Mechanics vs interpretation based RPG systems

[Ajustusdaniel]

a build

By interpretation based, I mean that anything you want to sensibly do, you tell the DM and roll to see if it can happen. "I use control water to suck the water out of my opponent", "I tackle my enemy and beat him with his own weapon".

But roll what? If there's no mechanics supporting the action, then is the DM just making up a probability on the fly?

[EggKookoo]

But roll what? If there's no mechanics supporting the action, then is the DM just making up a probability on the fly?

A good system will provide guidance for the GM to determine it. Some systems outright list probabilities, but no list is without gaps, so in the end the GM may need to come up with something.

[Vahnavoi]

But roll what? If there's no mechanics supporting the action, then is the DM just making up a probability on the fly?

You think the numbers put on paper by some third party are less arbitrary than a GMs? Please. Made up numbers don't become better by being written down. At best, you're hoping a writer did research where your GM didn't, but if you don't trust your GM to do their job, why are you letting them be a GM?

[Satinavian]

You think the numbers put on paper by some third party are less arbitrary than a GMs? Please. Made up numbers don't become better by being written down. At best, you're hoping a writer did research where your GM didn't, but if you don't trust your GM to do their job, why are you letting them be a GM?

Numbers put on paper by a third party have a couple of advantages :

1) They are consistent. The numbers on the paper don't change but a GM hardly remembers all their rulings and difficulties over several sessions

2) Those third party people had time to think about those numbers and talk with each other about them as well while the GM has to decide on the fly to not bog the game down. That doesn't make all the numbers on paper better but on average, they are.

3) Players can know the numbers and plan accordingly. Misunderstanding and miscommunication about expected difficulties are far rarer.

[Ajustusdaniel]

A good system will provide guidance for the GM to determine it. Some systems outright list probabilities, but no list is without gaps, so in the end the GM may need to come up with something.

Well, then that system is providing mechanics guiding what you can and can't practically do. Sure, there's a difference between "No, you can't do that," and "Well, sure, you can do that, if you roll three sixes (IE, in one out of 216 cases)," but in this case I would come down on the side of "Mechanics."

You think the numbers put on paper by some third party are less arbitrary than a GMs? Please. Made up numbers don't become better by being written down. At best, you're hoping a writer did research where your GM didn't, but if you don't trust your GM to do their job, why are you letting them be a GM?

Well, that might be an argument for systemless roleplay. Why use anything anyone else has written down?

But beyond that, if you can trust your GM to have done the research for every possibility any player might raise in a system with no mechanical limit on what characters have what kind of abilities or can attempt what kind of tactics, then, yeah, hang onto that GM.

Back to the initial question. Let's game this out. A player's sorcerer wants to knock someone off a bridge. What does he roll? What does the fighter have to roll to cast a firebolt?

[Vahnavoi]

1) So you are presuming a person with bad memory and self-consistency and are letting the be a GM because...? Why'd a person who can't keep their own rules straight be any good at running rules made by someone else?

2) How are you quantifying that claim at the end?

3) You have a living person at your table that you can ask directly for that information. If they aren't providing you with info, chances are you are not supposed to know your exact chances to begin with. This is why you shouldn't confuse issues of mechanics versus interpretation with issues of available knowledge. A GMs numbers aren't bad just because you don't or can't know them beforehand.

---

Well, that might be an argument for systemless roleplay. Why use anything anyone else has written down?

No, it isn't. You are committing a non-sequitur where you think "not written down" means "no system". All systems start as thoughts in someone's head; the reason to put those thoughts on paper is because someone thinks they're good or useful enough to record for further use. But the process of being written down is not what makes them good or useful.

But beyond that, if you can trust your GM to have done the research for every possibility any player might raise in a system with no mechanical limit on what characters have what kind of abilities or can attempt what kind of tactics, then, yeah, hang onto that GM.

A red herring; nobody does research for everything, not your GM, not book writers. You have a GM at your table precisely because having a living human invent solutions to one-off problems is more practical than trying to write absolutely comprehensive mechanical rules.

Back to the initial question. Let's game this out. A player's sorcerer wants to knock someone off a bridge. What does he roll? What does the fighter have to roll to cast a firebolt?

These aren't the initial question and there isn't a point to answering them outside an actual game being played; that's the point of interpretative GMing.

[Ajustusdaniel]

No, it isn't. You are committing a non-sequitur where you think "not written down" means "no system". All systems start as thoughts in someone's head; the reason to put those thoughts on paper is because someone thinks they're good or useful enough to record for further use. But the process of being written down is not what makes them good or useful.

Writing things down opens up a ton of uses for them- for example, I as a player can read them, and therefore have a sense of what is and is not feasible within a system, whereas I have difficulty doing that with my GM's thoughts.

A red herring; nobody does research for everything, not your GM, not book writers. You have a GM at your table precisely because having a living human invent solutions to one-off problems is more practical than trying to write absolutely comprehensive mechanical rules.

I'll note you're the one who brought up the expectation that the GM should have done the research in order to have the answer to given questions.

These aren't the initial question and there isn't a point to answering them outside an actual game being played; that's the point of interpretative GMing.

So, in your interpretative system, at what point do I, as a player, find out whether or not my fighter can cast firebolt?

[Cluedrew]

1) So you are presuming a person with bad memory and self-consistency and are letting the be a GM because...? Why'd a person who can't keep their own rules straight be any good at running rules made by someone else?

Because maybe one is a brilliant story-teller, world-builder and can create what feel like deep preplanned NPCs on the fly, but this person has no sense of probability, in fact is terrible at math in general, can't quite figure out how all these different abilities come together and is now trying to extrapolate from the time John's cyborg knocked his evil win off the cliff and the time Crystal dropped the vending machine on the mob boss and now Ryan's really short but absurdly muscled mutant is trying to knock a robot off a roof top.

2) How are you quantifying that claim at the end?

Time spend and number of people who looked at in. A system that was made up, tested, updated and so on until the creators thought it was up to snuff is - all other things being equal - going to be a lot better than one that was made up just now. In fact if the GM picked up the system and decided it worth running it probably passed their inspection. Unless you just play D&D (or some other system in theory) because its the only one you know.

I got nothing to say about point three. Both sides can work just fine.

[EggKookoo]

Well, then that system is providing mechanics guiding what you can and can't practically do. Sure, there's a difference between "No, you can't do that," and "Well, sure, you can do that, if you roll three sixes (IE, in one out of 216 cases)," but in this case I would come down on the side of "Mechanics."

Okay, if you don't mind clarifying. Are we saying an "interpretation" game has no mechanics (i.e. rules) at all?

[Pex]

You think the numbers put on paper by some third party are less arbitrary than a GMs? Please. Made up numbers don't become better by being written down. At best, you're hoping a writer did research where your GM didn't, but if you don't trust your GM to do their job, why are you letting them be a GM?

The mechanics game was bought precisely so no one has to make up the numbers. Of course those numbers the game designers come up with are arbitrary, but that's their job. The numbers have to come from somewhere, and we pay the game company for them. Now the DM is free to run the game instead of doing statistcal analysis.

[Telok]

I'm just saying D&D currently is more akin to the "interpretation" model Drache64 is talking about than the "mechanics" one.

See, I don't know about that. Drache64's "interpretation" thing sounded more freeform or rules-light. Although maybe the Amber Diceless system would also fit, and that was pretty "mechanics first" and not rules-light even if it did tie it's fiction and mechanics very tightly.

I think OD&D and 1e AD&D were pretty heavy on interpretation and DM discretion. Classes were pretty basic, no skill system as many gamers today would recognize it, and all those seemingly random tables and "sub-systems" in AD&D were pretty obviously the result of someone coming up with a rule to address something a player wanted to do. And all those random bits were also "optional".

AD&D 2e, and after that the WotC D&Ds, are all just mechanic based re-writes of the previous edition. People only appeal to a completely abstract and disconnected sort of basic die roll method when they run out of guidance from the system. Anywhere there's actual rules people use them (when they remember they exist) in preference to, and often overriding any player asks or fiction layer action.

The Paranoia anniversary edition is much more fiction driven. There aren't these D&D-esqe sub-systems and variant rules scattered around. It's all "you have stuff written on your character sheet, roll a d20 under that number but higher is better". Everything else not covered by the character sheet abilities (which is very little) is a question of "is it funny" and maybe flipping a coin. The system enacts the fiction without bodging, hacks, or the DM having to make up rules and sub-systems.

Pendragon is mechanics heavy and has several sub-systems, notably the winter court rotine and inheritance stuff, but it matches the fiction you're running beautifully. However it's also quite explicit about the bounds of that fiction and the fact that deviating from Arthurian legend knighthood is not supported by the rules.

So my understanding of the original question is more about the "rules-light vs rules heavy in play" rather than "mechanics vs free-form during design", because while there's significant overlap in some systems I've met other systems that don't correlate that way.

The mechanics game was bought precisely so no one has to make up the numbers. Of course those numbers the game designers come up with are arbitrary, but that's their job. The numbers have to come from somewhere, and we pay the game company for them. Now the DM is free to run the game instead of doing statistcal analysis.

Sorry for the double post but I'm on the phone and need to hit this one up.

I recently did research for game numbers. Specifically I combed through published research papers for graphs and numbers relating to the probability of people seeing things. Then I mapped that to a game, with numbers. I took what I learned from the research and the stats of what I thought the people involved would be. Then took the odds of those stats rolling different %s (not a flat math system) and mapped them to the graphs. The resulting "did they see it" stuff should work for bunnies in a field, people in a shop, jets in the sky, and battleships on the horizon.

The numbers don't need to be arbitrary or made up. The designers should know the % outputs of their system. They absolutely can map those %s to the kind of stuff that actually does happen in RL.

[EggKookoo]

See, I don't know about that. Drache64's "interpretation" thing sounded more freeform or rules-light. Although maybe the Amber Diceless system would also fit, and that was pretty "mechanics first" and not rules-light even if it did tie it's fiction and mechanics very tightly.

Yeah, I backed up and cast about for a definition of these terms.

[Vahnavoi]

Writing things down opens up a ton of uses for them- for example, I as a player can read them, and therefore have a sense of what is and is not feasible within a system, whereas I have difficulty doing that with my GM's thoughts.

Again, it's a live person at your table, you can ask them directly for information. The only time you need to guess is when you're not supposed to know in the first place. Interpretative GMing is not synonymous with hiding knowable information.

I'll note you're the one who brought up the expectation that the GM should have done the research in order to have the answer to given questions.

You are correct; no-one's still doing or expecting anyone to do research on everything.

So, in your interpretative system, at what point do I, as a player, find out whether or not my fighter can cast firebolt?

At the point at which you think to ask, provided it is something that you, as a player, are meant to know. Or, in the case you never do think to ask, at the point where the GM deems the answer to be immediately obvious and relevant to your decisions.

---

The mechanics game was bought precisely so no one has to make up the numbers. Of course those numbers the game designers come up with are arbitrary, but that's their job. The numbers have to come from somewhere, and we pay the game company for them. Now the DM is free to run the game instead of doing statistcal analysis.

Paying money for something is no guarantee of quality. So you paid a company to make up numbers instead of your GM making up numbers; this means those numbers are less arbitrary because...?

---

Because maybe one is a brilliant story-teller, world-builder and can create what feel like deep preplanned NPCs on the fly, but this person has no sense of probability, in fact is terrible at math in general, can't quite figure out how all these different abilities come together and is now trying to extrapolate from the time John's cyborg knocked his evil win off the cliff and the time Crystal dropped the vending machine on the mob boss and now Ryan's really short but absurdly muscled mutant is trying to knock a robot off a roof top.

You just described:

A) A GM who ought to improve a facet of their GMing

B) the writers of the World of Darkness systems.

But in case it's A) : why do you think a math heavy mechanized system would be an answer to this person's problems?

Time spend and number of people who looked at in. A system that was made up, tested, updated and so on until the creators thought it was up to snuff is - all other things being equal - going to be a lot better than one that was made up just now. In fact if the GM picked up the system and decided it worth running it probably passed their inspection. Unless you just play D&D (or some other system in theory) because its the only one you know.

That looks like an empirical claim. You'd like some statistics about numbers in published game systems versus numbers decided by individual game masters to support it. Now, it's reasonable to expect that adults who've been taught probability do better than 10-year-olds, but it's equally true 10-year-olds picking up their first game system are incapable of analyzing or even appreciating what the adults did. For adults playing, I'd say it's a toss-up, because state of professional tabletop game design often isn't all that hot compared to the amateur side of things. See aforementioned comment on World of Darkness. I'd be happy to be proven wrong.

I got nothing to say about point three. Both sides can work just fine.

Part of the reason why I'm asking to consider availability of information separately.

[Ajustusdaniel]

At the point at which you think to ask, provided it is something that you, as a player, are meant to know. Or, in the case you never do think to ask, at the point where the GM deems the answer to be immediately obvious and relevant to your decisions.

---

So in my case, that would be at or before character creation, along with follow-up questions as to "How does that work, in play?" And "Could I be better at that if I did X?" And "Then that would make me worse at Y?" So what we wind up with looks, from my perspective, an awful lot like a mechanical system, wherein the details of the system have been transmitted to me orally rather than textually.

[Vahnavoi]

---

So in my case, that would be at or before character creation, along with follow-up questions as to "How does that work, in play?" And "Could I be better at that if I did X?" And "Then that would make me worse at Y?" So what we wind up with looks, from my perspective, an awful lot like a mechanical system, wherein the details of the system have been transmitted to me orally rather than textually.

The system would just basic semantic logic based on game premises. F. ex., D&D doesn't usually say "Fighters can't cast firebolt" ; instead, that is a statement you can arrive at using basic eliminative logic: the game has a specific mechanic for firebolt and fighters typically don't get it.

[Satinavian]

1) So you are presuming a person with bad memory and self-consistency and are letting the be a GM because...? Why'd a person who can't keep their own rules straight be any good at running rules made by someone else?

Because everyone should be able to GM ? If you can claim superior DMs, i can claim superior designers. But more importantly we should make running a game easy to get enough DMs.

2) How are you quantifying that claim at the end?

I don't need to quantify it. Qualifying is enough. Several people collaborating with enough time to think stuff over tend to produce better results than one person under time pressure. Do you disagree ?

3) You have a living person at your table that you can ask directly for that information. If they aren't providing you with info, chances are you are not supposed to know your exact chances to begin with. This is why you shouldn't confuse issues of mechanics versus interpretation with issues of available knowledge. A GMs numbers aren't bad just because you don't or can't know them beforehand.

When the PCs start to ask the DM for chances for everything they are only considering to, it bogs down the game to some absurd degree. Especcially with multi-step plans and difficult preparations. We already get hour long planning sessions as it is.

[Ajustusdaniel]

The system would just basic semantic logic based on game premises. F. ex., D&D doesn't usually say "Fighters can't cast firebolt" ; instead, that is a statement you can arrive at using basic eliminative logic: the game has a specific mechanic for firebolt and fighters typically don't get it.

If I have a system where I go in having a good idea what my character is capable of doing, and how to build my character to be better at certain things than others, I'm happy. We could quibble about the best way to present that information, but at the end of the day, there are still mechanics underlying it.

[Vahnavoi]

Because everyone should be able to GM ? If you can claim superior DMs, i can claim superior designers. But more importantly we should make running a game easy to get enough DMs.

Maybe everyone should be able to, but not everyone is. Mechanics are a dubious way to make running a game easier. Again, consider the historical example of Kriegsspiel. The umpire (functionally, game master) was given more control because the mechanics were taking too long to resolve things; this put greater pressure on the umpire to be an expert in warfare, but consider the implication of reversing the process. You can eliminate the need for the umpire to be an expert, even eliminanate the need for an umpire alltogether, but this requires rules that are complete enough that knowing them is effectively sufficient to make the players into experts.

The same applies to roleplaying games. There's a point beyond which mechanical design can only make things easier for the GM by demanding more from the players.

I don't need to quantify it. Qualifying is enough. Several people collaborating with enough time to think stuff over tend to produce better results than one person under time pressure. Do you disagree ?

I do partially disagree, but I'm struggling to find a good way and example to explain my disagreement. Watch this video and pay special attention to the point where it talks about decisions in strategy games. The generalized form of the point being made is: groups of humans only defeat individual humans in some types of decision problems. It is an open question whether the parts of game design material to our topic fall in that group. That's why you'd want to quantify it.

This matters more in theory than in practice, because in practice, mechanics of even a lot of published game systems are basically made by one person. The number of tabletop games which had their mechanics actually designed as a coherent group effort is limited.

When the PCs start to ask the DM for chances for everything they are only considering to, it bogs down the game to some absurd degree. Especcially with multi-step plans and difficult preparations. We already get hour long planning sessions as it is.

Intricate mechanics can bog down gaming to equally absurd degree. You'd really want to quantify the time taken for your plans in different kinds of systems to see whether mechanization is actually helping you at all.

---

If I have a system where I go in having a good idea what my character is capable of doing, and how to build my character to be better at certain things than others, I'm happy. We could quibble about the best way to present that information, but at the end of the day, there are still mechanics underlying it.

As noted by myself, absence of writing doesn't mean absence of system. If that was the extent of it, this discussion would end here, but I'm not sure you understood what I was talking about.

Let me try to find a better example: in context of 3rd edition D&D, some cheeky wanker noticed that the system doesn't "technically" define "dead" or "death" as a conditions and thus the system doesn't prohibit you from taking actions when dead. This argument was obviously nonsense; the system had just used "dead" and "death" in plain English, with the expectation that whoever reads actually understands English as a natural language. This is the most basic way in which interpretation comes in in all systems written mostly in a natural language. They rely on a human to get over naturally occurring semantic ambiguity and incompleteness. The "mechanics" are interpretative powers of the human brain, and thus what you can get out of written rules is always limited by the brains of the actual people at your table.

[NichG]

Because everyone should be able to GM ? If you can claim superior DMs, i can claim superior designers. But more importantly we should make running a game easy to get enough DMs.

Personally, I'm not really a fan of this direction. I don't want systems that protect me from bad GMs, I want systems that help good GMs and players shine and reach new heights. I'd much rather play a game with a 25% chance of being transcendant and a 75% chance of being awful than a game with a 100% chance of being decent but not spectacular.

I think its fine if everyone can try to GM. They can do badly, they can do well, they can even say 'I want to improve' and get better. But if you had a system that guaranteed that all GMs would be equal, by virtue of taking responsibility out of their hands, I wouldn't want to play it.

[Ajustusdaniel]

As noted by myself, absence of writing doesn't mean absence of system. If that was the extent of it, this discussion would end here, but I'm not sure you understood what I was talking about.

Let me try to find a better example: in context of 3rd edition D&D, some cheeky wanker noticed that the system doesn't "technically" define "dead" or "death" as a conditions and thus the system doesn't prohibit you from taking actions when dead. This argument was obviously nonsense; the system had just used "dead" and "death" in plain English, with the expectation that whoever reads actually understands English as a natural language. This is the most basic way in which interpretation comes in in all systems written mostly in a natural language. They rely on a human to get over naturally occurring semantic ambiguity and incompleteness. The "mechanics" are interpretative powers of the human brain, and thus what you can get out of written rules is always limited by the brains of the actual people at your table.

Sure- I wouldn't want to replace my GMs with a GM-atron even on purely a rules level, setting aside the storytelling and worldbuilding aspects of the role. But the OP defined a mechanics based system as

Most of what you want to do has to be interpreted through available mechanics like feats and class abilities.

And what I'm trying to work out it was they meant by an interpretation based system being defined in opposition to that. In which case, the question of whether or not the 'mechanics like feats and class abilities' should be written down is a moot point¹. If the question "can I do X," is not defined by available mechanics, whether those are created by the GM or a third party, whether they're written down in books or related verbally by the GM, how do I develop a sense of what my character can do?

[Vahnavoi]

The same way you'd do in a game with hidden rules: through trial and error, using your natural language understanding of the game situation as a starting point. The difference is largely on the GM's side, not the player's.

[Ajustusdaniel]

The same way you'd do in a game with hidden rules: through trial and error, using your natural language understanding of the game situation as a starting point. The difference is largely on the GM's side, not the player's.

Right. So by the end of character creation, I've arrived at what

looks, from my perspective, an awful lot like a mechanical system, wherein the details of the system have been transmitted to me orally rather than textually.

Since the OP seems to have been asking us to evaluate the benefits of mechanics vs interpretation from a player's perspective, are you going with "there is little to no difference?"

[Cluedrew]

On Rules as Tools: This is in partially in response to Vahnavoi and NichG but also just covers a framing decide. Whenever I design or think about systems I start with no rules. Then I add and later change rules to try and improve the situation. So if I have created or just picked a system to run (outside of a play-test/test-run) than I have decided that its rules are an improvement over not having rules. What makes it an improvement? It improves my ability to run the game. And at a certain level (possibly not exactly and often framed differently) I think that is what all GM, by that title or not, all looking for, rules that help them run the game that is better than sitting down with some blank pieces of paper.

I just realized I don't know how this relates to the main topic anymore. I might have to re-focus.

[anthon]

i come from the "Make a Dex/Con check at -8" school of DMing, Class of '92.

Back in those days, everybody had stats of around 12-19 (except dump stats) and if you rolled on a d20, you had around a 75% chance of succeeding overall.

So if it was meant to be easy, we might add +2 or +4 to the check.

if it was something you sorta could fail at, we would make it a flat check.

if it was hard, -2 to -4

and if it was insane, daring, or outlandish, -6 to -8.

Sometimes we would wed a Dexterity Check to an Attack or whatever. Like trying to Balance on a tree limb in a wind storm while trying to move silently into position to leap atop and backstab a goblin captain.

or trying to run up the back of the neck of a dragon to get into position to stab/lasso them.

Rules were there to provide examples of where difficulties were, and we use those as numbers of a thermometer to gauge what we guessed was "hot" or "cold" for difficulty.

Not having a particular skill or feat or proficiency only meant a higher difficulty. If your stat was amazing, or your background said you could probably do it, or would at least try, then we would assign a difficulty.

Let the dice/Kismet decide whether you were worthy.

[anthon]

Well, then that system is providing mechanics guiding what you can and can't practically do. Sure, there's a difference between "No, you can't do that," and "Well, sure, you can do that, if you roll three sixes (IE, in one out of 216 cases)," but in this case I would come down on the side of "Mechanics."



Well, that might be an argument for systemless roleplay. Why use anything anyone else has written down?

But beyond that, if you can trust your GM to have done the research for every possibility any player might raise in a system with no mechanical limit on what characters have what kind of abilities or can attempt what kind of tactics, then, yeah, hang onto that GM.

Back to the initial question. Let's game this out. A player's sorcerer wants to knock someone off a bridge. What does he roll? What does the fighter have to roll to cast a firebolt?

The purpose of game mechanics is to inform the reader that fire from a cutting torch is hotter than fire from an oven.

that's the short and skinny of all tables, rules, and mechanics:

you are comparing things and you can't be sure your reader will know the differences, so you jot down some quick comparisons, so they understand things like bamboo are strong, balsa is weak, and hardwoods are brittle.

You let the reader know that obsidian is sharp, steel is strong, and copper conducts heat as well as lightning bolts.

When the player enters a fictional setting, the GM may have things like multi-eyed argos blobs or tentacle acid monsters. But how acidic? How strong is that blob? What if two wizards each cast one charm monster spell, and the Argos Eye Blob has to fight the Tentacle Acid Monster?

You now have to arbitrate how these two creatures might interact.

Mechanics are there to help people resolve the many arguments you see in Free Form roleplaying. Their tool is number.

[Pex]

Paying money for something is no guarantee of quality. So you paid a company to make up numbers instead of your GM making up numbers; this means those numbers are less arbitrary because...?

Naturally. As I said, see 3E Truenamer or 4E. The successful games will have mechanics that do work well being fun to play. Sometimes it's just a matter of personal taste. Some people very much like 4E after all. People today are still playing 3E. It's the game designers' job to come up with the rules. They're being paid to do so. The DM spends time setting up the campaign. Having defined rules means he can focus on the scenery and plots. I would trust game designers who spend 40 hours a week working on the rules to come up with better arbitrary numbers more than a DM who works 40 hours a week at some other job, may have a family, can only spend an hour prep time for the game. For the youngsters, better than the DM who has to study for Finals or even only in high school going through puberty.

[Ashiel]

Mechanics vs interpretation based RPG systems.

People may nit pick on my terms here, and they'll probably all be right. So please focus on what I'm trying to say not any wrong terminology I accidentally use.

By mechanics based I mean an RPG system like D&D. Most of what you want to do has to be interpreted through available mechanics like feats and class abilities. Do you want to disarm your opponent with your Sorcerer staff? Too bad, you're not a battle master. Do you want to blast a spell at your enemy to knock him off a bridge? Better hope you're a warlock with repelling blast.
Pros:
Tactical mechanic combat
Some people need structure in their imagination

Cons:
Limits to what you can do.
Able to break the game mechanics
Players encouraged to optimize a build

By interpretation based, I mean that anything you want to sensibly do, you tell the DM and roll to see if it can happen. "I use control water to suck the water out of my opponent", "I tackle my enemy and beat him with his own weapon".
Pros:
Can do anything you can reasonably think of.
Free to think outside the box solution, no cookie cutter approach to combat, builds, spells.

Cons:
You might suck at imagination, no cookie cutter approach to combat, builds, spells.

What do you guys think? I'd love to read some discussions on this.

I prefer the more mechanical system like D&D 3.x/d20, because IMHO it facilitates better roleplaying, because it provides a stable framework for what is reasonable, and as a direct result you can quickly learn how things relate to one-another. You can understand what you can do, why you can do it, and so on, and that allows you to engage in the world more as if you are living in it and have a more immersive experience.

A simple example I use most of the time when discussing topics like this is, in D&D if you understand that the DC to climb a tree is 15, and you have a +5 climb modifier, you can climb trees. You don't have to ask the GM if you can climb a tree. If the difficulty is higher, such as being heavily covered in wet moss (say a +2 to the DC), or someone oiled the tree by casting grease on the it (say +10 to the DC), you can easily understand why you can normally climb trees easily but this one is harder and understand how big a deal it is to be able to casually climb those trees relative to what you're capable of now. It also means that you might intuitively understand how to solve such problems: if there's thick moss grown on the tree, you might spend a bit using a shortsword to scrape the moss off as you're climbing it (effectively negating the +2 to the DC in exchange for climbing up slower) or attempt to foil the lubricant by patting the tree with crushed chalk or flour; for you understand why there is a difference and can interact with it naturally.

This leads to immersive emergent gameplay. If you're wandering through a forest with a thick canopy (dim light, things are difficult to see clearly) on the way to find a tower somewhere in the forest and you're not sure which way you are going, but you have a +5 climb modifier, you don't say "Hey GM, can I climb a tree to look around? If I get a better view from up there, will I be able to see any better? Will I...?". Instead, what you say is "I can climb trees, so I climb up a tall one and have a look around for the tower." You know that you can climb trees, and you know that getting clear line of sight in bright light above the canopy will allow you to see the tower without trouble and orient yourself. You have, naturally, intuitively, interacted with the world as if it were in a sense real and as a result makes it easier to roleplay without pauses, for both the player and the GM. The GM likewise benefits because s/he doesn't have to decide whether or not you can climb this tree and can instead focus on describing the results of your actions and further the story from there.

It opens avenues for more pro-active gameplay and decision making. Knowing how the world works allows you to act accordingly. Knowing that it's harder to see you at a distance (-1 penalty to Perception per 10 ft.) can influence your decisions when tailing a suspected spy through a city, as you decide whether or not you want to risk them getting out of sight because they are far enough away that they could turn a corner or two before you could see which way they went, or if you want to stay further back to ensure you aren't spotted in a crowd.

There is functionally nothing lost in a crunchy system since nothing is off limits if the GM and group wants it so. You can always simply ask if you can use a thing in a way not specified, but these sorts of considerations are actually made easier to handle for the players and GMs by a consistent framework that you can reference. For example, if your party is ambushed by some invisible stalkers or rogues with greater invisibility cast on them, you can ask to do something not covered by the rules such as "Can I scatter powdered sugar around the room to reveal the invisible people?" - "Sure, but it won't reveal them completely, and the cloud will make it hard to see anything in the room for a bit, so they'll loose invisibility and everyone in the room gains concealment 20% for...two rounds, then only the invisible people will have concealment as the dust settles".

In a similar vein, being able to compare things at different scales can help decide on things that aren't already covered in the rules. If you know that iron has hardness 10, you know you need to have a fire that's hotter than 10 points of fire damage (on average) after halving it to melt the metal, so if a minotaur bull-rushes a PC into the coals of a giant furnace, you can say "Oh, well that's probably at least 6d6 fire damage since it's gotta be hot enough to actually get the metal soft".

So I like the mechanical stuff because I find it supports roleplaying and doing interesting things more than flavors of "mother-may-I".

[Pex]

It's great when people say the same thing I've been saying for years but with better verbiage.

What Ashiel said.

[Vahnavoi]

Right. So by the end of character creation, I've arrived at what Since the OP seems to have been asking us to evaluate the benefits of mechanics vs interpretation from a player's perspective, are you going with "there is little to no difference?"

No, but to make the difference noticeable to the player, they'd have to try out the same actions under different rulesets and different GMs. I'd advise you to go back and read my first post to this thread, 'cause I feel this will turn into me repeating myself otherwise.

Shortly: there's a difference, firstly in how long it takes for a game master to process a player move and secondly in accuracy of resolution, but this metric doesn't always favor one type of resolution over the other. It depends on what you're trying to model, to what degree of fidelity and how long you want to spend time on it. That's why sane game design uses both styles.

---

On Rules as Tools: This is in partially in response to Vahnavoi and NichG but also just covers a framing decide. Whenever I design or think about systems I start with no rules. Then I add and later change rules to try and improve the situation. So if I have created or just picked a system to run (outside of a play-test/test-run) than I have decided that its rules are an improvement over not having rules. What makes it an improvement? It improves my ability to run the game. And at a certain level (possibly not exactly and often framed differently) I think that is what all GM, by that title or not, all looking for, rules that help them run the game that is better than sitting down with some blank pieces of paper.

I just realized I don't know how this relates to the main topic anymore. I might have to re-focus.

That's all fine. I'm not against "rules as tools" mindset. It does relate to the main topic in a very simple way: when testing out mechanics, you'd also want to test them against freeform judgements by living humans. To use my prior example, you don't need to spend ink on detailing what "dead" and "death" means when everyone over age 4 understands it means you don't get to do things as your character anymore.

---

. It's the game designers' job to come up with the rules. They're being paid to do so. The DM spends time setting up the campaign. Having defined rules means he can focus on the scenery and plots. I would trust game designers who spend 40 hours a week working on the rules to come up with better arbitrary numbers more than a DM who works 40 hours a week at some other job, may have a family, can only spend an hour prep time for the game. For the youngsters, better than the DM who has to study for Finals or even only in high school going through puberty.

The average tabletop game designer doesn't spend 40 hours a week on game mechanics. Locally, I know for fact that majority of games were done by individual people as a sideprojects to their real work as book authors, videogame designers, artists and what not. The tabletop roleplaying industry is niche and amateurish enough that this is honestly quite rare to find a person crunching numbers as their actual day job even in big companies like Paizo or Wizards of the Coast.

So what you're actually paying for in a published game design, is only tiny part of effort spent on rules, and mostly marketing and logistics costs, graphic design etc.. It doesn't take much for an interested amateur who doesn't have to worry about the business side of things to do better on the numbers side.

Also, for the youngsters crunching it for the finals or an adult mired in work and family life... I heavily doubt a mechanics-intensive game design is actually in their interests.

[PhoenixPhyre]

I find that mechanics-heavy games cause way more jarring problems than they solve.

Take, for instance, that DC 15 "climb a tree" check. An average human child commoner 1 has at most a +0 to Climb. So they not only can't take 10, they fall out of the tree 70% of the time when climbing under any pressure.

Whereas in real life, kids climb trees all the time--falling out of them is the exception. Even under pressure. I'd say that child-me (who was anything but athletic) is a better climber than current-me. More coordinated, way less fearful, and most importantly a whole heck of a lot lighter.

So these mechanics claim to simulate reality...but instead make things worse. While at the same time introducing substantial overhead to every single interaction. Now the DM must either keep a table open (ugh, table lookups are a drag) or keep dozens of tables constantly in memory including interpolating between tables of modifiers. And no, the players can't take that load entirely because they don't (and can't) know all the details. Unless you only ever use "stock" elements (if it's not in the table, it doesn't exist). Which is bland and repetitive.

Beyond even that, why are we rolling to climb an average tree in the first place? Either it's a special tree with a special place in the world and it's do or die (and thus won't use the book numbers at all) or (much more likely) it's a relatively inconsequential part of something else. No substantial chance of failure, no particular interesting results of failure or success...why roll in the first place? Just narrate that you climbed the tree and move on. Maybe cost extra movement if you're under pressure and need to do it faster than someone else.

Edit: I will say that more "interpretive" games have other issues as well. Just different ones. So IMO, there's a balance to be struck. Call it "rules medium" or "rules as shared toolkit". The rules aren't in control, but they're building blocks for tables to use. You don't get the mathematical precision[1], but you also get turns that don't take 10s of minutes. You get the much better narrative fit, but you don't get the constant need to discuss how to translate a player action into a character action.

[1] which I'd say is mostly false and unrealistic--you can't pin down the success probability of most things beyond "these things I can always do (unless something radical intervenes", "those things I can often do" and "those things I can only do when I get really lucky"). People are bad with probabilities in real life. Horribly so. And most probabilities aren't even meaningful for things you do only a few times.