how bad would this mess up the game?

[gr8artist]

So, here's my thoughts. It rarely seems to fit, thematically, when a build includes a lot of multiclassing for dips and low-level abilities. I've always imagined that every level was one step closer to your desired outcome. But, many of the cross-role PrC's practically require skipping back and forth between drastically different classes.
So what if we consolidated the whole build into one class, a custom class? Would this throw game mechanics way out of whack?

To build such a custom class, we take a look at the end results of the desired build tree. For an example, lets go with a rogue 10/wizard 10.
At level 20, he'll have SA 10d6, he could cast up to 5th level spells, and he'd have 80+20+(20*Int) skill ranks (100 for ease of math). He'd have 10d4+10d6 HP and a BAB of +12. His base saves would be +6/+10/+10
So, when building the custom class, we round each of these values to one that averages out to be the same. A d5 HP/level hit-die would work, but unless we want to use d6(reroll 6's), this is irritating. So we bump this up to d6. Since we rounded that up, we'll round his BAB down to the caster progression (1/2 levels). His saves would round up to poor/good/good so we'll bump his 5 skill ranks/level down to 4. We'll use the bard casting progression, letting him get up to 6th level wizard spells, and he'll have 1d6 sneak attack at every even level. Since his spells got bumped up from 5 to 6th, we'll give him fewer class skills than the average. Scatter the bonus feats and special stuff (like evasion) into the dead levels (ones where he doesn't get a lot more spells or SA) but make sure everything comes at a level later than its original. And viola'! We have the Shivmage.
Hit Die: d6, Skill ranks/level: 4+int mod, BAB 1/2 levels, Fort poor, Ref good, Will good, Spells/day as a bard, SA 1d6/2 levels, and not a lot of class skills. Most additional class features of a 10th level rogue, 10th level wizard.

So... thoughts?

[ZamielVanWeber]

A bunch of the "mixed" classes will cry at this, but I do approve of the idea. You are giving magic (because people will almost always pick magic) to classes that lack it without the "restart your character" syndrome

[gr8artist]

the system obviously stutters at multiclassing caster classes. Just think how many spells/day a sorc/oracle gets at first level. They'd both advance slowly (bard style) but blast it all you'd be casting for days.
For days.

[Grod_The_Giant]

the system obviously stutters at multiclassing caster classes. Just think how many spells/day a sorc/oracle gets at first level. They'd both advance slowly (bard style) but blast it all you'd be casting for days.
For days.
Which is the point of a theurge build, isn't it?

I kind of like the idea. Although you're going to have to take dual-advancement PrCs into account. Your hypothetical Rogue/Wizard, for example, isn't going to be a Rogue/Wizard-- he'll be a Rouge 3/Wizard 5/Arcane Trickster 10/Something Else 3, with 7d6 sneak attack and probably 9th level spells.

[kestrel404]

This is a neat concept, but kind of tough to implement. You've basically got 2 choices:
You can front-load the work and make a new 'character building' system where all the various class features can be taken 'a la carte' and scaled to fit the 20th level build;
Or, you back-end the work, offloading it to the player/GM so that when they make a new character they are required to create their whole 20th level build in advance, then have the GM distribute the class features and abilities appropriately.

It's a neat idea, but both methods involve a LOT of work. It would probably be simpler to come up with an entirely new (non-class-based) system than do this.

[gr8artist]

Well, I realize that it has some flaws, but I think most people have a good idea what they want when they roll up a character. The only exception is when a new book is released and my group sees a cool new PrC they could qualify for. Other than that, most of the people I play with can tell you what a 20th level version of their character can do. They know what classes they'll take and when, and everything is planned accordingly.
If someone is taking a more natural character growth pattern, taking each new level in whatever class would benefit them at that time, then this wouldn't work for them.
But for the people who look at the Arcane Trickster and think "I want to play that" but they look at the multiclassing and missed levels, odd save calculations, BAB hiccups, and odd skill rank distribution, and they cry... This solution would work well for them.
It also eliminates the "I'll take class X for my first level for extra ranks (exploiting the x4 at 1st level) or health (starting as a barb for the max'd d12)" rules exploits. No longer would the character go a short time learning all the skills, only to spend a 5th of his career doing nothing but practicing magic. Now he does it all the same.

Like I said, it's a subjective, flavorful fix to make odd character builds make more sense. It makes character growth linear, direct, and rational instead of erratic, irrational, and convoluted.

[Grod_The_Giant]

But for the people who look at the Arcane Trickster and think "I want to play that" but they look at the multiclassing and missed levels, odd save calculations, BAB hiccups, and odd skill rank distribution, and they cry... This solution would work well for them.
It sounds nice, sure, but the issue is that a Rogue 5/Wizard 5/Arcane Trickster 10 is going to be much more capable than your hypothetical "Rogue/Wizard 20"

[gr8artist]

Why do you say that? If using my combination class system, he'd have roughly the same stuff as the same build using a multiclass system. It's hardly fair to compare an optimized 3 class build to a 2 class build.

The Rog/Wiz/AT build would be treated the same way as another multiclass set-up. Total the SA from Rog and AT, and distribute evenly among levels. Find the highest level spell that a wiz 5/AT 10 could cast, they get that at about 16/17. Class features come along at about the same time, and they'd have the average health and skill ranks that a multiclass character would.
The system simplifies character progression. Instead of going A>B>C, you figure out what the total result would be and stretch it to fit 20 levels.

[Yogibear41]

Seems like a lot of work to come out at the same outcome as just normal muticlassing. If from what I understand your primary concern is that someone being a 5th level wizard or something then taking a 1st level of rogue to earn some stealthy abilities doesn't seem to fit roleplaying wise, when all along they should just have been a Nightblade 6 (seems like a fitting name) or something?
(Or perhaps more fittingly rogue3/wizard 3)

If I am missing your point somehow please forgive me, and feel free to correct me.

[Kristinn]

I was reading this, and I couldn't figure out the purpose of this exercise. It seems as though it accomplishes the very same thing as multiclassing does anyway. Do you want to make it less powerful to dip for goodies? Or are you trying to make it more convenient in some way?

[Hand_of_Vecna]

Its an interesting enough idea, but why not just use the partial BAB/saves rules and maybe allow people balanced homebrew base classes before entering their distinctive build?

Basically, when homebrewing I like to stick to tiny tweeks or make something totally new and this falls too quarly in between for my taste. On the other hand I've done something similar for players that were adverse to book diving and playing something with four classes; writing up a a 20 level progression with classes, feats and skill ranks decided in a format similar to a base class entry.

[Crake]

there was a post on this forum not too long ago that did much the same what you're trying to achieve here, but I didn't bookmark it, I'll see if I can find it again.

Edit: Found it here (http://www.giantitp.com/forums/showthread.php?t=293875)

[Studoku]

Have you considered gestalt? It's significantly higher powered than this fix but it's straightforward and allows players to play two classes at once without faffing around with PrCs.

[gr8artist]

I have never played Gestalt, but I know the rules, and you're right about it being a lot stronger.
My idea here was just as Yogibear said, it didn't make sense to me for characters to change careers partway through their character growth, or get all of a single type of benefit front-(or back)-loaded. (Such as a character that takes all the Rog levels he'll ever need first, before anything else. All of his character's sneak attack is front loaded.)
The blended-class character shouldn't be noticeably weaker or stronger at the end than his multiclass counterpart. The only difference is in the distribution of his features and the uniformity of his BAB/HD/Saves.

[erikun]

For an example, lets go with a rogue 10/wizard 10.
At level 20, he'll have SA 10d6...
A Rogue 10/Wizard 10 will have +5d6 Sneak Attack.

Besides that, I sort of like this idea. The character can level up skills to max level, rather than trying to get them cross-classed or only when taking Rogue. The character can cast spells as a full spellcaster, rather than being stuck at CL 10.

However, I would recommend some "in-between" charts for BAB and saves. Your character who is supposed to be interested in sneak-attacking is stuck with a terrible poor BAB, rather than the 3/8 they probably should have. Similarly, you'd want to work out how a prepared caster would use the Bard's spells-per-day list, because if a Rogue/Wizard and Rogue/Sorcerer have the same number of spells per day, then there is little reason to go with the Sorcerer side.

Heck, you might be best just to create a spells/day chart just for use for gish-casters that specifically goes to 5th level (based off the Adept, perhaps?), and one with increased spells/day for people interested in theurging.

[Roog]

Iit didn't make sense to me for characters to change careers partway through their character growth

Why does it not make sense to you?