Under Consolidation

This is "under consolidation until all ideas are reviewed" - i.e. the YANI archive has caught up to the present, which it has. This list is temporary and the consolidated ideas should go off to some other list.

Buffing boomerangs

• L's Magic Boomerang patch (they teleport back to your hand when they hit stuff) http://l.j-factor.com/nethack/magic.diff
• there's a patch somewhere that makes boomerangs fly in a straight line like Link's boomerang, trading away flavor for the gameplay benefit of making it a lot more usable
• #2031: Enchanted boomerangs can continue flying after hitting a monster, with an enchantment/7 chance. (Add 2 to the effective enchantment if the boomerang is blessed; subtract 2 for each successive monster it hits.) If the boomerang is cursed, enchantment doesn't matter - it will always fall on the first hit.
  • The reason behind this is twofold: because it makes boomerangs a more attractive weapon, and for the slapstick comedy value.
• #658: An artifact boomerang which is a guaranteed first sacrifice gift for Archeologists.

Buffing cavemen

• lorimer's Less Boring Cavemen patch (flint tipped arrows, taming mastodons, harder to read spellbooks, extra sling multishot) (bilious 285)
• ignore: all nemeses fail to respect E, sparking flint, Keolewa, decreased chance of god niceness, possibly the part about restricting edged weapons (the goal should be to make Cav more attractive)
• probably fold in fourk sling buffs (185253e)
• probably fold in fourk increase flint stack size (put in cav-home) (7b0f71d)
• fourk also places a lot of random rock piles around, which is not a great idea. consider producing flint, or whatever the ideal sling ammo is, from more sources (rock moles drop flint stones, as they have them in their stomach to break things down, and perhaps monster digging can generate some flint along with the rocks)
• fourk make flint stones d6/d8 (d8283e9)

Sacrifice gifts

• evilhack book gifts (8b0ebb7)
• fourk nonartifact sacrifice gifts (87ed523, 0113478, 1875083) (also implemented in evilhack)
• #1403: Barbarians and cavemen get weapons/ammo instead of spellbooks as a prayer boon, or enhancements to their current weapons.
• #2201: Don't gift a character a spellbook as a prayer boon if the character is illiterate or certain roles (like cavemen or barbarians), and if a spellbook is given limit its level somehow so that a character doesn't get a spellbook of much too high level.
• #2227: Gods can grant you (unique?) pets as a sacrifice gift. Example here was Sleipnir for valkyries.
• #4023: Your god may give you useful non-artifact items as sacrifice gifts, such as a stethoscope or bag of holding, so long as you aren't carrying one of them already. A slightly more advanced system that wouldn't require carrying a bunch of items to avoid getting duplicates would be to track the last time you picked up or put down each type of object, and if you have never touched one or it's been a sufficiently long time, you are eligible to get that item as a gift.
• #2644: Remove the feature where being gifted an artifact in a restricted skill unrestricts it to Basic; compensate for this by fixing the balance of artifact base types and the gifting logic so that your god's first gift to you is never an artifact whose skill you're restricted in.

Quest redesign / role differentiation

• fourk quest stocking (06911f7, e91af58)
• wait on / fold into quest redesigns (note: this covers Arc, Bar, Kni, Sam)
• also: fourk add some shuriken to the Monk quest
• Guarantee a bunch of flint, also regular rocks, scattered around Cav home.
• #2839: Allow quest monster generation to be specified more precisely than having a fixed set of probabilities for two specific monsters and two monster classes (e.g. the Wizard quest has a certain distribution of vampire bats, B, xorns, W, and random monsters; it is not possible to change this distribution without changing it for every quest).
  • This would be better done by taking Un's level-specific monstergen implementation and then customizing them as necessary for quests.
• #2137: New "hedge" terrain #. It blocks movement but not line of sight (like iron bars), can be chopped down with an axe, and has a chance of blocking ranged weapon attacks passing through it. Possibly it can be jumped over. (Not a requirement if quest redesign ends up not needing this at all.)
• #2776: Level that contains a structure with a lot of gargoyle statues. Some of the statues are statue traps. (This should go into some redesigned quest but no candidate is chosen currently.)

Buffing lance

• fourk knights won't break lance when jousting under certain conditions (good alignment?) in order to incentivize doing it earlier

Magic traps

• this has no actual ideas. fold things in.

Speed system

• fourk speed formula tweaks (1eb7c60)
• Speed only grants speed bonuses when moving.
  • Perhaps the best idea for this so far is to make move actions that were taken under the influence of speed-giving things deduct less than 12 movement points, rather than giving the hero bonus movement points.
• new monster speed system array of 12 bits idea
• #1482: Various ideas to give the bonus from certain sources of speed only as a bonus to movement, not to getting more turns to attack, use items, etc.
  • First, to classify: most agree that speed boots, intrinsic speed, and steed speed should be sources that only give extra movement opportunities, while polyform speed should give extra full turns. Less agreement on magical sources of very fast speed (the potion and haste self), but general opinion seems to be that these should also not be limited to movement.
  • Add a second pool of movement points that coexists with regular movement points, but can be used ONLY for moving. Speed boots and its fellow sources add to this pool, and any unused points in it are discarded at the end of the player's turn. When you make a movement, points are deducted from this pool before being deducted from the "core" pool. Note that this approach can cause core points to accumulate to the point where the player can eventually get multiple non-movement actions on their turn.
  • Reduce the movement point cost if you are taking a movement action and have sources of movement-only speed. (This might interact oddly if general-purpose sources of speed continue to add movement points as usual.)
• #1789: Monster speed system that tries to keep all the good qualities of 3.6.1's while also not letting randomization make monsters super fast or super slow: add a 16 bit int to the monst struct, and every 12 turns set (speed % 12) of them to 1. Then on each turn, the monster gets (speed / 12) moves, plus 1 if the bit for that turn is 1. This 12-turn cycle would need to be phase shifted randomly per monster (probably using hashing based on monster ID) so that you can't move-count based on the turn clock.
  • Research on an efficient algorithm to generate the series of 12 bits has found https://stackoverflow.com/questions/17010857/efficient-random-permutation-of-n-set-bits, though it might be more entropy-efficient in the long run simply to generate all 4094 permutations of bits and randomly pick one with the correct number of bits whenever needed.
• #1898: Speed boots' speed is gently nerfed to be better than intrinsic speed, but worse than temporary sources.
• need to publish the Speed System Ideals for comment at some point

Moar randomized appearances

• note: these usually need new tiles
• unnethack make ring probabilities non-uniform (e3e872)
  • This is either unused in unnethack for making some rings meaningfully different probs, or is necessary for adding randomized ring descriptions.
• fourk (2d2f661)
• unnethack additional appearances too (9a0eb19, a1f88fb, c627088, c89c528)
• splice mithril wand would be a good randomized appearance (also look for others. bone wand? just go through splice objects.c in general)
• a good time to periodically update these would be with all the other data files per release
• L has a patch adding appearances too (http://l.j-factor.com/nethack/shuffled.diff); tiles not provided

Controlling amount of random wishes

• unnethack remove throne wishes (a0d62cd)
  • beware becoming nerfhack mk II...
• this needs to look at other sources of random wishes. removing throne wishes doesn't do much by itself if you can still fountain quaff for a water demon wish, or get a smoky wish, or a bunch of magic lamps

Wizard mode enhancements

• unnethack get shoptype from SHOPTYPE by name (7c6b6)

Regeneration

• unnethack regeneration is only hunger inducing when actually regenning HP (642d3)
  • This should wait on other regen changes; if regeneration is buffed to more than 1HP/turn, cutting its hunger penalty when not in effect is probably making it too powerful.
• fiqhack regen is based on con and xl plus healer role bonus
• regen is faster the lower your HP% is or make it scale better by multiplying your own regen rate by 1.5x or 2x
  • possible scaling formula: you heal at (2 - (HP/maxHP)) times your normal rate

Unicorn horns

• Vanilla horns now don't cure attributes, consistent with current xnh; however, it's probably still a problem that they're a cure-all for hallu, stun, blind, and other effects making those effects less interesting after a horn is acquired.
• #1558: In an effort to make unicorn horns not infinitely usable but also preserve their other good qualities: Split the unicorn horn into two items, the horn, a simple tool which now has a number of charges, and a "unicorn spear" / "hardened unicorn horn" which is a weapon.
  • Both items have the same base damage, are two-handed, and use the unicorn horn skill, but reading a scroll of enchant weapon while wielding a unicorn horn will transform it into the weapon version like how crysknives work. The weapon cannot be used to cure ailments.
• #1551: ais523 also proposed an interesting idea - each time a unicorn horn cures a given ailment, you gain a tiny shred of resistance to unicorn horns curing that ailment (or perhaps just unicorn horns working at all) in the future. However, FIQ doesn't like this approach because it's not very visible to the user. Perhaps a message that the horn failed because you're used to it is warranted. (SAVEBREAK)
• fourk make unihorns finite (2b6fa3a) (however this is not the likely chosen solution)
• Remove the ability of unicorn horns to cure anything but nausea and sickness. This is a mega-nerf and should be treated with care.
  • Requires at least the addition of something that can cure stunning or the removal of Rider stunlocking.
• #743: Cancelling a unicorn horn makes it non-magical and removes its curative properties.
  • maaaybe. might mesh ok with the 'unicorn spear' proposal
• #875: Give unicorn horns a to-hit penalty: they are not a manufactured weapon, and are not designed to be wielded in your hand.

Monster stomach contents

• #1793 is the gist of this; however, there are problems with that implementation. Notably, pet monsters with reasonable stomach contents can still pick things up. Perhaps the correct solution is to give monsters another object chain representing their stomach contents (this would not be trivial: may have to add another where type and rules for stomach objects pointing back to the carrying monst).
• Strictly speaking this is a vanilla issue: dropping items in a purple worm's stomach then escaping leaves the items stealable by nymph
• It also raises symmetry questions. Items the player eats aren't stored in their stomach. What if they polyself into a monster that is supposed to store stomach items due to slowly digesting them? Probably just delete them like normal food, since it's not really relevant what happens to the player's stomach contents if they die.
• #1644: Bookworm, a bright blue w monster that eats books (they end up in its stomach/inventory and aren't destroyed, though they may lose a read charge). Levels up, gaining power and speed, for each book it eats (higher level books give it more power), eventually gaining more attacks. Possibly can also eat scrolls. Initializes as having already eaten a few books. There must also be interactions/YAFMs with eating blank books, novels, and the Book of the Dead. (SAVEREAK, NEW TILE)
• rock moles eat gems as well as gold
• gelcubes should put all engulfed items into their stomach
• If a monster (like a purple worm) eats another monster, that monster's inventory should be added into the eater's stomach (minus anything made of FLESH or VEGETABLE, which get instantly destroyed).
• #1829: Rock moles and other item-eating monsters drop any gold or gems they have eaten (digesting them can't be a fast process), or they destroy 10% of all gold they eat and have a 10% chance of destroying any other items they eat; otherwise those items will be placed in the mole's inventory.

What To Do With Slow Digestion

• #679: Merge the rings of sustain ability and slow digestion into a "ring of stasis". Double-edged effects: slows down digestion like normal, but also slows down healing, energy regeneration, timer effects like confusion and sickness, any timing-out intrinsics, prevents exercise from happening, blocks or reduces to 1 effects that directly change attributes.
• #727: Merge the rings of slow digestion and hunger into a single chargeable ring of digestion. Negative values increase hunger faster, positive values slow it down. Duplicating the effect of a vanilla ring of slow digestion would require getting a +10 ring or stacking up to 10 with two rings. (Possibly the negative/positive values should be switched, but that would complicate the generation code.)
• Or don't merge slow digestion with anything, but do make it chargeable, with +10 giving the same effect as currently. E.g. if the ring is +N, there's a 2N/20 chance per turn of not hungering normally.
• #2023: Make the ring of slow digestion's effects totally opposite those of the ring of hunger: rather than cancelling your normal digestion entirely, it only stops you hungering on every other turn, effectively increasing the hunger rate while you wear the ring from 1/20 to 1/2.
• #1256: Non-cursed rings of hunger don't drain excess nutrition when you are Hungry, and slow digestion rings don't slow your digestion when you're oversatiated.
• #2760: Merge the rings of slow digestion and regeneration. Keep the name "regeneration"; make it chargeable. A positive rate boots your regeneration and your hunger rate, up until they are doubled at +10; a negative rate decreases both until they are halved at -10, with the intermediate values scaling as appropriate.
  • Possibly, since the extremes of +10 are hard to attain, make the doubling/halving threshold at +7 instead.
  • This one is unlikely, due to other regeneration changes being more likely.

New shop types

• The main decision to be made here is how much probability the new shop types should steal from existing shops (if
• fourk new shop types music, gift, rare goods (c56cf30)
• unnethack add canned food shop (1cdf4, 8fc675)
  • there is no Town level in xnethack and one isn't planned, but perhaps a Minetown variant or two could have one. also perhaps they could rarely sell tinning kits and tin openers.

New spells

• #1592: Additional priest spells: raise undead, floor to lava, ball lightning (not a ray and thus not reflectable), summon elemental.

Acid-corpse alchemy

• note that this was slated to be written up as a reddit mini-proposal
• newt and electric eel: gain energy
• stalker: 50/50 invis or see invis
• chameleon: polymorph
• ghoul: sickness
• nurse: full healing (problematic; while it would incentivize carrying acid around, it eliminates problems with cannibalism and preparing nurse tins. Maybe instead it should make extra healing with small chances of regular/full, or a 1/3 chance of each, or a 6/2/1 distribution or something)
• nymph: obj detect
• homunculus: sleep
• violet fungus: hallu
• quantum mechanic: speed, rarely
• ???: blindness (candidates aren't good. ravens?)
• do identical pre-eating taint check: if dipped corpse is tainted, turns into sickness

Trap scaling

• in particular, falling rock traps dumping multiple rocks farther down in the dungeon. when that happens, should probably reset rock trap damage from arbitrary 2d6 to "standard ROCK dmg" like comment says,
  • possible range depth/6 to depth/3
• other traps need to scale too. there was some idea about gehennom pits being demon-made and designed for lots of pain
• Probably, augmenting struct trap with one or more fields is a good idea, so damage isn't just dependent on depth and can be controlled more finely, making this a SAVEBREAK
• #2840: Make arrow traps multishoot arrows, and falling rock traps dump multiple rocks, at higher level difficulties (also increase the chance their ammo is poisoned)
• #3725: Some falling rock traps (particularly deeper in the dungeon) contain one or more boulders instead of rocks.
• #4122: On deeper levels, land mines cause a fiery explosion in addition to their existing effects. Player-placed land mines get a bit trickier; a simple implementation will just consider the depth of the level the trap is on, but a more correct one would track via a field on the land mine object (in case it is picked up and carried around to another level) whether or not it will produce an explosion when triggered.
• #3977: Falling rock traps can dump multiple rocks onto you at once, up to (depth/4). Damage is taken for each rock. If a trap would drop 4 or more rocks on you, it may instead drop (rocks/4) boulders on your head instead. Helms should be less effective at reducing damage from falling rocks, and almost useless at reducing damage from falling houlders. If multiple boulders are dropped, they might be dropped on nearby spaces as well, so that a pathway might end up blocked.
• #3978: Pits get deeper the further down in the dungeon you are, which makes them deal more damage, might cause wounded legs with a chance that increases with depth, and take longer to escape from (or are so deep you can't climb out at all). In Gehennom, they might deal extra damage on account of being designed and put there by demons. The extra depth could be conveyed to the player by "You fall into a [very] deep pit!"
• #3979: Deeper in the dungeon, rust traps might target your pack instead of your armor or gear, and could possibly be renamed to "water trap". On deeper levels and particularly in Gehennom, rust traps phase out to be replaced by acid traps. These deal corrosion and acid damage (and quite a bit of it, like having multiple potions of acid dumped on your head).
• #3983: Container traps' effects scale based on dungeon level (ideally on the level a container generated, not on the current level, so a player can't bring a chest up to level 1 to make it use the weakest possible trap.) In addition to adding some more effects, also add some effects which cosmetically are the same as other ones, but are more or less damaging. This would, however, make untrapping containers even more useless than it is now; perhaps a trap should stay armed even after you set it off. Another idea is to give certain traps an arming item, like a potion of oil for the fire effect and a land mine for the explosion effect. Successfully untrapping it will deposit the arming item into the container. Also, make container traps independent of Luck, because currently they encourage the player to ignore containers until Luck is maximized. #4174: Make container traps deterministic (either as a hash of their object ID, or stored as a field on the object). Untrapping them successfully can then yield an item related to the trap, which then gets deposited inside the container. The items for each type of trap are: Hallucination gas = potion of hallucination Paralysis = potion of paralysis Electric shock = 0:1 wand of lightning Tower of flame = potion of oil Poisoned needle = nothing Poison gas = potion of sickness Explosion = land mine
• part of #3981: anti-magic fields in gehennom should be capable of cancelling you or your gear
• #4453: Convert the sleeping gas trap to just "gas trap", and make it scale based on the difficulty. Each individual trap releases a single type of gas; which type is partially dependent on the level difficulty. At low levels, they all emit sleeping gas like normal. At higher levels, they emit stunning hallucination gas like container traps, clouds of poison gas, and maybe even nastier gases that reduce speed or miasmas that cause illness.

Dealing with covetous

• #37: If the player does not pay the demon lords, they remain on the downstairs and will not move until the player has paid, which allows the player to return with the requisite amount of gold. However, they will get angry if you get below their lair in any way (possibly measured by testing your lowest level reached each time you enter the lair).
  • This requires something be done with covetous AI; it won't work as long as they warp every time after being disturbed. Not sure this is even slated as a want-to-do idea.
• #1543: Covetous warpers can only warp to the location of the item they covet. This means that if you are carrying it, they can warp to you all they want, but they can't warp back and forth to the stairs.

New poison system

• #716: Condensing all of NetHack's poison mechanics into one unified mechanic:
  • Dexterity and Constitution poisons are removed. Quasits (the only AD_DRDX attack) might still directly harm Dex, but it's no longer flavored as poison. Rabid rats (the only AD_DRCO attack) don't cause poison, they cause disease, which hits Constitution. Poison will now only affect Strength.
  • Poison is now implemented as a timing-out trinsic. The amount of Str loss is a direct function of how much timeout is left, so Str will gradually recover as the poison goes away.
  • Passing Con saving throws could give small bonuses to decreasing the timeout faster than normal.
  • While poisoned, HP regeneration is either impeded or stopped entirely, or else the player's HP is damaged every turn, or every several turns.
  • It could also heavily abuse Str while it lasted instead of being directly tied to Str.
• #818: Overhaul poison so that its effects are less random: some might still cause instadeath, but it would only be found in the late game or via bosses like Demogorgon and Scorpius; early game poison shouldn't drain multiple stat points at once.

Nutrition system changes

• consider reverting the extra monsters at level creation and fewer over time thing
• #758: The monster spawn rate gradually decreases the longer you spend on a level. (May not apply if you have the demigod flag set or are carrying the Amulet.)
• #847: Most corpses have sharply decreased amounts of nutrition (from the current values). They may also be too mangled to be good for a sacrifice, randomly based on their object ID. Their intrinsic-conveying chances are unchanged, though.
• #1197: Instead of being perpetually rechargeable (that opens up a lot of exploits and farming like smoky potion farming and infinite altar farming with the Platinum Yendorian Express Card), horns of plenty now stop being rechargeable after 7 recharges.

Viability of thrown potions

• potion viability as splash weapons: vapors should have a 100% chance of hitting anything adjacent, 50% in next largest radius, 25% in next
  • possibly waiting on meaningful vapor effects on monsters? need to investigate that a bit
  • #1915: Potions of acid have a splash radius effect when broken (could probably be implemented as an acid explosion).
• #1958: When acid breaks on someone's head and spills on them, it actually does more damage than quaffing it, because it is spread out over more surface area.
• #2247: Traps that shoot out potions. Generate with a variety of negative potions such as sleep, acid, lit oil, hallucination.

Phasing out Luck

• #2001: The chance for writing an unknown scroll or spellbook is based not on luck but on how many scrolls and spellbooks you already have identified (and thus, how familiar you are with their sort of magic). Later written up as https://old.reddit.com/r/nethack/comments/d8nihe/
• #2543: Add "magical writing" as a nonweapon skill, which is trained at a fast rate by writing scrolls and spellbooks, and increases your odds of writing unknown scrolls and spellbooks successfully.

Religion, alignment, and god stuff

• #2230: The more gold you pay to a priest, the stronger his god becomes. Newly generated priests of that god could get better equipment, or altars of that god might be harder to convert. In Moloch's case, Gehennom gets slightly harder if you pour a lot of gold into his temple. This applies with money donated to your own god as well, but the effects are naturally less visible since you won't go around fighting your own priests or converting your own altars.
  • Requires at the least separate god tracking.