Rate the Spell-Building mechanics you've seen

[Quertus]

So, 3e Epic Spells get some (somewhat deserved) flak, usually around how everyone with a few Determinator brain cells builds all their spells for free. But this got me thinking, just what alternatives are out there, and how do they stack up.

So, here's what I've got, off the top of my head:

Spoiler: 3e Epic Spells

Show

Power Limit: Limited by Spellcraft skill.
Cost: Exorbitant costs in time and money.
Effects/Samples: Generally Snoozeville. It's not that it's mostly just more of the same, it's in many ways more limited than pre-epic spells.
Mitigating Factors: Power, in the form of other people casting spells. Probably others, but who cares, nobody uses them.
Gatekeeping: One odd limit on Knowledge:Religion for Resurrection, and probably restrictions on using banned schools; otherwise, none.

Spoiler: 2e True Dwoemers

Show

Power Limit: Limited by casting time.
Cost: Time, and one Common component.
Effects/Samples: Largely more of the same vs existing spells, but often clearly different in scope or power.
Mitigating Factors: Components, valued by rarity (common components add nothing; goddess tears or a pinch of the sands of time, OTOH, can reduce the cost of a spell). Maybe others?
Gatekeeping: Probably specialists are restricted from using opposed schools; otherwise, none.

Spoiler: WoD d20 Mage

Show

Power Limit: Limited by Spellcraft skill.
Cost: "Components" (mana), and Exhaustion (adds to DC to cast spells).
Effects/Samples: Well done, with spells with built-in scaling, spells with custom effects, and spells covering a huge range of power.
Mitigating Factors: Additional casting time, plus a few oddballs, like "harmful healing" and "harmless minions". Fairly uninteresting.
Gatekeeping: None (which is really odd, since the original WoD Mage had LOTS).

Spoiler: "Normal" WoD

Show

Power Limit: Choice of effect is limited by ranks in 9 (or 10) "Spheres"; power of effect is limited by ranks in "Arete" (basically, "magic power") and chance.
Cost: none (besides risk of Paradox...).
Effects/Samples: Fairly well done, with multiple examples of each Sphere Level, and several multi-sphere examples.
Mitigating Factors: Component/Focus (required by default, becomes a mitigating factor for # Spheres = Arete or something; depending on the type of Mage, could be anything from a program to an orgy).
Gatekeeping: Choice of effect is limited by ranks in 9 (or 10) "Spheres".

Then, for other ones I've seen, there's the original WoD spellcasting system(s), someone proposed an alternative to 3e Epic Spells, the Lina Inverse system, and I'm sure Gurps has a magic system, alongside how one can flavor effects in point-buy systems like Mutants and Masterminds as "spells". And how could I forget Ars Magica (other than the fact I've only read, not played it)?

At a glance, IMO, never mind the 3e Mitigation, 3e Costs are terrible; 2e Mitigating Factors are cool; "Normal" WoD Gatekeeping makes the most sense, and WoD d20 Mage probably has among the best effects and sample spells (if they're a smidgen less interesting than the much harder to adjudicate original WoD effects).

So, what systems have y'all seen, can y'all explain, and how would you rate them? What should a build-a-bear magic system look like?

EDIT: Systems mentioned in this thread so far: 3e Epic Spells, 2e True Dwoemers, "normal" D&D spell research, WoD d20 Mage, WoD, M&M, Ars Magica, Words of Power, Bad Latin, Rune Circuits, Elder Scrolls, Infosphere,
Exalted Fair Folk Grace Magic, Amber Diceless, The Feel of Magic, Spire: the City must Fall, Dresden Rituals, Shadowrun, Dragonlance SAGA system, 3e Unearthed Arcana incantations, TORG, Rifts Technomagic devices.

I'll try and write up "stat blocks" for these, and get approved ones into the OP for reference.

[clash]

I looked at a lot of build a Bear magic systems before I ended up iterating on my own that I call words of power.

The basics is your choose a word of power element and effect and combine it with a shape to create a spell so you might choose fire element with the effect of causing the burned condition and the cube shape and the table shows how much damage and mana each part costs. If you want more info on it I can DM you but that's the core behind it.

[Grod_The_Giant]

What should a build-a-bear magic system look like?

Mutants and Masterminds. The answer is Mutants and Masterminds (3e, specifically).

[NichG]

Well, I had a DM run a game where you could wave a wand and make up something in fake latin (not real latin - that would automatically fail), and he'd make the spell do what he thought that sounded like it should do - it'd continue to consistently do that in the future too. That's probably the most freeform one I've seen.

On the other extreme of something highly structured - same campaign, same DM, but there was a system of symbols that each had a narrow sort of modulatory effect and could be chained together in various ways. So you could make runeworks like electric circuits, that you'd have to design in detail. Things like 'this rune transmits the effect of the other attached runes to a complementary receiver rune', 'this rune acts like a button to allow runic signals past it or interrupt them', 'this rune converts magical energy to a force applied perpendicular to its surface, with a maximum capacity based on the material its inscribed on', etc. Between those and magical metals, the party invented reactionless drives, designed and built our own aircraft in game, which ultimately became a flying time machine.

Of course I played a character who ended up practicing both...

Elder Scrolls RPG has a mess of a spell-building system. There are a few funny exploits but on the whole I probably cannot recommend it. Its mostly 'discrete effect + shape + intensity', but the shape stuff is kind of either useless or strictly better than all other possibilities for the price. There are other issues with that system, between actually casting spells being unreasonably hard to succeed at for anything but an end-game character, to being able to totally bypass all of that by instead using the Enchanting subsystem that effectively lets you cast with 0% failure rate (because you roll when you try to enchant, not when you use the item) and then ignore your own magicka pool and just cast off of soul gems, which have five to fifty times a character's magicka pool and you can carry as many as you'd ever need... And since there are shape parameters you can scale with spell cost without increasing the difficulty of crafting the item, you can make city-killer bombs that eat a single grand soulgem and put out a fire effect with a mile radius and a duration of a day, or stuff like that...

How I would do it... I think I'd want to integrate it with the metaphysics. Rather than just abstract blocks you can put together or wildcard 'what do you want the spell to do, and I'll assign a level' stuff, I'd start with 'why does magic work?' - what are the new physical principles or conceptual principles that create these effects? Then within that, there would be some logic for what can be done, what can be tried, etc. So it wouldn't really be build-a-bear picking off of a combinatorial list per se.

For example, in a recent campaign the party found a sort of abstract conceptual space called the Infosphere that represented all of the possible statistical inferences between things in the world that could exist. E.g. if you wrote something down such that someone could infer that it was you who wrote it, the infosphere would contain a connection between the thing you wrote and you. The infosphere was non-spatial, in the sense that 'moving' was more like choosing a level of abstraction or framing - you could 'move' in the direction of interpreting a person as a person, or 'move' in the direction of interpreting them as a collection of atoms. When looking in the future direction, the infosphere would show all possible futures given what you knew about the world, whereas the past was more concrete. And at high levels of power, someone could directly change the links - delete the connection between a letter and its writer to make it impossible to tell who wrote it, create a connection to make the characteristics of one thing have causal influence on the details of another thing despite there not having been a physical meeting, etc.

So within that metaphysics framework, players could just... try things. Tracking people was pretty standard stuff to do with it. One player figured out how to use the panoply of futures feel out 'if I think about doing this, does the future change?'. A different player created a thing that would basically act as a hub connecting people's misdeeds, which would also (via creating links) causally influence events around those people to give them karmic comeuppance. It would be hard to, as a player, know that e.g. 'to do this effect, I need Insight 4, but for that effect I would need Insight 7'. But a player who understood real-life Information Theory could readily come up with a lot of weird stuff to try that would likely work as they imagined. And a player who didn't could feel their way through it.

That kind of thing would be my preference, but if you want something more structured, maybe the thing I did for Mythclad would work as a design. The idea there was that 'big spells' were extended weaves that you could hang a large number of metamagic-like smaller spells onto from round to round. So rather than the main spells being designed to be self-contained complex effects, an actual effective combat spell would be this thing you'd build on round after round, and it would get stronger and more complex in its consequences as you did so - you'd start with a vector of delivery like a 'turret' or 'spellblade' or whatever perhaps, and then add a metamagic to give it an energy type, a different metamagic to have it inflict a status condition, a different metamagic to have it automatically strike out at enemies rather than you needing to spend an action to have it do so, yet another metamagic to make it automatically get additional free strikes against enemy casters when they try to start their own spell, etc. Weaves could be taken over, multiple casters could add to a single weave, an enemy caster could even try to force their own metamagics into a weave to disrupt the effect.

This was an action-point system, and the big spells would usually cost 75% of a full round of points to cast, whereas the metamagics might cost 20-50% depending on the metamagic.

That said, I didn't see it get all that much use in that campaign. The players went for the Strategist, Inventor, and Warrior paths instead (there were 9 subsystems, 3 of them had this weave+metamagic structure).

[meschlum]

Ursa's Guiding Light

The Fair Folk are not known for their restraint, be it in love or wrath. So when a city god ordained the elimination of ursines within his territory, one of the Raksha took it amiss. The Light was woven from the hair of the city's nobles, leaving their heads exposed. It was set on the back of a city priest, who had many sins and burdens to carry. It traveled through the city, north to south, east to west, never losing its waying.

And so the Guiding Light was raised, and all it touched became that which was forbidden - though they could speak, and think, and act, it was a simple truth that they were now impossible to put up with.

Mechanics
5 dot Outward Facing Oneiromancy
Assumption of the City's Heart - once the Light has been invoked upon a city, it takes the shape of a beacon on a desolate space, casting its transformative light throughout the city.
Mad God Mien - the power of the Light is too great for mere spells to break it, even the vaunted counter magics of the Anathema
Bestial Transformation - those who are touched by the Light and do not invoke the hidden word become strange and alien beasts whose names cannot be spoken.
Fall of Night Shadows the Truth - within the reach of the Light the name of the things that are not permitted is silenced as it is spoken, erased as it is written, forgotten as it is thought.

And there you have it!

Power Limit: this is the Exalted setting. So utterly terrifying to mundanes, and requiring creativity from PCs. Or just big enough numbers.
Cost: crippling if you are trying to sustain the curse on your own (15 committed motes out of a pool of 10 * Essence is not small), quite manageable if you have minions to do it for you.
Effects / Samples: a lot more than the designers might have expected, in part due to poor editing.
Mitigating Factors: this is fairytale magic. There will be ways to get past it (or there should, anyway). Tends to summon rules lawyers as a side effect.
Gatekeeping: No? Yes? Avocado? A starting character can have the Light, but it's not exactly going to help them survive. Once you have been created, building new effects comparable to the Light is a major endeavor, though it can be cheated into being much easier with a specialized build (or a minion with the same...)


Ars Magica has sort of defined effects for each level (they get vaguer and less well edited as the power tier goes up), and a rather high chance of botching rolls (a natural 1 on a d10 is twice the odds of a critical threat in D&D). Spontaneous magic is hard (unless you cheat, and even if you do it's not easy), so while you need to be a powerful spellcaster to use potent effects, you might not have many - and may need years to develop new ones.


Amber Diceless has magic, which is not all that impressive in setting, as well as conjuration, which is essentially free power if you invest time into it (and, technically, stay in one place, but you can make the effects permanent by investing more time. And you can modify the rate at which time passes...). Effects tend to cap out at leveling a few city blocks or so, and more esoteric effects are open to debate. Or you just skip the magic and start tearing holes in reality to consume the local universe, but there isn't much in between space.

[Vahnavoi]

I have seen some inventive spell lists, but no good spell generation mechanics for a while now. So if I need any, I end up making my own. This said, there are few core ideas that appeal to me the most:

Scrabble using weird letters:

Basically a player has some limited around of tokens, quite likely exactly like the tokens their character would have, that then have to be arranged to make a legible word or sentence in target language. The language could just be a real one players are familiar with, but for extra fun, it can be invented or foreign. The effect is then either interpreted from the word/sentence in a freeform manner, or checked from a list of valid combinations.

Plan an operation:

May be in play independently or in addition to the above. The basic rule is that each spell on some entity: a person, a spirit, a god, etc.. A new spell consists of a sequence of operations applied to the entity. Example operations would be: Invoke (call in), Evoke (call out), Bind (affix entity to an object or place), Banish (cast out/remove) and Destroy (kill/delete). What is achievable by an operation, depends on the entity. For example, to have a telepathic conversation with someone, you might Invoke them (call them into your mind) to exchange ideas, the Banish them (cast them out of your mind) to end the conversation.

Obey the Rules:

A spell has to obey some rules that are traditional to real magic. The two most common are Law of Sympathy (like affects like, symbol affects reality) and Law of Contagion (things that have been in contact remain connected, there is a "taint" that can be moved from object to object, etc.). I typically add Law of Intent or Will (the magician has to understand the symbols and reaffirm they wish the effect of the spell, for example, by stating it three times) and Law of Impermanence (spells and their effect end when no conscious being is left to Intent/Will them into being).

---

Stringing the above together, as far as I'm concerned, gives sufficient framework for creating spells that feel intuitively, well, magical. They can then interact with other tabletop mechanics such as spell points, but I find those to border on trivial. If material costs of spell components are well-established, you may not need any other point mechanic to limit them beyond counting money.

Anyways, to give an example of how to build a spell in accordance of the above:

You want to make a spell to torture a person. Materials required: a doll in the likeness of that person (to obey Law of Sympathy) and a strand of hair or piece of a personal belonging (to obey Law of Contagion). You wish to Evoke (call their mind out) and Bind (affix their mind to) the person to the doll. From weird runic stones your game master has supplied you, you now have to arrange the sentences [Evoke][Person] and [Bind][Person][Object]. Whatever mumbo jumbo you now have before you, you have to repeat it three times for the spell to take effect.

Congratulations, now you have a magic effigy to abuse in order to harm your person!

[Eldan]

I come to hte conclusion that I like it simple. Very simple. All those complex systems don't really do it for me anymore.

My second favorite is how Spire: the City must Fall recommends you do it: just roll a skill check to achieve what you want. In Spire, a skill check consists of two parts, a skill describing what you want to achieve (like Fight to fight things, or Fix to repair or heal things, or Investigate) and a domain describing how (on what, with what) you want to achieve it (like Technology to use or affect a machine, or Crime if you're dealing with bandits, etc.) So how do you cast a spell? You roll the appropriate skill + the occult or divine domain. That's it. There's a specific stress table for occult consequences, to represent the horrific outcomes dark magic can have, and that's it.

Probably my favorite is Ritual casting in Dresden Files accelerated. You describe what you want your ritual to achieve and together with the DM, sent a difficulty, based on how long the effect would last, how strong it is, how many people it affects, etc. Then you roll a skill check. The scale of the ritual also determines how many conditions your ritual has. If you succeed on the check, you set the conditions. If you fail, the DM sets them. Conditiosn are things like "can only be cast at a certain time (like sunrise)" or "needs the help of a supernatural creatre" or "has an expensive magical component". So you want to conjure a citywide storm to cover up your activites? Well, you failed the skill check, so your GM determines you need the help of a wind spirit, and to cast the spell on a mountaintop, so you need to find some ritual offerings first and then go on a hike of several hours.

It's really very thematic if done right.

[Lord Torath]

Shadowrun IIe had a pretty decent system for creating new spells in the Grimoire. Mostly it revolved around determining the casting time/duration and drain level.

[Quertus]

Mutants and Masterminds. The answer is Mutants and Masterminds (3e, specifically).

M&M 3e isn’t the bear, it’s the the bar. That is, M&M, and its DC spiritual ancestor, are a good answer for almost anything, so they set the bar for “if you’re gonna make something new, it had better be at least this good”.

Spoiler: Mutants and Masterminds

Show

Power Limit: Limited by level, or “group limited” by level (accuracy + damage limited as a group), or not at all.
Cost: None (braids build points).
Effects/Samples: A few sample characters. Shrug. Abilities are all over the place.
Mitigating Factors: various limitations and drawbacks can make the power cost less build points via division rather than subtraction, but this does not affect power level limits.
Gatekeeping: None?

When your Jedi is using the same system for their lightsaber and their Force powers as your Wizard is for their magic and their tank (Because what Wizard don’t want to ride around in a tank?), it doesn’t feel very magical. But the math definitely sets the bar for what “good” looks like.

Ursa's Guiding Light

The Fair Folk are not known for their restraint, be it in love or wrath. So when a city god ordained the elimination of ursines within his territory, one of the Raksha took it amiss. The Light was woven from the hair of the city's nobles, leaving their heads exposed. It was set on the back of a city priest, who had many sins and burdens to carry. It traveled through the city, north to south, east to west, never losing its waying.

And so the Guiding Light was raised, and all it touched became that which was forbidden - though they could speak, and think, and act, it was a simple truth that they were now impossible to put up with.

Mechanics
5 dot Outward Facing Oneiromancy
Assumption of the City's Heart - once the Light has been invoked upon a city, it takes the shape of a beacon on a desolate space, casting its transformative light throughout the city.
Mad God Mien - the power of the Light is too great for mere spells to break it, even the vaunted counter magics of the Anathema
Bestial Transformation - those who are touched by the Light and do not invoke the hidden word become strange and alien beasts whose names cannot be spoken.
Fall of Night Shadows the Truth - within the reach of the Light the name of the things that are not permitted is silenced as it is spoken, erased as it is written, forgotten as it is thought.

And there you have it!

Power Limit: this is the Exalted setting. So utterly terrifying to mundanes, and requiring creativity from PCs. Or just big enough numbers.
Cost: crippling if you are trying to sustain the curse on your own (15 committed motes out of a pool of 10 * Essence is not small), quite manageable if you have minions to do it for you.
Effects / Samples: a lot more than the designers might have expected, in part due to poor editing.
Mitigating Factors: this is fairytale magic. There will be ways to get past it (or there should, anyway). Tends to summon rules lawyers as a side effect.
Gatekeeping: No? Yes? Avocado? A starting character can have the Light, but it's not exactly going to help them survive. Once you have been created, building new effects comparable to the Light is a major endeavor, though it can be cheated into being much easier with a specialized build (or a minion with the same...)

I probably should have explained my system (and asked for feedback!). Let me try to redo that…

Power Limit: big enough numbers.
Cost: crippling amounts of Essence (mana), “major endeavor”.
Effects / Samples: “a lot more than the designers might have expected, in part due to poor editing.” - really? What I remember of Exalted felt very… “static magic”, to use WoD terms for it.
Mitigating Factors: Specialized build, Minions (can pay mana upkeep costs and/or possess the specialized build)
Gatekeeping: Number of ranks in abilities like “Oneiromancy“.

I may not know exactly what all those words mean, but I think they’re in the right places now. (Note to self: go back and bold those keywords in my examples.)

Ars Magica

Amber Diceless

I don’t really know those systems, but I’ll make a 1st pass at these, that you can laugh at and correct if you like… in my next post (to make sure I don’t lose this).

[Quertus]

Ars Magica

Power Limit: Level/Tier?
Cost: Time (and threat of botching).
Effects / Samples: ???
Mitigating Factors: Yes?
Gatekeeping: None?

Amber Diceless

Power Limit: Time.
Cost: Time, stay in one place.
Effects / Samples: ???
Mitigating Factors: Manipulating time.
Gatekeeping: None?

I looked at a lot of build a Bear magic systems before I ended up iterating on my own that I call words of power.

The basics is your choose a word of power element and effect and combine it with a shape to create a spell so you might choose fire element with the effect of causing the burned condition and the cube shape and the table shows how much damage and mana each part costs. If you want more info on it I can DM you but that's the core behind it.

Can you explain it in terms of my categories? I’ll make silly guesses if it’ll help.

[LibraryOgre]

So, I can think of three off the top of my head: Ars Magica, Dragonlance SAGA, and Mage: The Ascension.

Ars Magica uses a verb-noun system... your ability to cast a spell is based on your skills in the respective verbs and nouns that govern what you want to do. Someone with a high Muto score can cast a lot of different spells that change shapes; someone with a high Terram score can do all sorts of things with dirt. If you have high Muto AND Terram, you can make a lot of different spells based on those two themes, and your most powerful spells will change rock and dirt. Each particular combination has benchmarks for range, duration, and area of effect; if I do a Creo Ignam effect to blast someone with fire, the base is the damage, and I have to increase the power to increase range or damage or how many people I hit. Spells can be designed to be formulaic (a spell you have designed in a lab and practiced until it works well) or spontaneous ("I want to do this thing right now and I don't have a spell that will do it"), with spontaneous spells being pretty much cantrips except for really powerful magi.

Overall, it is a very good, very crunchy, system. It allows for a broad variety of magi with a lot of different types, depending on their specialties and other quirks (such as Virtues that add bonuses in specific circumstances, against certain targets, etc.).

*

Dragonlance SAGA system is a "noun" system, if you will... your various schools and spheres determine what sort of effects you can do. With Cryomancy, you can do cold magic. With Necromancy, you can mess with dead bodies. You put together the spell based on how long you're going to cast, how long it will last, how far away it will work, how many, what, etc. It's pretty flexible, and if you and the GM are on the same page about things, you can get a good idea of how hard things will be to pull off. However, it's also somewhat hobbled by the core mechanics of SAGA, which are underwhelming, to say the least.

I've used it as a base for a couple different systems (Psionics and Illusions), however; the bones are good, they were just manipulated with some weak muscles and very stupid ligaments.

*

Mage the Ascension was a "noun" system, with 9 nouns encompassing all of reality. Correspondence, Entropy, Forces, Life, Matter, Prime, Spirit, and Time... and Mind, which I had to look up for some reason. Each level of a given "sphere" dictated what you could do with that sphere... level 1 usually meant "you could sense it", and level 5 meant "you can do whatever you want with it", for the most part. Mage had a secondary limit of Arete... an attribute which determined how good you were at making magic that worked with the world.

It wasn't a bad system, but was very freeform... whereas the other two specifically spelled out how your spell was put together (SAGA is "to get X effect at Y range and for Z time, you need to have a total of X+Y+Z"; Ars Magica is similar), Mage more or less said "You can do X; figure out what that means", and depending on your SG, that could be very easy or very hard. I played with one who tried to make Correspondence (space) and Time part of everything. I played with another determined if spells worked with a Tarot card draw. IME, oWoD was about style far more than substance.

[Quertus]

Well, I had…

Lingua Occultis

Power Limit: Voidus
Cost: Signus Twigus (oritus Riddikulus).
Effects / Samples: ???
Mitigating Factors: Voidus
Gatekeeping: Voidus

Rune Circuits

Power Limit: None / by material (maybe space?)
Cost: negligible materials cost?.
Effects / Samples: this might be a rare instance where the system / GM didn’t provide some sample spells…
Mitigating Factors: none
Gatekeeping: none

Elder Scrolls

Power Limit: by skills.
Cost: ???.
Effects / Samples: ???
Mitigating Factors: Enchanting, Soul gems.
Gatekeeping: Various Magic skills

EDIT: I thought I missed one!

Infosphere

Power Limit: ???.
Cost: ???.
Effects / Samples: ???
Mitigating Factors: None.
Gatekeeping: Insight; Existing knowledge for Future Divinations.

[NichG]

Lingua Occultis

Power Limit: Voidus
Cost: Signus Twigus (oritus Riddikulus).
Effects / Samples: ???
Mitigating Factors: Voidus
Gatekeeping: Voidus

Tier of wand was a cap against some spells working if the interpretation would be too powerful, which in turn was controlled by a skill+attribute thing. Cost was basically just the materials for the wand and then you cast forever, and action economy. Effects were all over the place, anything from standard fireball type stuff to really weird things at the very high end like cracking the shell of the universe and letting things from outside in. Practically speaking, it was most useful for off-the-cuff 'I need a utility effect' improv because the high variance in aiming for a big thing meant you basically had to find the big spells in the setting rather than just toss them off.

Rune Circuits

Power Limit: None / by material (maybe space?)
Cost: negligible materials cost?.
Effects / Samples: this might be a rare instance where the system / GM didn’t provide some sample spells…
Mitigating Factors: none
Gatekeeping: none

Power limit was material and energy source - the 'aetheric generator' rune you could use like a solar panel for small stuff, but if you want to make anything significant you're going to be looking at some massive installation on a leyline or finding the magical equivalent of antimatter fuel or things like that.

Cost: Time - its slow to scribe these things accurately; and not-so-negligible material costs if you want it to actually sustain high-powered effects. For a runic flashlight, sure, just sketch something out on a piece of paper. For a spacecraft, you're looking at least to titanium as the supporting material, probably actually needing magical alloy hijinks to support the currents unless you make everything really big...

What basic runic elements you could 'discover' were skill gated, but then you could teach others who could use them with less skill (but they would still need some). Also practically speaking, there were some complex runes you needed to use machines to draw accurately - getting an angle wrong by a tenth of a degree would make it not work, etc.

As far as sample effects, I'd consider each single-rune effect to be one. Write the explosion rune on a piece of paper and you have a simple noise-maker, carve it into a piece of steel and you have a basic grenade, etc. Write the aether->light conversion rune on something and power it and you have a flashlight.

Elder Scrolls

Power Limit: by skills.
Cost: ???.
Effects / Samples: ???
Mitigating Factors: Enchanting, Soul gems.
Gatekeeping: Various Magic skills

If you play it like you're supposed to, power limit is by skill and magicka supply, cost is magicka, effects are really straight-forward - fire, shock, etc - and generally speaking aren't broadly utility. However the 'Detect' effect is utterly open-ended and broken; you can freely add 'noun' after Detect to make a spell. Detect Thief? Sure! Detect Truth? Sure, just write down your questions and the universe will answer! Detect 'People who in the future we will have wished we killed now'? Go for it!

But most things were like 'restore health', 'deal fire damage', 'drain attribute score', etc.

If you go the enchanting route, its basically filled soulgems and to some extent enchanting skill, though you can sub money for skill in that case since its a percentile dice system where skills go 0 to 100, so your skill is never really more than the variance of your roll anyhow. There are modifiers so you can have a zero chance of success, but just like its hard to get a success chance worth actually risking in combat, its hard for anything you'd ever be intended to be able to do to be so hard that you really literally have a zero percent chance of doing it.

[Quertus]

There's so many things my 5-category system doesn't cover, but here's what the categories were meant to encapsulate (which is often multiple things in each category):

Power Limit: Is there something that regulates the upper bound of how powerful your spells can be, and/or how powerful a particular casting of a spell might be? Whether it's Spellcraft for 3e D&D, Arete for WoD Mage, or Level for M&M, it goes here.

Cost: Is there any cost to research or cast or maintain a spell? Things like Time and Gold and XP for 3e D&D, risk of Paradox for WoD, and mana (components) for d20 Wod go here.

Effects / Samples: This silly (and likely soon to be obsoleted) category was meant to answer the question, "how well can one understand the scope of the magic system (and how it will look to use in practice) by looking at the example spells?". The dissonance between 3e sample spells and the free spells at an actual table or between Elder Scrolls spells and enchanting, vs how well (I think) d20 Wod handled their examples would go here.

Mitigating Factors: This is the things that make the spell easier, lest costly, and/or more powerful than "normal", with an explanation of what they affect. So getting X people to cast Y levels of spells (or taking damage from the spell) reducing the Spellcraft DC and therefore the Cost in Time GP and XP on top of chance of success and maximum power in 3e D&D, cool thematic rare components reducing the research time and casting time in 2e D&D, and unnecessary foci (as well as spending Quintessence and/or Willpower) making casting easier / more powerful in WoD are Mitigating Factors.

Gatekeeping: I think this is an interesting category: is there anything that says arbitrary mage can't just create and cast any spell possible under this system? If they can't, if there's some prerequisites beyond raw power holding them back, like Spheres in WoD, Oneiromancy and company in Exalted, or wand tier in bad latin (or is that just a power meter?), this goes under Gatekeeping.

I know I'm missing some stuff with these categories (and not just good descriptions and flavor), but I can't quite seem to put my finger on specifically what is missing from this list that should be there.

Thoughts?

(I'll get back to replying to posts and maybe adding systems to the OP some day. Maybe even tomorrow.)

[NichG]

One thing about some of the systems I posted about that isn't true of others is the degree to which they represent something that you could spend a year thinking about, and then realize 'oh I could have done X!'. Or like, if the internet as a whole spent 10 years theorycrafting, they'd still be discovering new tricks.

If I want something that actually feels like what being a researcher is like (but for fantasy stuff like 'how does magic work?'), that's kind of an important element. Yeah once you realize how Fourier transforms work they're not that hard to just do, but it took hundreds of years to think 'oh this is actually quite useful!'. Those systems tend to have less character-resource-based gatekeeping, because the true gatekeeping is that figuring out how to do a thing is actually a research-grade inherently difficult (but still usually solvable) problem. This is hard to design, but its a joy when someone lands on a subsystem that demonstrates some of those properties.

[meschlum]

Exalted Fair Folk Grace Magic

Power Limit: Big numbers. Also creative ways of getting around them. A technically limited set of effects is available, but the in setting and editing exploits are... impressive.
Cost: Big effects cost a lot of essence (mana), and are a major endeavor to create.
Effects / Samples: “a lot more than the designers might have expected, in part due to poor editing.” - really? What I remember of Exalted felt very… “static magic”, to use WoD terms for it.

Exalted abilities are very static. Fair Folk Grace Magic is highly customizable, see the sample spell I created!

Mitigating Factors: Minions to cover essence costs. Specialized builds (self or minion) to create new effects. Character creation allows access to insane amounts of power (and minions).
Gatekeeping: Be a Raksha or friends with one for the casting. Have the right ('static') Charms (powers) to create new effects (largely available at character creation, if you're willing to specialize or have a minion do so for you).

Amber Diceless

Power Limit: Time. Weak against greater powers (of which there are many: Pattern, Chaos, and sheer greater ability)
Cost: Time, stay in one place, maintain your magic once prepared.
Effects / Samples: summon an army of nigh-invulnerable animated armors with swords that will cut through most things. Unleash a dozen lightning bolts at once (then need to recharge them). Travel to a random world (where magic might not work, air may be poison, and the floor is lava).
Mitigating Factors: Manipulating time. Access to greater powers (Chaos makes minor magic easier, Order can help with greater forms)
Gatekeeping: Be in a universe that allows magic, or travel to one. Spend time learning magic.

[clash]

Can you explain it in terms of my categories? I’ll make silly guesses if it’ll help.

Power Limit: mana limit per turn
Cost: mana, components
Effects / Samples: fireballs, mind control, necromancy etc
Mitigating Factors: ??
Gatekeeping: Class abilities for learning words of power and obtaining mana.

[Bohandas]

I feel 3e Unearthed Arcana incantations should be mentioned here. I don't have the time at the present moment to write up a full synopsis of them but I want to make sure they're at least mentioned so I'm mentioning them

[SimonMoon6]

I remember the 1st edition AD&D system of "try designing a spell and assigning level and everything else to it. The DM will look at it and decide if that seems reasonable or not, but he won't tell you. Then, you spend a bunch of time and money trying to make the spell. Then, you roll dice to see if you succeeded, but if the DM didn't approve of the spell, you automatically fail."

The other system that I will always remember trying to use was the system in TORG (first edition). The first time I read through it, I was completely confused and had no idea what it was all about, but the second time reading it, I figured it out and couldn't figure out why I didn't understand it the first time.

There are tons of numbers to fill in, in a complicated looking form. You have to consider your skills that apply to someone casting this spell as well as some other things representing magical knowledge but aren't technically skills. You consider the difficulty of making your spell, involving what kinds of things you're trying to do and what sort of range, duration, etc you want the skill to have. And then you look at another table to see what sort of modifiers you get for how powerful your effect should be. And to figure out the difficulties, you're considering how all five possible magic skills might be applicable to the spell (usually, several of them will apply), each adding extra difficulties to overcome.

Then, you roll on your five skills to see how well you've done and add the results and that's another modifier to keep track of. And then, you have more rolls to make involving magic "theorems". And basically, these rolls take a week of the character's time; you can re-roll some rolls but then, you have to spend all that time all over again.

And at the end, a number gets spit out of the formula on the worksheet. That number is split into two parts: a difficulty to cast the spell and a damage value for "backlash", meaning how much damage you take when you inevitably fail to cast the spell. In the example spell, they get a result of 30, which they break into a difficulty of 9 (pretty easy) and a backlash of 21 (pretty severe).

Power Limit: limited by skill, luck (ameliorated by spending possibilities (if you really want to) and time), and the other knowledges that aren't skills). You could probably give any spell a really low difficulty but then you'd have a really huge backlash, so you might just explode if you roll a 1 when casting the spell.

Cost: Takes weeks to make a spell.

Effects / Samples: Anything imaginable might be possible, but the difficulties (or backlash) could become ridiculous. The game doesn't come with a lot of spells, so this is almost necessary for anyone playing a wizard for a long time, I imagine. The sample spell created is to turn a man into a mouse.

Mitigating Factors: In the setting, magic isn't supposed to work everywhere, so using it would be a "contradiction" which is... usually not a big deal for a PC but it can be annoying.


Gatekeeping: You have to know the main four magical skills to create a spell in the first place. Some wizards might get by with only one or two of them in their adventuring careers if they don't want to create spells, so they have to invest a bit more in skills if they do, but it's not that big a deal.

[Quertus]

Time to roll up my sleeves and try to respond to some of these things.

I have seen some inventive spell lists, but no good spell generation mechanics for a while now. So if I need any, I end up making my own. This said, there are few core ideas that appeal to me the most:

Scrabble using weird letters:

Basically a player has some limited around of tokens, quite likely exactly like the tokens their character would have, that then have to be arranged to make a legible word or sentence in target language. The language could just be a real one players are familiar with, but for extra fun, it can be invented or foreign. The effect is then either interpreted from the word/sentence in a freeform manner, or checked from a list of valid combinations.

Plan an operation:

May be in play independently or in addition to the above. The basic rule is that each spell on some entity: a person, a spirit, a god, etc.. A new spell consists of a sequence of operations applied to the entity. Example operations would be: Invoke (call in), Evoke (call out), Bind (affix entity to an object or place), Banish (cast out/remove) and Destroy (kill/delete). What is achievable by an operation, depends on the entity. For example, to have a telepathic conversation with someone, you might Invoke them (call them into your mind) to exchange ideas, the Banish them (cast them out of your mind) to end the conversation.

Obey the Rules:

A spell has to obey some rules that are traditional to real magic. The two most common are Law of Sympathy (like affects like, symbol affects reality) and Law of Contagion (things that have been in contact remain connected, there is a "taint" that can be moved from object to object, etc.). I typically add Law of Intent or Will (the magician has to understand the symbols and reaffirm they wish the effect of the spell, for example, by stating it three times) and Law of Impermanence (spells and their effect end when no conscious being is left to Intent/Will them into being).

---

Stringing the above together, as far as I'm concerned, gives sufficient framework for creating spells that feel intuitively, well, magical. They can then interact with other tabletop mechanics such as spell points, but I find those to border on trivial. If material costs of spell components are well-established, you may not need any other point mechanic to limit them beyond counting money.

Anyways, to give an example of how to build a spell in accordance of the above:

You want to make a spell to torture a person. Materials required: a doll in the likeness of that person (to obey Law of Sympathy) and a strand of hair or piece of a personal belonging (to obey Law of Contagion). You wish to Evoke (call their mind out) and Bind (affix their mind to) the person to the doll. From weird runic stones your game master has supplied you, you now have to arrange the sentences [Evoke][Person] and [Bind][Person][Object]. Whatever mumbo jumbo you now have before you, you have to repeat it three times for the spell to take effect.

Congratulations, now you have a magic effigy to abuse in order to harm your person!

Until you give me a better name, I'm going to dub this "The Feel of Magic". It's an interesting sort of meta-system that could be bolted on top of other systems. By itself, it tells me the flavor of the spell, but nothing of the mechanics of "can this particular mage cast this particular spell?". It's a very different focus, and I think that the flavor is comparable to 2e D&D True Dwoemers or 2e D&D Item Creation (dagnabbit, that's another one I need to add at some point) or maybe Dresden Rituals done right. So, top tier feels, IMO.

One thing about some of the systems I posted about that isn't true of others is the degree to which they represent something that you could spend a year thinking about, and then realize 'oh I could have done X!'. Or like, if the internet as a whole spent 10 years theorycrafting, they'd still be discovering new tricks.

If I want something that actually feels like what being a researcher is like (but for fantasy stuff like 'how does magic work?'), that's kind of an important element. Yeah once you realize how Fourier transforms work they're not that hard to just do, but it took hundreds of years to think 'oh this is actually quite useful!'. Those systems tend to have less character-resource-based gatekeeping, because the true gatekeeping is that figuring out how to do a thing is actually a research-grade inherently difficult (but still usually solvable) problem. This is hard to design, but its a joy when someone lands on a subsystem that demonstrates some of those properties.

I'm gonna turn this on its head, and describe systems thusly: how "complete" does the system feel, counting just effects that are formally detailed; how much does it feel like one can really sink their teeth into the system and thereby can McGyver up a spell/item/solution without outside assistance or GM judgement call? Then, how much is there to the system beyond this (potentially nonexistent) framework, and what do the access methods to "outside the box" look and feel like ("Mother May I")?

So, for example, 2e True Dwoemers and 3e Epic Spells let you do the math, and the effects feel close to complete if lacking in comparison to the breadth of 1st-9th level spells, but inventing new effects to fill in any missing "outside the box" concepts feels almost like cheating. M&M feels perfectly complete, and "outside the box" is all but a foreign concept. Normal D&D spell creation, OTOH, has a few set rules in 2e, where "Faerie's Grace: Grants Flight + Invisibility" is a 5th level spell, and has a few reasonable guidelines, so one might expect "Coldball: Deals d6 damage per level (max 10d6) in a 20' radius" is a solid 3rd level spell, but everything else is full-on "Mother May I" territory. WoD d20 felt like you could play it entirely within the provided framework, yet at the same time strongly encouraged the existence of effects outside that framework, as multiple of their sample spells included custom components. WoD felt like it ought to be complete several times over with its "more than one way to skin a cat" attitude, but in play felt like "Mother May I" with a side of "No.", at least under my GMs. Rifts had a system (technomagic, maybe?) for creating items that was very formulaic; the only "GM intervention" necessary was the caveat that some items may have unintended side effects.

Then, looking at the systems you presented, it seems none of them came with an instruction manual. So they all start out at 0% user friendly, and it's up to interactions with the GM to make them useful. Or at least that's true of Bad Latin (which grew by psychology, knowledge:GM, and context) and Infosphere (which grew by information theory and trial-and-error (although I came up with about 3x the effects you listed just from the name, so I found this conceptually approachable to guess what could be done, even if I couldn't guess whether a given character could do it)); Rune Circuits might have had all the approachability and consistency of programming. Elder Scrolls was pretty static, right? It just did what it said on the tin, no fancy bells and whistles requiring interpretation?

So, 2 potential new categories: Completeness, and Incompleteness. Completeness measures how much this feels like a full set of player-facing rules vs how much it feels like pieces of this jigsaw puzzle are missing; alternately, it measures how similar 2 games with different GMs might look using this system. Incompleteness measures the type of flavor of outside the box thinking the system is designed to support.

Completeness

0%: Bad Latin, Infosphere
20%: "normal" D&D spell research
80%: 3e Epic Spells, 2e True Dwoemers
100%: M&M, WoD Mage, WoD d20, Elder Scrolls... Rune Circuits?

Incompleteness

Huh?: M&M, Elder Scrolls
Feels like cheating: D&D 2e True Dwoemers, D&D 3e Epic Spells
Mother May I: "normal" D&D spell research, WoD Mage
Phlebotinum: Rune Circuits
Encouraged: WoD d20
Research: Infosphere

Hmmm... It almost feels like there should be a "depth" category, or a "breadth" category, even if they have some overlap with these categories. Kinda like "base stats" and "derived stats", I may be hitting this at multiple levels / it may be important to be able to think about these systems as multiple layers.

And, at some point, I need to look at all the other systems mentioned: 3e Epic Spells, 2e True Dwoemers, "normal" D&D spell research, WoD d20 Mage, WoD, M&M, Ars Magica, Words of Power, Bad Latin, Rune Circuits, Elder Scrolls, Infosphere, Exalted Fair Folk Grace Magic, Amber Diceless, The Feel of Magic, Spire: the City must Fall, Dresden Rituals, Shadowrun, Dragonlance SAGA system, 3e Unearthed Arcana incantations, TORG, Rifts Technomagic devices. Just a few magical bear-building factories out there.

[NichG]

Then, looking at the systems you presented, it seems none of them came with an instruction manual. So they all start out at 0% user friendly, and it's up to interactions with the GM to make them useful. Or at least that's true of Bad Latin (which grew by psychology, knowledge:GM, and context) and Infosphere (which grew by information theory and trial-and-error (although I came up with about 3x the effects you listed just from the name, so I found this conceptually approachable to guess what could be done, even if I couldn't guess whether a given character could do it)); Rune Circuits might have had all the approachability and consistency of programming. Elder Scrolls was pretty static, right? It just did what it said on the tin, no fancy bells and whistles requiring interpretation?

Well, aside from Elder Scrolls, I'd say they all require GM interaction, even the rune circuits. But I suppose where they might differ is the extent to which there's some sort of 'argument the GM has to accept' - of course they don't *have* to, but as you say 'it would feel wrong if the GM said no'. On the one hand you could call that entirely 'Mother May I?' On the other hand, if the GM just said 'no it doesn't work', there was a presumption enforced by table norms that the GM either spotted or privately knew a reason it shouldn't work, which further experimentation by the characters could systematically find.

To put it another way, there's a difference between a game situation where if a ruling surprises you it indicates 'the GM and I are not on the same page', and a game situation where if a ruling surprises you it faithfully indicates 'there is some reason in the world that I am not aware of, I should take this as evidence and push further to understand why'. The rune circuits thing was very dependent on the GM very reliably keeping things in the second category. If you had a GM running it who made things not work 'because of balance concerns' or whatever, it would have had a very different feel (and probably not been as viable).

Like, when we used the stuff to make a reactionless drive for an airship, we presented the GM with a physics calculation. If they said 'no, it doesn't work', for that GM we would have trusted that we misunderstood what the rune and metal combination we used actually did - that we thought aluminum was 'reduce inertial mass' but instead maybe its 'speed up the relative passage of time' or something.

[Quertus]

Well, aside from Elder Scrolls, I'd say they all require GM interaction, even the rune circuits. But I suppose where they might differ is the extent to which there's some sort of 'argument the GM has to accept' - of course they don't *have* to, but as you say 'it would feel wrong if the GM said no'. On the one hand you could call that entirely 'Mother May I?' On the other hand, if the GM just said 'no it doesn't work', there was a presumption enforced by table norms that the GM either spotted or privately knew a reason it shouldn't work, which further experimentation by the characters could systematically find.

To put it another way, there's a difference between a game situation where if a ruling surprises you it indicates 'the GM and I are not on the same page', and a game situation where if a ruling surprises you it faithfully indicates 'there is some reason in the world that I am not aware of, I should take this as evidence and push further to understand why'. The rune circuits thing was very dependent on the GM very reliably keeping things in the second category. If you had a GM running it who made things not work 'because of balance concerns' or whatever, it would have had a very different feel (and probably not been as viable).

Like, when we used the stuff to make a reactionless drive for an airship, we presented the GM with a physics calculation. If they said 'no, it doesn't work', for that GM we would have trusted that we misunderstood what the rune and metal combination we used actually did - that we thought aluminum was 'reduce inertial mass' but instead maybe its 'speed up the relative passage of time' or something.

To flip that on its head again (that seems my classic move), imagine if (your memory was erased as necessary and) you had to create spells under the 3e Epic Spells, 2e True Dwoemers, WoD d20 or WoD Mage systems without any knowledge of the underlying mechanics. Everything would still be trial and error; the only difference (as far as I can tell, and this is the point, so definitely correct me if I'm wrong) would be whether things were hung on abstract things the character understands, on things that have a GUI-icon equivalent for the Player (rune X, material Y, "hocus pocus" verbiage, etc).

Which brings me to another "stand it on its head" concern: in D&D, I can have my Wizard spend downtime attempting to research how to do something; however, if we're on a space flight for 5 days, and I tell the GM, "I have my character say lots of different things in bad latin" while you tell the GM, "I try out <rune X> on different composite materials / different alloys / materials created different ways (cold-forged, replicated, summoned, instantiated from platonic ideal, etc)", I don't expect the GM to be able to either adjudicate that satisfactorily, nor for us to learn as much as we should from our characters doing that / if we had actually played through all the details. In other words, these systems seem to require building Player skills - which I think is the specific feel you said you were looking for out of a magic system.

[Quertus]

Looking at various systems, another difference I've noticed is something I'll call Variance, akin to "replay value", that measures how different multiple characters by the same player can be. Some Variance is just due to increasing player skill - you can tell the Epic Spells you made as a total noob vs the ones you made once you got a clue and started using Mitigating Factors, for example (at least, for those who didn't just start using Mitigating Factors straight out of the gate, of course). And then there's the Variance between a WoD Forces/Prime Mage and a WoD Correspondence/Entropy Mage and a WoD Life/Entropy Mage, or between a WoD Virtual Adept, Akashic, and Cult of Ecstasy member. WoD d20, there tends to be very little Variance, not just between my characters, but between the characters of all the people I've seen use it - there's a fairly optimal path that everyone IME takes, making everyone's characters feel somewhat samey at a baseline. I'm guessing the only thing close is how apparently every Elder Scrolls character would put their spells in items?

So many bears to poke with all these different pokers. Senility willing, I'll get further in understanding (or just making wildly inaccurate guesses about) these bears.

[meschlum]

Since I'm the only one to have raised them...

Grace Magic

Completeness 100% to 80%. The available pool of effects is very versatile, and made out of relatively few building blocks. Doing things outside the expected range takes work, but can often be achieved. More 'completeness' can be obtained by creating new Charms to fill in gaps or expand thematic options - or by getting specific splatbooks and exploiting the content there.

Incompleteness comes in two broad flavors: fairy tales because it's supposed to be the theme, and incomplete / cheating because you can just go for the numeric benefits rather than expend time and effort into cursing cities into being inhabited by Mowgli's friends (singing can be made compulsory).

Variance very high. If you plan on hunting Exalts and other Things Bigger Than You, focus on numbers and cheats, and you can be sort of in their weight class. If you want to create a network of teleportation gates throughout Creation, you can have the core seeds for it at character creation instead (and build more as you go). If you want to be a sentient flower pot that spawns magical girls and monsters of the week, you can also do that at character creation...


Amber Diceless

Completeness 20% to 60% Magic is a collection of ~20 spells or so, anything else is up to the GM. Conjuration (creating items of power) is more complete, but still has a limited range of broad effects (e.g. 'be a better weapon', 'shapeshift across a few predetermined forms', 'travel between universes') which can probably be tinkered with to a greater extent. Different GMs will, as a rule, make for completely different games (Magic is not worth many points, so it's only locally powerful. What locally means is a matter of viewpoint).

Incompleteness Mother May I due to limited base effects, or Encouraged because you're all about using your pool of limited effects to create something powerful enough to stand up to reality shapers.

Variance Fairly high, due to differing GM interpretations and the wide variety of core characters you could play who end up with the same flavor of power (whether you think your power comes from nanomachines or calling on elder gods (or both) does not matter in how it is applied).

[Lord Torath]

I remember the 1st edition AD&D system of "try designing a spell and assigning level and everything else to it. The DM will look at it and decide if that seems reasonable or not, but he won't tell you. Then, you spend a bunch of time and money trying to make the spell. Then, you roll dice to see if you succeeded, but if the DM didn't approve of the spell, you automatically fail."

2E AD&D changed this so the DM is supposed to go over the spell and make any changes he felt were required, then give the spell back to the player. If the player decides they still want to research it, great! If they no longer like it after the DM's modifications, they csn further negotiate or just drop it.

[jjordan]

GURPS hands down. Thaumatology contains everything you need to create a magic system or add on to one. GURPS also provides a good model for balancing spells via the benefits/drawbacks model inherent to the system.

The Spell Law magic system used by Iron Crown Enterprises provides a good framework for a magic system.

[Vahnavoi]

Until you give me a better name, I'm going to dub this "The Feel of Magic". It's an interesting sort of meta-system that could be bolted on top of other systems. By itself, it tells me the flavor of the spell, but nothing of the mechanics of "can this particular mage cast this particular spell?". It's a very different focus, and I think that the flavor is comparable to 2e D&D True Dwoemers or 2e D&D Item Creation (dagnabbit, that's another one I need to add at some point) or maybe Dresden Rituals done right. So, top tier feels, IMO.

"Can this particular mage cast this particular spell?" is the kind of question you answer after a player has drawn their scrabble blocks and can see if they can spell "Cthulhu" with them or not (etc.). Trying to answer the question without knowing particulars of the mage and their situation is, obviously, fruitless.

To make this a bit more concrete, let me give you a simple vocabulary. There are 14 runes - 4 entities, 5 operations (verbs), 4 targets (objects) and 1 meta-operation (amplifier). You get to draw seven out of a bag whenever you try to cast a spell, roll dice on which ones for the sake of trying this out. The amplifier rune (Pargon) has four copies, others just one.

What did you get?

Ignoring the canonical spell list of source game and extending the principle to cover all possible combinations, there are 80 valid sentences (4 entities times 5 operators time 4 targets) with three degrees of amplitude each (depending on whether a spell uses 2 or 4 amplifier runes), for a total of 240 spells. In order to cast a spell, the runes have to be placed in a specially made circle - 1 entity rune, 1 operator rune and 1 target rune, accompanied by either 0, 2 or 4 amplifier runes. The rules of magic have to be obeyed in that either the tools used to prepare the circle, the circle and the runes themselves, the person casting the spell or the place where the circle is laid has to have a physical, historical connection to the entities. The spellcaster also has to know who or what these entities are, you can't cast a spell in an entity's name if you have no idea of that entity. Higher amplitude spells cost more spell and sanity points, if you care of such things.