Thought process: a "mana die"

[Buufreak]

Alright, hear me out. Everything you gain a level, you get a hit die, you roll it, you add a mod, then you add that total to your max hit points. What would happen if we did the same thing with ability to cast spells, ie mana?

[Doctor Despair]

I believe Complete Arcane has a spell point variant. Also see: psionic power points.

[Vaern]

I believe Complete Arcane has a spell point variant. Also see: psionic power points.

It's in Unearthed Arcana, and I believe it's also available on the SRD. I believe what he's suggesting, though, is rolling on level up to randomize the amount of spell points gained per level, rather than having static numbers pulled from a table as in the existing spell point variant.

I think the main balancing issue with this is that it would result in a linear progression of spell points. The two most common ways to implement spell points involve making a spell cost points equal to its caster level, or making a spell cost points equal to its spell level. Both of these methods basically require that casters gain spell points per level on a gradually increasing curve to retain casting abilities on par with core Vancian casting.

With a mana die variant, one of three things is going to happen:
1) Casters end up with underwhelming spellcasting abilities at high level
2) Casters start out much stronger at lower levels so the end-game portion of their linear progression lines up with where it should be
3) The DM has worked out some compromise to balance the system in some other way

Regardless of which of these comes into play, this change would also result in inconsistency among casters. Two casters of the same class, level, and ability scores may end up with significantly different spell point pools if one of them got lucky rolls and the other rolled poorly. Spell points aren't quite in the same boat as HP here. Spells are a class feature. When a fighter levels up, he knows his attack bonus is increasing and whether he's going to get a new feat. A rogue knows when his sneak attack is going to increase. Druids know when their wild shape progression is going to come into play. And when it comes to spellcasters, it's nice to know how many spells per day they'll be able to cast next level so they can plan their character progression accordingly.

It may be fun to spice things up a bit, but I certainly don't think it would be for everyone. That being said, I've already had a few homebrew ideas surrounding this theoretical mana die system while writing this post...

[Buufreak]

Vaern, I appreciate the feedback, and will continue with the general idea I was going for, but first wanted to go over some thoughts.


I think the main balancing issue with this is that it would result in a linear progression of spell points. The two most common ways to implement spell points involve making a spell cost points equal to its caster level, or making a spell cost points equal to its spell level. Both of these methods basically require that casters gain spell points per level on a gradually increasing curve to retain casting abilities on par with core Vancian casting.

I would be going with spell level = mana cost. Otherwise we end up with a situation similar to Truenamers, I would think, especially if using effects that increase caster level. And really, I don't want "on par with core Vancian casting." My group hates full casters, partially because of the book keeping that comes with vancian and full casting, partially because they know the kind of power they can bring to the table. This is an effort to bring some ceilings down a notch or two and make them more appealing. Plus, as I have found out, people just like throwing dice. Given, it is just 1 extra throw per level, but in the theory and "hey guys, what do you think" stages, it sounds more favorable.


With a mana die variant, one of three things is going to happen:
1) Casters end up with underwhelming spellcasting abilities at high level
2) Casters start out much stronger at lower levels so the end-game portion of their linear progression lines up with where it should be
3) The DM has worked out some compromise to balance the system in some other way

Alright, lets use the numbers I have written down and see what happens. Wizard, lets say level 15. I've got them down as a 1d4+cha for MD. Assuming mediocre charisma, giving a +1, full die at level 1, average all the others. That gives us 54 at level 15. That... well damn, is about 1/3 of what the spell point table would suggest. Okay, die size might need adjusting. I'll keep working on that.

(Ran the numbers again with 1d8, came much closer to what I had expected but didn't test. 86 points would likely give all the spellslinging needed for a day, right?)


Regardless of which of these comes into play, this change would also result in inconsistency among casters. Two casters of the same class, level, and ability scores may end up with significantly different spell point pools if one of them got lucky rolls and the other rolled poorly. Spell points aren't quite in the same boat as HP here. Spells are a class feature. When a fighter levels up, he knows his attack bonus is increasing and whether he's going to get a new feat. A rogue knows when his sneak attack is going to increase. Druids know when their wild shape progression is going to come into play. And when it comes to spellcasters, it's nice to know how many spells per day they'll be able to cast next level so they can plan their character progression accordingly.

But there is still variance in those same casters. Did they roll the same ability scores? Did they pick the exact same race, features, ACFs, feats, gear, so on and so forth? Every character is going to have variance, even among 2 wizards, because not everything is precise and perfect. Not in our world, and not in the game world. Even something as simple as 2 wizards casting the same spell with the same stats at the same target has variance, because dice still need to be rolled, be it an attack or save. And I am rather alright with that. I like the variance. I like the possibility of failure. I don't play to win! (Sorry, there is someone who lurks that thinks that is the only thing that matters, and I'm rather annoyed at this point.)


It may be fun to spice things up a bit, but I certainly don't think it would be for everyone. That being said, I've already had a few homebrew ideas surrounding this theoretical mana die system while writing this post...

I would like to know about those, if you get them ironed out. Compare notes and what not.

As for my ideas that I don't think were covered above: The idea came when considering how many games exist with a mana bar, and how you are able to consume said bar as you wish. Got 40 points? That's 40 1pt spells, or 20 2pt, etm. And while I am fully aware (and wish people would stop suggesting it) that 3.x has a spell point variant, it is far too static. This level gives this many points, guaranteed. I'm just not a fan of that. A hit die size is consistent, but the value that comes off of it is not. The only part that remains somewhat consistent would be the Con mod that is being added to is. So, similarly, I wanted to add a static modifier to this mana roll. I put some fair thought into it, and as gimmick as it might sound, I accept the terms and conditions of making it MD + Cha mod. It removes a dump stat. It doesn't let wizards go completely out of control by basing it off of Int. Same with divine casters. I've always been a fan of magic being a force of personality kind of thing, and this helps enforce that idea. Further, I understand that different classes have different magic access naturally built into their kits. It is accounted for. Sorc MD > Other full casters > half casters > non casters. Simple system for a simple effect. It also would drastically lower nova'ing, which I think would drastically reign in some things.

Regardless, I sincerely thank you for your opinion and perspective, it is only helping me further fine tune and balance things!

[pabelfly]

Decided to do a thought experiment with Psion. Every time you level, you roll a D10 to dictate how much PP you get from the level.

If you roll a 1, you get 20% of the normal amount of PP (rounded to nearest number),
If you roll a 2, you get 40% of the normal amount of PP (rounded to nearest number),
If you roll a 5, you get the normal amount of PP
If you roll a 10, you get twice the normal amount of PP

Between level 1 and 2, Psions get, on average, 4PP (before stat bonuses). With this system, there can be a range of 1PP and 8PP. That could be up to seven more first-level psionic abilities. From level 19 to 20, this can be the difference between 6PP and 64 PP from level 19 to 20. That's quite a few augments or extra abilities, or ninth-level psionic abilities.

So, a huge difference each level. I don't mind some randomness in my games, but whether that's something you or your players want is another question entirely.

[InvisibleBison]

One difference that comes to mind is that, since non-casters still get a mana die, taking a level or two in a non-casting class isn't quite as painful for caster builds.

Also, there's room for non-casters to get mechanics that let them do something with their mana points.

[Ramza00]

It changes the style of play (not that this is a bad thing)

I argue ever permanent randomness like rolling for hit dice, and rolling for spell points increases the feeling of horror and fickle fate.

Likewise if we do this everday (rolling for hit points and spell points, like an 8th level cleric rolls 8d8+Con every day for hit points) it becomes ordinary, especially the more die rolls you have. At the same time the hidden fickle fate is always there behind it all even if you forgot the ordinary thing.

[Vaern]

Alright, lets use the numbers I have written down and see what happens. Wizard, lets say level 15. I've got them down as a 1d4+cha for MD. Assuming mediocre charisma, giving a +1, full die at level 1, average all the others. That gives us 54 at level 15. That... well damn, is about 1/3 of what the spell point table would suggest. Okay, die size might need adjusting. I'll keep working on that.

(Ran the numbers again with 1d8, came much closer to what I had expected but didn't test. 86 points would likely give all the spellslinging needed for a day, right?)

Keep in mind, though, that you're going for spell level = spell point cost, while the Unearthed Arcana table assumes that caster level = mana cost. They can't necessarily be balance against each other. Using your system, burning hand at CL 5 would only cost 1 point and fireball at the same level would cost 3 points. Using the UA system, both would cost 5. Your system should give casters a much lower number of mana points, especially if you're hoping to throttle your wizards as an extra bonus.

I might recommend a flat d6/level for wizards. On average, this gives them more spellpower per day than core casters until 8th level, then they begin falling behind and eventually end up significantly weaker than core. By 15 they can expect to have roughly 50 mana, versus core's 100 total spell levels. This looks a lot weaker at a glance, but the added flexibility more than makes up for it. The core wizard only gets one 7th level spell slot, but this wizard could technically go nova and burn all of his points on 7 7th-level spells since he is not restrained by spell slots. Given that level of flexibility, I might even try aiming a bit lower, but trying to find a good fit works.

I'd also suggest a flat 1 point per level starting at 4th for paladins and rangers, 1d3/level for bards, and 1d6/level for druid and cleric. This should put them about on par with the wizard, being stronger early game but then having a much smaller amount of raw spellpower that they can use much more flexibly later on.

I don't know what I'd do with sorcerers, though. Saying a d8 or a d10 would be easy... but, with everyone else's casting becoming much more flexible, the sorcerer's increased spell per day count is really all they have going for them. Just doesn't feel right to leave them with the ability to roll a 1. It might actually be better to give them the same d6 as the other full casters, but compensate with class features that interact with mana points in a way that allows them to consistently cast more spells per day.

As for the charisma modifier... it would certainly be nice to have some scaling off of ability scores, but when spell level = mana cost it'd probably just be too strong to include a bonus on every mana die roll, especially with partial casters. If anything, I'd give each character a flat bonus to their mana points equal to the total level of bonus spell slots they'd normally get for having a high ability score modifier. This gives them a safety cushion against a low roll or two, but without allowing their mana pool to grow to an unreasonable size.


I would like to know about those, if you get them ironed out. Compare notes and what not.

For starters, feats that are essentially the equivalent of toughness and improved toughness for mana points. These become much, much more desirable if you reduce your mana die sizes.

For a prestige class, a rework of blood magus that gains the ability to spend hit points in place of mana points. As-is, the blood magus's only ability to utilize blood as a part of their spellcasting consists of taking up to 4 points of damage per spell to almost make up for the caster level progression that the class misses out on. Allowing blood to actually fuel spellcasting (while buffing spells cast in such a way) would make the class feel much more meaningful, and maybe even worth delaying spell progression. This again becomes much, much more desirable if mana points are made more scarce, and would likely be sought out as a shortcut for wizards seeking to match sorcerers' naturally higher levels of mana (or even by sorcerers who fell short due to unlucky rolls, or who are especially reckless in their use of magic).

[Quertus]

Did you know UA already has… oh, you do. Well, why not just use that… oh, because it’s too many, and y’all like rolling dice. Ok. Sounds good. And it gives muggles / muggle classes mana. And makes caters MAD. Ok.

But, if y’all like rolling dice so much that “X% of the UA table” loses to a mana die, why not do the same for skill points, BAB, saves, bonus feats, and dice of Sneak Attack Damage? That would be some amazing variance between classes, and so many rolls on level-up.

[RandomPeasant]

I would ask what the advantage of this is supposed to be. The random rolls that go into character creation are an artifact of versions of the game that expected a sort of Darwinian selection whereby characters with above-average rolls survived. Very few people play that way at this point, so adding more of this sort of randomness seems undesirable. Beyond that, spell points aren't really a good idea for traditional casters, as the tradeoff between spell levels is very rarely linear. At low levels, being able to cast two color sprays in place of a web dramatically increases a characters sustainability, at very little cost to their offensive output. At high levels, you would trade far more than merely nine magic missiles (or even silent images) for another shapechange. Even discounting that, the game already has point-based casters in the Psion and friends. The benefit of reducing mechanical variety is unclear to me, particularly when the proposed system isn't even a direct parallel to the Psionic one.

[Chronos]

Yeah, almost nobody any more plays with random HP, because it's needlessly and pointlessly fickle, and results in clear imbalances between characters for no real benefit. The same would be true of this (in addition to the fact that you'd need to be using "mana points" of some sort to begin with, which the system really isn't designed for).

[Maat Mons]

Not to pile on the hate-train, but I too think rolls should happen only in play, and never in character creation. If you like rolling dice, you might try the UA variant where casters make a check to see if the spell affects a target, instead of the target making a saving throw.

I my perpetually-unfinished homebrew game, everyone has the same amount of mana. High level, low level, caster, non-caster. The way some people have greater endurance than others is that you make a roll when casting a spell to see how much mana it cost you. More powerful spells influence the roll towards higher costs. And your spellcasting ability and level in casting classes influence the roll towards lower costs. Considering doing a thing where you can only attempt to cast a spell if you have enough mana to pay the maximum cost you could roll. So as spellcasters tire out over the course of the day, they’re restricted to less powerful spells.

Personally, I agree with the idea that force of personality should govern spellcasting. But I disagree with the idea that Charisma should represent force of personality.

Edit: In order to keep HP and MP from feeling too same-y, I suggest making them recover at drastically different rates. I like the idea of MP recovering quickly, and HP recovering slowly. Fast-recovering mana does, of course, argue for a smaller maximum pool.

[Vaern]

Edit: In order to keep HP and MP from feeling too same-y, I suggest making them recover at drastically different rates. I like the idea of MP recovering quickly, and HP recovering slowly. Fast-recovering mana does, of course, argue for a smaller maximum pool.

I'd imagine mana would just fully recovers over 8 hours of rest, just like spell slots do. Though I suppose slower recovery could balance the flexibility granted by a spell point system allowing you to empty your mana bar on just your highest level spells - if you burn everything at once like that, it may be a couple days before you've fully recharged. It also adds a bit more demand for the potential mana potions that could come with a spell point system.

[Melayl]

I did work out a full spell point system. It would be relatively easy to use die mechanics to gain their mana point increase instead of by level. It should be in my signature.
Feel free to steal whatever you'd like from it.

Edit: nope, wasn't in my sig. It's here:

https://forums.giantitp.com/showthread.php?140759-Melayl-s-mana-based-spellcasting-system-3-5-PEACH

[Maat Mons]

I was thinking maybe make mana recover every encounter, but have a smaller pool to compensate. It would eliminate the incentive to fight only one battle every day.

In retrospect however, the speed of hit point recovery is irrelevant. As far as adventurers are concerned, natural healing may as well not exist. You do all your healing via magic.

Edit: So how do specialist Wizards interact with this? Do the gain bonus mana that can only be spent on spells from the school they specialized in?

[SangoProduction]

The primary issue with mana as a concept (introduced as an adaptation of the system, rather than its own system, ala Spheres of Power):
Let's say they have 4 level 1 slots, and 1 level 2 slot. A not-uncommon allocation.
Well, if level 1 slots used 1 mana (and so on), they actually have the mana for 3 level 2 spells.
Sure, sure "they are just burning through resources faster. They aren't actually more powerful." Unless... for example... they are not normally running through all their spells in a day. "So just give them less mana..." Is the correct response. Same with spell slots too, honestly. If they are supposed to be a careful resource, then you should either have enough encounters to expend them, or they should have less to expend.
But of course the "fun spells" which are cast socially (outside of prestidigitation) also use spell slots. You could create a separate pool of them. Or maybe even say "Outside of combat, puzzles, or other such 'encounters,' you may freely cast non-buff spells." Or something the sort.

[ToranIronfinder]

I believe the Dark Eye, the German system that the Realms of Arkana system used something like this kind of approach. Makes got Astral points at level up, or had to separate their HD on level up between Astral points and hit points.

I think, though, a change like this would require a complete redo of the magic system, as that system doesn't use Vancian spellcasting at all. Part of what works in that system though was that you didn't receive all of your Astral points at the beginning of the day, it took a while to recharge, so going Nova (or say strengthening your magic wand) wasn't adviseable.

[bekeleven]

I feel like this conversation springs from a pretty reasonable desire, the desire to equalize the resources gained by mages and martials. And the (valid) concerns people have with this plan are excellent demonstrations on how martials need a higher-scaling, non-random resource gained each level.

That could come in a lot of forms. For example, fighters could gain [Fighter level] hit points every level. You'd need to write up some stuff about it (off the top of my head: Make it continue into full-BAB prestige classes, like how casters can continue casting into their prestige classes) but it would help fighters a bit. It wouldn't address their primary weaknesses, which are utility-related...

...An example of a mundane class with a more utilitarian resource would be ninja (and PF monk) with their Ki pools. So those should scale more aggressively.

[ciopo]

What if also spell themself costs a variable amount of mana points? Let's say Xd4-Y (min X), where X is the spell level and Y is the casting ability modifier? And on level up casters get an increase of Xd4+Y*X mana points, where X is the highest spell level they have access to and Y is their casting ability modifier?

Call it the vagaries of mana or whatever

[lesser_minion]

I think the main balancing issue with this is that it would result in a linear progression of spell points. The two most common ways to implement spell points involve making a spell cost points equal to its caster level, or making a spell cost points equal to its spell level. Both of these methods basically require that casters gain spell points per level on a gradually increasing curve to retain casting abilities on par with core Vancian casting.

The problem with common spell point variants (that try to convert most of a caster's allotment of spells into points) is that they give casters credit for spells that aren't very important to their power at a given level. For a spell point system, if we want to be conservative about balancing relative to core, then the number of spell points ought to be such that a caster cannot ever cast more spell of a given level using spell points than they could have using slots.

For a druid of level/casting ability 20 (+1 spell at each level from 1-5, and +1 extra 1st-level spell), the maximum number of spells of a given level castable per day are as follows:



Spell Level
No. Casts
Casts x Level


IX
4
36


VIII
8
64


VII
12
84


VI
16
96


V
22
110


IV
28
116


III
34
102


II
40
80


I
47
47


-
53
-



This suggests that prepared full casters should start with about 2-3 spell points and finish with no more than 44, including any points that need to be reserved to their domain/specialist spells.

(EDIT: I put a first attempt at a spell point system here, but I've refined it, cleaned up the explanation, and moved it to my other post)

[ciopo]

using the mana accumulated from having had lower level spell access to cast highest level spell is indeed a fair problem of a spell point system.

It's not easily solvable tho, that's more or less always going to be the case in a spell point system, unless you do things like separate pools for each spell level.. but if you do that, you're basically back to being a spontaneous vancian with extra steps.

A truenamer-lite solution could be that with each cast of a spell of a particular level, the cost for that day for that spell level increases in some manner

[lesser_minion]

using the mana accumulated from having had lower level spell access to cast highest level spell is indeed a fair problem of a spell point system.

It's not easily solvable tho, that's more or less always going to be the case in a spell point system, unless you do things like separate pools for each spell level.. but if you do that, you're basically back to being a spontaneous vancian with extra steps.

A truenamer-lite solution could be that with each cast of a spell of a particular level, the cost for that day for that spell level increases in some manner

If you're using spell mana cost = spell level, then mana pool = caster level + lesser of key ability modifier or highest castable spell level will be weaker at all levels than the core spell slots system. It's certainly not impossible to come up with a simple design for spell points that doesn't overpower things -- I'm actually more worried about this sort of approach nerfing casters too much.

I think overall, AB + (CL * multiplier), with the ability bonus capped at 1 higher than your highest castable spell level, is still probably what I'd suggest, at least prior to playtesting. Even when the number of casts you can get of your best spells is higher than spell slots would allow, getting four 9ths at 17th level (instead of three), even with how powerful those spells are, doesn't sound like it would make you much stronger than a 17th-level spell slot user. And it's certainly a lot better than UA spell points, where a 17th-level wizard or druid gets enough points to cast 16 ninth-level spells every day.

Ultimately though, while this is a spell point system that I think is reasonably simple without being horrendously overpowered (with a few caveats, some of which I've already mentioned), it needs some adaptation before it's what the OP wants, and a lot of playtesting to make sure that casters are still having fun despite these numbers being so small.

[Maat Mons]

If you take the base Wizard spell slots, not counting bonus slots from ability score or specialization, and look at the equivalent mana cost, you can then get pretty close to that with the following progression:


LevelMana Gained
1 - 31
4 - 51d6
6 - 72d4
8 - 92d6
10 - 112d8
12 - 133d6
14 - 153d8
16 - 174d6
18 - 204d8


The effect the casting ability score has on effective mana is weird. I'll try to come back later to address how to approximate that.

[Remuko]

i like the idea of an MP system (spell points isnt that) and Id think the benefit of it would be you could assign varying costs per spell instead of having all the same level spells cost the same. Make the MP pool big for more granularity in the costs that way you can make the strongest spells of each level cost more. You could even soft ban some "broken" spells by jacking their cost way up, if you saw fit. Its not as simple but this customizability (for the DM) seems like the real benefit you could get out of such a thing.

[Glimbur]

Did you know UA already has… oh, you do. Well, why not just use that… oh, because it’s too many, and y’all like rolling dice. Ok. Sounds good. And it gives muggles / muggle classes mana. And makes caters MAD. Ok.

But, if y’all like rolling dice so much that “X% of the UA table” loses to a mana die, why not do the same for skill points, BAB, saves, bonus feats, and dice of Sneak Attack Damage? That would be some amazing variance between classes, and so many rolls on level-up.

I made that class for a PrC contest. It got the most hate mail of any class I have made, and I would not endorse using it in a real game. http://www.giantitp.com/forums/showpost.php?p=14588327&postcount=8.

[Quertus]

I made that class for a PrC contest. It got the most hate mail of any class I have made, and I would not endorse using it in a real game. http://www.giantitp.com/forums/showpost.php?p=14588327&postcount=8.

Great class. Not *quite* what I was describing, but sounds like loads of fun. Not sure if I could get a GM to go for it.

I was saying, why not give Rogue a 50% chance of getting +1d6 SA each level, all classes 50% chance of improving good saves each level beyond the first, etc. Maybe replacing 2 skill points from each class with a d3 (or 3 from 6+Int / 8+Int classes with a d5). 1-in-4 chance of stat boost and 1-in-3 chance of getting a feat on level-up. Etc.

Lots of fun unpredictability on level-up.

For a Chaotic class, I like what you did with saves, HD, BAB, and skill points. I think I would have made the abilities simpler and more Chaotic. For example, each encounter you roll for an encounter Power, which might be Turning, Rage, etc. Each day you roll a power, which might be Sneak Attack, spells, etc. And many such tables would have a blank that is filled in unique to this particular character. So sometimes this particular character starts battles with Stoneskin active, whereas sometimes that one knows a martial stance.