Sending Magic Back to School (Long)

[FireLance]

I like the idea that magic can do anything, but no one wizard can do everything.

Really, one of the key advantages of a Vancian system that is closer to the original books (and to the OP, and to 4e ) than what has been seen in earlier editions of the game is that spellcasters have access to only a handful of powerful spells at any one time. So a high-level wizard is powerful, but not more powerful than a high-level fighter or rogue.

 

[dd.stevenson]

Can Magic Do Anything? What are its limitations?

The problem is, the answers to these questions are campaign-specific. Traditionally, the question has gone unanswered in the CRB, and in the interim all sorts of official and home-brewed settings have come up with their own answers. Which means DDN can't suddenly answer them without stepping on toes.

And keeping the CRB officially agnostic on this matter really has no downside, as long as we're keeping vancian memorization in--spells are just "specific things we know how to do" and the details of the underlying theory can be left to the campaign settings. Which, IMO, is where they rightfully belong.

CRB agnosticism also frees us to create spells based solely on gamist criteria. (Within the constraints of backwards compatibility.)



Fixed Number of Spells Per Level

I believe what Mearls has said is that there will be a hard cap to the number of spells casters can memorize. So, for example, if the limit is set to 10, then the progression might look like this:

	1st	2nd	3rd	4th	5th	6th	7th	8th	9th
1st	1	—	—	—	—	—	—	—	—
2nd	2	—	—	—	—	—	—	—	—
3rd	2	1	—	—	—	—	—	—	—
4th	3	2	—	—	—	—	—	—	—
5th	3	2	1	—	—	—	—	—	—
6th	3	3	2	—	—	—	—	—	—
7th	4	3	2	1	—	—	—	—	—
8th	2	3	3	2	—	—	—	—	—
9th	—	4	3	2	1	—	—	—	—
10th	—	2	3	3	2	—	—	—	—
11th	—	—	4	3	2	1	—	—	—
12th	—	—	2	3	3	2	—	—	—
13th	—	—	—	4	3	2	1	—	—
14th	—	—	—	2	3	3	2	—	—
15th	—	—	—	—	4	3	2	1	—
16th	—	—	—	—	2	3	3	2	—
17th	—	—	—	—	—	4	3	2	1
18th	—	—	—	—	—	2	3	3	2
19th	—	—	—	—	—	—	4	3	3
20th	—	—	—	—	—	—	2	4	4

This would allow the limits to be easily house-ruled: the hard cap could be set very low to create a nearly 4E experience, or it could be set very high in order to create a nearly 3E experience.

 

[Tony Vargas]

I think a fundamental question about magic that D&D has answered differently (or declined to answer) in different versions of the game is: What /is/ magic?

For instance, Divine magic is very clearly has the power of a Deity behind it (but where do Deities get their power?), Psionics is trained (or trainable or enhanceable) psychic phenomena. In AD&D, magic-users siphoned formless energy from some other plane, projecting their material components or even breath into that plane to maintain some sort of balance - a very scientific, conservation of math/energy take. In 3e and 4e, arcane magic is left undefined - how it is done is related, but what it actually is, not so much.

If magic is simply arbitrary, than there's no reason not to balance it by putting equally arbitrary limits on what it can accomplished or how dependably it can be accesses. If there is a reason behind magic, then bounds and limits may suggest themselves. For instance, the power of a deity might or might not be limitless and the expression of that power through mortal vessels might depend upon the faith or strength of the mortal, while what might be accomplished with divine power could be bounded by the Deity's portfolio (Domains).

Thus, whatever magic is, it probably shouldn't be able to literally 'do anything.' If magic is undefinable, it can be slaved to game balance, since there's nothing to simulate or model. If magic is defined and explained, then limits would be implied that could be worked with to create magic that works within the game in a balanced and playable way.

 

[steenan]

There is one more question to be asked: How do we want the setting (with its magic) to interact with gameplay?

Do we want a world that is, in itself, consistent? Then, magic can't allow one to "do anything". It needs clear rules - and the people in the game world will use and abuse these rules. Players can and should use all their ingenuity in such setting - and the rules of the game, while simple, should reflect the rules of the world close enough to allow for creativity without breaking.

Or, we can have a world that follows a specific genre. Then, magic can be allowed to theoretically "do anything". But at the same time players should be asked (openly, in big letters, on the first page of the rulebook) to limit themselves and play by the genre, not by what RAW allow. Either you buy in the genre and want to abide by it, or the game will break.

D&D often mixed these two and it blew many groups in their faces. Sometimes, things were limited mechanically, for balance, without any good reason in setting. Sometimes, things were allowed that either required completely rebuilding the setting, or allowed players to break it in awful ways.

Distinction between these two approaches is important in many aspects of a game, but completely crucial in designing the magic system.

 

[slobo777]

A while ago, I had an interesting discussion with a friend who wanted to homebrew a spell that "gave you a feat". Not a specific feat, but a feat slot, where the caster could decide which feat was gained. In terms of power, it wasn't major - in D&D Next I could see this as maybe a 2nd-level spell.

I hated the idea, and this is why: It spoke directly to game mechanics, and had no rationale in the game world.

Whether or not spells can "do anything", I definitely expect what they do to be based first on a descriptive elements (e.g. create a burst of energy) as opposed to a gamist element (e.g. does X amount of damage of a specific type). There is no problem with having the latter, or tweaking it for gamist purposes (such as balance or picking out a favourite damage type), but IMO if you start with mechanics first in spells, you can quickly end up in abstract nonsense land, spells called "Defense IV" etc.

 

[Crazy Jerome]

I don't see any inherent problem with "magic can do anything." Rather, the problem emerges in D&D when you start with "magic can do everything"--and then this means that a wizard player that games the system, games the DM, gets lucky, etc. can do "anything that needs doing" by casting a spell that takes a few seconds and has no appreciable cost or risk.

I'd rather that game balance efforts be focused on the costs and risks, rather than than the magic, which I agree with previous posters is really a campaign issue, not a system issue. The game should, in a perfect world, allow "magic to do anything" so that in your campaign you can decide that magic can do whatever subset of "anything" that makes sense.

Part of the problem is that 7 to 10 levels of spells (depending on how you count it) doesn't scale very well with D&D assumptions and Vancian casting. That is, you can easily set up a Vancian casting system that will work pretty well for 3 to 4 spell levels, but the more you range from that, the harder it is to balance.

You can't, for example, fix 3E casting by saying that a spell takes 1 round per spell level to cast. That's unreasonably tough for most 2nd or 3rd level spells, yet doesn't really account for the immense power of 6th+ level buffs.

I'm not sure if it would surmount the level scaling problem, but I'd like to see an attempt to put spells into tiers, an independent dimension from level. So say you have lesser and greater spells. The lesser spells of each level scale about like non-magical character abilities do. They work like D&D magic has usually worked. A wizard who has prepared a 6th level lesser spell in a 6th level slot can pop that thing out in a round, no real risks. But it is not any more impressive than what the fighter or rogue can do--merely different. In contrast, the greater spells have risks and costs on top of that, and these drawbacks scale rapidly with each level. They take a slot same as the lesser spells, but you don't cast one of these babies lightly. There are checks involved, and when you fail you risk hit point damage, ability score damage, temporary insanity, etc.

Think BECMI teleport versus 3E/4E teleports. A "lesser teleport" lets you pop over to that hill that you can see, easily, no risk. A greater teleport lets you go anywhere you know well on the planet, but you may arrive off-target and risk being entombed.

This would also be independent of cantrips/spells/rituals. Both lesser and greater spells can exist in all three. That gives three dimension to spell casting (with level). Now you can have things such as: Greater higher level cantrips - can do anytime you please, but cost time and gold, and backfires in your face if you screw up.

 

[Tony Vargas]

Balancing power with risk or cost is difficult. Cost turns the game into an accounting exercise, which, while I'm sure I'm not alone in finding such things engrossing to a degree, can't have too broad an appeal. Risk, OTOH, gets very swingy, at best. When you bet and win, the payout is huge and the consequence minimal, when you bet and lose, there's no payout, and you're hosed. Since the idea is to pull out the big guns in desperate situations, that can turn a difficult, dramatic encounter into a cakewalk or a TPK. Either way your game can be spoiled - a lose-lose. Further, there's almost certain to be ways of minimizing that risk. Investing in some skill or attribute or feat or ability, for instance, or getting the right item or whatever. If such things are not in core, they'll eventually see the light of day in a supplement or something - the demand for them would just be too strong to ignore. And, once there are ways around the risk, powergamers will erase it.

Balancing what magic does is a much firmer foundation than balancing how often, easily, or economically it does broken things.

 

[Crazy Jerome]

[MENTION=996]Tony Vargas[/MENTION], you are reading "cost and risk" in game terms there. I meant them broader than that, then applied to the game. So a "cost" can be gold, but it can be anything that you must give up or pay to do the thing. It can be extra time, automatic but minor damage to yourself, the spell "burning out" for a day or a week after you cast it, etc. Meanwhile, a "risk" is something severe enough that you would not take it automatically, but might take a chance on in a pinch.

In normal Vancian casting, this doesn't work, because if you put those kinds of drawbacks on all spells, people complain (and with some reason). Or, people immediately start watering them down with feats, magic items, etc. And as you say, if you are going to negate them, why bother in the first place? (The same thing applies to this business in the other thread about having special magic armor for wizards to wear to cast in. If you can pay gold to avoid the problem, at a certain level, it no longer applies. So might as well say that at Level X you can cast in armor.)

Thus the point of having lesser and greater spells, which was the main point. You want easy, risk-free casting? Fine, you can focus on lesser spells, and you don't even need to waste feats and other such choices on uncrippling your casting. You want cosmic power? Fine, you can take your chances with the risks and/or pay the costs involved. But no grumbling, since it is your choice.

You'll note that this cuts to the heart of the disagreement about wizard power, separating those sincere about flavor options from those trying to keep the wizard powerful so that they can dominate the game.

 

[pemerton]

On the one hand, you could go the 3e route and say "magic can do these specific things, and everyone plays by the same rules." On the other hand, you could go the 4e route and say "magic can do anything, but the PC wizard class can only do these specific things (and no one in the multiverse except you is a member of the PC wizard class)." The latter is convenient in a rounded-corners, Chris Perkins kind of way: It's easy to make up crap on the spot, but it destroys the possibility of a coherent world where magic makes sense (which I guess was the point).

I like your characterisation of the 3E/4e contrast, except for the last clause - I don't think that the 4e approach destroys at all the possibility of a coherent world where magic makes sense. It's just that the coherence is delivered by the way the participants (perhaps most importantly the GM) narrate the fiction within whatever constraints the mechanics establish. The coherent story just can't be read of the mechanics themselves.

 

[GX.Sigma]

I like your characterisation of the 3E/4e contrast, except for the last clause - I don't think that the 4e approach destroys at all the possibility of a coherent world where magic makes sense. It's just that the coherence is delivered by the way the participants (perhaps most importantly the GM) narrate the fiction within whatever constraints the mechanics establish. The coherent story just can't be read of the mechanics themselves.

Can you elaborate on this? I'm confused as to what you mean. I don't think the mechanics establish any constraints.