Psychochild's Blog

A developer's musings on game development and writing.

28 October, 2011

Weekend Design Challenge: Looking at smelly feats

When last we did a weekend design challenge… oh, has it really been three months? Yeah, I slacked a bit here. To be fair, I’ve been busy. :)

But, let’s get back on the horse and take a look at our “coverting D&D 3rd edition to classless” project, shall we?

Finding our place

Okay, so where were we? Last post we had separated out the class-specific abilities into a spreadsheet. Remember the handy hyper-text d20 SRD. Now we need to come up with a way for players to define their own class.

Remember the goal was to re-design the d20 system as a classless system for a single-player computer RPG. That’s our boundary for this work. Again, yes, it’s arbitrary but it’s like a “real” game design problem.

As a reminder, I’m going to go through my own design process here. I’m not going to claim this is the one true way, or even a good way, but it’s how I generally work. This is a glimpse into my design process, not a proscription for how it should be to be a “real game developer”.

Beating feats

As I said last time, let’s use feats as the core of our system here. The basic idea is to give the player so many feats per level and let them buy up their abilities. This brings a lot of questions:

  • How closely will we emulate the class-based system?
  • How many feats are granted per level?
  • What class-specific abilities, such as Rangers being able to take a Combat Style, ignoring typical prerequisites while restricting character to light armor?
  • What about non-feat abilities like hit die, bonus to hit, etc?

Let’s take these one at a time.

How much deviation?

Our goal is to make the classless system for a computer game. So, we’d like to have as few “exceptions” to the rules as possible. We’d like things to be a mostly “one size fits all”, so we’re going to use the existing rules as a guideline rather than a straight-jacket. We’ll get something where you could mostly build a character based on the existing classes, but you won’t necessarily be able to build one exactly.

Note that there might be a lot of other issues that influence us. If we were only going to go up to 10th level in the game with no expectation of a sequel, we might balance things a bit differently. But, let’s assume we’re not going to have a limit on the game quite yet; we’ll work up to level 20 to keep it simple for now.

How many feats?

Okay, determining the number of feats to give is a really tough problem. The basic d20 setup doesn’t balance things out very well. In fact, some of the initial research I found about deconstructing classes was intended to show, for example, that Fighters get a bit boned on their special feats and should get a bonus feat every level.

So I’ve added a new page to the spreadsheet where I’m going to count up each ability that a class gets. Abilities that increase each level get counted per level. The snag comes from spellcasting… how should that work? I’ll add a few variables under the page with a level of spellcasting equal to a certain number of abilities, and we can play around with that a bit. For now, we’ll count lesser spellcasting as 1 ability per level, main spellcasting as 2 abilities.

Looking at this, we see a lot of disparity between the classes. Fighters are left behind quite a bit even compared to the other non-spellcasting classes. Monk has a ton of abilities, and the spellcasters are have fairly elevated totals for abilities given their spells. But, we just picked an arbitrary number for our calculations; we might need to adjust those figures later.

For now, let’s pick some numbers to work from: 4 feats at first level, 2 feats every level after that. These include the bonus feats granted at first and every third level; it gives us an average of 42 total abilities, which matches a few classes in the middle of the pack. We will almost certainly adjust the number of feats later, though.

Dealing with special cases

Since we want to keep this simple, we’ll ditch the special cases like the ranger bonus feats. This takes away a bit from the game since re-creating a ranger will be harder, but it’ll still be possible to create a lightly-armored fighter character if you want.

Okay, what haven’t we covered?

Weapon proficiencies

Trying to fit this into the existing feat-based system provides a few snags. For example, weapon proficiencies can be taken as feats, but that means that the warrior type classes would have a huge amount of added feats at the beginning; there are 35 martial weapons in the core SRD alone before we go to any additional books.) Requiring feats for each weapon also opens things up to quite a bit of min/maxing, where people could just take the popular weapons. One option would be to have a special “Martial Weapon Proficiency” feat that is only available at level 1 for people who want to build a fighter type. Another option would be to give a discount at first level, so that 1 feat equals, for example, 10 weapon proficiencies. Or, allow the character to pick five “free” proficiencies at level 1, then buy other (perhaps at a discount).

This also depends on how many weapons we’re going to implement. If we aren’t going to do polearms, for example, then this diminishes some of the power of the warrior-type classes and makes spending feats on weapons less useful.

Armor proficiencies

We have a similar situation here, where proficiencies could be taken, but that would require the heavy armor types (the Cleric, Paladin, and Fighter) to spend a lot of proficiencies at first level. Perhaps give light armor as a free proficiency, but require others to be bought. Fighters would require 4 proficiencies: Medium, Heavy, Shields, and Tower Shields.

Attack bonus

This could be pretty easy, a feat = +1 Attack bonus. But, again, the fighter types would have a tough time at first level with yet another feat required and it would eat into other abilities.

Savings throws

Here everything goes to hell. There are three save categories, but not every class is the same. Some have one strong save, others have two, some have all three. How do you balance that out with just feats?

Rethinking feats

Well, okay, maybe feats won’t work quite as well as we had hoped. So, what alternatives should we consider?

Consider that, and we’ll discuss some alternatives next week. This week, ask questions and contribute your thoughts as to how to use feats for class features.


  1. One reason feats are hard to balance is simple math. If you have the very conservative number of 20 feats out of which 5 can be picked by the player, the number of combinations is already 1.86 mio.

    I’m not a proponent of Blizzard’s new talent system in the context of current World of Warcraft. However, it is still interesting to look at. The intend is to allow a player a choice of one out of three talents/feats every some levels. This selection process happens up to six times while developing the character. This quite conservative approach already creates 729 combinations for every spec. That is 729 combinations the devs need to think through in some way.

    In my opinion, one needs to distinguish between the two reasons for character customization:
    1) players like to grow a character
    2) hard choices are content and allow players to have unique characters even in the presence of min/maxing.

    To satisfy point (1) you can add some choices that turn out to not be choices at all. The widely used talent trees are a good example. Whenever you can distribute one talent point you can only choose one talent/feat from the current tier. And later you often end up filling out almost the entire talent tier. There is some choice here, but the real benefit is that players like to grow a character while they are entertained with the illusion of choice.

    To satisfy point (2) you need a very conservative approach to be able to think through all the possible combinations. For example, the one with ‘just’ 729 combinations. It makes players think really hard, which means it keeps their mind busy, which means that it is for all practical purposed content. And at the same time this approach can lead to unique characters, because players believe that their choice of six times one out of three is the best, even though differences in efectiveness may be almost non-existent. An interesting question now is whether and if, how often players can undo a prior choice.

    Comment by Nils — 29 October, 2011 @ 4:29 AM

  2. I have a system that could maybe fit this (hat-tip Elder Scrolls). Everything is arranged around factions . Each faction has entry requirements and each faction can train certain skills for faction points earned by performing duties for the faction.

    Example: Human town, village outside. Human players start in the village.

    Village contains various NPCs
    - sergeant of the village watch faction
    - deacon of the temple faction
    - witch of the witch faction
    - hedgewizard of the hedgewizard faction
    - shady character of the theif faction

    (The Kingdom itself has an overall faction – the above are all subfactions within it.)

    Say the requirements for the base factions are two stats above average
    - village watch, strength and constitution
    - deacon, faith and determination
    - witch, intelligence and empathy
    - hedgewizard, intelligence and determination
    - shady character, any two out of strength, agility, dexterity and intelligence
    - poacher of the hunter faction, constitution and dexterity

    Player joins the faction they choose and gets access to those skills that seem to fit that faction. They have to earn faction points to buy them though. Taking the fighter route as an example and using my skills as i can’t recall the feats. The pattern will be similar.

    Say the village watch trains club1, spear1, shield1, light armour

    In the town is the town watch faction. say their requirement is four combat skills/feats. Say their faction trains dagger, club2, spear, shield, shield bash, light armour, dodge, parry etc plus other combat skills in tiers based on certain conditions e.g. min 8 combat skills, and some non-combat skills we’ll come back to.

    The town also has a mageguard faction which is a subfaction of the mage guild for fighters attached to the mage school, a temple guard subfaction for temple warriors and the soldiers of the royal army subfaction. Each of these factions requires 8 combat skills as a requirement and teachs further combat skills. The mageguard could lead to fighter-mage path. The templeguard could lead to a paladin path. The soldier path could lead to the higher royal guard and king’s guard factions as a pure fighter.

    Also the town watch teaches some skills that overlap with the thieve’s guild. And having 2-3 thieve’s guild skills like streetwise and haggling means a player can switch to the theive’s guild. Inside the thieve’s guild there’s a theif path and an assassin’s path. Also via the thieve’s guild the player could join the brigands faction for a mixture of light armour combat skills like dual wielding.

    Another example, a priest player might pick the deacon path in the village and after gaining a few skills meet the requirements for the temple in the town and switch to that. At the temple he picks up some priest spells but also enough combat skills to meet the requirements to join the town watch as their chaplain. At the watch our priest picks up the gambling skill and other bad habits which makes him able to join the thief guild which eventually leads to the brigands and the player as a kind of dual mace-wielding Friar Tuck character.

    There’d be a lot of different mage schools which would all train a collection of general mage skills and plus a collection of spells unique to that school so after a time at their starting mage school a mage player may want to wander around finding other schools to learn from.

    Similarly with priests. There’d be general priest skills taught at each temple and a list of specific ones for each god but also each pantheon would have minor gods or allied demigods scattered around the world with their own temples training their subset of spells. A dark elf priest might level up and get a bunch of spells from his god of power but his god is also neutral or allied with some demi-gods like a snake god in one zone or a rat-god in another who have their own collection of spells and skills.

    You get the idea.

    So it would be the same system with feats. Each faction has prior stat and skill requirements and a list of skills, spells, feats that suit that faction they can train the players in (except if its levels then its the faction you’re in when you level instead of buying skills individually with exp you earned with that faction). The player hops from faction to faction according to taste as long as they meet the entry requirements.

    “Weapon proficiencies”
    In mine there’s only five weapon skills, dagger, axe (covering clubs, maces etc), spear, staff and sword (in three tiers), with two-handed as an optional extra for axe and sword. So i’d suggest either grouping the proficiencies so one choice covers all the weapons in that group or have a cascading skill e.g. weaponskill1 gives three initial weapon proficiencies, weaponskill2 five more, weaponskill3 seven more etc.

    “Armor proficiencies”
    Fighters could get light armour earlier and potentially be in medium armour when priest and thieves were in light (depending on skill choices). I think fighter being forced to picks lots of very basic combat skills can be compensated by having particularly cool feats that have a requirement of lots of combat skills that only pure fighters will meet i.e. cool feats that require 12 combat feats or 24 combat feats etc.

    “Attack bonus”
    Some could be on the weapons themselves. So Staff1, a good weapon for a mage, could get defense +1. Sword1, which required Dagger1 and either Axe1 or Spear1 could have attack +1, defense +1. So the fighters wouldn’t need to buy a bonus it would come from the equipment that others didn’t get because they spent their points on fireballs.

    “Savings throws”
    Players start with all weak saves, factions train strong saves in things that seem appropriate e.g. mage guild, save vs magic.

    Comment by bloob — 29 October, 2011 @ 10:19 PM

  3. Slowing down and looking through the d20 list a bit more.

    Levels, feats per level, buy feats as you describe. Some thoughts:

    1) Stats start low and players can use feat points to add points. Say stats all start at 8 and first feat point spent as a stat boost gives +8 points to spend on attributes, second gives +6, third +4, fourth +2, after that one stat point per feat point. Then make hitpoints a function of con instead of class e.g. con < 10 = d4, con 10 = d6, 11 = d6+1, 12 = d6+2, 13 = d8, 14 = d8+1, 15 = d8+2, 16 = d10, 17 d10+1 etc.

    2) Bundle some skills up in collections like combat1 which gets to choose three skills from the bundle of weapon proficiencies, shields, attack bonus etc. Combat2 allows a choice of another three. Or cascading to benefit specialization so combat1 gives one choice from the long list of combat skills, combat2 would give three more, so a player only using one of their starting feats on the combat meta skill would get one point to spend on staff or dager or mace and a player using two of their initial feats on combat meta skills would have four combat skills to choose from the actual list, say sword, shield, light armor and +1 attack. Combat3 would give another three, or five or whatever.

    3) Fighter bonus feats could still be bonus feats based on number of mundane combat feats selected or the number of meta combat feats so in the above case combat1 might not get a fighter bonus feat as well but combat2 and beyond would get a fighter bonus feat on top of the other choices. If the proficiencies were chosen individually then a player could get a bonus feat per two or three martial skills.

    4) Savings throw. All have weak saves at start getting a strong save for each type is a feat.

    5) Things like Rage would have the same pre-requisites. One feat to get first at level 1. Upgrade at level 4 requires the level one skill bought already.

    6) Bard, Monk and some Druid shape changing feats might need to follow a chain progression.

    Comment by bloob — 30 October, 2011 @ 1:57 AM

  4. Recently the armor proficiency structure has been bugging me. A first-level fighter starts out with the ability to use plate, but not the means. Unlike a mage, she can’t use all the abilities her class gives her without the cooperation of the GM. Usually, you don’t get that good plate until level 3 or so, but it depends a lot on the GM. That’s the 3rd Edition and prior solution. 4e fixed this, and I think that was a good choice.

    So I’m completely ok with someone not taking medium or heavy armor proficiency right away. Really, a development scheme would give medium armor more of a reason to exist. In my opinion, medium armor really didn’t have much of a viable niche in 3.0-3.5. Unless it was a mithril breastplate or something.

    I don’t know, weapon proficiencies may be similar. Give everyone simple weapons, I’d think. Then it’s just one feat to take martial ones, and maybe you skip that until second or third level, even if you’re planning a heavy melee type. Lord knows you might not have a longsword.

    Comment by Toldain — 1 November, 2011 @ 9:20 AM

  5. Weekend Design Challenge: To the points

    [...] Last challenge, we were looking at the problems with using feats for making a classless d20 system. This week, we'll look at some solutions. [...]

    Pingback by Psychochild's Blog — 13 November, 2011 @ 2:04 PM

Leave a comment

I value your comment and think the discussions are the best part of this blog. However, there's this scourge called comment spam, so I choose to moderate comments rather than giving filthy spammers any advantage.

If this is your first comment, it will be held for moderation and therefore will not show up immediately. I will approve your comment when I can, usually within a day. Comments should eventually be approved if not spam. If your comment doesn't show up and it wasn't spam, send me an email as the spam catchers might have caught it by accident.

Line and paragraph breaks automatic, HTML allowed: <a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <strike> <strong>

Email Subscription

Get posts by email:

Recent Comments


Search the Blog


November 2019
« Aug    



Standard Disclaimer

I speak only for myself, not for any company.

My Book


Around the Internet

Game and Online Developers

Game News Sites

Game Ranters and Discussion

Help for Businesses

Other Fun Stuff

Quiet (aka Dead) Sites

Posts Copyright Brian Green, aka Psychochild. Comments belong to their authors.

Support me and my work on