragon´s head was so big that even the several dozen liters of disintegrated volume would not necessarilly kill it.
And the spells description offered no measure of how much damage this spell would do to a victim that was not instantly killed, so it was totally up to the GM how much damage that would be or when a victim would be completely disintegrated.
So when D&D 3.5 came out WotC would go and change the rules of the spell to "does 40W6 points of damage" (or so) and only disintegrating a victim if all of it´s hitpoints were diminished.


Now I guess I would put it like that - you can mathematically balance the figures but you can´t mathematically balance fun.
D&D 4th Edition shows what can happen if you try to hard to "get your rules straight".
Previous editions had featured certain sub-rulesystems for e.g. spellcasting, but now those systems were "balanced" out and by doing that lost much of their appeal.
Now every character class had loads of special abilities, and because there were so many of them, they all somehow seemed to be pretty much alike. They, too, lost appeal.

So for effectively balancing a game you also have to take into account how things affect the player, how much fun a certain rule or element is, etc., and that won´t be done by following a guideline or relying on mathematics.
Cause in the end balancing only serves the purpose of making the game more fun, and a mathematically balanced build that is no fun to play contradicts that.

I however know of a book that covers RPGs, and it even is freely available as a pdf.
It´s called "Design Patterns of Successful Role-Playing Games" and is written by Whitson John Kirk III.
You can find it here: http://rpg-design-patterns.speedykitty.com/doku.php

I personally only had a short look at it, because it was to analytic for me.
It did seem like a pretty thorough work though, with it´s over 260 pages and lots of graphs, so give it a try!

FES

Thanks for the good info. I haven't had a chance to mess with any of the newer editions of the AD&D rule set. I stop playing when I started college and am just getting back into it several years after graduating. Thanks for the link Design Patterns of Successful RPGs. That sounds like an interesting read.