Quote from: BoxCrayonTales;1087942While leveling systems may have advantages in particular areas, they also introduce their own problems. Like the mechanics for epic levels and mythic ranks, which were generally bad and quickly discarded by the creators. 5e engages in a bit of this with its "legendary" monsters, whose shtick is that they break the Challenge Rating system.

A problem I find with the the CR/leveling system is that it arbitrarily places monsters behind a level-based paywall, for lack of better terminology. As a side effect, this results in designers creating variations of the same monster for different CRs because it is difficult to keep monsters relevant outside of their CR bracket.

Another problem with leveling systems is that they affect a character's combat ability regardless of their backstory. This is particularly prevalent in Pathfinder's NPC compendiums, where the kings of countries are 10th level aristocrats and as such have CR ~9 durability even though in real life any king could die of a simple stab wound or poisoned chalice. The only way around this is to design NPCs differently than PCs, so that they can have high skills in their desired area of competence while still having low CR (called the 0th-level NPC in pre-3e editions).

A leveling system works for abstraction purposes, but it breaks down when you start trying to simulate a world that isn't a comedy where the D&D rules are literally the laws of physics (like the Order of the Stick comic) or a satire that explores how leveling systems aren't realistic (like the Overlord anime).

You're thinking about how levels work in D&D rather than thinking of what could be done with them.  In particular in regard to epic levels (apart from D&D do you really need them - do you really need them for D&D?)

I've been running the Swedish game Symbaroum for a year now.  It's classless and levelless - but is clearly influenced by D&D.  And I've been acutely conscious of how much it needs some kind of tiering over the long run (particularly as it's a fantasy game with escalating threats).

At it's most basic a level/tier system could work something like this.  Until you have earned 50 xp total your max skill rank is 3.  From 50 -100 Xp it's 7.  From 100-150 it's 9.   You could also link exception based powers to tiers.  Symbaroum has powers that come at Novice/Adept/Master level.  If I was running the game again I'd specifically make sure I tiered access to those powers to a certain point, so you couldn't just spend all your xp and go straight to master but had wait until your total xp reached a certain level before you could get a level in adept etc.

As for combat power based on level - again that depends on how you build all the other elements of the system around it.  There's no reason to necessitate level automatically increases combat power (In Earthdawn going up a circle opens up new abilities for purchase but you have to decide to raise your existing abilities with XP).  The most important function of levels is to cap power.

But yes, overall I'll concede levels are probably always going to be less inherently simulationist than a pure point buy system.  But of course there's a lot of conceptual room in between D&D and GURPS at all ends of the spectrum.  It depends what you care about.

Quote from: Jaeger;1088030Your experience is different from mine. I find my players do a little focusing at first, but they quickly branch out into other skill sets that have little to nothing to do with the original character 'template'.

I do have some players that care little for a "build" and basically just wing it. That can be refreshing and fun and matches what I like to do when I (rarely) get to be a player. On the flip side I have the munchkins. I have one player in particular that will find the most optimal way to essentially build batman, regardless of genre or system. Every game. Years ago I found that annoying, but now I've long since just come to accept it as who he is / what he enjoys.

Quote from: BoxCrayonTales;1087942While leveling systems may have advantages in particular areas, they also introduce their own problems. Like the mechanics for epic levels and mythic ranks, which were generally bad and quickly discarded by the creators. 5e engages in a bit of this with its "legendary" monsters, whose shtick is that they break the Challenge Rating system.

A problem I find with the the CR/leveling system is that it arbitrarily places monsters behind a level-based paywall, for lack of better terminology. As a side effect, this results in designers creating variations of the same monster for different CRs because it is difficult to keep monsters relevant outside of their CR bracket.

How do you make a monster that is equally relevant at all tiers of play?  Having different levels of Dragons makes more sense then having one type of Dragon that is always challenging to fight at all character levels.

Quote from: TJS;1087935I think it's interesting that we see what are basically class systems without levels, but not the opposite, levels without classes.  I think this is probably because a clear sense of classes helps make a game more immediately marketable.  But I think it's a shame because I think a lot of the benefits of levels, such as in particular being able tier character abilities, guard against the tendency toward over-extreme specialisation and incommensurate niches, and have some clear sense of character power and ability to handle threats over the long run, would be really beneficial even for systems that don't wont to encode classes in their systems.

They exist, at least for some definitions of "levels".  Savage Worlds, for example, has characters progress through the tiers of "Novice", "Seasoned", "Veteran", "Heroic", and, finally, "Legendary" based on the number of ability improvements the character has gained.  Some (but not all) special abilities also have minimum tier requirements, so, for example, a character can't start with the Block ability, they have to reach Seasoned first, and then progress to Veteran before taking Improved Block.

Straight point-based games such as Hero and GURPS can also do tiering based on character point totals, but they generally don't have any kind of tier requirements for gating abilities, so I don't think that really fits what you're talking about here.

Quote from: estar;1087939RPGs are not about moving defined game pieces around.

Yes, exactly.  And, to me, classes and levels make things feel a lot more like "moving defined game pieces around" than a more freeform, customizable method of character development.

Quote from: estar;1087939What people forget about non class based system that characters are not random hodge-podges of abilities. That there are logical patterns arising out of how the system works or how it relates to a genre or setting. Patterns that in terms of mechanics make the freeform character creation system as predictable as a class based system.

There are logical patterns, sure, and they arise in the real world, too, but I don't see that making things as predictable as a class based system.  You could define real-world-me as a dual-classed Sysadmin/Programmer, both of which are pretty well-defined packages of related skills and abilities, but neither of those classes includes skills in teaching ballroom dance.  Classes and levels are enough to handle "defined game pieces", but you need something which goes beyond that (which may be a system that's skill-based from the start, or may be something added on top of a class-and-level-based core) if you want to model the complexity of an actual person.

Quote from: estar;1087939Something that I learned when I played RPGs like the Hero System since the mid 80s and GURPS since the late 80s.

Perhaps our experiences with those systems differed, then.  If I went through one of the published books of Champions characters and picked out five Bricks or five Energy Projectors with a given point total at random, I would expect to see more variety between them than I would if I took five random same-level D&D Fighters because Hero has so many more knobs to adjust.  The general resulting patterns may be similar, but the details are another matter entirely.

Quote from: Shasarak;1088067How do you make a monster that is equally relevant at all tiers of play?  Having different levels of Dragons makes more sense then having one type of Dragon that is always challenging to fight at all character levels.

That's why I'm listing this a flaw inherent to leveling systems.

D&D and its derivatives suffer from a severe case of monster bloat due to imposed quotas, both those imposed by the leveling system itself and by a desire to have as many monsters as possible. When monsters are being written to fill quotas rather than as labors of love, the quality suffers. Creative and inspiring bestiaries like the Creature Collection series are few and far between (given the expense in making them), while first party bestiaries and monster manuals often feel like shovelware.

While you can create multiple versions of the same dragon for different CR brackets, the problem is that all too often they feel homogeneous and uninspired (Pathfinder's mythic monsters suffer this problem in spades). However, this approach is actually quite rare for monsters other than dragons because other monsters generally don't have age categories or some other easy means of representing advancement (hence the name Dungeons & Dragons). A more common approach is to write the other versions as being different monsters with different backstories to make them feel fresh, but this generally results in the monsters being redundant due to the imposed quota (e.g. humanoids like goblinoids and beastmen originally being created to fulfill a hit dice quota).

I think the key to more flexible class design is to break them down into more discrete chunks. A good chunk of multi-classing is just that the classes in D&D are too course-grained to allow players to always fully realize what they want to play. If you break down the classes a bit though you can eliminate a lot of the need to multi-class.

My system deals with what would in D&D be a single class by breaking it up into three parts; archetype (superset), class (subset) and background (anything non-combat related).

Archetype provides very basic features that set the feel. Class provides specific abilities that move the archetype towards a specific expression. Background, as mentioned above provides skills and non-combat abilities.

So, for example, there is the "Skilled" (or in OD&D terms a "Fighting Man") archetype. This archetype provides a number of basic weapon and armor proficiencies and provides you with combat stances and specialization options. It also asks you to pick a specific combat style (Strong, Swift or Berserker) and combat focus (Daring, Tactical or Wary) that provide additional elements to refine your basic "Fighting Man" style (Tactical makes your use of "Aid Other" more effective for example).

Then class is a specific set of abilities that define how you use those basic features. The Brigand, for example gets "clever tricks" they can use to gain advantages for themselves in a battle. The Ravager gians abilities to let it rapidly close with its chosen opponent. The Defender gains abilities that make them a virtual wall between opponents and their allies while Sentinels protect their allies using cover fire. The Sentinel The Sharpshooter gains abilities to disable and hinder their foes using ranged attacks, etc.

Finally, the background provides non-combat traits to round it out. So you might be an Arcanist who supplements their martial prowess with utility spells, a Barbarian with special abilities for wilderness survival, Military with more tactical options (ex. Combat Engineering lets you coordinate your allies to dig trenches and errect fortifications more quickly, Forced March lets your group travel further in a day), an Outlaw (gain abilities related to infiltration, deception and other criminal endeavors) or Religious (abilities related to both swaying the faithful and performing non-combat miracles).

So putting the elements together you might go Strong Tactical Military Defender for a classic D&D Fighter or Swift Daring Outlaw Brigand for a D&D Rogue, or Berserker Wary Barbarian Ravager for a stereotypical D&D Barbarian rage machine. But you could also mix it up with a Swift Wary Religious Defender who fights with a polearm in light armor (or with the right options; unarmed and unarmored and you a Monk who specializes in protecting his allies.

Or the Barbarian might instead be a Strong Daring Captain and be more like Conan in the books or a Swift Wary Skirmisher with a beast companion like the D&D Ranger.

Between just the Skilled Archetype/classes and Backgrounds;

- Style: Strong, Swift, Berserker.
- Focus: Daring, Tactical, Wary.
- Class: Brigand, Captain, Defender, Disabler, Ravager, Sentinel, Sharpshooter, Sidekick, Skirmisher.
- Background: Arcanist, Aristocrat, Artisan, Barbarian, Commoner, Entertainer, Military, Outlaw, Religious, Traveler.

... you've got hundreds of possible combinations to express a Fighting Man that are mechanically distinctive from each other even without a multi-classing system.

ETA: The other archetype is "Spellcaster" with a choice of spellcasting path (and subpath); Astral (faithful, militant, zealous), Gadgeteering (big lug, mad genius, monkeywrencher, troubleshooter), Primal (covenant, sorcery - with a secondary choice for both of a clever, insightful, potent or swift patron spirit) or Wizardry (lore, social, war) and the class options of Abjurer, Benedictor, Empowered, Interdictor, Maledictor and Summoner.

Taken together you can go anywhere on the spectrum between pure fighting man (of multiple varieties) to fighting man who uses utility magic (astral, gadgeteering, primal or wizardry), to spellcaster who's good at fighting to pure spellcaster with no multi-classing needed to get you there.

TL;DR the problem with classes is they're often more all encompassing than they need to be.

Quote from: nDervish;1088147Yes, exactly.  And, to me, classes and levels make things feel a lot more like "moving defined game pieces around" than a more freeform, customizable method of character development.

The point raised by the OP is that class/level has flaws which I countered not if you think of it like X.

Also I raised in my previous posts is that a system has to work with the way one thinks or it feels off or it is disliked. Nothing wrong with that but it doesn't make a class/level design flawed from a design standpoint. And I have no problem with folks continuing to like alternatives to class/level like points because of preference.

Quote from: nDervish;1088147There are logical patterns, sure, and they arise in the real world, too, but I don't see that making things as predictable as a class based system.

Come on, you need to be more specific. D&D 3.X, and D&D 5e have plenty of customization feature and both support enough freeform character generation that you can't assume a character is just X on the basis of class alone.

Quote from: nDervish;1088147You could define real-world-me as a dual-classed Sysadmin/Programmer, both of which are pretty well-defined packages of related skills and abilities, but neither of those classes includes skills in teaching ballroom dance.

And class based design have long since accommodated the player who wants his character to have ballroom dances.

Quote from: nDervish;1088147Classes and levels are enough to handle "defined game pieces", but you need something which goes beyond that (which may be a system that's skill-based from the start, or may be something added on top of a class-and-level-based core) if you want to model the complexity of an actual person.

Perhaps our experiences with those systems differed, then.  If I went through one of the published books of Champions characters and picked out five Bricks or five Energy Projectors with a given point total at random, I would expect to see more variety between them than I would if I took five random same-level D&D Fighters because Hero has so many more knobs to adjust.  The general resulting patterns may be similar, but the details are another matter entirely.

That argument only work if you view characters in a RPG campaign to defined by mechanics alone. A view I don't share or use as an assumption in my points. I been quite clear that I view RPG character to be defined by their description aspects of which may defined by mechanics like a 10 strength. But other aspects are description only.

When I make rulings irregardless whether it is GURPS or OD&D 3 LBB I look at the entire description of the character. Of course the ruling is more straight forward in GURPS if the character has Dancing (Ballroom) - 14 i.e. roll 3d6 if you roll 14 or lower you succeed, while in OD&D 3 LBB I would make a decision based on level and attributes. OK you described your character having experience in ballroom dancing, your character has a Dexterity of 15, and is level 2, so roll 15 or better on a d20 and add +2 for dex, and +2 for level.

Again a system has to work with the way you think in order to be fun. If class/level makes character feel like game pieces to you then not liking class/level is understandable. And why a skill based or point based design would work better for you.

Quote from: Chris24601;1088163I think the key to more flexible class design is to break them down into more discrete chunks.

This assumes people want to get into that level of detail in their mechanics. Other are fine in just making a Fighter with a higher dexterity and saying they are swashbuckling fencer while another opts for a higher strength and describes their character as a brute smashing their opponent.

The only issue if the referee ignores the description when it would be a factor. Or the player wants to describe their character in a way that doesn't make sense with the mechanics that does exist.

Quote from: Chris24601;1088163I think the key to more flexible class design is to break them down into more discrete chunks. A good chunk of multi-classing is just that the classes in D&D are too course-grained to allow players to always fully realize what they want to play. If you break down the classes a bit though you can eliminate a lot of the need to multi-class.

My system deals with what would in D&D be a single class by breaking it up into three parts; archetype (superset), class (subset) and background (anything non-combat related).

Archetype provides very basic features that set the feel. Class provides specific abilities that move the archetype towards a specific expression. Background, as mentioned above provides skills and non-combat abilities.

So, for example, there is the "Skilled" (or in OD&D terms a "Fighting Man") archetype. This archetype provides a number of basic weapon and armor proficiencies and provides you with combat stances and specialization options. It also asks you to pick a specific combat style (Strong, Swift or Berserker) and combat focus (Daring, Tactical or Wary) that provide additional elements to refine your basic "Fighting Man" style (Tactical makes your use of "Aid Other" more effective for example).

Then class is a specific set of abilities that define how you use those basic features. The Brigand, for example gets "clever tricks" they can use to gain advantages for themselves in a battle. The Ravager gians abilities to let it rapidly close with its chosen opponent. The Defender gains abilities that make them a virtual wall between opponents and their allies while Sentinels protect their allies using cover fire. The Sentinel The Sharpshooter gains abilities to disable and hinder their foes using ranged attacks, etc.

Finally, the background provides non-combat traits to round it out. So you might be an Arcanist who supplements their martial prowess with utility spells, a Barbarian with special abilities for wilderness survival, Military with more tactical options (ex. Combat Engineering lets you coordinate your allies to dig trenches and errect fortifications more quickly, Forced March lets your group travel further in a day), an Outlaw (gain abilities related to infiltration, deception and other criminal endeavors) or Religious (abilities related to both swaying the faithful and performing non-combat miracles).

So putting the elements together you might go Strong Tactical Military Defender for a classic D&D Fighter or Swift Daring Outlaw Brigand for a D&D Rogue, or Berserker Wary Barbarian Ravager for a stereotypical D&D Barbarian rage machine. But you could also mix it up with a Swift Wary Religious Defender who fights with a polearm in light armor (or with the right options; unarmed and unarmored and you a Monk who specializes in protecting his allies.

Or the Barbarian might instead be a Strong Daring Captain and be more like Conan in the books or a Swift Wary Skirmisher with a beast companion like the D&D Ranger.

Between just the Skilled Archetype/classes and Backgrounds;

- Style: Strong, Swift, Berserker.
- Focus: Daring, Tactical, Wary.
- Class: Brigand, Captain, Defender, Disabler, Ravager, Sentinel, Sharpshooter, Sidekick, Skirmisher.
- Background: Arcanist, Aristocrat, Artisan, Barbarian, Commoner, Entertainer, Military, Outlaw, Religious, Traveler.

... you've got hundreds of possible combinations to express a Fighting Man that are mechanically distinctive from each other even without a multi-classing system.

ETA: The other archetype is "Spellcaster" with a choice of spellcasting path (and subpath); Astral (faithful, militant, zealous), Gadgeteering (big lug, mad genius, monkeywrencher, troubleshooter), Primal (covenant, sorcery - with a secondary choice for both of a clever, insightful, potent or swift patron spirit) or Wizardry (lore, social, war) and the class options of Abjurer, Benedictor, Empowered, Interdictor, Maledictor and Summoner.

Taken together you can go anywhere on the spectrum between pure fighting man (of multiple varieties) to fighting man who uses utility magic (astral, gadgeteering, primal or wizardry), to spellcaster who's good at fighting to pure spellcaster with no multi-classing needed to get you there.

TL;DR the problem with classes is they're often more all encompassing than they need to be.

Agreed. This is no more obvious than in the class bloat that afflicts every D&D derivative. 2e had a bazillion kits. 3e (including Pathfinder) had a bazillion classes, prestige classes, archetypes, class features, feats, etc. Even ignoring supplements, every edition of D&D has gradually increased the number of core classes. 3e added the barbarian, monk and sorcerer. 5e added the warlock. Presumably inspired by Pathfinder, 5e went to the trouble of giving every class unique specialization trees.

There's a series of 3pp for Pathfinder 1e, The Genius Guide to the Talented [Insert Class] series of books, that breaks down all the class bloat that accumulated over the years into modular class features or "talents" that players can pick and choose to build their PC. It borders on a point buy system without really being one. Of course, it only solves the class bloat by converting it into class feature bloat.

Class bloat goes hand in hand with niche protection. A lot of class bloat can be traced back to developers trying to create new niches.

Another problem I have with the all-encompassing style of design is that it is ironically restrictive. Some character concepts are simply impossible to implement without resorting to homebrew or 3pp, such as a divine spellcaster whose primary casting attribute is intelligence (e.g. Heroes of Horror's archivist) or charisma (e.g. Pathfinder's oracle, Kobold Press' shaman) or an arcane caster who relies on wisdom. Even when they did get published, they suffered from bloat because writers usually tried to invent a niche for them.

So long as the default classes aren't flexible enough, class bloat will forever remain a problem. 4e addressed this problem by using universal guidelines for inventing class features rather than every class running on its own logic. 5e throwing out the baby with the bathwater frustrates me to no end.

Quote from: estar;1088168This assumes people want to get into that level of detail in their mechanics. Other are fine in just making a Fighter with a higher dexterity and saying they are swashbuckling fencer while another opts for a higher strength and describes their character as a brute smashing their opponent.

The only issue if the referee ignores the description when it would be a factor. Or the player wants to describe their character in a way that doesn't make sense with the mechanics that does exist.

I don't have a problem with the idea of classes having flavored talent trees like sorcerer bloodlines or monk arts or witch hexes or whatever they're called. The problem comes in when you have to write loads and loads of unique exception-based powers to populate those lists, which inevitably leads to power creep. This is why I like Risus' cliche mechanic and any similar mechanics in other systems.

Subtraction by addition.

I would argue that any Pathfinder, 3.5, 4e, or 5e fighter is less Versatile than a DCCRPG Warrior with his mighty deed die.

Quote from: BoxCrayonTales;1088172I don't have a problem with the idea of classes having flavored talent trees like sorcerer bloodlines or monk arts or witch hexes or whatever they're called. The problem comes in when you have to write loads and loads of unique exception-based powers to populate those lists, which inevitably leads to power creep. This is why I like Risus' cliche mechanic and any similar mechanics in other systems.

My observation while nearly all design choices can be made to work, but there are consequences.  

You can represent Middle Earth like ToR does, or AiME, or use a 4th edition exception based design. But if you go the 4e route, you now have to write all the exceptions. With a further consequence is that now becomes harder to house rule your system. It easy to come up with a single new exception mechanics, not as easy when you have to come with dozens to make some high level element of the system like a class.

And then there are sentiments like yours. Some folks will think it the greatest thing ever and others will think "really!? why do I have to do all this?"

Quote from: BoxCrayonTales;1088172Class bloat goes hand in hand with niche protection. A lot of class bloat can be traced back to developers trying to create new niches.

So what do you mean by niches?

If a setting has a thousand professions each represented by a classes, I argue the problem isn't the design of the system but rather that the author thought there were a thousand elements that players would be interested in playing as a character. The problem lies with the design of the setting or the author's view of a genre.

Now I am going to assume that niche in this case is more of a wargaming element than roleplaying because that the usual context for a criticism of niche. You have your tank, healer, controller, damage dealer, and so on. All things that make since in terms of the wargame that the mechanics of an RPG create but often have little to do with how the genre or setting works.

In which case, my criticism is why there are niches at all? I will be looking to see if the "niches" match anything in how the author view the genre or setting. Or it is an artifact of treating an RPG as a wargame and the campaign as a combat scenario generator.

If it is latter then I would say the author needs to get their head out of their ass and focus more on having the mechanics reflect how they think the genre or setting works. If that makes for something boring or uninteresting then they need to rethink their approach and design something that is exciting enough that hobbyist would want to spend their time experiencing as their characters.

Finally my view that this issue afflict all system designs.

Quote from: Rhedyn;1088175Subtraction by addition.

I would argue that any Pathfinder, 3.5, 4e, or 5e fighter is less Versatile than a DCCRPG Warrior with his mighty deed die.

This isn't without consequences either. Because there nothing for the player to reference for specifics. Because without such a reference, some players feel like they are just throwing darts in the darkness at a target. Playing an elaborate game of "Mother may I" with the referee.

As a result for the above the work for a broad spectrum of hobbyists, the referee has to be aware enough and skilled enough as a teacher and coach to explain how Mighty Deeds works. Then work at establishing trust with the players that they are fair and impartial in their decisions on what constitute a specific kind of mighty deed.

It not as complex as it sounds me writing it. But it is something one needs to be aware of if you want to be successful referee in system with minimalist mechanics.

Quote from: estar;1088168This assumes people want to get into that level of detail in their mechanics. Other are fine in just making a Fighter with a higher dexterity and saying they are swashbuckling fencer while another opts for a higher strength and describes their character as a brute smashing their opponent.

The only issue if the referee ignores the description when it would be a factor. Or the player wants to describe their character in a way that doesn't make sense with the mechanics that does exist.

And that latter one is precisely the problem with most editions of D&D. The high Dex fighter gets gimped by the mechanics.

I was never able to get a satisfying swashbuckler prior to 4E (even the 2e kits were lackluster, the 3e classes were abysmal) because mechanics were always geared around physical armor (particularly when enchanted) being exponentially superior a high DEX (particularly the TSR-era versions were the Dex bonus to AC was always a straight add whether you were in padded or full plate). 4E was the first edition to let AC just scale with level and let light armor with high Dex be even in the ballpark of heavy armor. Similarly, 4E was the first edition where Strength wasn't the only determinator of weapon damage bonuses (3e had finesse, but it applied only to attack rolls, not damage) so you were always gimping yourself as a warrior to not prioritize strength and the heaviest armor you could get.

Now, this is certainly realistic. Real world plate is categorically superior to a gambeson or brigandine vest/coat even if the guy in gambeson is a lot more agile and with a lot of melee weapons striking power is as important as accuracy for bringing down a foe.

But the agile hero being a match for the guy in heavy armor is absolutely a fantasy trope, so if your world is intended to be a fantastic one (replete with dragons, elves, gods and wizards) and not a medieval simulator then there needs to be sufficient room in the mechanics to enable these tropes.

And that's where; short of "all warriors have the same capabilities and appearance is just fluff"; that having some finer grained mechanics for the classes can come in handy. Because I wouldn't want the light and agile warrior to feel like the heavy and strong warrior even if I want them to be about as useful in a party. They can do it in different ways.

So the strong warrior gets better AC from their heavy armor, but the quick warrior gets better mobility and less fatigue outside of combat. The strong warrior has an easier time battering through armor, the quick warrior has an easier time hitting small and quick targets.

It doesn't have to be an extreme advantage; it might be a point or two of AC for the strong warrior in their full armor and an extra pace of movement for the quick warrior; just enough to let them feel different.

It all comes down to how fine grained you like your mechanics. The Mighty Deeds from DCC was mentioned. We could go a step further and just replace every class with single feature of "roll a d6 vs. TN 2 (very easy), 3 (easy), 4 (moderate), 5 (hard), 6 (very hard). If you succeed, you narrate what happens. If you fail, the GM narrates what happens."

But I don't think that would be very satisfying to many people.

There's something to be said for the concept of "Adventure as Problem-Solving Exercise" where your particular bag of race/class/background/gear is your toolkit. For players who enjoy that, a more specific list is a benefit to their creativity rather than a hindrance.

My brother-in-law is a chef. He has a much easier time coming up with a meal when he's got a list of specific ingredients on hand (i.e. specific list of abilities the PC has) than if you told him "come up with something and we'll see if we can get you the ingredients." (i.e. "GM, can I do this with my Mighty Deeds?")

My ideal game falls in the "problem-solving toolkit" level of detail so my comments reflect this line of thinking. As I explained above, I found every edition of D&D prior to 4E woefully inadequate for playing the types of PCs I was interested in (prior to 4E, Palladium Fantasy 1e was my go-to for fantasy precisely because physical prowess and/or enough levels in weapon proficiencies and/or kobold/dwarven crafted weapons that gave bonuses to parry DID let you play a light armor warrior without unduly gimping yourself).

It's probably why I've never had the slightest interest in OSR games. My experience with the genuine article was that its an utter failure at emulating anything I gave a crap about (i.e. light-armored heroes, non-vancian casters and priests who behave like priests not mace-wielding pagan sorcerers in plate armor)... i.e. stuff you see in just about every non-D&D inspired presentation of the fantasy genre)... so why bother with games attempting to emulate that feel?

Quote from: estar;1088187This isn't without consequences either. Because there nothing for the player to reference for specifics. Because without such a reference, some players feel like they are just throwing darts in the darkness at a target. Playing an elaborate game of "Mother may I" with the referee.

As a result for the above the work for a broad spectrum of hobbyists, the referee has to be aware enough and skilled enough as a teacher and coach to explain how Mighty Deeds works. Then work at establishing trust with the players that they are fair and impartial in their decisions on what constitute a specific kind of mighty deed.

It not as complex as it sounds me writing it. But it is something one needs to be aware of if you want to be successful referee in system with minimalist mechanics.

In practice, you are much more likely to get a satisfactory martial via DCCRPG than D&D3.5.

Yeah you run into the "mother may I" approach but in these "flexible class design games" the paradigm is often "father says no" instead which is worse.