how about this mana point version?

[evilbob]

I have to agree; it seems like too many points off the bat, and the regen rate is just too high. Assuming a starting Cha of 16, +stat points at level 4 and 8, and at least one +2 item by an appropriate level, a sorc could cast her highest level spell every round until at least level 12 or so. And anyone tweaked out could extend that to ...well, the entire game.

<shameless plug> Anyone have thoughts about the one I posted above? Tiny spell pool, and points regen as a function of the highest spell level you have... </shameless plug> I'll say I like it because you can really only get off about 2 high level spells before you'll have to wait. The regen rate may still be too high, however, since you only have to wait 2 rounds to cast your highest level spell again. It could be changed to 1 point / 5 levels, rounded down... Although you'd end up with a rate of 4 points / round at level 20 as above, you'd gain them more slowly.

As for "wait until after combat," that's an interesting idea, but how would you write it out in game terms? Combat is rather subjective, unless you specified that you had to be counting initiative, and then it'd be rather strange to explain... Although come to think of it, I remember a mana point system that allowed 1 point to regen per round in combat, and 1 per hour outside of combat. Of course, it used a system in which 4 or 5 points was a huge mana pool, and spells only went up to level 6.

 

[Pyrex]

If you're looking for small pool with quick refresh, let me expand on the comments I posted above.

Casters Should:
--Have enough mana to cast their top-level spell and at least two lower-level spells per combat.
--Regenerate fast enough to be recovered before the next combat.
--Not be hurting for [MaxSpellLevel]-3 level spells.

Casters Should Not:
--Have enough mana to cast their top-level spell more than twice per combat.
--->I'd go so far as to say a second top-level cast should deplete their mana.
--Regenerate enough mana to cast a 3rd top-level spell over the length of an 'average' combat.

If we hold all these to be true, then an 18th level caster should have less than 27 mana (probably 20-24), and should regenerate no more than 2-3mana/round.

Similarly, a 1st level caster should have no more than 4-5 mana, and not regenerate more than 1mana/3rounds.

This suggests to me that the mana pool should be StatMod+(CasterLevel*0.75); and that we define regeneration rate to be MaxSpellLevel/3 per round.

Or more simply,
-->1mana every third round while you are limited to 1st level spells,
-->1mana every other round when you gain 2nd level spells,
-->1mana/round when you gain 3rd level spells,
-->2mana/round when you gain 6th level spells,
-->3/round when you gain 9th level spells.

That comes out to 2-4 mana for a first level caster, regenerating another 1st level spell every three rounds, up to ~25mana for a 20th level caster regenerating another 9th level spell every three rounds.

 

[evilbob]

I really like your regeneration rule - I would have never thought about extending the number of rounds before regen'ing at low levels.

In order to simply the math (.75 * anything is messy), I wonder if total points = stat mod + (caster level) with spells costing (spell level +1) points would work out nicely or not... You'd also eliminate the annoying zero spell rules. Must... do... math...

 

[Pyrex]

Hmm, that results in a 20th level caster having having a 30 mana pool, casting 3 9th level spells in a row, casting an 8th level spell during round 4 and topping out at a 2nd level spell the 5th round. Still seems a little too front-loaded. The manapool and/or regen rate still needs to come down a bit.

What about StatMod+(0.5*CL) with a cost of SL?

That gives a 20th level caster 20 mana, meaning at best they get two 9th level casts at the top of combat and would have to wait until round 4 before casting a 3rd?

Hmm, regen still feels a little fast. I'd rather make that 2nd 9th-level in-a-row cast a hard choice.

What if we further stipulate that mana only recovers on rounds when you're not casting spells?

And cantrips should simply not cost mana. Which, if it bothers you, might require banning Cure Minor Wounds.

 

[evilbob]

Damn it, I'm going to have to make a chart.

Well, I don't like the idea of 1/2 CL because that makes leveling up more meaningless and aquiring +stat gear more important. (Eagle's splendor suddently becomes worth 4 sorcerer levels.) That's just something I feel needs avoiding.

As for regen, another thought I had was that a level 18 sorc suddenly has no incentive to gain those last two levels. (+2 mana points? Pff.) So, instead of "access to spell levels," I thought why not throw it back to caster level again? I was thinking about this type of regen rule, modeled on yours above:
regeneration rate = (caster level / 5) per round (aka the "sorcs need a d6 to count with" rule)
--> lvl 1 = 1 / 5 rounds
--> lvl 2 = 1 / 4 rounds
--> lvl 3 = 1 / 3 rounds
--> lvl 4 = 1 / 2 rounds
--> lvl 5 = 1/round
--> lvl 10 = 2/round
--> lvl 15 = 3/round
--> lvl 20 = 4/round

Add to this making it a class feature and not inherent to "spellcasting," and that gives one hell of an incentive to straight-class up to 20 (since you'd only get your regen rate from your straight-class, even if you received more spell points from PrCs).

What if we further stipulate that mana only recovers on rounds when you're not casting spells?

That seems like the only way it'd be possible to get close to your ideas about casts per battle. I was thinking about this recently too, because I realized I'd been taking it for granted in the calculations I've been making... 27 points + (cost = spell level) + regen 3/round = four level 9 casts in a row, not three. And a fireball every round, ever. Yeah, no regen on rounds you cast... that's really the only way it could work. (Plus, that makes sense in an RP sort of way, right? ...Right?) And that counts for any round you would normally count toward regen as above - meaning, level 1 guys can't cast 5 magic missles and then regen a point, they'd have to wait 5 more rounds. (Although I'm not opposed to allowing staggering - meaning, you could shoot a missle, wait 3 rounds, shoot another missle, then wait 2 more rounds and get a point back. Plus it sort of follows that way.)

As for cantrips not costing mana... Lemme check... Yep, I still don't like it.



Ok, so here's my chart, based on "average" sorc progression (btw, I have no idea how to do the nifty "insert text" trick where you can control text font and make it all courier - any tips on that are appreciated):
Sorc lvl + stat mod = total points (regen rate / rounds)
1 + 3 = 4 (1/5)
2 + 3 = 5 (1/4)
3 + 3 = 6 (1/3)
4 + 4 = 8 (1/2)
5 + 4 = 9 (1/1)
6 + 4 = 10
7 + 4 = 11
8 + 5 = 13
9 + 5 = 14
10 + 6 = 16 (2/1)
11 + 6 = 17
12 + 6 = 18
13 + 6 = 19
14 + 7 = 21
15 + 7 = 22 (3/1)
16 + 7 = 23
17 + 7 = 24
18 + 7 = 25
19 + 8 = 27
20 + 8 = 28 (4/1)

If spells cost (spell level + 1) mana, that means a level 18 sorc can get 2 level nines, and a level 4 off before bottoming out. Getting back a third level 9 spell would take 4 rounds, or 2 rounds if she didn't fire the level 4 (meaning round-by-round: 1: 9th, 2: 9th, 3: none, 4: none, 5th: 9th). Considering that a sorc should have 3 level 9s and a huge host of lower spells (but then again due to regen always be pre-buffed) at that level, it seems a little overly restrictive. ...Certainly within the guidelines you stated above, but perhaps a bit restrictive. (Another rule I just thought of: time stop doesn't allow regen during the spell's effect.)

On the other end, a level 2 sorc would have 2 firsts, and then 4 rounds later a 3rd first. Considering that same sorc would normally get 5 firsts including bonus spells, that seems about right.

In the middle, we have a level 6 with 2 3rds and another 3rd after 2 rounds. A level 10 would have just barely 3 5ths, but 1 point left over and it would take 3 rounds to get another 5th. Not too shabby, but still perhaps a bit on the restrictive side. A level 20 would be able to fire 2 9ths, then wait just 1 round for another one. If they were really maxed out, they could get 3 in a row before bottoming out. This seems fine, or if anything a bit low-powered (compared to having 6-8 of every level spell that must last the whole day).

Overall, I'd say that method would at best cause the sorc to lose power overall, especially in "boss fights," but still retain full usefulness at least two or three rounds into any battle, especially against random trash mobs. (And at level 20 they can shoot a fireball every other round. ) At worst it makes them very useless very quickly.


More questions: does regen work while stunned/incapacitated/etc.? I'm thinking the old "while you're awake" rule works here, which helps because stunning a sorc for a round doesn't hurt him as much as other chars.

 

[Pyrex]

Add to this making it a class feature and not inherent to "spellcasting," and that gives one hell of an incentive to straight-class up to 20 (since you'd only get your regen rate from your straight-class, even if you received more spell points from PrCs).

Ouch. That totally hamstrings PrC's. What if a Sor wants to take all 10 levels of a 10 level PrC? That *halves* his mana regen compared to a straight-class Sor...

meaning, level 1 guys can't cast 5 magic missles and then regen a point, they'd have to wait 5 more rounds. (Although I'm not opposed to allowing staggering - meaning, you could shoot a missle, wait 3 rounds, shoot another missle, then wait 2 more rounds and get a point back. Plus it sort of follows that way.)

Yep, that's pretty much what I had in mind. Though to be clear I'd count rounds where Sor is using Scrolls/Wands/etc as "not casting".

Ok, so here's my chart, based on "average" sorc progression (btw, I have no idea how to do the nifty "insert text" trick where you can control text font and make it all courier - any tips on that are appreciated):

You use the {code} tag to preserve spacing thusly:

Sorc lvl + stat mod = total points (regen rate / rounds)
 1 + 3 =  4 (1/5)
 2 + 3 =  5 (1/4)
 3 + 3 =  6 (1/3)
 4 + 4 =  8 (1/2)
 5 + 4 =  9 (1/1)
 6 + 4 = 10
 7 + 4 = 11
 8 + 5 = 13
 9 + 5 = 14
10 + 6 = 16 (2/1)
11 + 6 = 17
12 + 6 = 18
13 + 6 = 19
14 + 7 = 21
15 + 7 = 22 (3/1)
16 + 7 = 23
17 + 7 = 24
18 + 7 = 25
19 + 8 = 27
20 + 8 = 28 (4/1)

Considering that a sorc should have 3 level 9s and a huge host of lower spells (but then again due to regen always be pre-buffed) at that level, it seems a little overly restrictive.

A Sor should definately be able to cast at least 3x 9th level spells per day, but I really don't want to see them casting 3x 9th level spells in each and every combat.

Ideally, I'd like to see them top out at a 9th, an 8th, a 7th and two 6th level spells during combat given all their precombat buffs will be operating constantly*.

(Another rule I just thought of: time stop doesn't allow regen during the spell's effect.)

Amusing concept; casting Time Stop then hoping it lasts long enough to get you back more mana than it cost. Given that we already ruled that you don't regen mana on rounds when you cast a spell I'm not sure this ruling is required; but neither am I strongly opposed to it.

In the middle, we have a level 6 with 2 3rds and another 3rd after 2 rounds. A level 10 would have just barely 3 5ths, but 1 point left over and it would take 3 rounds to get another 5th. Not too shabby, but still perhaps a bit on the restrictive side. A level 20 would be able to fire 2 9ths, then wait just 1 round for another one. If they were really maxed out, they could get 3 in a row before bottoming out. This seems fine, or if anything a bit low-powered (compared to having 6-8 of every level spell that must last the whole day).

With this rapid-regen a Sor should never have more than 1/4th the total daily spellcasting capacity of a standard Sor, who is expected to get 4 daily encounters out of his daily spells.

Like I said above, I believe that casting 2x top level spells back-to-back should generally empty your mana pool.

Overall, I'd say that method would at best cause the sorc to lose power overall, especially in "boss fights," but still retain full usefulness at least two or three rounds into any battle, especially against random trash mobs. (And at level 20 they can shoot a fireball every other round. ) At worst it makes them very useless very quickly.

Not when you consider the massive amount of buff spells they'll be slinging around.

More questions: does regen work while stunned/incapacitated/etc.? I'm thinking the old "while you're awake" rule works here, which helps because stunning a sorc for a round doesn't hurt him as much as other chars.

I'd go the other way and enumerate which conditions prevent regen. Dazed and Stunned should clearly prevent regen, but Paralyzed should not.

Another idea (requires more bookkeeping):
You seem to want a system whereby Sor's have a higher potential for high-level spells, whereas I'm intentionally trying to reduce that potential as a balancing factor for their huge out-of-combat casting capacity.

What if we give them a slightly higher manapool (CL*1.5 or even CL*2), but declare that you can't recover the mana spent on spells with duration longer than Instaneous until that spell expires? i.e., Walking around with Mage Armor up all day reduces your manapool by 1. Walking around with Mind Blank all day reduces your manapool by 8. Keeping a party of 4 buffed with Bear's Endurance all day reduces your manapool by (2*4) 8.

That way the 'cost' of walking into a fight with 4x 9th level spells available is that you can't do as much buffing; oppositely, the cost of walking into combat with the whole party buffed-to-the-9's is that your spell capacity is expended and you need to wands & scrolls that battle.

 

[evilbob]

Ouch. That totally hamstrings PrC's.

Exactly! Otherwise why the *bleep* would you take a straight-classed sorc? (Why does anyone do that now?) But that may be solving a different problem. I digress...

Though to be clear I'd count rounds where Sor is using Scrolls/Wands/etc as "not casting".

Totally with you there.

You use the {code} tag to preserve spacing

Thanks!  :)

Ideally, I'd like to see them top out at a 9th, an 8th, a 7th and two 6th level spells during combat given all their precombat buffs will be operating constantly*.

I agree! The hard part is giving them "an 8th, a 7th, and two 6th level spells" without giving them the ability to choose two 9ths instead... Well, without lots of math and tables and stuff.

As for time stop, I guess since it's 1d4 + 1 rounds, at any level before 20 you'd only make it up less than an average of the time (4 rounds), it might be ok. At level 20 you'd make it back on more than average occations (3 rounds). Dunno, just struck me as a possible exploit.

Not when you consider the massive amount of buff spells they'll be slinging around.

This is an interesting statement, considering that sorcs aren't known for their buff spells. Given the limited number of spells they can know, usually sorcs go for offense over anything else... But I can see how a regen sorc would be a GREAT utility device, basically giving all party members +4 to a stat if they wanna use their spells known that way... Then again, so few sorcs take buff spells because they get useless after a while... It's an interesting shift. I wonder if it would be better to rely on the "constant casts" when you only get a few, or if most folks would still go for the tried-and-true boom-count.

I'd go the other way and enumerate which conditions prevent regen. Dazed and Stunned should clearly prevent regen, but Paralyzed should not.

Actually, I disagree. I think if you're conscious, you're regening. Like auras. Dazed and stunned shouldn't matter. (But then I think they need all the help they can get.)

You seem to want a system whereby Sor's have a higher potential for high-level spells...

Actually, no I don't. I completely agree with your idea that casting several high-level spells shouldn't be possible, and that about two per battle is right on-target. The problem comes in - as I said above - with the fact that one 9th level spell does not seem to equal two 8ths + two 7ths + two 6ths + two 5ths etc. If you cast two 9th level spells as a standard level 20 caster, you've not even used remotely CLOSE to 1/4th your capacity, because you've got soooo many lower level spells to fall back on. The issue isn't that I want more high-level spells, it's that I want more low-level spells. What I really want is the ability to cast more like 1 highest level and 2 of all lower level spells per battle. Unfortunately, reducing the potential to cast additional high-level spells (as you were saying) completely removes the ability to cast multiple low-level spells. I guess that's why I was thinking before of a system that allows spells to cost fewer points as you level-up; that way, you could cast more of them without giving access to too many high-level spells.

And of course, the more I think about it, the more I'm wondering how badly a sorc is getting nerfed by not being able to blow their entire wad in boss fights...


As for the "keep spell points used until spells finish" idea, that might be a step in the right direction... I'll have to think about that one. I think it's addressing the correct issue, but it probably wouldn't work with the numbers we've been working with, because then a sorc would be able to do basically nothing. If you're buffed at all, you pretty much have no additional casting capability. Adding more spell points also doesn't really work toward solving that, because then we're right back with the "I can cast five 9ths in a row as long as I'm unbuffed" problem.

 

[evilbob]

perhaps I'm not that dumb after all

So I've been thinking about this whole situation quite a bit, and doing lots of math and going over things in my head, and I believe I keep coming back to the same conclusion. This is annoying to me, because I know I've come to this conclusion before, and I can't believe I essentially forgot it and started down the same path again, only to go around in circles and come out at the same place. It reminds me of about a year or two ago when I also tried for quite some time to come up with a spell-point system and failed. However, both to preserve this conclusion and to put it in a form that can be easily shared with others, I have decided to fancy it up a bit, and use bold text so it will be harder for me to forget my own ideas.


evilbob's 1st Law of Spell Point System Creation for DnD Magic

- The difference in power between each successive spell level cannot be expressed as a simple linear progression. This is because the associated strength of each spell level increases at a pace that is greater than linear.

evilbob's 2nd Law of Spell Point System Creation for DnD Magic

- Any spell point system in which (cost of spell) = (spell level) + (some constant number) WILL NOT WORK. This is because any spell cost expression that can be expressed as above may also be expressed as (current spell level cost) = (previous spell level cost + 1), and this violates evilbob's 1st law.


So there you have it. Cost = spell level or cost = spell level + 1 is simply never going to work, regardless of the system you throw around it, because it is fundamentally flawed. There HAS to be a non-linear progression. You simply cannot account for the "one 9th level spell is not worth more than one 8th + two 7ths + two 6ths etc." paradigm any other way.

And, as long as I'm making up rules and being fancy about it, here are a couple of postulated theorems as well.


evilbob's 1st Theorem of Spell Point System Creation for DnD Magic

- Any system that is very complex will not be adopted or used by anyone. The definition of "very complex" is difficult to pinpoint, but it is assumed anything much more complicated than the current "fire and forget" system fails that test.

evilbob's 2nd Theorem of Spell Point System Creation for DnD Magic

- Any system that has a spell point cost per spell level equal to or greater than 2^(spell level) puts too much of a difference between spell levels. The associated strength of each spell level increases at a pace less than that. (i.e. one 9th level spell is not worth 256 1st level spells.)


So, if you accept these ideas, there's more of a space to work within - as opposed to trial and error without bound. More on my additional guesses to come soon...

 

[Pyrex]

I completely agree with both Laws and both Theorems.

To be accurate though, the 2nd Law is actually a corollary of the 1st law because it's directly derived from the 1st.

Conceptually speaking, how do you feel about Recharge Magic? It's really the closest thing WotC has published to "small pool / rapid regen".

Hmm, that gives me an idea, which may be too complicated:

First, we need non-linear costs for spells that are easy to calculate and don't get out of hand; so I propose something like the following:

SL 1-3 cost SL*1
SL 4-6 cost SL*2
SL 7-9 cost SL*3

I'd like something cleaner like SL^1.5, but that violates the "too complicated" axiom. I suppose something with a steeper ramp like SL*(SL*0.5) could also be used, but you're trading a smoother gradient for increased complexity again.

Thusly, a 6th level spell is worth 4 3rd levels; a 9th level is worth 2.25 6th level spells or 9 3rd level spells.

A 9th level spell is also worth 27 1st level spells, but I don't really care if a 18th level caster is casting 1st level spells all combat (or all day) long; that possibility is effectively balanced by the opportunity cost of wasting combat rounds casting 1st level spells.

I'm also pondering some form on non-linear regeneration (i.e., mana spent for higher level spells takes longer to regenerate than mana spent on 1st level spells), but I'm not sure if there's a simple enough way to track/express it.

Hmm, how about if after casting a spell you can't regen mana for SL rounds?

 

[Malhost Zormaeril]

I really like your regeneration rule - I would have never thought about extending the number of rounds before regen'ing at low levels.

In order to simply the math (.75 * anything is messy), I wonder if total points = stat mod + (caster level) with spells costing (spell level +1) points would work out nicely or not... You'd also eliminate the annoying zero spell rules. Must... do... math...

Not wanting to sound rude, but sounding, anyway, there's a general trend in D&D that math must be continually dumbed down. I've never heard that a Cleric's BAB was hard to calculate: that's exactly .75 * level. "Balanced" systems should be thought of at least from a 1st level and a 20th level vantage point, and to meet the requirements set forth by Pyrex, no other progression will do. Of course, you may disagree on the premises, and get therefore to a different progression. As a rule, I think it's fine that a wizard may cast a top-level spell, sit for five rounds, and cast it again. Others may not.

Well, truth is I'm more attracted to Difficulty/Drain kinds of spell systems, but they're admittedly difficult to do. True Dweomers back in Skills and Powers were very cool, but wouldn't work on less-than-epic levels. Since spell points are the next best thing, I'm all for them.