Statistics (Str, Dex, Int) General Feedback

General principles

This is again from collaboration between evktalo and dpeg. A quick rundown of the current state (immediately before release of 0.6):

  • Int is relevant if and only if you're a caster.
  • Str is very relevant if you want to use heavy armours, although the effect is capped. Otherwise, everyone will value Str at least somewhat, because of the carrying capacity it provides. Str provides a little damage for some weapons, and accuracy for some (the heavier ones), but that seems to matter little.
  • Dex is relevant for EV purposes. Also for a myriad of other things, including stealth, avoiding traps etc. All of this is unclear and probably not very relevant.
  • All stats are relevant for stat death purposes, but we don't consider that to be interesting. Stat deaths should be dealt with as indicated on this wiki page, ideally for 0.7.

Here are our proposals:

  • Remove all effects on damage from stats. The reason is that damage is already governed by a plethora of parameters (weapon type, weapon enchantment, skill, stats, brand, racial prefix) and it is impossible to make all of these meaningful.
  • Either remove all effects of stats on accuracy or make them much more meaningful. (See below for how that could be done.)
  • Separate stat effects from skill effects as far as possible. In cases where we don't want full separation (e.g. Int's effect on spell hunger), it should be simple. Simple enough that it could be mentioned in the manual with a few words.
  • There are various moves (cleaving, stabbing etc.) which depend on the weapon (and perhaps some more). Chances for triggering these should not come from skills, but only from stats. (I.e. a very strong fighter could use his axe to cleave several nearby opponents, even if his axe skill is still lacking — that just means that each of those attacks does not as much damage as it could. Likewise, a very dexterous character gets more stabbing attempts — she may blunder them all because of no Stabbing skill. Note that in this example, an axe user may enjoy none, one or both moves, depending on Str and Dex.)

Int

Int is decently implemented right now, because it has an important role with spell hunger. However, let's look at the full list of magic-related effects:

  • Spell levels: only from XL and Spellcasting — this is good and clear
  • MP: the same
  • Spell hunger: - Int*Spc
  • Spell success: school skills, Int, Spellcasting
  • Spell power: school skills, Int, Spellcasting

The last two lines indicate that matters might be improved: all three (of schools, Spellcasting and Int) are factored in, and it's more difficult for the player to know which of the three to raise in order to build the character. There are various conceivable scenarios, here is one of them:

  • Spell success: school skills, Spellcasting
  • Spell power: school skills, Int

There are endless variations: We could take out Spellcasting out of the power formula, but have Spellcasting and/or Int provide a minimum power. Perhaps spell schools would help with success, but only if your Int wasn't high enough, and that help would increase the hunger costs if it was needed. Perhaps Int could keep helping power, but only to a certain point (i.e. a two-tier power cap); after that, you'd need strong spell school skills. On the other hand, the low bonus to the success and power of spells from Spellcasting fits within the “flexibility” idea.

keskitalo: Affinity with the schools could represent raw power. Spellcasting could represent flexibility (the ability to take on lots of spells, and having MP). Maybe Int could represent the ability to channel this power (success and hunger)… Not sure what the separation should be. One way to support this could be new spell staves, e.g. a staff that helps with casting success but increases hunger, or increases power but reduces success). In general though, Int already works okay… if we could introduce more interesting character development choices (and say, differentiate kobold casters from elf casters more), it might be worth looking into this.

keskitalo: It should be noted that in the current system, Spellcasting matters fairly little (clarification: for success and power), and Int act as a multiplier for success and power. Spellcasting giving a small boost for all the spells regardless of school does fit in the “flexibility” idea.

keskitalo: We could also add in miscast severity, so that you could build a caster that has a low success rate with mostly non-dangerous miscasts, but casts at high power when succeeding. Int could give MR (as suggested by b0rsuk below), and MR could mitigate miscasts. Thus, you could have high-Int, low-Spellcasting casters that could attempt casting high-level spells with low success but less harmful miscasts. Contrast with high-Spellcasting casters, that could output spells in a more stable manner, but would suffer more from failed attempts. Differentation between these types would probably be mostly done in the species level, but it would be great if you could steer say a human caster in either direction.

borsuk: Make MP come from Int. This makes it useful for non-casters who need MP.

squashmonster: MP from Int is good. I don't think Int should affect spell power. Spell power often determines damage, so if Str and Dex are being removed from damage calculations, it doesn't make sense for Int to matter for damage. Likewise, if the other stats end up getting removed from accuracy, Int should be removed from spell success. I would like to see Int end up determining “special moves” like is proposed for weapons: it could be one move per school to keep it simple-ish.

I don't think equating spell power with melee damage and spell success with accuracy is useful. Special moves for casting on the other hand sounds interesting. What sort of effects could there be? — evktalo 2010-04-06 21:00
Would “special moves” for casting really even work with one move per school? In transmutations, for example, how would the special effect of Alter Self apply to Polymorph Other, Blade Hands, and Fulsome Distillation (which would also somehow get necromancy specials)? Though, perhaps something…
* General to all spellcasting: such as an intelligence-dependent short-term memory effect boosting the success/power of successive castings, though I guess this encourages “rehearsing” castings before an encounter, which is probably bad, so a different “move” may be desired
* Grouped based on spell usage, or similar to spell power: say, with high intelligence you can “improve” spells to do nifty things, but this is probably too powerful, and too similar to spell power, so other “moves” would probably be better here too
…could work? — mrmistermonkey 2010-04-07 01:24
I think the best ideas for casting “special moves” don't have to be related to the spell being cast. For example, having an effect near you that applies while you're casting the spell works: earth can have increased AC, ice can slow enemies near you down, necromancy can have fear. Other good ideas are side-effects, for example conjurations could make clouds near you (perhaps ones opaque to enemies but not you, so it's always good), and fire could make additional fire clouds in empty spaces near the target. Others might have to be more related to the spell effects - most poison spells could benefit from duration, while with summoning one idea is to add a delay before summoned monsters react, and let summoning's special ability be lowering that (another idea is to add short-term bats to every summon). A school like transmutation is difficult, but note that most transmutations are either dual-school or form spells. Form spells could give you a generic stat boost while in that form; the dual-school ones don't need a special Transmutation effect. — squashmonster 2010-04-07 05:22

Int vs confusion

I agree with b0rsuk that Int should be noticeable not just for casters. One idea is that Int has an effect on not getting confused. Here is the current status:

  • Confusion spells by monsters are blocked with chance depending on player's MR.
  • If the confusion goes through, it is a plain duration (no further attempts to get rid of it).
  • Confusion cast by player can be blocked using the monster's HD.

Here is a somewhat specific proposal: Whenever the player is affected by confusion, check the player's Int. I am thinking of a non-flat chance to get rid of the confusion similar to this:

Int    1     5      10     15     20     30
 %     5%    10%    50%    75%    90%    99%

(If we want to, we can also have a cap on the duration.) greensnark says that high Int should not provide good chances to escape confusion; but that low Int should give confusion penalties (because confusion against players is currently rather too weak). This is a matter of numbers, of course. Note that if MR keeps trying to block confusion right away, as now (which is sensible; and is more useful against confusion spells than mephitic clouds), then there'd be two lines of protection — in other words, confusion would need to get stronger in return. E.g. longer durations.

For monsters, use HD as a substitute for intelligence. However, more could be done, and this is a chance to distinguish confusion-type spells. Some rough ideas:

Mephitic Cloud

  • power influences duration and/or size (not “strength” of cloud)
  • intelligent monster get a bonus to their HD
  • spellcasting monsters get a bonus of level-of-best-spell to their HD
  • non-confused monsters in the cloud could try to leave it

Confusing Touch

  • Works at most once per monster.
  • Has good chances. For example, there could be no boni to HD.
  • Very short duration.
  • These three points make it the stabber's confusing spell of choice. (Would still be useful in all melee, of course.)

Analysis of using Int here:

  • With the above, it suddenly would matter if your Troll has Int 3 or 7. If you don't want to invest into Int, then you could either use !healing generously, or dedicate yourself to “clarity.
  • The Mephitic Cloud change aims to solve the obvious brokenness of the current spell…
  • …and to make potentially susceptible monster behave noticeably different (compare plain elf vs elf annihilator in Elf:7, for example).
  • greensnark mentions that the name “mephitic cloud” may not fit anymore after these changes. I don't really see why, but the name could be changed, of course.
That's a lot analysis for what I fear is little benefit. With Crawl as it currently is, would deffinitely avoid spending X points on Int just to shield against confusion. I would rather cross my fingers and look for Clarity, boost my MR or just plain be careful. I don't think confusion is widespread enough to be such a danger. If Intelligence boosted MR instead, I might consider improving Int of my character without spellcasting. You know how in folk tales and beliefs you can ofter counter a spell or curse with the right words ? Int could simulate that - characters with higher Int would be less likely to be affected, because they know the right words to recite. — b0rsuk 2010-03-27 06:23
I like the idea about Int helping against confusion, but I love the idea about Int giving magic resistance. (We could then remove the MR gains from the overloaded Enchantments skill.) Numbers? — evktalo 2010-03-30 19:26
Enchantments skill no longer directly gives MR - you need to wield the staff of enchantments to get the benefit from skill. This has been a really good change, probably more interesting for casters, but less interesting for non-casters than getting MR from Int, since a caster wants Int anyway, but a non-fighter can't choose to boost it. However, non-casters still have MR items so the situation might be pretty good. — evktalo 2010-08-28 09:01

Str

High strength should helps making heavy armours and heavy weapons useful. Does not provide additional damage, but you'll need a certain strength in order to make full use of the weapon (this should be transparent, i.e. we should mention this number in the weapon's or armour's description: “You will need a strength of 24 in order to make full use of this item.” The number itself may depend on size.) The drawback of this approach is that Str effects on armour/weapon are effectively always capped.

This has two side effects:
* races with very high Str but few armour slots don't get these benefits. Back to square one.
* 30 Str character deals about the same damage with mace as a 8 Str character would. This essentially makes race design item-centric. I like this as much as diversifying weapon categories by magic brands which might appear on them. Which is to say, not at all.
To elaborate, this design puts ogres on mercy of Random Number Generator. If you don't find a decent mace, midgame is getting impossibly hard. For ogres, early game is hard enough outside few specific combinations. Effectively you are not stronger until you find better items - maces or armour.b0rsuk 2010-03-27 06:23

However, Str should also help with certain weapon moves (cleaving for axes etc.). The idea is that the chance to get these increases (uncapped) with the stat. If all heavier weapons get access to Str-dependent moves, that would automatically make Str meaningful for melee fighters, beyond caps.

Str also helps with carrying capacity, of course. (But probably nobody would get Str beyond 30, say, for carrying capacity purposes.)

Perhaps the reduction to weapon delay could be capped at half your weighted strength or so? 7-delay exec would require about 32 strength at 8 dex. — tgw 2010-03-22 01:02

Personally, I'd like to see a a change of the way EV penalty is calculated with strength. Currently you need 3x the armour's EV penalty to offset the extra EV penalty, strength beyond that doesn't matter. I'd rather this were a soft cap, with potential to even LOWER the EV penalty with high strength. So the current AEVP would change from

(EVP + MAX(0, 3*EVP - Str)) * (45 - Armour) /45

to something like

MAX(0, (EVP + (4*EVP - Str)/3)) * (45 - Armour) /45 (Just a quick example)

This specific equation would make dragon armour a bit more difficult to wear for weaklings, but that's probably a good thing since you tend to want 9ish strength for encumbrance anyways, making it a moot decision. — greepish 2011-12-06 07:43

AEVP=(3evp^2)/str would be a good formula for eliminating the breakpoints. AEVP would increase/decrease according to strength but would never reach zero (as long as EVP>0) and heavy armour would be affected by strength more than light armour, so it would make strength always desirable for armour and more so for heavier armour. — some12fat2move 2013-16-02 18:17
I really like this idea. Link to tavern thread. Galefury made graphs to compare the new AEVP to the current one depending on base EVP and str both with a factor of 3 (some12fat2move's original proposal) and a factor of 3.5 (curve much closer to the current one). I think we should test this in trunk with a 3.5 factor as soon as possible. After 0.12 fork or maybe even before? — galehar 2013-03-28 11:58

Str and (base) damage

TGW's idea above is great, and b0rsuk's concern is valid. Below is my new idea.

Str requirement for heavy weapons is handled like this: You need Str equal to the base damage of a weapon to get the full benefit from it. If your Str is below the base damage, the weapon is treated like the base damage was your Str. That is, if your Str is 14, exec axe has a base damage of 14 (normally 20). Slaying bonus to damage also needs Str to get the benefit to apply: an exec axe and a +0,+8 slaying ring need Str of 28 to fully benefit from.

The above doesn't fundamentally change combat numbers, but only introduces a Str requirement. It is a slight nerf to melee overall.

Additionally (in line of b0rsuk's comments), the excess Str over base damage and slaying requirements becomes a simple +dX to damage, not affected by skill. That is, a character with the exec axe and +0,+8 slaying and 35 Str, gets a +d7 to damage rolls. With only a dagger, 35 Str would give +d31 to damage (4 base damage from dagger). This sounds big, but base damage (inc. slaying) is always better.

I suspect the +dX is problematic in the early game (too strong), but ok in the late game (not too strong, but pumping Str still gives a benefit). Str/Dex distinction for weapons would be removed. Str plainly gives damage, and Dex plainly gives accuracy (as below) for all sorts of melee weapons. No idea yet about unarmed combat and ranged weapons. This goes against the stated principle of stats not affecting damage, but I think it's worth considering. A boost to pure melee (high Str, not Int) wouldn't be a bad thing and hopefully this wouldn't also boost crusaders. — evktalo 2010-08-28 09:01

Effect of Strength on armour EVP

Formula

(2/5) * evp^2 / (str+3) instead of evp+max(0,3*evp-str)
The EVP has a bigger scale about *3 (GDA is 27). The penalty to dodging bonus is adjusted to use EVP/3 instead of EVP. The bigger scale should make thing easier by giving us more room to balance and differentiate armours.

Graph

Link

  • Brown is leather and WDA (-1 and -2)
  • Green is fire (and ice and pearl) dragon armour (-3)
  • Light blue is chain, splint and plate (-4, -5, -6)
  • In yellow are CPM and GDA (-8, -9)

Armour stat changes

  • Splint mail is removed
  • Chain gets +1 AC (7→8)
  • -1 → -4
  • -2 → -7
  • -3 → -11
  • -4 → -15
  • -5 → splint is removed, SDA -17?
  • -6 → -19
  • -8 → -24
  • -9 → -27
  • After some playtesting, we can adjust the dragon armours a bit

Comments

  • The new chain mail is a bit like a merge of the old splint mail and chain mail. We can keep splint instead, to avoid having to change any AC value, but I find the name a bit confusing. I never remember which is which between scale and splint, so I'd be happy to see one go.
  • Chain (7/-4), Splint (8/-5), Plate (10/-6) become Chain (8/-15) and Plate(10/-19).
  • It might be good to merge ring and scale too.

galehar 2013-04-18 15:56

Dex

Dex effect on EV to be different from Dodging

Dex always helped with EV, and this should stay, in our opinion. However, it could be modified:

  • The EV coming from Dex would depend on how many open squares are around you; the EV coming from Doding skill would apply everywhere.
  • This could be combined with the zig-zagging EV idea where you get an EV bonus if you're moving (unpredictably). This might make EV fighters really change their positioning around. If this was done, the EV displayed should always be the actual value — i.e. not some “base” EV number, but change according to your surroundings. (Or, display as base EV plus situational EV, e.g. EV: 14+4.)

Thus, both Dodging and Dex would give you EV, but in a different manner. Ideally, the player can make strategical (stat/skill development) and tactical (positioning) decisions about which way to go.

I suggest another solution: many situations could be approached either with Strength, or with Dexterity. A net trap falls on you. If you try to just move away, you attempt to shred the net, which is faster, but requires some Str to work well. Or you can try to disarm the trap, in which case it's slower, quieter, but you obtain an undamaged net. b0rsuk
Sure, this is what goes on with nets at the moment. held_and_restrained has more effects with Str, Dex or both helping you out of them. I'm thinking it's ok if stats also have a purpose of “helping you out of predicaments” (such as Int helping against confusion). Ultimately this is a small effect though, and I don't think anyone would boost Str specifically so that they get out of spider webs faster. It will (hopefully) make differently statted builds respond differently to threats (and perform against them differently), though. Maybe you even want a might potion or a Str ring in case you get caught in something. — evktalo 2010-08-28 09:01

No more Dex effect on Stealth

Dex should not have any effect on Stealth. First, Dex already directly interferes with one skill (Dodging). Next, Dex affecting Stealth (if it was meaningful) would doubly reward certain (small) species, who're heavy on Dex and good at Doding and Stealth anyway, and it would likewise doubly punish other (e.g. large) species.

I suggest to give large and small races Stealth aptitude of 100. Make Dexterity affect stealth in a significant way. Additionally, some static stealth modifiers would come from race (like centaurs having hooves), or from size (ogres, spriggans). But none of this “Ogres have bad stealth aptitude because they're big” bullshit. This is just bad for your head. If I play an ogre transmuter and cast a spell that shrinks me, now what ? What If I become a spider and my Dex skyrockets ? What if I morph my centaur and no longer have hooves ? — b0rsuk 2010-03-25 16:27

Stealth should be a function of species, armour, skill alone. There was an idea that Stealth could be higher if adjacent to walls — this could be done, but that's a completely separate question.

Accuracy as a function of Dex (and Str), not skill

Currently, accuracy of a weapon is determined by weapon type, first enchantment, Str/Dex (depending on the weapon), skill. Since weapon skill already gives damage and speed, it could be a good idea to determine accuracy in a different manner (which the player can choose to invest in). As a minor side effect, this would give potions of agility a nice perk.

I thought about something similar: Damage is no longer affected by skill, only by Str. Skill already affects accuracy (which I think it should) and weapon speed, which is effectively equivalent to damage, because monsters' toughness scales linearly with HD and general “danger”. There are unused niches among monsters, such as fragile monsters dealing a ton of damage. A couple of them exist in early game, but they become less and less common. b0rsuk

Looking at http://r1gm.free.fr/Crawl/crawlSS_052_weapons.txt, the Str/Dex distinction affects which stats are required for an accuracy bonus for melee weapons. Dex weapons only use Dex for the accuracy bonus, while Strength weapons use both equally. The idea behind this is good (Dex affects accuracy in general, Str helps you handle the heavier weapons), but to-hit doesn't seem to be considered to matter.

Whatever the mythical combat overhaul turns out to be, we do want it to make accuracy meaningful. So here's one approach:

  • To-hit is not influenced by skill anymore. (Skill continues to improve damage and speed, so it something you really want.)
  • To-hit is instead influenced by weapon type (one of the three magic numbers); first enchantment; Dex or Dex/Str (depending on weapon).

If doing this, we take the weapon skill out of the accuracy-design process (in a natural way which indicates interesting implications for gameplay). Of course, we still have to tweak numbers so that to-hit becomes important (but not too much so). This will be a problem of the magnitude of the AC change, I guess.

By now I think it should only consider Dex, never Str/Dex even for heavy weapons. — evktalo 2010-08-28 09:01

Dex as a playing style

The design goal is to enable the short blade/stabbing build without external crutches. For this, either of the following two proposals, or both could be employed.

  1. High Dex allows to stab enemies even under unfavourable conditions. (Here I am thinking of 0% stab chance for Dex 18, and perhaps 10% stab chance for Dex 36. The stabbing itself should not use Dex, that's what the skill is for.) One aim of this is to enable the stabber template without crutches like lucky slaying rings or Okawaru's Might (which is the reliable version of the former).
    In this sense, “stabbing” would be one of our new weapon moves. Something you can do with all weapons (as opposed to other moves which only work on special weapon types).
  2. Assuming that Dex governs accuracy (as proposed above) and there's an increased minimum damage for good to-hit rolls (as proposed below), then Dex and high accuracy could get a kind of “critical hit” potential/feel, without introducing actual critical hits (such as stabs).

High accuracy and mimimum damage: A good hit-to roll could increase the minimum outcome of the damage roll. The accurate hitter would have more guaranteed damage per hit, but the possible maximum damage would be the same. (This idea is a priori independent of stats.)

Less special cases (borsuk)

Let's discuss the effect of stats on players.

Currently Crawl uses a lot of special cases for effects like Might or Statue Form. These make game mechanics opaque, and potentially misleading:

I almost always use shortblades and I have been debating whether I should place all my stat points in strength or dexterity. 
I know shortblades say they are better for the dexterous, but the five strength from might potions seems to add a lot of damage.

http://forums.somethingawful.com/showthread.php?threadid=3155652&userid=52755

Actually, +5 Str from Might has neggligible effect on damage. What the player noticed was +1d10 (extra 1-10) hidden damage bonus. Similarly, many players try Statue Form and notice nothing special. It's not obvious that it deals 12 + Str base damage when fighting unarmed, especially that players in statue form can wield items. I think a better solution would be to link damage dealt in melee to Strenght, and temporarily modify Strenght. That way, player can see how big the benefit is without reaching for spoilers.

A list of effects which could be replaced with stat bonuses:

  • Might: hidden +1d10 damage (appears to only give +5 Str)
  • Statue form: Massive damage bonus to Unarmed Combat (Only +3 Str is visible, but uses special case for damage: base is 12+Str)
I'm pretty sure a strength bonus comparable to Statue Form's extra power would exceed 72 by very, very much. — tgw 2010-03-06 15:03
b0rsuk I agree stats need a bigger focus (besides Int), but assuming few other things change, your example limits heavy armour users from being effective with dex focused weapons (mostly ranged, small penalty to some melee). — vandal 2009-12-30 03:50
Some ranged weapons can favour Str more than Dex - Slings and especially Throwing come to mind. That way, stat training will have meaningful effect on gameplay. I don't agree with dpeg that killing an orc warrior with 5 shots, using highly enchanted ammo is overpowered. A conjurer of comparable level could probably kill 5 orc warriors with one shot. b0rsuk
Transparent damage has been needed across the board imo, spells especially hide this info, like how much damage does magic dart do in comparison to mystic blast if both are at 5 ticks? Very confusing to new players. — porkchop 2010-03-06 13:48
We're getting a little off-topic, but as a player I don't want to know if ogre deals 5-20 or 0-25 damage. I want to know how much damage can he deal relative to my health. How many hits, approximately, I can take before dying, and how many times I have to hit the enemy. I think the game should do that (display simple simulation of ods) at least for melee combat - would be very helpful for newbies. Large part of being good at Crawl is knowing when to fight and when to run. This is hard to determine without experience, thanks to scarcity of escape tools, teleportation delays etc. Even I (an experienced player) would benefit from it. I think relative strength of me and a monster gives more information than simple damage rolls could provide. b0rsuk
There won't be damage numbers listed, ever. We don't want players to agonise over long lists of pointless numbers. What I could imagine is a statement like this (for example in the monster's description): “If this ogre hits you twice on with the giant spiked club the head, you're probably dead.” — dpeg 2010-03-06 15:47

Stat Death

Currently, the player character dies instantly if any of their stats reaches 0. This can be surprising and extremely offputting to new players who are not used to having to look at their stats and who suddenly find a promising game snuffed out. We recently discussed this on ##crawl-dev, and came up with a rough set of proposals to eliminate stat drain instakill.

The general consensus is that stats drained to 0:

  • Should not kill immediately.
  • Should have immediate negative effects that make it undesirable to have the stat at 0.
  • Should lead to death if not addressed within a period of time, somewhat like the Starving effect.

In addition, we propose that the current stat recovery mechanism (where a drained stat is restored in one jump from say 1 (9) → 9) be changed to restore stats one point at a time.

  • Totally, yes. That has always bothered me. — jpeg 2010-01-07 18:47
  • Clarification: the potion of restore abilities will keep working as before. — dpeg 2010-01-07 19:01
  • Let me (dpeg) add some notes:
    • Quite a number of stat deaths (the majority?) come from players forgetting about their stats. Those deaths will not occur anymore and that's good.
    • It can happen that a player gets stat 0 and is unable to do much about it. Just like with starvation, she should be able to struggle for a bit but we should be generous; it shouldn't last too long. Note that with starvation, waiting is never a good idea — with stat 0 it could be. What to do?
    • The current colour coding of low stats should stay.
    • We don't want permanent effects. So not permanent stat loss etc.
    • We don't want to make 0 stat trivial, so even if you can counter by drinking a potion of restore ability, there should be some bad effects carried over. This should prevent a player ignoring stats completely, relying on !RA after the fact. The proposals for this include:
      • it is always undesirable to have a 0 stat with nearby monsters (e.g. the paralysis)
      • carried over Slow
      • rotting HP/MP

Getting to 0

The problem with stat death seems to hinge on insta-death. Stat loss could work like poison, accruing over time. This avoids insta-death for all but the most careless players who roam Zot with Int 1. — brendan 2010-04-04 13:16

Effects of a stat at 0

When any stat goes to 0, the player receives a strident warning message about it having negative effects. A stat going to zero should produce (immediately)

  • a short term paralysis effect, perhaps 1d3 turns, independent of the stat involved.
  • a Slowing effect (dpeg) that persists while the stat is at 0. Active berserk and haste effects will collapse at this point. The slowing may be countered by fresh haste from potion or spell, but hasting effects will have a much reduced duration while the player is at stat 0.
    • Actually, I proposed that the Slow keeps going even if the stat 0 is repaired. Thematic reasoning: things get hard at 0 Dex/Str/Int and even if you recover, it takes a while until you're back to normal.
  • other effects based on the stat involved (see blow). These are cumulative, so if the player has multiple stats at zero, the effects for all of the zero stats apply simultaneously.
  • a counter counting down to death, perhaps at 150 turns from the initial stat going to zero. Multiple stats at zero, and attempts at draining stats that are already at zero will accelerate this counter. The player will be warned about impending death as the counter ticks on
  • dpeg: I suggested that 0 stat should be very taxing on the organism, reflected in rotting. I.e. a Rotting effect as long as a stat is 0.

What about this: slowing only on 0 dex, rotting only on 0 str and death counter (“forgot to breathe”) only on 0 int? Those effects are extremely harsh already. 0 dex means that almost every attack is going to be a hit. With slow this means that every monster can hit twice with no effort… Death counter could be short (to balance it with 0 dex) and rot very fast (slow-able by potions of healing?). This would make all effects dangerous, but not immediately deadly. Int drain would make it an emergency, because you can't just stand there and fight with the ticking counter, nor can you blink away easily. Str drain would force you to use wands or spells to defend and would root you in one place (collapsed on the ground). Dex drain would make you incredibly bad at both spells and physical combat, so it wouldn't need any special danger on top of that. — someone-else 2010-02-15 19:44

In general, any stat that goes to 0 is probably a stat that the player does not depend heavily on, so the ill-effects of zero stats need to be strong enough to push players to take action.

Int 0

The player cannot cast spells or read scroll or books while their Int remains at 0.

In addition, there could be a chance per turn of losing an action altogether (in addition to the Slow effect) because the PC is having trouble thinking. Potential messages:

You pause for a moment, so as not to slip on that banana peel.
Where's your head at right now?
You feel the situation is grave. Trying to count to ten only makes your fear worse.
Str 0

The player's carrying capacity naturally collapses with Str at 0. In addition, they will have extremely low damage output in melee and ranged combat.

HP regeneration could be greatly reduced with Str 0 (i.e. treating Str as a constitution stat)

  • What do you mean by “collapse”? Unable to pick up anything, or just easily encumbered? — jpeg 2010-01-07 18:47
    • Reduced CC beyond what Str 0 already does, I say.
Dex 0

The player may strike themselves when they attack. Their to-hit will be very low (missing an attack may direct at themselves), and they will suffer an EV penalty. The to-hit penalty applies to all forms of combat: spells, melee, ranged.

Make the player fail stealth checks and unable to stab. — tgw 2010-03-06 15:03
Additionally, movement could be actively loud, and maybe all item manipulation as well. — og17 2010-03-06 21:20

Comments

  • All of this sounds great! I think we could use it as-is, though probably only for 0.7. One question: what steps can the player take to increase lost stats again? Restore ability, I guess. I'd also like to keep the current restrictions of transformations and equipment of items known to reduce stats being prevented if this would reduce stats below 1, even if this is no longer directly fatal. We'll need to handle negative stats resulting from equipment correctly, so the player doesn't gain stats when removing the item. Should negative stats affect the maximum stats instead? — jpeg 2010-01-07 18:47
  • Well, now is the time to use that ring of Dex+5 :) — dpeg 2010-01-07 19:01
  • Something that might be nice is a warning when a stat is reduced by, or to, a certain amount, similar to the low hitpoint warning/ouch, that really hurt! messages you get for hp. Right now, it becomes red if it is 3 or lower, which is easy to miss if you're not paying extremely close attention. — ogaz 2010-05-02 22:38
Logged in as: Anonymous (VIEWER)
dcss/brainstorm/mutation/stats.txt · Last modified: 2013-04-26 00:27 by galehar
 
Recent changes RSS feed Donate Powered by PHP Valid XHTML 1.0 Valid CSS Driven by DokuWiki