What insights into the rules of D&D could a character gain through experimentation?

[obryn]

I remember a random little thought experiment I read somewhere as the basis for this, but wanted to see how far down the rabbit hole we could go. I'm personally very comfortable with games where the rules are not the world's physics, and instead represent kind of a narrative layer on top of the fiction...

...But in higher-sim games, this is not the way of things. And, if the laws of physics are reflected in the rules of the game, I started thinking about wizards. This is intended kind of tongue-in-cheek.

Okay. So there's some wizards of varying levels in a wizard's tower somewhere. They are indubitably intelligent - smarter than we are - so they start experimenting with the laws of the universe.

(1) They know that their members can prepare X, Y, Z spells of I, II, III levels, and that these are always distinct quantities in a regular progression. They have at this point figured out the spell chart.

(2) The ranges and areas of effect of their spells increase in distinct, quantized measurements - usually 5' or 10' each* - and that this increase corresponds with the number of spells prepared.

(3) These gains only come in specific units. At this point, these Wizards have figured out Levels as a fundamental feature of the universe, and can even engineer out their specific levels through use of (for example) spell ranges.

(4) By blasting damaging spells at uniform targets of specific kind - like walls and blocks of wood - they can determine the percent of the time the item is blown up. By letting Wizards of different levels do this, they can further chart the efficacy - the block of wood is destroyed X% of the time by a Fireball spell by a wizard of Y level. At this point, they can figure out Hit Points as a fundamental feature of objects.

(4a) If they're particularly unethical, they can figure it out about goblins or kobolds, too.

(5) By the same token, they can determine that the probability of destroying targets is shaped in a curve best represented by Xd6, where X is their Level.

(6) With repeated uses of spells like Charm Person on each other, they can determine (a) saving throws, (b) Wizards' Will save progressions, (c) the caster's Intelligence bonuses, and (d) the targets' Wisdom bonuses.

(7) By doing this against (for example) town guards, they can further work out non-Wizards' Will Save progressions and Wisdom scores.

(8) The Wizards can get people to lift heavy things. They notice the maximum amounts their subjects can lift are, in turn, quantized into distinct units of measurement instead of smooth. By casting buffing spells, they can work out the progression and notice that it matches at every step. They have just worked out possibly as much as the entire 3-18 or 3-22 scale for Strength and can refer to it by a number.


Okay. So at this point, these Wizards can speak about Levels, Saving Throws, Wisdom Bonuses, Strength, and Hit Points. What else can they do? Can they - from the spell list in the PHB - figure out the entire rules of the game in which they are characters?

-O

* As an aside, "five feet" is a fundamental measurement of the universe, bound into its metaphysical laws.

[PetterTomBos]

The problem with quantum mechanics is that it is quantized into discrete little bits, that seemingly work in weird ways...

DUN DUN DUNNNNN!!!

(Tongue in cheek of course ;) )

But seriously, that would be a fun scenario. Hmm, about lvl.s and spells, yup, they could figure it out. But, given that NPC's don't lvl. that often, feats that help spells, and such, it would be a really hard task. But, Einstein was probably lvl. 6 ;)

(This could esily be "fixed" by the DM ruling that NPC' gain lvl. continously, and that some may not gain all features and so on)

[obryn]

Yep, and going further...

(9) Bonus spells for high intelligence, as soon as they figure out the progression system. They can now determine their own Intelligence scores based on how many spells they can prepare when they get their first spell of X level. They now know who's smarter.

(10) Their actual numerical level - in case I wasn't clear before - because it's used in spell range and area formulae. So they could (and maybe should) refer to themselves as a Level 5 Wizard and have it mean the same thing as when a player says it.

(11) Get their cleric buddies into the mix; priests of a magic or knowledge god should be amenable. Their levels and bonus spells from Wisdom should likewise be apparent. So now you can refer to "James, the Level 7 Cleric." They also know there's a Wisdom statistic, since the same thing that makes Clerics learn more spells also contributes to the % chance a Charm spell will affect them.

(12) Can they at this point figure out ... characters' or monsters' hit points, the amount of damage done by daggers vs. swords, and how much a Cure X Wounds spell heals? This is less immediately obvious, but...

For certain, we can add to the list (1) That there are further variations in Intelligence and Wisdom based on bonus spells, (2) Cleric levels, and (3) a Wisdom statistic beyond the mere bonus.

-O

[Kazyan]

This is extremely handwavable with "they're Wizards; they smart enough to get around all the problems for knowing this stuff!" as an ether variable. But I do have to bring up something: a lot of this would require massive sample sizes to determine anything. The poor shmuck who gets hit with Charm Person 1000 times resists it, what, 213 times? Uh-oh, there goes that granularity for Int and Wis mods; it's not a clean 5% multiple; better hit him with another 4000 and wait why is he crying.

You have to be able to quantify that saving throws work in steps of 5% before you get determine Int mods and Wis mods, but even then, there's a lot of confounding variables that could throw out otherwise accurate hypotheses. Don't even get me started on circumstance modifiers throwing off all the data.

[obryn]

Well, keep in mind - this is a tower of bored wizards, all of whom are very smart on account of being wizards. I will posit that these don't rely on any probability distributions and are immediately "visible" with a pool of a few dozen Wizards:

(1) Quantifiable caster levels for wizards and clerics, including the actual level numbers, based on both casting capability and experimentation with known spell ranges.

(2) Ditto for Bards, Paladins, etc - really anyone who can advance in level and gains both observable spells and caster levels. (The existence of character class as a concept in and of itself may be notable.)

(3) Sorcerer levels, likewise, with their progressions - including the knowledge that they get higher-level spells later than you might expect given the efficacy of their lower-level spells.

(4) Stepped increases for Bonus Spells, giving insight into Intelligence, Wisdom, and (for Sorcerers/Bards) Charisma as definite stats.

(5) The Strength stat and at a minimum its Max Lift progression

(6) The fact that an increase through a known spell (Bull's Strength) is (for example) +4 steps on the table they worked out in (5).

(7) The fundamental workings of the universe are quantifiable in 5' increments, and that this number has a metaphysical importance.

All of these would be directly observable to a character if rules of game = rules of physics.

(Hit points and saving throws do require large repetition, but they're far from outside the realm of experimentation. For example, if a Wizard invents a 0-level spell "Make a Will save or say, 'I FAILED MY WILL SAVE'" then with enough castings, the distribution becomes pretty clear.

-O

[The Grue]

This is a magnificent thread and everyone in it should be proud. I salute you.

I'm going to suggest that my party wizard start such experiments to delve into the fundamental laws of the universe.

EDIT: Regarding number 2 above, they may be able to infer the existence of a fourth characteristic, by noticing that the saving throws of any two Paladins of a given level vary by a hard value that is tied to neither Wisdom nor Intelligence nor Strength. Charisma might be to them what the graviton is to our scientists; they know it has to exist, but they don't know much beyond that.

Likewise, they might observe another mystery variable affecting the number of extra spells a Sorcerer gets over and above his base progression and the save DCs of his spells, which they might mistake for Intelligence given Sorcerers are arcane casters like Wizards. This would inevitably result in a Wizard with an apparent Int score of 16 being demonstrably more intelligent than a Sorcerer with an apparent Int score of 18 (perhaps through some sort of standardized problem-solving test), and yet the Sorcerer's spells are stronger by all observations. Conclusion: The characteristic that governs Sorcerer spells is another mystery variable like Charisma, but there would be no direct evidence linking it to the variable that affects Paladin throws.

Unless they got their hands on a Paladin/Sorcerer multiclass and were able to separate and independently account for level-dependant variables from each class on the overall character scores.

Also if any of their test subjects have ranks in Open Lock they would learn that, in the absence of any threats or time constraints, an individual can always open locks that he otherwise would fail to open 50% of the time - in effect, they have discovered the Take 10 rule, and with that knowledge could catalogue the DCs of various common lock types and determine the number of effective ranks a person has in Open Lock.

[obryn]

Also if any of their test subjects have ranks in Open Lock they could work out the approximate DCs of various common lock types. They would also learn that, given adequate time and in the absence of any threats or time constraints, an individual can always open locks that he otherwise would fail to open 50% of the time - in effect, they have discovered the Take 10 rule.

Hah! Perfect! Another good one.

-O

[Sith_Happens]

This is extremely handwavable with "they're Wizards; they smart enough to get around all the problems for knowing this stuff!" as an ether variable. But I do have to bring up something: a lot of this would require massive sample sizes to determine anything. The poor shmuck who gets hit with Charm Person 1000 times resists it, what, 213 times? Uh-oh, there goes that granularity for Int and Wis mods; it's not a clean 5% multiple; better hit him with another 4000 and wait why is he crying.

You have to be able to quantify that saving throws work in steps of 5% before you get determine Int mods and Wis mods, but even then, there's a lot of confounding variables that could throw out otherwise accurate hypotheses. Don't even get me started on circumstance modifiers throwing off all the data.

Pretty sure you could get around at least some of this with a sufficiently rigorous knowledge of statistics.

[obryn]

Pretty sure you could get around at least some of this with a sufficiently rigorous knowledge of statistics.

Yep, and let's face it - if these guys have sufficient knowledge to put reality in an armlock and force it to cry for mercy using only the powers of their brains, basic probability is child's play.

-O

[ReaderAt2046]

In Morithias's Grimoire of The Rift Campaign, they've figured out the level-up mechanic in-universe (though admittedly that one's not quite a normal example. The premise is that the DM of a group of gamers stumbled across a real spellbook and accidentally did something that caused us to wake up inside the campaign we were starting.)

[valadil]

I'm with you until #4. It takes a huge leap of faith to assume that all walls and blocks of wood take the same amount of damage to destroy. I don't se why scientifically minded wizards would make that assumption.

[NichG]

One slight complication with using large blocks of material - objects do not heal. So while a 5ft block of wood has such and such a number of maxhp that is always the same, any damage that wood block has suffered in its entire history as an object will subtract from that.

Combine with the fact that a gaussian distribution and a 10d6 damage distribution look really similar, and that you aren't actually getting the damage distribution, you're getting the time-to-x-hp-damage distribution (which means you may be doing several 10d6 blasts in a row) means you're even closer to gaussian. So I think the wizards would likely figure out the central limit theorem from this before figuring out discrete damage and health.

The clerics get Cure Minor Wounds though, which gives you discrete damage quite nicely.

[nedz]

Well, in 3E, Wizards have access to skills such as Knowledge(Arcana) and Spellcraft which abstract all of this. In 1E and 2E these were handwaved.

In no version of the game, AFAIK, do Wizards have access to skills such as Knowledge(Mathematics), Knowledge(Scientific Method) or Statscraft.

Even in 3E Sorcerers get to handwave knowledge of all of this, since you don't actually need Knowledge(Arcana) or Spellcraft to cast spells competently; which of course means knowing level dependant things like range.

[Doug Lampert]

I remember a random little thought experiment I read somewhere as the basis for this, but wanted to see how far down the rabbit hole we could go. I'm personally very comfortable with games where the rules are not the world's physics, and instead represent kind of a narrative layer on top of the fiction...

...But in higher-sim games, this is not the way of things. And, if the laws of physics are reflected in the rules of the game, I started thinking about wizards. This is intended kind of tongue-in-cheek.

Okay. So there's some wizards of varying levels in a wizard's tower somewhere. They are indubitably intelligent - smarter than we are - so they start experimenting with the laws of the universe.

(1) They know that their members can prepare X, Y, Z spells of I, II, III levels, and that these are always distinct quantities in a regular progression. They have at this point figured out the spell chart.

(2) The ranges and areas of effect of their spells increase in distinct, quantized measurements - usually 5' or 10' each* - and that this increase corresponds with the number of spells prepared.

(3) These gains only come in specific units. At this point, these Wizards have figured out Levels as a fundamental feature of the universe, and can even engineer out their specific levels through use of (for example) spell ranges.

(4) By blasting damaging spells at uniform targets of specific kind - like walls and blocks of wood - they can determine the percent of the time the item is blown up. By letting Wizards of different levels do this, they can further chart the efficacy - the block of wood is destroyed X% of the time by a Fireball spell by a wizard of Y level. At this point, they can figure out Hit Points as a fundamental feature of objects.

(4a) If they're particularly unethical, they can figure it out about goblins or kobolds, too.

(5) By the same token, they can determine that the probability of destroying targets is shaped in a curve best represented by Xd6, where X is their Level.

(6) With repeated uses of spells like Charm Person on each other, they can determine (a) saving throws, (b) Wizards' Will save progressions, (c) the caster's Intelligence bonuses, and (d) the targets' Wisdom bonuses.

(7) By doing this against (for example) town guards, they can further work out non-Wizards' Will Save progressions and Wisdom scores.

(8) The Wizards can get people to lift heavy things. They notice the maximum amounts their subjects can lift are, in turn, quantized into distinct units of measurement instead of smooth. By casting buffing spells, they can work out the progression and notice that it matches at every step. They have just worked out possibly as much as the entire 3-18 or 3-22 scale for Strength and can refer to it by a number.


Okay. So at this point, these Wizards can speak about Levels, Saving Throws, Wisdom Bonuses, Strength, and Hit Points. What else can they do? Can they - from the spell list in the PHB - figure out the entire rules of the game in which they are characters?

-O

* As an aside, "five feet" is a fundamental measurement of the universe, bound into its metaphysical laws.

Inflict and Cure Minor Wounds allows them to realize that HP are quantized and what the size is more easily than what you suggest. That's FAR better than using inanimate objects to figure out that HP are quantized.

The spell Trap the Soul SPECIFICALLY states that the value of a component needed for a particular target can be researched, the cost is 1000 GP/HD, so they can find ANY creature's HD.

You can't make items if you don't have the XP. Wizards have scribe scroll at level 1. It costs such a character 1 XP per level 1 spell, and 0.5 XP per level 0 spell, round down total costs for the scroll, but IIRC there's a minimum of 1 XP burried somewhere in the rules. They can figure out that XP is quantized. Since a character can delay advancement to craft they can probably figure out the XP table.

Note that the XP table might be too difficult due to how rare leveling is and characters being reluctant do delay it and the possible complication of Level Adjustment, but that XP exists and is quantized is relatively trivial. Same for HP given cure and inflict minor, the existance, quantization, and that there is SOME relation to level and HD will be figured out, but the size of a HD may escape them unless they're clever with Con boosters and experiments.

Int boosters will be vital in figuring out the bonus spell table and which spells are bonus and which from class levels.

[obryn]

Inflict and Cure Minor Wounds allows them to realize that HP are quantized and what the size is more easily than what you suggest. That's FAR better than using inanimate objects to figure out that HP are quantized.

That's true! It would certainly give that insight about living creatures faster - especially when you have a 5th-level Cleric and a 1st-Level Cleric both doing it.

I was going with the idea that inanimate objects (1) put up less of a fight, and (2) have standardized HPs and Hardness ratings based purely on their physical makeup.

(I'm not worried about those objects being damaged before; there's always Mending and the like.)

-O

[obryn]

Well, in 3E, Wizards have access to skills such as Knowledge(Arcana) and Spellcraft which abstract all of this. In 1E and 2E these were handwaved.

In no version of the game, AFAIK, do Wizards have access to skills such as Knowledge(Mathematics), Knowledge(Scientific Method) or Statscraft.

Even in 3E Sorcerers get to handwave knowledge of all of this, since you don't actually need Knowledge(Arcana) or Spellcraft to cast spells competently; which of course means knowing level dependant things like range.

In that case, they would be able to reach the remarkable insight that not a one among them is able to do math no matter how hard they hit the books.

But honestly - in a setting where magic operates in an exceptionally orderly fashion, much of this follows more or less immediately. With 16+ Intelligence, I think we can credit Wizards with being perfectly capable of logic, reason, and simple math.

-O

[The Grue]

I'm with you until #4. It takes a huge leap of faith to assume that all walls and blocks of wood take the same amount of damage to destroy. I don't se why scientifically minded wizards would make that assumption.

They might not know directly the hardness of wood or how many HP it has per inch, but they could certainly discover the relationship between an object's thickness and how many fireballs it takes to destroy. If they know that the power of a given wizard's fireball varies within a range between x and x(6) where x is the wizard's level, they can find approximate values for those things through repeated experimentation and observation.

[endoperez]

The wizards wouldn't want to blast random chunks of wood or stone, but homogenous chunks of wood or stone or something. Wall of Ice would be level 4, Wall of Stone and Fabricate level 5.

What is the least random way of dealing damage, and would the wizards come across it in a normal way? For example, if there's a Cantrip that deals exactly 1 point of damage, and a wizard decides to use that for testing since he can spam that more than other spells, it shouldn't take very long to figure out some damage rules.

Clerics' Inflict Minor Wounds is negated by a saving throw and can't be used on an object.


They know that spells can be resisted. They also know that you can fail your saving throw voluntarily. If they know both of those things, and they want to test Daze, a harmless 0-level spell, they'll notice it doesn't work on creatures with more than 4 HD/levels. Like a Wizard who knows a third-level spell.

Spell durations would be another easy way of measuring one's level. Ghost Sound is 1 round/level, which gives them "round" as a measurement of time.

[The Grue]

Actually, least random way of dealing damage is to Maximize a damage-dealing spell.

[TuggyNE]

Actually, least random way of dealing damage is to Maximize a damage-dealing spell.

Maximize a cantrip.

[PetterTomBos]

I don't think they would call it lvl.s, at least not at first. They would find some really funny unit to measure it in, like:

Every wizard has a fireballian number, this is the range (in feet) he can fling a fireball, without any metamagics attached to it. This number is usually related to other caracteristics of his spells, like the damage of a shocking grasp, the number of spells he can carry in his head, and so on. There seem to be a linear relationship, yet some deviations are known to occur [Gorukan/Fligian "On the existance and relations of the spell-adder"].

This is fun Way funnier than exam-reading!

[The Grue]

Maximize a cantrip.

I bet this is the first time in D&D history someone has suggested that.

[Jay R]

All wizards do this automatically, as proven by the fact that your wizard already knows the range of her spells at sixth level, even before she casts those spells at sixth level.

[headwarpage]

Awesome thread. I think that, if wizards could figure out all the casting stats, and maybe Strength as well, they would be able to infer the existence of Dexterity and Constitution. Reason being, once they've figured out the nature of the quantifiable stats, it should be fairly easy to figure out the effect of the stat-boosting spells. So then they've got four stat-boosting spells that add equally to known stats. Since they have two more stat-boosting spells, I suspect that they would posit that there were two more stats.

So, what kind of experiments would be necessary to figure out the nature and effect of these stats, which I assume they would dub "grace" and "endurance?"

[Randomguy]

Well, they could always find a bunch of volunteers of the same level (which they would have to pay a lot) and hit them with maximized cantrips until they fall over, thus determining their HP. Then heal them back up again, hit them with Bears endurance and repeat the experiment, and see how many more it takes. That would determine the effects of Constitution.

They can also work out the falling damage rules by dropping things onto other things.

[obryn]

Well, they could always find a bunch of volunteers of the same level (which they would have to pay a lot) and hit them with maximized cantrips until they fall over, thus determining their HP. Then heal them back up again, hit them with Bears endurance and repeat the experiment, and see how many more it takes. That would determine the effects of Constitution.

They can also work out the falling damage rules by dropping things onto other things.

Very nice!

[headwarpage]

Well, they could always find a bunch of volunteers of the same level (which they would have to pay a lot) and hit them with maximized cantrips until they fall over, thus determining their HP. Then heal them back up again, hit them with Bears endurance and repeat the experiment, and see how many more it takes. That would determine the effects of Constitution.

They can also work out the falling damage rules by dropping things onto other things.

I think that would work, but how would they know their "volunteers" were all of the same level? It's easy with wizards, but what's the observable difference between a level 1 commoner and a level 2 commoner? And how would they observe that HP increase with level, anyway? I don't think we've figured out a way to measure levels in non-spellcasters yet, and I doubt they'd be eager to experiment on themselves.

Also, you could skip maximizing by researching a cantrip that does a minimal amount of (exactly 1) damage every time. Or by finding a cleric to cast inflict minor wounds.

[endoperez]

Sleep puts 4 HD of creatures into sleep. They can probably experiment with that, and figure out "normal" HD values, but I don't think they'd have an easy way of telling how many HD a certain creature has. Some spells let them know that the target either resisted or is above a certain threshold, but that's it.

[tyckspoon]

It's easy with wizards, but what's the observable difference between a level 1 commoner and a level 2 commoner?

BAB is probably the most significant and amenable-to-measurement difference, since *every* level 2 Commoner will have better BAB than *every* level 1 Commoner; the 2 significant problems I can think of are figuring out that 'ability to hit stuff' is *also* a fundamental quantized number of your world (distinct from Strength and Dexterity) and that it does not let you pin down an exact level on its own- by BAB, a Commoner 3 appears the same as a Commoner 2 or even a Commoner 1/Expert 1, but each will have different HD and HP.

Edit: Although if the Wizards are using attack-roll using spells for their experiments (Maximized Acid Splash/Ray of Frost?) they might come across BAB on the way- at every level where your spells get more powerful but you do not gain access to a new tier of spells, your casting gets more accurate. If they think to test it, they will find that this also holds true for other weapons both melee and ranged, regardless of your skill (proficiencies) or natural talents with said weapons (Strength/Dex mods.) Examination of Clerics and sufficiently leveled Paladins/Rangers could reveal the existence of the 3/4 and 1/1 BAB progressions as well.

Sleep puts 4 HD of creatures into sleep. They can probably experiment with that, and figure out "normal" HD values, but I don't think they'd have an easy way of telling how many HD a certain creature has. Some spells let them know that the target either resisted or is above a certain threshold, but that's it.

If you permit spell research for this purpose, you could create spells specifically to explore this kind of thing.. make a series of Daze-like spells, for example, that only affect creatures of up to 1/2/3... HD/Levels. It's harder if you have to work with only existing printed spells, but there are probably a few decent 'tools' to be found somewhere.

[TuggyNE]

but I don't think they'd have an easy way of telling how many HD a certain creature has.

I don't know about easy, but there is a RAW way to do it, as already mentioned: the research needed to find the right size of gem for trap the soul will tell you precisely how many HD a creature has. Whether they'd think to use that at first I don't know, since it's not directly connected, but it is a discrete value that can be observed.