Brainstorm 3.x Multiclassing
- Thread starter Najo
- Start date
Aus_Snow
First Post
You're right of course, Runestar - and personally, this is one of the features that first impressed me about 3e.
However, many people seem to find, or have found, that the balance is lacking, especially when diluting any given full casting class (either with another full casting class or anything else.)
'Patches' like Practised Caster are there because of these issues, but they only go part-way to solving them. IMO, anyway.
However, many people seem to find, or have found, that the balance is lacking, especially when diluting any given full casting class (either with another full casting class or anything else.)
'Patches' like Practised Caster are there because of these issues, but they only go part-way to solving them. IMO, anyway.
Jeff Wilder
First Post
If you really want to do this -- and I for one would find it interesting -- then there's no sense in reinventing the wheel. You'll want to do some research on some sites (I'd suggest Paizo, RPG.net, and here if you've got search permissions) and present the most common solutions and the arguments for and against them. That'll give a much more viable starting point.
For instance, I think fractional saves, BAB, and CLP (caster level progression, works like BAB) seem like they'd all be part of the solution, but I've read posts alluding to problems with all of those. I couldn't tell you what those problems are perceived to be, so I can't make any judgments as to whether the objections are valid, or look for ways around them if so.
Maybe multiclassing ought to provide full abilities from both classes (i.e., wizards in plate armor with no arcane spell failure), but at a cost similar to ECL, perhaps scaling upward with level? E.g., you wouldn't be a "wizard 1 / fighter 1", you'd be a 1st level fighter/wizard, and (due to ECL) a 2nd level PC. As a 5th level fighter/wizard, perhaps you'd be ECL +2, and so on.
This is obviously somewhat similar to how multiclassing was handled in 1E, and in practice would work out somewhat similarly to the hybrid prestige classes like mystic theurge. There are obvious problems with the concept (what about three classes, or four?) and it reduces flexibility some (no such thing as "mostly a fighter/wizard with a splash of rogue;" you're just a fighter/wizard/rogue), but the truth is that the hybrid classes like mystic theurge are really pretty well balanced, and something like this would allow similar results with any classes.
Anyway, I really do think the place to start would be with summarizing ideas that have been presented, supported, and/or shot down. You'll get much more response if you do that.
For instance, I think fractional saves, BAB, and CLP (caster level progression, works like BAB) seem like they'd all be part of the solution, but I've read posts alluding to problems with all of those. I couldn't tell you what those problems are perceived to be, so I can't make any judgments as to whether the objections are valid, or look for ways around them if so.
Maybe multiclassing ought to provide full abilities from both classes (i.e., wizards in plate armor with no arcane spell failure), but at a cost similar to ECL, perhaps scaling upward with level? E.g., you wouldn't be a "wizard 1 / fighter 1", you'd be a 1st level fighter/wizard, and (due to ECL) a 2nd level PC. As a 5th level fighter/wizard, perhaps you'd be ECL +2, and so on.
This is obviously somewhat similar to how multiclassing was handled in 1E, and in practice would work out somewhat similarly to the hybrid prestige classes like mystic theurge. There are obvious problems with the concept (what about three classes, or four?) and it reduces flexibility some (no such thing as "mostly a fighter/wizard with a splash of rogue;" you're just a fighter/wizard/rogue), but the truth is that the hybrid classes like mystic theurge are really pretty well balanced, and something like this would allow similar results with any classes.
Anyway, I really do think the place to start would be with summarizing ideas that have been presented, supported, and/or shot down. You'll get much more response if you do that.
Well, the main issues are this:
1. Caster levels not stacking making spell caster spells weaker and weaker as the character levels.
2. Dipping into a class for 1 or 2 levels to create broken combos.
3. Different results coming from the order you take two different classes in. For example, going rogue 1/fighter1 verses fighter 1/rogue 1 is not the same character.
4. Saving throws being min/maxed by clever multiclassing so all of your base saves by level 20 are +12/12/12 or greater. Even using multiclassing to get +14 on some.
5. Overcomplicating characters with dabbling in lots of classes or taking 1 level in every class.
6. Class features not progressing properily to be useful at high level (i.e. turning undead)
These are some of the major issues with the current rules.
My favorite solution is Generic Classes (not quite the same as the ones in UA, but the same idea). But I don't think you're asking about a change quite that drastic ...
In other ideas, I've been quite impressed with how well feats like "Daring Outlaw" and "Swift Hunter" can work if they're well-designed. Of course the major problem with these is that they haven't yet been well-designed for the purpose of combining caster and non-caster classes (or multiple caster classes). Still, I think a lot more good feats of this style could be made.
In other ideas, I've been quite impressed with how well feats like "Daring Outlaw" and "Swift Hunter" can work if they're well-designed. Of course the major problem with these is that they haven't yet been well-designed for the purpose of combining caster and non-caster classes (or multiple caster classes). Still, I think a lot more good feats of this style could be made.
4 can be resolved using the fractional bab/save rules in UA (which also resolves the problem of 3/4 bab classes haemorraghing their bab inadvertantly unless they stopped at 4 lvs of each.
I am fine with 2 and 5 though. To me, it is more of something that DMs should adjudicate accordingly to the requirements of their game, rather than a limitation hard-coded into the game's mechanics. Multiclassing does not always produce broken characters, nor do you necessarily need to multiclass to obtain a powerful PC. Druid20 is the best example.
Some may be fine allowing a dip in barb for pounce, arguing that it is a small step towards keeping melees in line with spellcasters (and others less so). The best solution, IMO, would be to have the DM vet through everyone's build prior to running the game and ensure that everything is in order. If you are unhappy with his barb/fighter/bear warrior/warshaper/frenzied berserker build which you feel will upset game balance, then have it rebuild it. If you think the party can handle it, allow it by all means.
I don't think there is a clear 1-size-fits-all solution here, nor would one be desirable.
[sblock=Example
![:]](http://www.enworld.org/forum/images/smilies/devious.png "Devious :]")
A long while back I went to the Character Optimization board with a build request, and I ended up with a character build with eight classes -- two base, six prestige. My request wasn't for the most power available short of Pun-Pun, or to heavily focus on one thing so as to be unbeatable in that area, nor was I just looking for the most multiclassed build possible. My request was more along the lines of "My werebear barbarian PC and his redeemed succubus wife have children, how can I assemble a character that demonstrates the capabilities of that mixed heritage, for when the kids grow up into playable characters?" After a lot of discussion a build was worked out: Bard 4/Barbarian 1/Spellsword 1/Dragonslayer 1/Rage Mage 2/Bear Warrior 1/Sublime Chord 1/Eldritch Knight 9. This grants BAB +17/+12/+7/+2, Charisma-based spontaneous casting of a small number of level 1-9 spells on the bard and sorcerer/wizard spell lists, a limited ability to ignore arcane spell failure, limited bardic knowlege, rage, the ability to cast spells while raging, and the ability to transform into a bear while raging, as well as some other minor abilities that aren't so important.You might look at that class combination and cringe, thinking "What an overcomplicated, unfocused, dipped mess of a class collection." I look at it and think "This is Rachel Lovato, an energetic and outgoing young woman who's a capable warrior. She can draw on the bestial strength passed down from her father David, the compelling presence and magical talents of Seneca her mother, and the fiery temper she inherited from both. She prefers to get by with her cunning and charm, but she's a bit of a tomboy and likes a good brawl more than is proper for a lady. She hasn't seen as much of the multiverse as her parents, but she learned a little about everything from the stories they told her as a child, and it serves Rachel well in her own adventures." (I also think that from a raw power perspective, a single-classed Druid 20 would defeat Rachel easily.)[/sblock]
While I agree somewhat with 1 and 6, I am not sure if it should not be the case. Why should a cleric who takes lvs in fighter be better at casting spells or turning undead?
Jeff Wilder
First Post
There are two separate caster level issues: (1) The power of spells you can cast, and (2) the ability to cast higher level spells.
The problem with caster/caster or caster/non-caster multiclassing is that (2) is already easily enough of a price to pay for the versatility you gain. Penalizing the multiclasser with (1) is pushing the combination into ... well, I won't say "unplayability," but certainly into a markedly lower competence level. If a DM's or player's concern is some level of power parity between players, multiclassing sucks.
(For some DMs and players, that won't matter. either due to player personality or due to the nature of the campaign. But for some it will, and IMO it's understandable. Most people IME don't want to their PCs to be gimped compared to other PCs.)
In RAW, BAB -- "skill at arms" -- is already a function of experience. Even the most cloistered wizard gets better at fighting as he gains levels. Adding in a similar mechanic for caster level is nicely symmetrical ... one gets better at casting the spells one can already cast with experience, regardless of which class one is pursuing when the experience is gained. You don't learn new, more powerful spells ... you just learn to use the ones you already know with greater proficiency.
Or, to put it another way, a fighter who switches to wizard continues to get better at fighting (albeit slowly), why can't a wizard who switches to fighter get better at casting, even if slowly?
4 can be resolved using the fractional bab/save rules in UA (which also resolves the problem of 3/4 bab classes haemorraghing their bab inadvertantly unless they stopped at 4 lvs of each.
I am fine with 2 and 5 though. To me, it is more of something that DMs should adjudicate accordingly to the requirements of their game, rather than a limitation hard-coded into the game's mechanics. Multiclassing does not always produce broken characters, nor do you necessarily need to multiclass to obtain a powerful PC. Druid20 is the best example.
Some may be fine allowing a dip in barb for pounce, arguing that it is a small step towards keeping melees in line with spellcasters (and others less so). The best solution, IMO, would be to have the DM vet through everyone's build prior to running the game and ensure that everything is in order. If you are unhappy with his barb/fighter/bear warrior/warshaper/frenzied berserker build which you feel will upset game balance, then have it rebuild it. If you think the party can handle it, allow it by all means.
I don't think there is a clear 1-size-fits-all solution here, nor would one be desirable.
[sblock=Example
You might look at that class combination and cringe, thinking "What an overcomplicated, unfocused, dipped mess of a class collection." I look at it and think "This is Rachel Lovato, an energetic and outgoing young woman who's a capable warrior. She can draw on the bestial strength passed down from her father David, the compelling presence and magical talents of Seneca her mother, and the fiery temper she inherited from both. She prefers to get by with her cunning and charm, but she's a bit of a tomboy and likes a good brawl more than is proper for a lady. She hasn't seen as much of the multiverse as her parents, but she learned a little about everything from the stories they told her as a child, and it serves Rachel well in her own adventures." (I also think that from a raw power perspective, a single-classed Druid 20 would defeat Rachel easily.)[/sblock]
While I agree somewhat with 1 and 6, I am not sure if it should not be the case. Why should a cleric who takes lvs in fighter be better at casting spells or turning undead?
I think removing the need for DM's approval as much as possible takes away punishing creative players. It also makes less work for a DM. A DM should only have to worry about the concept, backstory and goals of the character, not the mechanics so much.
Likewise, I think character optimization is a good thing and building synergy into a character should be rewarded. But all of the choices for a player should be comparable, with a choice leading the character down a different path, not some choices obviously better than others.
As for multiclass spell casters needing to be effective, that is the heart of the issues with 3.x math and something 4e is attempting to fix. When a spell caster doesn't gain caster level, the few offensive spells they do have become useless at high level. This is the reason prestige "band-aids" like the mystic theurge were created, to counter the math imbalance. In 4e, they removed it all together by leaving character level out of the multi class issue.
Jeff brings up the point I was refering to, he says it much clearer though. The solutions I've seen suggested for this thing is a) caster level = character level or b) have a caster level BAB in every class. Fighters, Barbarians etc are +1 every other level. Paladins and Bards are +3/4 every level and Clerics and Wizards +1 every level.
In both cases, the level cleric 10/ wizard 10 casts as a 20th level character but has the spell selection of a level 10 cleric and a level 10 wizard.
In both cases, the level cleric 10/ wizard 10 casts as a 20th level character but has the spell selection of a level 10 cleric and a level 10 wizard.
