Time attacks and HOW much did I lose stats?

[Derakon]

ToME 2x has an option "stats are represented in a linear way" that basically does what fizzix proposed. I don't think it changed any game mechanics, though; stats still behave the same way behind the scenes (so stat gains after 18 are mysteriously much more useful).

[Timo Pietilä]

I don't have any knowledge of the current D&D systems, but I think if there's any motivation to make headway on this, the first solution is to make a system that still is able to use everything in tables.c

So I'd propose going from 5-40. 5 would correspond to 3, 20 corresponds to 18, 30 corresponds to 18/100 and 40 corresponds to 18/200. Starting stats are at 12.

This would essentially be the same results as currently exist, except everything is shifted by 2. (moving from 3-38 seems arbitrary)

Max effective stat is 18/220 not 18/200, so starting from 3 and ending up to 40 with baseline at 12 is exact copy of current system for tables.c point of view. No need for shifting. You could show over 40 values with ** instead of 18/***.

Something like time attacks and stat-potion fractional increases to XXX part of the stat would need rework to compensate that change.

Also I think there should be "bell curve" effect on how stats affect things. At both ends effect should be small, and in the middle effect should be big one. That would make maximizing your stats less tempting and "stat-gain" less important part of the game.

[Magnate]

Max effective stat is 18/220 not 18/200, so starting from 3 and ending up to 40 with baseline at 12 is exact copy of current system for tables.c point of view. No need for shifting. You could show over 40 values with ** instead of 18/***.

Something like time attacks and stat-potion fractional increases to XXX part of the stat would need rework to compensate that change.

Also I think there should be "bell curve" effect on how stats affect things. At both ends effect should be small, and in the middle effect should be big one. That would make maximizing your stats less tempting and "stat-gain" less important part of the game.

I share Timo's wish for non-linear gains from stat intervals. Something like:

-0.25 per point below 8
-0.5 per point between 8 and 12
+1 per point between 15 and 20
+0.5 per point between 21 and 30
+0.25 per point above 30

That would be a very significant change, but rewriting tables.c is among the easier tasks on the 327 outstanding tickets ...

[bebo]

Also I think there should be "bell curve" effect on how stats affect things. At both ends effect should be small, and in the middle effect should be big one. That would make maximizing your stats less tempting and "stat-gain" less important part of the game.

Excellent Idea!

[Hariolor]

excellent idea!

+1 .

[Tiburon Silverflame]

fizzix is right: the ONLY reason for the 3-18 scale is because you rolled 3d6 for stats in 1st edition D&D. Everything else...like "common animals have an Int of 1 or 2" was an afterthought.

Bell-curve is probably a bad idea. Flat-curve is Good in the sense that the +4 =Int means the same to you regardless of your current rating...and so does -4 from whatever kind of attack. Also, with bell curve you just make DAMN sure, with a caster for example, that you max out Int or Wis, and it gives a huge advantage to optimal race/class combinations by pushing you so far into the high-benefit section of the bell curve. It's also being done at relatively low cost, because the very low Str isn't costing you much more than it does now.

That said: not all stats behave the same. When they're per level, which is to say the mana table (int or wis) and the hit point table (con)...I think those need to be flat curves. That doesn't have to hold true elsewhere. Even here, the stats don't behave the same. The impact from +5 damage for Str != +5 AC from Dex. That argues against a pure 3E-style one-size-fits-all solution of using the *same curve* for all stats and all applications. One could have a staged curve for, say, Dex bonus to AC...if 10's the baseline +0, then a table like:

3-5 == -2
6-8 == -1
9-11 == +0
12 == +1
14 == +2
16 == +3
18 == +4
20 == +5
22 == +6
24 == +7
26 == +8
29 == +9
32 == +10
36 == +11
40 == +12

One pretty commonly hears that bonuses to AC are worth less, which is why I'm making this a broader scale; that +12 won't affect balance all that much. I actually like that you can get a small AC boost early on.

[PowerDiver]

I like flat curves for all stats, with one exception. Let stats range from 3 to 38, but count 38 as 40. That gives real incentive to increase from 37 to 38, the way it is now. RDH used to argue loudly in favor of breakpoints, and while I disagree in general I do find them interesting at the top.

So for something like mana which tops out at 8 per level, I would make it INT/5 except that it would be 8 for INT 38 or above.

[Hariolor]

fizzix is right: the ONLY reason for the 3-18 scale is because you rolled 3d6 for stats in 1st edition D&D. Everything else...like "common animals have an Int of 1 or 2" was an afterthought.

Yep, and the 18/percentile was, I suspect, merely a way for Gary to implement some degree of statistical control over stats that advanced beyond 18. It was *his* solution, but hardly the best one...this is the guy that invented THAC0 as well, you know?

*snip* Also, with bell curve you just make DAMN sure, with a caster for example, that you max out Int or Wis, and it gives a huge advantage to optimal race/class combinations by pushing you so far into the high-benefit section of the bell curve. *snip*

I am not sure I follow this argument. A bell curve I took to mean a bell to the interval of increase to the bonus conferred by the stat, thus the lowest and highest ends would be less critical to achieve. Thus on a scale from 3-38, an increase from 3 to 4, or from 37 to 38, would improve your bonus(es) by +1, while the change from 20-21 would improve the bonus by maybe (+5).

Thus there is a big incentive to get a stat to the midpoint, but there's diminishing returns once you cross the center point...this makes a lot of sense to me from a game mechanics perspective.

The amusing flipside is that even creatures with absurdly high values, say STR=80, would only be slightly stronger than a creature with STR=40...but alas.

[Zikke]

I thought the reason was that you figured out the stats by rolling 3 six-sided dice. I mean, everything you said is true, but it could also be true by replacing 3 with 5, right?

Yeah you're right also. The original way of rolling D&D characters was 3d6 per stat, which naturally led to 3 to 18 ranges. One of my DMs was evil and made us roll 1d20 and discard 19s and 20s... >_<

[buzzkill]

Yeah you're right also. The original way of rolling D&D characters was 3d6 per stat, which naturally led to 3 to 18 ranges. One of my DMs was evil and made us roll 1d20 and discard 19s and 20s... >_<

A truly evil DM would have had those 19's and 20's roll over.

I seem to recall a humorous poem from back in the day. I probably read it in either a Dungeon or a Dragon magazine. I believe the title was something like "Egad, what a cheap referee". Anybody know this? Can anyone find it? My google search turned up nada.