Alt. Spell-learning rules (some math involved)

[Dannyalcatraz]

I'm designing a campaign in which, due to a global catastrophe, most written texts- spellbooks included- have been lost or destroyed along with civilization. Thus, anyone looking to learn a spell will have to research it themselves, using Spellcraft, since there will be few written sources from which to learn, and as a result, any spellbook or scroll found will be incredibly valuable...

Base Spellcraft DC to learn a new spell from the PHB: DC =15 + 3/spell level adjustment. A die roll of 20 is a success, regardless of the DC.

When the PC is created, the player can choose his spells freely, but for every increase in spellcasting power, instead of choosing spells normally, he must make Spellcraft checks. Instead, he picks a spell and rolls a spellcraft check- if the check is unsuccessful, he cannot learn that spell that level, and must wait until next level, and must choose another spell to research.

Non-PHB spells: +6/spell level instead of +3/spell level adjustment. This modifier does not apply to non-PHB spells that are part of a class' base spell list, like the Warmage has spells from Complete Mage, or Wu Jen have spells from Oriental Adventures. (The campaign will allow non-PHB spells from the Completes, Spell Compendium, and the Book of All Spells- and possibly more as well.) Rationale: PHB spells were more common, and there weill be more scraps and rumors of what they did, making the research much easier.

Specialist modifier: -2d4/spell level Specialist School. Rationale: Specialists find it easier to learn spells within their specialty. Even non-PHB spells within their specialty are only moderately more difficult for them to learn than a non-specialty PHB school.

Opposed-school modifier: + 3d4/spell level. The inexperience of Specialists with their opposed schools means they are unable to accurately comprehend variables unique to the school. Each opposed school they try to learn is a mystery to them. Rationale: Specialists find it more difficult (but not impossible in this campaign) to learn spells from Opposed Schools. They treat an opposed school spell as being at least as difficult as a non-PHB spell. However, a Specialist may only learn one opposed-school spell per level. Thus, a 20th level mage can only know 10 opposed-school spells, 1 each levels 0-9.

PCs with levels in classes with limited spellcasting treat all spells not on their class list (from any source) as an opposed school. Thus, a 20th level bard, with only 6 spell levels, could learn 6 "opposed-school" spells, 1 each levels 1-6.

Metamagic Adjustment: Treat LA of Feat as part of the spell's level for purposes of level adjustment. Spellcasters may research a spell in an already modified form- essentially creating a new spell- since they are working with only minimal research sources. They may still take metamagic feats and apply them normally, even to spells they learned in modified form. Example: a spellcaster who researches a version of Magic Missile (a PHB spell) that does 5 points per missile (as if it had been Maximized) would have to beat a DC of 27 to learn the spell, and it would be a 4th level spell. If that same spellcaster tried to learn a Maximized version of a 1st level spell from the Book of All Spells would have to beat a DC of 39 to learn the spell. Rationale: The research process will introduce some happy discoveries, so some spells may be created in a more powerful form.

Weakened Version of spell- level adjustment by mutual agreement, a mirror image of the Metamagic adjustment, above. Rationale: The research process will introduce some errors, so some spells may be created in a less powerful form.

I'm looking for a little math-fu to see if my formulae are reasonable.

 

[Jack Simth]

When the PC is created, the player can choose his spells freely, but for every increase in spellcasting power, instead of choosing spells normally, he must make Spellcraft checks. Instead, he picks a spell and rolls a spellcraft check- if the check is successful, he cannot learn that spell that level, and must wait until next level, and must choose another spell to research.

(Emphasis added)

I'm guessing that's a typo.

A couple of things....

Does this apply to all spellcasters, or just the "learned" spellcasters? Does mean things to Sorcerers, after all, with their much lower Spellcraft check. Clerics/Druids, too, who get a doozy of a nerf - not only must they specifically learn their spells, but they have to learn the healing spells, the remove bad stuff spells, (potentially - unclear in your write up) spend money to research them, and Intelligence isn't their good stat.

 

[Drawmack]

Ohh, I love analyzing game math so here goes.

Base Spellcraft DC to learn a new spell from the PHB: DC =15 + 3/spell level adjustment.

This does not make much sense to me as the spells get easier to learn the higher their level.

It is easily fixed though DC = 15 + spell level/3 (round up)

I’ll demonstrate

Modifiers by level

Level   Orig. Frac. Orig. Dec. Orig. DC My Frac. My Dec. My DC
1       3/1         3          18       1/3      1       16
2       3/2         1.5        16.5     2/3      1       16
3       3/3         1          16       3/3      1       16
4       3/4         0.75       15.75    4/3      2       17
5       3/5         0.6        15.6     5/3      2       17
6       3/6         0.5        15.5     6/3      2       17
7       3/7         0.43       15.43    7/3      3       18
8       3/8         0.38       15.38    8/3      3       18
9       3/9         0.34       15.34    9/3      3       18

This switch extrapolates easily to all of you fractional level modifiers except for the 3d4/spell level. Again increasing your denominator as spell level increases will make higher level spells easier to learn then lower level spells.

For example a worst case scenario on a first level spell is 15 + 12/1 = 15 + 12 = 27. A worst case scenario on a ninth level spell is 15 + 12/9 = 15 + 1.34 = 16.34.

You could, pretty easily, do one of two things for this. Leave the 3d4 roll as is in the numerator but make the denominator 10 – spell level.

For example: A worst case scenario on a first level is 15 + 12/(10 – 1) = 15 + 12/9 = 15 + 1.34 = 16.34. A worst case scenario on ninth level is 15 + 12/(10 – 9) = 15 + 12 = 27.

See how this turns it around. (You could use this for the base spells too just to keep it the same)

The other thing you can do is, drop the 3d4 for and just make the DC 15 + spell level. This makes spells a little easier to learn but the mechanic is simpler as well.

 

[Jack Simth]

I think the slash is for "per" - so a 9th level spell would require a spellcraft check of 15+3*9=42.... so basically, no 9th level spell for you.

 

[Dannyalcatraz]

First- thanks all for responding!

1) One thing I forgot in the original post- a roll of 20 is, as per the normal rules, a success, so it would still be possible to learn a spell- even a tough one- 5% of the time. Edited into original post

2) Yes, " if the check is successful, he cannot learn that spell " was a typo, and once I finish this post, I'll correct it- thanks! Corrected in original post

3) For some reason, I can't read that coded text...but yes, as Jack Smith pointed out, it means "per" rather than "divided by."

4) Unless a PC is trying to learn a spell not normally on their list, this mechanic applies only to the "learned" spellcasters, since only they are affected by the collapse of civilization by not having access to tomes or teaching schools from which to learn spells.

OTOH, those who aren't learned spellcasters won't be learning as many of the metamagic-inherent spells.

5)

The other thing you can do is, drop the 3d4 for and just make the DC 15 + spell level. This makes spells a little easier to learn but the mechanic is simpler as well.

Why would I want to make the spells easier to learn? You do realize that the +3d4 modifier only applies to Specialist types trying to learn Opposition-school spells, or limited casters trying to learn spells from beyond their class list (like a Bard trying to learn Fireball).

OTOH, perhaps dropping the /per level aspect is advised...

6)

a 9th level spell would require a spellcraft check of 15+3*9=42.... so basically, no 9th level spell for you

However, a maxed out mage should have enough spellcraft (23 ranks) and Int modifiers (+4 to +6 or so) to make a roll of 42 achievable.

OTOH, a non-PHB 9th level spell would be spell would be a 15+6*9=69...

Perhaps those mods for PHB spells should drop to 2/level, and to 3/level for non-PHB spells?

 

[Drawmack]

First- thanks all for responding!
5)


Why would I want to make the spells easier to learn? You do realize that the +3d4 modifier only applies to Specialist types trying to learn Opposition-school spells, or limited casters trying to learn spells from beyond their class list (like a Bard trying to learn Fireball).

OTOH, perhaps dropping the /per level aspect is advised...

I made the statement on a faulty read of the rules, which is why examples are sometimes important.

 

[Drawmack]

Okay, so the DC is 15 + 3/spell level so the progression is 15, 18, 21, 24, 27, 30, 33, 36, 39, 42.

I would probably house rule also that spellcraft is a class skill for any class that can cast spells and instead of tying it to int I would tie it to whatever ability the spellcasters bonus spells are tied to, just makes a little more sence to me.

Assuming that spellcraft is a class skill then the skill points progression is 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23. Under this system any spellcaster is probably going to keep their spellcraft skill maxed. Let's say they have a +2 (or 14, 15) in their ability to modify this skill. That means they'll be adding 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25 to their roll.

The Wizard class gets spells at the following levels (class level/spell level) 0/1, 1/1, 3/2, 5/3, 7/4, 9/5, 11/6, 13/7, 15/8, 17/9.

Following are the modifier, assuming +2 from ability / spell DC at the level a spell level is acquired: 6/15, 6/18, 8/21, 10/24, 12/27, 14/30, 16/33, 18/36, 20/39, 22/42

So a first level caster has a 45% chance of learning a first level spell, but a seventeenth level character only stands a 5% chance of learning a ninth level spell this seems a bit wonky to me. So I think the 3 per spell level is a bit off center.

Learning theory tells us that the closer you get to mastery of a skill the longer it takes you to progress in your learning of said skill. So the learning chance should probably decrease but, this seems a bit drastic.

This line: “if the check is unsuccessful, he cannot learn that spell that level, and must wait until next level, and must choose another spell to research.” Doesn’t seem to fit the system very well. It seems you’re attempting to create a moderate magic world, yet the PCs are guaranteed the standard level of spells, just not necessarily the spells they choose. This will not create a lower level of magic, just a less tamed one. If this is the way you want it to play out then it can be done very simply with role playing where they can only learn spells if they find a tome containing said spell. This would add great plot devices to the game.

For non-phb spells the dc modifier is doubled making the dc progression 15, 21, 27, 33, 39, 45, 51, 57, 63, 69. This means that learning any non-phb spell above 4 requires a roll of 20 at any level. This is extremely difficult, which may be what you want. I just thought I’d point that out to you.

With the modifier for Specialist School your roll makes very little sense. The mean is 5, there is a 25% chance that the player will roll a 5. This means that 25% of the time the player will be reducing the base modifier by 2 per spell level. At a ninth level spell this would reduce the base modifier to -3 25% of the time. Ouch. It gets even worse if you only consider the percentage of the time they will keep the modifier at 15 or less for any given level. 18% of the time the modifier to learn a spell will be 15 or less regardless of the level spell being learned.

I would think that a flat reduction of 1 or 2 per spell level would work out much better.

Also the modifier for Opposed school spells if very, very high. The mean value of 3d4 is 7.5. A majority of the time the only way someone can learn an opposed school spell is to roll a natural 20. Again this may be better done with a static modifier to the DC per spell level – but I’m not sure what number to use as the number would significantly change the flavor of the game.

I’ll stop with the analysis there and see what you think about it.

 

[Dannyalcatraz]

I haven't gone through everything on this, so this is only a partial response...

I would probably house rule also that spellcraft is a class skill for any class that can cast spells

That's a good idea.

...and instead of tying it to int I would tie it to whatever ability the spellcasters bonus spells are tied to, just makes a little more sence to me.

Well, this is about organized research and study...I think it will have to stay Int based. Tying a skill that models the learning process to a stat like Wis or Cha just doesn't make sense to me.

Besides, those PCs who don't use Int for spellcasting generally don't have to study their spells anyway- they just know theirs, or they are granted them by their gods. By and large, its not going to affect them much, unless they're trying to add spells from another class' list.

Let's say they have a +2 (or 14, 15) in their ability to modify this skill.

I don't think that's the best assumption. Almost any primary spellcaster is going to place his highest stat in his casting stat, if for no other reason than he can't gain access to the higher level spells at all without it.

A 14 or 15 would represent the absolute minimum for a caster intending to retain a full progression. 16-17 would be the middle guys, and 18+ would be the star pupils.

This line: “if the check is unsuccessful, he cannot learn that spell that level, and must wait until next level, and must choose another spell to research.” Doesn’t seem to fit the system very well. It seems you’re attempting to create a moderate magic world, yet the PCs are guaranteed the standard level of spells, just not necessarily the spells they choose.

Its not that the magic is moderate at all- the world has the same amount of magic in it as before, but there are no teachers or textbooks- every spellcaster is forging his own path. Originally, I was going to have each player design their own spells, but I didn't find a system that was flexible enough to do so without changing certain fundamental features of the game's magic system. One thing I can count on in my group is that several of the more influential (read: more likely to complain) players don't want major variations in the system. If it doesn't smell like D&D, they won't play it.

The idea is that a mage is researching several spells at a time at any given time in his life. If he is (mechanically) unsuccessful with learning one particular spell at this time, he is still likely to learn ONE of the ones he's researching. The assumption is that the player will choose the one he most wants for his PC, and that, even if he fails the first roll, the odds will still favor him getting a spell that he likes...even if its a stronger form of a lower level spell or a weaker form of a higher level spell.

If this is the way you want it to play out then it can be done very simply with role playing where they can only learn spells if they find a tome containing said spell. This would add great plot devices to the game.

Unfortunately, I can't depend on that. Think- if I do that, I'd have to ensure that there were a certain number of tomes and scrolls lying around for them to find each few adventures. That kind of screws the campaign's "most written texts, spellbooks included, have been destroyed, along with the civilizations that created them" theme. The way I'm seeing the campaign, there may only be a couple of scrolls or books lying around in a given 5 level arc...and I'm planning on having only one active school of wizardry in the world. Only in the subterranean empires that survived would there be any appreciable number of spellbooks lying around, and not only will they be jealously guarded, it will be some time before the PCs would be powerful enough to adventure in those areas...and if they don't have spells before then, they'll fail anyway!

I mean, we're talking about a world in which a scroll with a couple of 1st level spells might be as valuable (in GP terms) as a powerful staff, simply because it can be used to teach others.

With that kind of text rarity, depending on that plot driven "you only learn what you find" campaign mechanic would simply screw the spellcasters.

Instead, I'm allowing a meta version of that. I'm assuming the PCs are constantly finding bits and scraps, and piecing them together along with their own research...and I don't have to sprinkle actual tomes and scrolls across the landscape.

Overall, though, I think you're right- my per-level mods may be too high. What do you think of 2/level (PHB) and 3/level (non-PHB)?

As for the non-PHB, Specialist and Opposed school mods...I'm still mulling those over. In a sense, non-PHB spells are supposed to be "rarer" than the PHB spells, so there will be less known and knowable data on how to recreate those spells in this campaign world, thus they should be harder to learn. Specialists should find learning school spells fairly easy, since they're intimately familiar with the theoretical underpinnings of their specialty. Learning opposed school spells should be truly difficult- especially since it is a "gift" as compared to an outright ban in the base game... But I can concede that my mods may still be off.

Anyone else have an opinion on those in particular?

 

[Drawmack]

First, I know that 14 - 15 is kind of low. That's why I used it. It's something I learned about rule analysis a long time ago. If you're testing the feasability/flexibility of rules you have to look at them at the outeredges more closely then at the center.

Well a +2 for PHB spells would keep the difficulty level of current level spells linear with the maxed spellcraft and a +3 for non-PHB spells would decrease the chance of learning the spell by 5% with each subsequent spell level. So that could be okay, for an analysis of this lets look at how difficult it is to learn 1st level spells and taking the linear affects from there for extrapolation.

A first level spell has a learn DC of 15 + 2 = 17
A first level caster has 4 ranks (max) in spell craft

14/15 = +6 for ranks + int = 45% chance on PHB spell, 40% chance on non-phb spell
16/17 = +7 for ranks + int = 50% chance on PHB spell, 45% chance on non-phb spell
18/19 = +8 for ranks + int = 55% chance on PHB spell, 50% chance on non-phb spell
20 = +9 for ranks + int = 60% chance on PHB spell, 55% chance on non-phb spell

The % PHB chance in static as levels increase. I.E. It is no harder for a 17th level wizard to learn a 9th level spell then for a first level wizard to learn a first level spell, which makes sense to me.

The non-phb spells are more rare - so they get harder to learn as the levels in crease does not make much sence. Maybe instead of increasing the per level hit on non-phb spells you could increase the starting modifier.

So PHB spells 15 + 2 per spell level
Non-PHB spells 20 + 2 per spell level

Effectively Non-PHB spells are always 25% harder to learn then PHB spells.

 

[Dannyalcatraz]

First, I know that 14 - 15 is kind of low. That's why I used it. It's something I learned about rule analysis a long time ago. If you're testing the feasability/flexibility of rules you have to look at them at the outeredges more closely then at the center.

Grrr! I hate it when people point out something that I have forgotten. Push the boundaries of the game to what rattles and what shakes apart- that is my very credo when playtesting!

So PHB spells 15 + 2 per spell level
Non-PHB spells 20 + 2 per spell level

Effectively Non-PHB spells are always 25% harder to learn then PHB spells.

Simple. Linear.

I like it!

I guess that with this model, you'd then change the other modifiers to linear ones as well?

Specialist mods could be -2 and Opposed school mods would be +2...not per level, but just a straight adjustment. Or, if you prefer, the base Spellcraft check would be DC 13 (18 non-PHB) and DC 17 (22 non-PHB), +2/spell level respectively?