I think I finally cracked to code. Or, at least, came up with some numbers. Even though I bet somebody somewhere has suggested this before (there is always Delta and Hack/Slash, for example).
My first problem was with MAGIC - I like to use a non-vancian system with spell failures and fumbles, but also don't want to throw away the huge lists of spells we already have.
|Aaron Miller - Right-click for source.|
Usually, I would use half the wizards' level for any "Arcana" roll, which is good but a bit inelegant. It also makes the Intelligence modifier TOO important if you add it to the roll; for example, a 12 level MU could cast 6 level spells, but with Intelligence 18 he would be able to cast 9th level spells. A bit too much for my tastes.
Here is what I came up with to fix this, up to a point:
Spell DC is 10 + Spell level x 2.
Easier than it sounds, really. First level spells are DC 12, second level spells are DC 14, and so on:
12 - 14 - 16 - 18 - 20 - 22 - 24 - 26 - 28
But will people be able to multiply by 2 and add 10 effortlessly during the game? I bet they will, but it would be nice if we had an easier way to deal with that... let us say, some kind of simple chart, something with old-school cred... Hum...
No more spell books, memorization, spell slots, etc. Just roll, add wizard's level and INT, and meet the DC on the character sheet.
Failing by 10 or more is a spell catastrophe, just because I like spell failures! Also, succeeding by 10 or more is a crit. Keep this is mind because we can use it later.
Simple enough, right? The beautiful part is that it works well with ascending AC ranges I use, and even DCs in D&D 5e:
Numenera also uses an intuitive 1-30 range for targets numbers, where you basically rate the difficulty of the task from 1 to 10 and multiply it by 3.
Now, if I want to make a B/X-like system, compatibly is a big deal, because I want to play old modules without conversions.
So let us consider SKILLS. Most thieves' skills get about 5% better each level at least until level 10. The "X-in-6" solution found is LotFP and other games (see a simple method here) seems to be a popular method of skill resolution. I want to come up with a system that is compatible with this, AND still uses something similar to B/X.
Now, assuming Medium as the to-go difficulty (makes sense, right?), and adding Thief level to all skills, you'd have 35% chance of succeeding in the first level.
Guess what, that is about 2-in-6.
At level 14, you'd basically have 100% chance of success. Not unlike B/X.
B/X says there is a 5% penalty for pick pocketing victims with level greater than 5. So this would mean DC for level 5 or less, DC 16 for level 6, 17 for level 7.... Easy, huh?
I like to add ability score into the mix, which makes thieves a bit more powerful and lets other character participate more meaningfully too.
|nJoo - Right-click for source.|
The best solution I have seem for this is the NWP mechanic: get a +3 bonus (in the d20 roll) for each point you invest in any given skill. The skill list is consolidated in something like 4e, 5e or LotFP (hiding and moving silently become a single skill, for example); thieves get 2 points per level.
This also translates the X-in-6 mechanic to a d20, but you must use DC 20 to do that. For example, a 2-in-6 chance would become a +6 (2 x 3) bonus to the roll and thus a 35% chance. A 6-in-6 chance would become a +18 bonus and thus successful 95% (not unlike LotFP where a 6-in-6 skills means succeeding about 97% of the time, in, rolling 6 twice in a row).
Of course, the X-in-6 mechanics assumes an untrained chance of 1-in-6; if you take that out of the equation, you can use DC 15 and make things a bit easier for amateurs, giving automatic successes for level 14 thieves or other "experts" (the "6-in-6" guys).
If you don't want skills, you can use fractions instead.
Of course, having that "99%" chance can be more fun than always succeeding, or failing 5% of the time. One option of "extending" the d20 roll is adding 1d6 (count sixes az zeroes to be exact) to the roll if you get natural 20, or subtracting it from the roll if you get a natural 1. This way, you get 30 possible results (from -4 to 25).
What about COMBAT? Not much to add there. Using Fighter level as the BAB works well for me; the B/X Fighter can certainly use the boost, and, in any case, the B/X Fighter needs this to at least begin to compare with the RC, BECMI or AD&D Fighter. Here are some other ideas.
Just remember that, in this system, combat and magic are NOT skills; they are three times more "valuable" than skills, in a sense, since every point you spent in skills is a +3 bonus (in a d20 roll).
TURN UNDEAD? Of course, this works better with 2d6 if we want to emulate B/X. If you want to use the d20 for that (and now I kind do), just set the DC at 10 plus undead's level. If the margin of success in 10 or more, the undead is destroyed (told you I was going to use it!). It makes TU a bit under-powered (compared to B/X) but, hey, the Cleric is a bit overpowered anyway. Maybe just make it an skill (useful por faithful laymen!) or spell.
But I LIKED your point about 2d6 for everything... Well, no problem, really - 2d6 for skills works well too, just use it instead on 1-in-X. Turn undead becomes a skill, etc. But right now I intend to use 2d6 only for the GM. Reactions, events, morale, weather, etc.
Well, this is what I got for now. Sounds good to me. Who knows, I might even stick with it for more than a few months!
In any case, let me know what you think!