24 November 2017

Class(es) Dismissed Part II: The Mage

In my previous post we pared things down to just two classes as the foundation for all player characters: the Warrior and the Mage.  I then went over how I plan to structure the Warrior.  Today I want to do the same thing with the Mage.  It starts with the intro:

Whether a student of the arcane arts, a channeller of your god’s almighty will, a gambler who has made terrible pacts in exchange for terrible power, or even one touched by the spark of divinity, you wield a power granted to few.

The above is deliberately generic.  If I'm going to successfully fold all possible spellcasting classes into this, then I need to go out of my way to allow methods of "spellcasting" outside the normal spellbook-and-study type methods.  This may necessitate some messing around with the spell research sections as well, unless I choose to keep it like generic D&D regardless of one's power source, in the interests of simplicity.  However, offhand I feel that limiting all sources of magic research and progression to such an academic formula is going to be limiting from a worldbuilding fashion, setting up images of Cambridge-like spell colleges in the wilds of the frozen barbarian north or deep within Orcish holdfasts.  But this is something to worry about later.

Before we delve further into magic, I want to settle the use of arms and armour.  Traditionally, of course, this is simple: mages get bupkis.  The reason for that was one part genre simulation and one part pure gamism.  Inspiration for the original class came from robed wizards who didn't run around with battle axes (even though those certainly existed in fiction as well), and whether it made sense or not, the fighter was meant to fight and the wizard, well, was not.

But if our one spellcaster class is going to encompass the maximum amount of design room (i.e. to cover the most possible number of character class concepts), then I think it's necessary to allow them to use at least some weapons and armour.  I think I'll go with unrestricted weapons use, as you'll need your hands free to cast spells anyways.  Since Mages are going to lack a lot of the fighting goodies that Warriors have, I don't see this as really infringing on the Warrior's role.  As for armour, I will restrict this a bit: no heavy armour for the mage.  However, this is a bit more complicated; I'll explain below.

Okay, simple enough.  Now what about magic itself?  Of course, we need a way to bring out the cleric concept even if it doesn't exist as a formal class: it's too much a part of D&D not too.  But beyond that, I want some way of ensuring that not all Mages are cookie cutter.  With Warriors I gave them a variety of fighting styles to give players the opportunity to craft a variety of fighter types.  For both classes I'm going to be using a feat-like system (heavily fenced off to ensure you don't get all the problems of 3rd ed feats; that's another post) which will give a decent amount of customization opportunities for every player.  At the same time, I really feel like there should be something inherent to the Mage class that provides a sense of uniqueness to each caster.

I'm sticking to classic Vancian magic (because I like it and it will greatly help in maintaining compatibility with other D&D/OSR products), so I can't really innovate there.  After ruminating on this for a while, I decided to go with the concept of spell schools.  Introduced in 2nd ed (and used in 3rd and 5th--maybe 4th as well, but I never looked at that one), these break all spells into eight subgroupings.

The eight spell schools

By default mages could generally use spells from any school: the school structure mainly existed to allow for class specialization (e.g. your character could be a necromancer and get extra necro spells) and mechanical tweaking in various places (e.g. "this creature is immune to illusions").

Related to all this, though something I'll discuss in more depth later, is that I was heavily inspired by Delta's Book of Spells.  This was a project by him to boil spells down to their absolute mechanical minimums.  One of the major things that bothers me about spells is the oceans of space they take up in any book.  Delta's book uses only those spells found in the original OD&D set, so it doesn't even go past spell level 6 because the very concept of those higher spell levels was introduced later; this dovetails nicely with Expert's ending of their spell list at the same spell level.  By sticking to these limited spell levels and ruthlessly purging the text of each spell, he was able to get all spells of a given spell level on a single page.  While I wanted to use more spells than just the OD&D core list, the idea of making every spell of any given spell level fit on two facing pages still seemed just fine to me: that's the entire spell section in a dozen pages.  Having all spells of a given spell level on two facing pages would also make it so that each spell level was immediately accessible by a reader.  As I want a small, readily absorbable ruleset, this all seems great.

Limits as Character-Defining

Fitting this into my idea for using schools, I decided that, instead of giving every Mage access to all eight schools right at level 1, I'd make them start with only four schools.  At each name level (a concept I've mentioned in earlier posts; a name level is every 5th level), a Mage would get access to a new school of their choice.  As such, after reaching 20th level a Mage would at last have access to all eight schools, a good level to mark a true master.  Before then, you're deliberately choosing what you want versus not, shaping your Mage's capabilities and the way they play at the table and thus, hopefully, ensuring that casters don't feel cookie-cutter.

This is where my earlier digression into spell lists comes in.  If you look at most unrestricted spell lists (i.e. the ones where they just throw in every spell they can thing of), you'll see that 1) there's way too many spells to meet my goal of two pages per spell level, and 2) take Conjuration, Evocation, Transmutation and bugger the rest, both because of  the general effect types those schools encompass but also because those schools get the most spells.  If I'm going to make school choice a meaningful choice, I need to ensure that the spell list is tweaked to reflect this.  Again, a full discussion of spells is really a talk for later, but at its heart I decided to go with precisely three spells per school per spell level in the core rulebook.  This way, you're guaranteed to get at least the same number of spells to play with as any other Mage.  By limiting spells to just three per spell level, I can at least ensure that a school doesn't become the most popular choice through sheer number of options (though of course the spells actually available will play a big part, too).  This gives 24 spells per spell level to play with (well, except for level 6, because the increasing complexity of spells at that level meant that I could only cram 16 per level on two pages there).  For comparison, B/X grants a dozen wizard spells per spell level, plus another 6-8 clerical spells.  So, even folding the two spell lists together (as I have), players will still have a little more in the way of options compared to B/X.

Here's a PDF showing my spell list as it now stands:
Simulacrum Spell List

Okay, the concept of limited spell schools alone I think will be fun.  But I think we can go a bit further.  After mulling it over a bit more, I realized that spell schools could also be a sort of character building currency.  I want to avoid the character building mini-game so endemic to 3rd edition play, but at the same time, I wanted a bit more in the way of options than your typical B/X game.  We also need to be sure that we're allowing the most number of character concepts to be created out of this single class, because again, this is the only primary spellcasting class in the game.

As such, I've gone with three options for starting Mages.  These options only exist at the start of the game: once play gets underway, you're stuck.  At their heart, each allows you to trade away your general spellcasting flexibility for abilities elsewhere.

The first two are similar in concept: they allow a Mage to be more fighty, at the expense of their access to spells.  Option 1 is the ability to wear light armour.  Lose one school, and get +3 AC to compensate (and the ability to wear light magical armour later in game).  Option 2 is a little more in-depth: you gain the Warrior's attack progression (+1 attack bonus every level).  I could see this being an eyebrow-raiser.  At first I had this so that it was +1 for every two levels, but this didn't make much difference early on, and with the Warrior's main class feature coming in as extra damage (in the form of damage dice; see my previous post), giving out the same attack bonus didn't seem as infringing: they're still never going to remotely keep up with the Warrior in melee.  By giving both classes the same attack bonus though, it also promises to really help when balancing melee-based encounters, in that I can ensure that a mage will always be able to contribute in melee if their player made them that way.

Option 3, the final one, is the concept of specialization.  Again a 2nd ed idea, this lets you give up knowledge in one area of magic to increase your ability elsewhere.  I've tossed the original idea of oppositional schools (in 2nd ed, when you specialized in school X you automatically lost access to school Y, its "opposite school").  The character can toss whatever school they want.

So that's it: the Mage in full.  Hopefully this should let a player make wizards, shamans, clerics, battle-wizards, all kinds of specialist casters, and so on, filling most of the caster roles your average player would desire.  In my previous post I upped a copy of the Warrior's page.  For this one I'm instead going to up both the Warrior and Mage together, so you can see how they look alongside one another as facing pages (view it in a two-page view to get the full effect).

Simulacrum - The Classes

(All art by Dean Spencer)

With just two classes, I can have all the character class info on two facing pages across from each other, which I think looks sharp and again aids in assimilating information.  Thoughts?

Of course, there's more to character creation than just classes.  I think in my next post I'll dive into Simulacrum's planned feat system, and see if we can salvage something out of the famous feat boondoggle appropriate for more OSR-style play.

5 November 2017

Class(es) Dismissed

One of the many fun things about the OSR is that it often looks at the issue of classes in entirely new lights.  I'd say that, other than the default "keep it all the way it was", there's two other major schools we can group these views into: chop some of the standard classes out, and add new ones in.

For the first school you have things like Delta's OD&D house rules where he axes the cleric because he doesn't think it's proper fantasy, or those that eliminate the thief because it robs the other classes of the ability to do things that people feel all adventurers should be able to handle, and additionally serves as a Trojan Horse for that dread marauder, Sir Skill System.

For the second school, you have guys who decide that the standard classes don't adequately cover the range of decent character options.  These designers take the 1st-edition AD&D route and add more and more classes and subclasses, so that you have cavaliers and assassins and knights and jesters and mountebanks and thief-acrobat-turnip-twaddlers and whatnot.  It's very Rifts©®™.

This was my favourite search result for "class structure".

The Taxonomy of Classes

While I can see the appeal of all three ways, for my own system I'm cooking up (as I've discussed), I want short and sweet.  That itself is justification enough for me to get cutting, but there's a larger reason there.  I've never been happy with how D&D handles the literature it was actually based on.  The class structure that so defines the game is excellent for people attempting to get what their character is going to play like and generally do, but its rigidity means that, even with multiclassing and dual-classing and prestige classing, you have a hard time properly representing even someone as basic and foundational to the genre as Conan, while other classes come with bizarre abilities (like the thief and read scrolls) that are there because Gygax or whoever read one book where a thief was able to do that and so it became the template for thieves everywhere regardless of how much or little it made sense elsewhere.

My broad issue with method 2 (adding more classes)--besides the size of the resulting character creation section and the increased time it takes to parse options during character creation and gameplay--is that it never really solves this issue.  Like the chimerical quest for "the" transitional fossil, each class added only creates two further class gaps to either side of it.  Of course, after a certain point we can say that those gaps are no longer really meaningful ("now what comes between the gymnast and the thief-acrobat?"), but I also think that you have to create more classes than I want in order to reach that point.  So, while I need to keep in mind some way of creating role flexibility, I don't think piling on the classes is going to be the way to do it.  Role flexibility is thus something I'll have to tackle in another section of the game.

With the idea of adding tons of classes abandoned, I might as well dive in wholeheartedly into the opposite: "just how far can I cut?"

1) I don't want race as class, because it's a blocker for some and I'm going for a very setting-neutral thing here, to allow me to adopt the most material with the least effort (I'm not going to get that all the time, but with something like race as class, which I didn't like anyways, I can go with it).

2) As it's already been done I know I can axe the thief.  That will mean I'll have to ensure that thief skills are properly distributed in the ruleset so that all the other characters can do those things, but I don't think that will be an issue and in any case it's a matter for later on.

3) While I don't agree in the least with Delta's reasoning for getting rid of the cleric, it did get me to consider if it was a feasible thing to do it, and I think the answer is yes.  We'll take a closer look later.

That leaves us with just two classes: the Warrior and the Mage.  Now that's trim.  Let's give it a try.

The Warrior

What do we want the warrior to be able to do?

  • Be the best at beating things to death.  This is first and foremost.
  • Do more things in melee than non-warriors (i.e. not just higher combat numbers, but more options)
  • Slaughter hordes of mooks/peons
  • Protect the caster
  • Cast spells

The last one is perhaps a touch unintuitive, but this is a class feature of both the paladin and the ranger classes and several key fictional fighters (Karl Edward Wagner's Kane comes immediately to mind), so I think it needs to be possible.  Magic access just shouldn't be a no-brainer, and also shouldn't ever be comparable to the mage.

1) Be the best at beating things to death

There's three main ways to accomplish this, offhand: number of attacks, to-hit bonuses, and damage amounts.  Make any of these substantially higher or even exclusive to the warrior and they'll be noticeably better at carving up their enemies.

I've been struck by 5th edition's interest in keeping numbers low overall, and I'd like to be able for a caster to still be able to hit things even at higher levels.  As such, the fighter will get better, but I don't want them completely outstripping the mage in this regard; I'm not a fan of LotFP's "only the fighter gets an attack bonus" niche protection methodology.  Here's what I'm using:

Warriors receive a class attack bonus of +1 every level, starting at level 1, stopping at +15.

That's about equivalent to the AD&D fighter over the same level range.  Keeping roughly the same to-hit range means I can keep broad compatibility with modules, monster manuals, and any related threat assessment mechanics (i.e. if text gives some sort of challenge or encounter rating, or if a module says it's designed for characters of X level).

While some hate it and struggle against it, IMO the best part about working with D&D and trying to reach combat design goals is the abstract nature of combat and related mechanisms in the game.  There's no exact mapping of attacks and damage in D&D.  Most of this is due to the hit points mechanic.  Now, it's drawn a lot of flack over the years for being vague--how exactly does one tie injury to HP in a manner that's consistent, for instance--and it's created some rules oddities for sure (such as the higher level you are, the longer it takes you to heal; I'm fixing that one for sure).  Still, while most people work to reform HP and combat in a way that makes them more precise, I think that's a major mistake.  If instead we embrace the fact that combat and HP are abstractions (which they have always unambiguously been; see the 1st ed DMG, p. 82), we can also play around with abstract damage to meet our goals.

In short, as the warrior climbs in levels, we can give them more dice of damage to represent how much better they're getting in combat.  No, their weapons don't magically do more damage.  Rather, just as characters don't gain more meat and natural armour plate as they go up levels but gain more HP regardless, as warriors climb levels they become increasingly skilled at countering the advantages an opponent has that give them their higher HP totals in the first place.  Here's how I plan to do this:

At each name level, they receive an extra weapon die of damage (so a level 10 warrior would deal 3D4 damage with small weapons, 3D6+1 with medium weapons, and 3D8+3 with large weapons).

(Right now I'm tying a lot of bonuses to every fifth level, which is what I'm calling a "name level".  So a lot of, but not all, goodies kick in at 5/10/15/20.  I'll talk about this another time).

Weapon damage I covered in my previous post (and is why I started my design series with that post).  Other than helping out the warrior, higher damage also helps keep combats fast, which I like: as levels climb and HP totals with them, the warrior is going to be able to keep up to a degree

With the increasing damage, you'll see quite noticeable ramp-ups of warrior capability as they climb the level tree.  This needs playtesting, of course, to make sure warriors don't hack through opposition too well, but as I also mentioned in an earlier post, I'm writing with the assumption of smaller player groups and so smaller adventuring parties.  As such, a warrior needs to be more capable than normal--this meets that goal.

Though I toyed with it for a while, I ultimately avoided extra attacks because of the ways an action economy can rapidly break down when dealing with them.  It also can make it easy for a character to overshadow the others at the table by taking up most of the play time, which gets old fast.

2) Do more things in melee than non-warriors

This one is damn tricky.  For one, one of the major reasons warriors often suck in games compared to mages is that people's realism alarm doesn't go off for mages in the same way it does for "natural" classes--classes that obey the general laws of reality.  In some cases people have tried to get around this by bending those laws, looking to wuxia, myth, and so on to expand what "realistically" means for a warrior.  In other cases, rulesets have looked to add special maneuvers and the like to just give the warrior more to do; ACKS takes this route, for instance (and is the inspiration for what follows).

I'm not interested in the first approach (just not my thing) and the second often fails due to implementation.  Why?  Well, the general idea is that doing things other than beating a guy like a baby seal is special, and such attempts need to be penalized because otherwise the warrior will run amok (or will spam one attack like some a 12-year-old playing Mortal Kombat).  ACKS, for instance, applies a -4 to all such attempts, so that if you want to try and knock someone down or disarm them or what have you, it's damn hard.  The result, however, is that people don't tend to bother.  The other issue is that each special maneuver comes with its own special rules, and between the fact that they're rarely used and that they each tend to work somewhat differently, you get a major slowdown mid-combat due to the need to look up how they work, as few players will memorize them in the same way as the general combat rules.

I think I have a way of getting around this.  I want to give warriors some fighting styles to choose from.  This won't be like feats in graciously permitting them to do something they should already be able to do, but rather just something that encourages a warrior to do something they were already inclined to do.  The idea is you have a warrior fighting concept in mind, and then you get something that reinforces that concept, without generally preventing you from doing any of the other things.

In short, guy who wants to use giant-ass swords will be pretty good at them.  Similarly, I don't need an archer class when that's just a weapon choice (and also the warrior that decides to be a sniper will be able to do things with bows no non-warrior can).  The lightly-armoured fast-mover archetype is covered as well (again without the need for a class).  But Brawler is where most of the fun comes in.  I originally had this as just a melee-only to-hit bonus, but numbers are boring and also threaten to undo the bounded range of potential to-hit values we're playing with.  Brawler compresses the page of ACKS special maneuvers down to a few lines, and their freebie nature makes it so they'll actually show up in play.  Making it random means that a player won't be doing the same thing over and over, avoiding that spamming repetitive feeling.  Defaulting to damage is so that a person doesn't wind up rolling over and over if they can't get a result that fits the scenario.

I do wonder if it's a wise idea to wall off these maneuvers behind just the Brawler, I have to admit.  But otherwise I'm good with it.

I also used to have Marksman as providing a to-hit bonus, but again--boring.  Allowing them to be the one group of combatants that can fire into melee is unique, and eliminating range penalties is something that is rewarding without being unbalancing (i.e. it doesn't make you more accurate, just more able to hit things at long range, so as you get higher and higher in level you can make more legendary shots).  As you can tell, I'm planning on using much higher range penalties than base D&D (based on Delta's series of interesting range and missile weapon analysis articles):

Archery -- Field Experiment
(last in the initial series)
Slings and Arrows and Outrageous Formulae
(interesting add-on)

Points 3-5
We'll cover the remaining points another day.  This post is already getting long, and in any case addressing these involves more than just the base character class section.  I'll just end by noting I thought about co-opting a modern D&D innovation, the Power Attack feat:

A warrior’s combat expertise allows them to balance power and finesse as needed. Before making attack rolls for a round, a warrior may choose to subtract an amount from all their attack rolls and add this amount to all their damage rolls. This number may not exceed their available attack bonus for that round, and cannot apply to spells.

Early on I had made this something free for all warriors, based on an off-hand comment by Gronan/Old Geezer a long time ago on some random forum (which amounted to "why not just bloody well let them do that rather than making it a choosable feat?").  It would help reinforce warrior combat prowess, and at the same time give the warrior player something to play around with each round.  However, I felt it would constantly lead to juggling numbers each round, as each player so inclined takes the opportunity to calculate the optimum attack bonus vs damage scenario for that round.

Below is a one-page pdf of what the warrior looks like right now.  With a similar one-page mage treatment, I can get both classes (and so the entire class selection part of character creation) out of the way in two facing pages, which makes me very happy.

(Art courtesy of Dean Spencer)

I've deliberately left out any attribute score class restrictions, because I feel that assumption of a role does not automatically mean that you're good at it.  If people are really concerned about this being unrealistic, there's already a mechanism known as Darwinism which generally serves to enforce what you're after.  (Additionally, with only two classes, I can't really redirect people to other classes if their ability scores suck).


Next we'll look at the mage, and then move on to the challenge of ensuring that these two class archetypes can actually fulfill the character-building desires of most players and games.