CON and HP at near-max? (200 vs ***)

[Chud]

Where does CON stop adding to HP? I currently have my natural CON maxed, but changing equipment to raise it from 18/200 to 18/*** makes no difference in HP.

In general is 18/200 where benefits generally stop, or are there other factors that affect the calculation?

[Zyphyr]

18/200 is where the bonus maxes out (at 12.5/level)

[fph]

I find it odd that some things max out at 18/200 and some at 18/220, without any particular reason. For instance IIRC the last two points of wisdom add no mana points for priests nor reduce spell fail, but do affect the saving throw. Moreover, IIRC a couple of stats simply max out at 18/200 for all practical purposes (so why not setting them to /*** starting from this threshold?)

[takkaria]

I find it odd that some things max out at 18/200 and some at 18/220, without any particular reason. For instance IIRC the last two points of wisdom add no mana points for priests nor reduce spell fail, but do affect the saving throw. Moreover, IIRC a couple of stats simply max out at 18/200 for all practical purposes (so why not setting them to /*** starting from this threshold?)

I think we should really be looking at making the effects of stats more linear all round, and then this wouldn't happen.

[PowerDiver]

I think we should really be looking at making the effects of stats more linear all round, and then this wouldn't happen.

R. Dan Henry(*) used to argue in favor of breakpoints. I still think the 18/10 is too big a deal, but there is something to them.

I've suggested before that things should be linear, but I like the idea that it is a 0 to 40 scale where you get the effect of 40 when you reach 38 (18/200), so it's worth juggling to go from 37 to 38. Given the min of 3, it would be more symmetric to make 37 mean full power, but so long as things are presented in percentiles max should be 18/200.

Linear stats is a huge change in terms of mana for low level casters, since full mana is 8/level.


(*) Is RDH posting here and I just don't recognize him?

[Hariolor]

A consistent and appealing powercurve doesn't have to be strictly linear.

bonus=x^2 (where stat value = x) gives a nice accelerating feeling of growth

bonus=x^2-x tones it back a little bit

bonus=sinx (over the first half radian) gives acceleration to mid-level, then gradual deceleration

I think any of these could be argued-for, and would be preferable to the opaque and inconsistent AD&D-style system that V currently-uses

[kaypy]

log or sqrt are common starting points, to model diminishing returns

[Antoine]

I think we should really be looking at making the effects of stats more linear all round, and then this wouldn't happen.

I actually would like to see stat effects being less linear at the bottom end. A stat of 3 should be much worse than a stat of 10 - at the moment there is not a lot of difference.

A.

[Nick]

(*) Is RDH posting here and I just don't recognize him?

Yes, but not recently, and no.

[PowerDiver]

I actually would like to see stat effects being less linear at the bottom end. A stat of 3 should be much worse than a stat of 10 - at the moment there is not a lot of difference.

Regarding linear, "I do not think it means what you think it means. " In a linear system, a str value of 3 would let you carry only 30% as much as a 10. A spellstat of 3 would give you only 30% of the mana. That's a lot of difference.

In fact, I think linear is too much difference for mana. For hp, max con approximately doubles your hp compared to minimum con. That seems reasonable to me for spellstat affecting sp as well.

[Derakon]

"Linear" doesn't presume a y-intercept of 0. E.g. your carrying capacity could be given as 15 + 5 * STR. Then STR 10 has a carrying capacity of 65, only about 2x what STR 3 gives (and STR 40 has a capacity of 215, which sounds more or less accurate).

[PowerDiver]

"Linear" doesn't presume a y-intercept of 0. E.g. your carrying capacity could be given as 15 + 5 * STR. Then STR 10 has a carrying capacity of 65, only about 2x what STR 3 gives (and STR 40 has a capacity of 215, which sounds more or less accurate).

I just checked, and was surprised to see that you are right on the mark. Capacity at which you are slowed is precisely 6 * (STR + 2) for STR all the way from 3 to 18. The other tables are not as bad as I remembered either. I was completely off the mark. I should have harped on DEX or CON instead, but perhaps I misremember those tables as well.

For the pedants: The definition of linear in some circles requires that 0 has to go to 0, and what you are talking about is defined as affine. I'd say the function is not a linear transformation on [3,18] even though I'd also say it has a linear growth rate. Feel free to laugh at me.

[Magnate]

I just checked, and was surprised to see that you are right on the mark. Capacity at which you are slowed is precisely 6 * (STR + 2) for STR all the way from 3 to 18. The other tables are not as bad as I remembered either. I was completely off the mark. I should have harped on DEX or CON instead, but perhaps I misremember those tables as well.

For the pedants: The definition of linear in some circles requires that 0 has to go to 0, and what you are talking about is defined as affine. I'd say the function is not a linear transformation on [3,18] even though I'd also say it has a linear growth rate. Feel free to laugh at me.

Thank you. I've always wanted to know the proper term for "linear with constant offset".

While I'm here, I discovered a neat trick while reading a ten-year-old thread on randarts (myshkin occasionally links to old threads on IRC for a laugh). It was in a post by Leon Marrick (author of O and modern S), responding to Greg Wooledge (author of the original randart generator) who was complaining that it was impossible to value the power of +speed properly (fancy that!), because linear was too low and quadratic was too high. Leon said use (a + x) / (b + x) and adjust x to achieve the desired degree of linearity. Interesting.

[Antoine]

In fact, I think linear is too much difference for mana. For hp, max con approximately doubles your hp compared to minimum con. That seems reasonable to me for spellstat affecting sp as well.

Whereas I think you should have virtually no SP if your spell stat is below (say) 6.

A.

[PowerDiver]

Capacity at which you are slowed is precisely 6 * (STR + 2) for STR all the way from 3 to 18.

I guess I'll try changing this some day to first slowing at 5 * STR for all values of STR from 3 up to 40. The main drawback is that heavy armor will become even less useful. Perhaps this idea belongs in one of the "make the game harder" threads.