Two challenge gods

[archaeo]

I was procrastinating, so I wrote up some spitballed proposals for two of those challenge gods. They're probably not good proposals, but I wrote them, so I may as well post them.

H, God of the (Popcorn) Harvest. Likes exploration, has normal piety decay.

0*: H weeds the garden: all monsters with HD < the player's XL (or whatever metric succeeds in removing vast majorities of easy encounters) immediately disappear/implode when entering LOS. H passively feeds the player when it pulls a weed, but it also creates noise.
1*: passive ability Knowledge of the Culled, player gains (piety/10)% XP from culled monsters, and gains a +(piety/100) bonus to skill training.
2*: passive ability Sowing the Fields, culled monsters have (piety/20)% chance to be replaced with a monster that meets H's standards.
3*: no ability
4*: active ability Song of the Harvest, all monsters are HD drained (MR check), with H culling those that fall below the popcorn threshold (8-10 piety, breath).
5*: no ability
6*: active ability Song of Sowing, for piety/20 turns, H suspends the harvest and surrounds the player with 15-20 friendly (not allied) mighted and hasted popcorn, initially pushing any adjacent enemies away from the player (15-20 piety, red draining, breath).

Wrath: H can summon swarms of 15-20 popcorn monsters that are all hasted, mighted, and spawn under ddoor status. For the duration of H's wrath, Sowing the Fields remains active, but its rate of activation is increased dramatically (though it doesn't affect the popcorn he summons).

* * *

new!Wulndraste, likes going downstairs (gain piety dependent on amount of time spent on level, only applicable the first time you descend, must take stairs, trap doors, or shafts), hates standing still (W has no piety decay; instead, W grows impatient if you spend too long on a level (on the order of <1,000 turns), and responds by shafting you and halving your piety). You enter penance if W reduces your piety below 0.

0*: there are no more upstairs or exit portals, at all. You gain the ability to immediately leave a branch when you pick up a rune (the fifth Pan rune boots you out of Pan). W values your free will; you are immune to non-god created shafts.
1*: all actions reduced by .1 speed. Active ability: Cleaving Wind, you gain Wind status (with significant -Wind cooldown) that deals moderate piety-dependent damage to all adjacent enemies for piety/10 turns.
2*: active ability Squall, all enemies are pushed away such that nothing is closer than piety/25 squares, and they receive collision damage if applicable, (8-10 piety)
3*: all actions reduced by .1 speed
4*: active ability Downdraft, player is shafted, but arrives on the floor with free invis, a big stealth bonus, and receives a few rounds of Passive Mapping 3 (5-10 piety, you don't gain piety for going downstairs)
5*: active ability Wind of Return. You are immediately teleported to the entrance of the branch (40-50 piety, gain one permanent mutation from the set of weak/dopey/clumsy)
6*: W will bless a single weapon with the Speed brand, styled as [Weapon] of the Wind {speed}.

Wrath: the upstairs are returned, but they frequently crumble into nothingness, which can shaft you or banish you. If you are in Depths, being shafted in this way may drop you into the Vestibule; in Zot, you may be shafted into Pan. Being shafted in this way causes significant disorientation; you are slowed and confused.

Mapsize reduction god is harder; I guess it might work simply by being a map god that expects you to follow the shortest path, enforcing its will by closing off everything beyond that path. Maybe pair it with echolocation and blindness!

edited: now that this is an OP, I might as well not hide the content behind a spoiler tag.
edit2: replaced H's direct skill bonus with a training bonus per njvack's comment, set at +2 bonus at 200 piety, which may be too high/low.

[tabstorm]

Well, when most early game monsters are either unkillable or easy, and you have a god that dosen't let you backtrack, who dosen't do anything aside from increased stamina (you won't reach his 4* and have it usable until like XL14 or so, and it's useless if you have a bit of foresight wrt. teleporting. Also, what exactly is stamina?) and some kind of stealth thing (when did stealth become good?) I can see how he wouldn't be overly enjoyable. Actually increasing attack power would make more sense for this god since you will be in some areas earlier than you would like to be. But, backtracking is so ingrained into the game that it probably wouldn't be enjoyable to block backtracking, especially if the god's powers are not that good.

[Lasty]

On first reading, archaeo's god ideas strike me as considerably better than most. I'd be interested in seeing them get tested out.

Edit: tho +5 to trained skills on the harvest god is definitely stepping on Ash, and "all actions reduced by [x] AUT" is a problem for things like finesse'd quickblades.

Edit2: Moved to GDD because I'd like to discuss these as serious proposals.

[xentronium]

I really like the first idea, mainly because it would eliminate so much tabbing.

[njvack]

I'd try the harvest god for sure. Might want some number tweaks, but the concept is interesting to me.

Instead of a straight bonus to skills (which steps on heroism & ash) maybe give a training bonus -- like having a weaker version of every manual.

[archaeo]

But, backtracking is so ingrained into the game that it probably wouldn't be enjoyable to block backtracking, especially if the god's powers are not that good.

Well, that's the trick, right? A god that prevents backtracking is totally reasonable if it makes you really strong, and it just so happens that we don't really have a speed buff god. I'm not sure that my proposal above is the best version of that idea, but it seems like an ok start to me. One correction that occurs to me is that Wind of Return should probably just be a branchport, given its high cost.

The popcorn harvest god is harder, just because the main passive ability I proposed probably steps on Ash's toes too much. It'd probably be okay for Crawl to have more than one god that boosts skills, but I can't help but think there's a better idea that isn't occurring to me; njvack's training bonus idea is much better, for example.

edited to add:

Edit: tho +5 to trained skills on the harvest god is definitely stepping on Ash, and "all actions reduced by [x] AUT" is a problem for things like finesse'd quickblades.

yeah, njvack's training bonus is better than a straight skill bonus, I'm adding it to the OP. I'm not sure I understand the second issue w/r/t finesse; do you mean haste? I mean, you're the dev, but isn't there a floor on action speed, e.g. a hasted/finessed player swinging a quick blade of speed can't hit faster than 0.2?

[asdu]

Sorry, but have you people all gone crazy? Surely the way to solve the issue of "too much popcorn" is not to incorporate the problem into other design thus guaranteeing that the problem will never actually be addressed.

[archaeo]

I mean, if some hypothetical future group of Crawl devs decides to remove all the popcorn encounters from the game, they probably wouldn't have any problem with removing the god that revolved around them. I'm skeptical that "too much popcorn" is actually a problem anyway (I suspect the only actual problem is the game's length, which could be addressed a lot more directly than by trying to remove all easy encounters), but if we're going to discuss it, let's take it to CYC.

But to answer your question: "gone"? Isn't participating in the Tavern proof that we were already crazy to begin with?

[johlstei]

Introducing a god that gets rid of popcorn would also be a good way to get play-testing of crawl without popcorn, without forking or making a drastic change for everyone.

[reaver]

I've considered something similar to H's 0* ability independently. (which is probably an indication it's a good idea!)

Not sure about the rest of H's abilities, my version was that you automatically gained XP for culled monsters, no other effects ever.

2* being probably a drawback specifically strikes me as quite bad.

Making Wulndraste conduct force movement after a period of time seems like it would help make it work properly, although take that with a grain of salt as I haven't played Wulndraste.

[archaeo]

FWIW, I think you could tie that 2* "ability" into the 0* conduct; I just figure it'd be a good idea to replace a few of the popcorn monsters with dangerous ones instead.

Most of those abilities were totally off-the-cuff ideas, and I'm sure there are better ones. My main thought is that "prevent backtracking god" and "remove easy monsters god" are cool open design spaces. I figure that the abilities for the former should help players alleviate the inability to go back upstairs, and the latter's abilities should help make up for all the missing XP; thus, speed and training bonuses.

[wizzzargh]

These both sound like great ideas to me, I especially like the Harvest God proposal.

Does the Harvest god cull/sow summons? It probably should- Fiends could still be dangerous summons but Ufetubi being culled immediately on arrival would be fantastic. What about friendly things like player summons? From a balance standpoint it probably doesn't need the nerf, but not allowing WEAK summons/undead could be interesting.

I think players should get 50% or even 100% XP from culled things, but not get any skill bonus to avoid stepping on Ash's toes. Or just get 0% XP AND no skill bonus. I don't think popcorn XP is that relevant, but losing the XP from them in exchange for nutrition could be an interesting conduct.

I don't know if you intended this with 'disappear/implode' but if enemy equipment imploded unless they were sowed(in which case it would act like Polymorph), that would save a lot of sanity in Orc. That's another (mild) nerf Harvey probably doesn't need though.

A Reaping Brand that became available at 3* could be an appropriate early/midgame power boost- if you're creating giant monsters, it seems only fair that you could gain some sort access to them.

Spoiler: show

wrath is death cobs

----------------------------------------------------------------------------------------------
new!Wulndraste

Rather than active ability Cleaving Wind, how about permanent Cleaving via the Blade effect?
From a coding standpoint, it might be easier to have the god simply excommunicate you if you take an upstairs or exit portal.
I think the god needs to give you some sort of HP regeneration or a more powerful combat boost if it's going to shaft you every 1000 turns. Instead of increasing your speed, it could Slow all enemies in sight once you reach 3*?
And Wind of Return needs to have an even worse effect... maybe -2 to each stat, or the mutation is random and bad so every time you use it you're risking something really nasty like teleportitis.

The god should probably give you a warning if you're about to enter a branch with no rune, since you could easily become stuck and have to be excommunicated to escape from Lair8/Orc/Elf/a previously entered branch. Banishment might cause problems as well, especially if you have the Abyssal rune already- maybe you can leave the abyss with no problems, but abyssal downstairs don't give piety.

[Lasty]

Another consideration for W2: if portaling back from a branch end to the surface is free, then it's like an instant safety teleport, and it makes branch ends much less dangerous that other branch floors. If it's not free and it might ever be possible not to pay the cost, then you can get permanently trapped.

[archaeo]

You could always make it like teleporting with the Orb, Lasty; you pick up the rune, and W "gathers the wind to blow you back to where you started" over 10-20 turns.

[Lasty]

1) Not every branch end has a rune.
2) What if you go back to the branch end after picking up the rune?
3) Orc has disconnected layouts where you might be unable to descend from the particular piece of O:3 you're on.
4) Tomb's layout requires going and up and down stairs repeatedly before getting to the rune.

Ugh, working on Wulndraste was a pain in the butt.

[archaeo]

For 1 and 2, it seems like you could just make Wind of Return a 0* ability with harsher punishments (reduced to 1* of piety, more bad malmuts, etc.), whatever it takes to make it always accessible but suicidal to use more than a couple times per game. FWIW, I also imagined the rune pickup transporting you back to the nearest Dungeon/Depths stairs/portal, i.e. you pick up the gossamer rune and get transported back to the Lair staircase in D.

For 3, W will just shaft you if you wait around too long anyway; since your W code already "recognizes" these layouts, couldn't you just speed that up? ("W says, 'The wind cannot flow freely here...'") If not, W has a shafting ability; I could see an argument for making it a 1* or 2* ability instead. Either way, W should discourage the player from entering Orc, since the only way out of Orc/Elf is by using Wind of Return.

For 4, W has that shafting ability, and honestly, I can't remember the last time I used the Tomb:1 corridor anyway.

[Lasty]

Ah yeah, I'd forgotten the self-shafting. I was having a Wulndraste1 flashback.

[Igxfl]

Sorry, but have you people all gone crazy? Surely the way to solve the issue of "too much popcorn" is not to incorporate the problem into other design thus guaranteeing that the problem will never actually be addressed.

Ashenzari was built around curse scrolls, which were later removed except for followers of Ashenzari.

[milski]

I'm not quite sure how the "remove easy monsters" god is considered a challenge god considering that after a period of time he acts to essentially triple your rate of skill earning, and the wrath for leaving him is relatively minor (compared to people that actually summon strong monsters on top of you, having popcorn *become* strong monsters is much weaker).

As for the actual abilities, HD is a very weird way of identifying popcorn, and HD<XL removes everything from the game except Ancient Liches, orbs of fire, and select uniques (including all pan/hell masters, and most everything at a level you'd actually encounter it). Even HD<XL/2 removes several threats entirely, including but not limited to:

Dowan and Duvessa, level 7/9
Boggarts, level 9.
Prince Ribbit, level 13
Sonja, level 13.
Tormentors, level 15.
Swamp dragons, level 17.
Vault Sentinel's, level 17
Ironbrand Convokers, level 19
Everything in Tomb but Bennu, Greater Mummies, and Reapers, level 21
Death cobs, level 21
Flayed ghost, level 23
Unlucky Pan lords, level 25.
Fire and Ice dragons, level 25.
Curse skulls, level 27.
Ghost moths, level 27

Most of those monsters are at least reasonably threatening when they would generate or have unique attacks, but would almost certainly be considered popcorn for any non speedrunner (and speedrunners would almost certainly have 5x "typical" speedrun EXP, if not more), removing a large portion of actual threats from the game. HD is simply not a good way to judge whether or not a monster is popcorn.

[archaeo]

Like I said, I'm happy for other suggestions on how you'd identify "popcorn" to be removed. It seems like a good idea to avoid a special-cased list that would be hard to maintain, but I'm drawing a blank on some mechanical hook for this kind of culling.

That said, the goal is definitely to be "challenging," but only in the sense that Chei or Qaz is challenging. I don't think there's a great reason to make a god that is outright detrimental to the player; we have Xom for that!

[milski]

There isn't a mechanical hook for that sort of thing, because the current "harmless/easy/dangerous/extremely dangerous" categories are already pretty terrible. You'd have to special case almost the entire list, aside from maybe putting rules to determine if melee, speed <=10 monsters are threatening enough to not be popcorn.

[milski]

These both sound like great ideas to me, I especially like the Harvest God proposal.

I think players should get 50% or even 100% XP from culled things, but not get any skill bonus to avoid stepping on Ash's toes. Or just get 0% XP AND no skill bonus. I don't think popcorn XP is that relevant, but losing the XP from them in exchange for nutrition could be an interesting conduct.

Popcorn EXP is probably more relevant than you'd think, depending on how weak you define "popcorn."

For instance, by the time you reach the mines, Orcs, Orc wizards, Orc priests, Orc warriors, and possibly Ogres are all essentially popcorn. Their spawn rate is so much higher than Orc knights, Orc warlords, Orc high priests, and Orc sorcerers that I would say more than 66% of the branch EXP would come from popcorn, and probably more than 90% of the EXP would come from popcorn excluding the Orc 4 boss vault. This situation is probably similar for every branch except maybe the Vaults, since you can't call Vault Guards popcorn without removing V:5.

[wizzzargh]

Remember you'd be getting new tougher monsters to kill in place of the popcorn, which would be worth a LOT more than popcorn.
My last game where I died shortly after orc, I compiled some monster XP (some of the weaker ones obviously spawned in D, not Orc but oh well)

264 orcs giving 3 xp each 792
24 orc priests 42 XP each 1008
35 orc warrior 133 XP each 4655
30 orc wizard 37 XP each 1110
16 Ogre 119 each 1904
9469 total

4 2-headed Ogre 282 each 1128
9 orc sorceror 563 each 5067
3 orc knight 617 each 1851
8 Orc High Priest 842 each 6736
Warlord 1671 (actually it was Roka and he's worth 5,700 by himself but I didn't want to skew things away from an 'average' orc experience)

16453 total

So if you got zero experience for all the 'chaff' AND weren't getting any tougher monsters, you'd still only be losing out on about 30% experience. But in crawl, tougher monsters give much, MUCH more experience than weak stuff, so I think removing the chaff and getting 0% experience, but having lets say 5% of it turn into something tough would actually be worth more experience.

I mean, one orc Knight is worth almost as much as 264 orcs. If 5% of 264, lets say 40 of those lame orcs got turned into Orc Knights and the rest vanished into nothing, that would worth almost as much XP as the ENTIRE normal orc branch shown here. Even if they were just turned into Orc Warriors, that would more than compensate for getting rid of all those priests, wizards, and ogres put together.

In conclusion, we don't need no stinkin' popcorn

[ion_frigate]

On the W2 rune instaescape thing - how about just making it so that getting a rune restores the up staircases on that level, and said staircases immediately return you to the beginning of the branch a la hell portals? That removes the possibility of instant escape, while keeping intact the whole 'no backtracking' thing about as much as you can in a dungeon with multiple necessary branches.

[onget]

If you want to remove all of the popcorn, you will get an empty extend game.

[byrel]

I was source diving the XP-granting code the other day, and it seems to be a quite sophisticated take on estimating absolute difficulty, taking into account how dangerous their spells are, whether they use equipment, their speed, their melee damage, their hp... You might make the threshold based on how much XP that monster will grant. A ballpark guess would be if the monster's worth less than 2.5% of the next XP level, it's popcorn. This would make the following monsters popcorn at the following levels:
goblin 3
hobgoblin 5
orc 5
orc wizard 9
orc priest 9
orc warrior 11
orc knight 14
orc sorcerer 14
orc high priest 15
orc warlord 21
Saint Roka never

yak 12
death yak 16
hydra 17

ogre 10
2-headed ogre 12
ogre mage 16

centaur 11
centaur warrior 16
Nessos 17
yaktaur 13
yaktaur captain 17

ice beast 11

boggart 11
giant orange brain 15
shadow demon 15

Really, the only ones of those I have a problem with is the shadow creature summoners; but that probably means they should give more XP, and the formula should be tweaked to value shadow creatures more highly.

Edit:
To address the notion this would leave the endgame bare:
Fiends never
Executioners 25
Color draconian 19
Draconian annihilator 21
Draconian knight 22
Draconian monk 23
Draconian scorcher 22
Draconian shifter 24
Draconian zealot 21
Tiamat never
Orb of fire never
Tentacled monstosity never
Death cob 17 (lolololol)

Mummy priest 19
Greater mummy never

Flying skulls 13
Jiangshi 17
Revenant 23
Bone dragon never
Curse skull 21

lich/ancient lich never

These are mostly OK too. Zot will be harder if you replace a percentage of draconians with OoF, ancient liches and tentacled monstrosities.

AF_HUNGER is massively underrated in XP calculations; death cobs are much scarier than any single draconian. Hungry ghosts are worthy of more special attention than ice beasts and orc warriors (similar XP value).

Summoners are under-XPed. Curse skulls, mummy priests and revenants are much more dangerous than that (though my experience is in trunk, and the curse skull's speed may up their XP significantly. I'm running off wiki values for XP.)

[ion_frigate]

I'm running off wiki values for XP.

Those are current as of 0.15 (it was done by a bot so barring renamed monsters they will *all* be 0.15, but still) - you probably want to use something more recent.

Though looking through the commits for AF_HUNGER, it does seem like the XP for it was not increased when it was made to trigger even on no-damage hits (or at least, I didn't find a commit saying it was). So it is likely that death cobs are under-XP'd.

[byrel]

Those are current as of 0.15 (it was done by a bot so barring renamed monsters they will *all* be 0.15, but still) - you probably want to use something more recent.

I'd be delighted to; but not so delighted as to create a table for myself using the source. Are there any other available sources?

[ion_frigate]

Haven't been on IRC in forever, but I'm assuming Sequell queries still work? That should give you the version of the monster from the current trunk.

[crate]

disclaimer that I probably won't be around to read replies to this post, since I'm not visiting Tavern much these days

or whatever metric succeeds in removing vast majorities of easy encounters

There are some problems here. First, you need to define what you mean by easy encounter; the problem you run into is that even if we somehow assume that crawl has only one hundred encounters in a typical winning game you must come to the conclusion that almost all of them are less than 1% likely to kill a player who is playing reasonably well (see spoiler).

Spoiler: show

If we assume that there are 80 encounters that have a 99% survival rate, 10 that have a 95% survival rate, 5 that have a 90% survival rate, 3 that have an 85% survival rate, and 1 each with 80% and 70% survival rate, we get players winning less than one percent of the time. And this is assuming only one hundred monster encounters, which is assuredly much too few; crawl with only one hundred monster encounters would either feel very empty or you'd have giant hordes swarming you from across the entire level but not actually being a threat (swarms of "popcorn"). I mean, players tend to visit approximately fifty to sixty floors in crawl wins, so one hundred encounters is about two per floor.

So, removing the easy monsters without making crawl either impossible or really empty is a pretty mammoth task to begin with. You could solve it by changing the structure of the game (if crawl has only five floors then you can greatly increase the difficulty of the average encounter) but that's probably making the game into not-crawl. Making crawl floors very empty is not exactly a good solution, since you're just exchanging one problem for another. We don't want to make crawl into a collection of fifty floors of labyrinths where you have to go search for the one or two packs of monsters on the floor and then get the loot. Well, I certainly don't, and I think most people would agree with me there.

There's also the problem with selecting which encounters to remove. Clearly, different encounters are different threats to different characters. Minotaurs of Trog don't care much about ghost moths, but other characters find them dangerous. Okay, perhaps your goal is just to remove the encounters that, for any character you play, never are threatening at all. That's a good goal, and crawl does still have those encounters. But because of the scale of crawl you either end up removing not very many encounters or you make the game pretty empty. So it's pretty much an unfixable problem, unless you reduce the scale of the game dramatically. You obviously don't want "rat" or "bat" spawning in depths (and they don't do so), but even things like green rat packs in lair might be something you want to keep around.

---

As far as Orc branch goes, the monster generation in Orc branch (capital O) is pretty bad, but I suggest that you look at every other branch in the game (except maybe Elf might have the same problem) instead, and think about those branches. You can (try to) fix Orc later, since it has some unique problems (shared by Elf, but not by any other branch in the game).

One problem with Orc monster generation is that it can't decide if it is aimed at pre-lair adventurers or post-lair adventurers, in what is surely a holdover from the past. Everyone goes to Orc after lair, of course, so probably that would be the place to aim. Even if you do something like uniformly replace "orc" with "orc warrior" in Orc (surely no one is complaining about priests, though I could go either way with wizards) you don't actually end up fixing very much. The proper player response would be to instead just do Orc even later, when orc warriors are completely unthreatening. This exact problem was actually happening in the past with the lair rune branches: they were aimed at a particular time of the game, but no one actually bothered to get the runes at that time, because there was no reason to do so; instead a sizable number of players just got all three runes right before winning. This was a serious enough problem (according to the devs, at least) that it warranted a dramatic change in the structure of the dungeon (runelock)! So, at the least, before you even bother fixing the monster generation in Orc you first must make the reward either required for progression (unlikely to happen) or at least worth taking on a bit of risk for (Elf is probably in an okay spot even though it shares these problems; Hall of Blades is (or at least was? do people go there now?) probably what you get when you do it wrong).

Orc also has big problems from having such a limited monster selection.

---

other comments:

2*: passive ability Sowing the Fields, culled monsters have (piety/20)% chance to be replaced with a monster that meets H's standards.

While I recognize that this is supposed to be a challenge god, encouraging the player to avoid gaining piety is not a good thing (particularly with a piety system where you can reduce your piety at will, like via piety decay). If you want this to be an effect, it should be independent of piety. (Yes, chei has this problem also, and it's bad design, though I guess at least there you can lose a bunch of piety quickly through abilities if you want. Making worshipping the god at all have drawbacks is fine; you get wrath if you abandon so it's a real choice.)

Alternatives are to have it be an always-on effect independent of piety, or to have it ramp up automatically with no way to "ramp down" afterward. The latter is tricky, since you don't want to do things like punish the player for autoexploring a level instead of leaving obviously-uninteresting sections unrevealed, so I'd recommend the former.

---

You might make the threshold based on how much XP that monster will grant. A ballpark guess would be if the monster's worth less than 2.5% of the next XP level, it's popcorn.

This particular metric is bad because it punishes races with a good xp apt. Total xp gained is better to look at instead.

The monsters you give are probably actually a good argument against implementing your particular numbers; from Sequell I get that average lair entry (using playable recent chars) is at xl 11 (so Yaks are gone before the end of lair). Yaktaurs and draconians vanish from the game entirely, on average (V entry is at xl 19, Depths at xl 20, Zot at xl 25). But if you change the numbers this might be okay.

With your particular example of draconians, I personally think that draconians are one of the best encounters in crawl (you can tweak their numbers upward if you think they are too easy; their general behaviour is very good though) while tentmons and orbs of fire are bad (tentmons are speed-10 melee-only monsters, orbs of fire are just a floating gear check). One problem is that you're looking at individual monster xp, though pack monsters (like draconians) should be handled differently from solo monsters (like most things in crawl). Maybe use the entire draconian pack xp as its xp value for your method?

---

new!Wulndraste

Unfortunately crawl inventory management is a huge pain in the ass with these limitations and I don't see a way to fix that. This was the biggest problem with "old" W, based on my experience. If you can fix inventory management and items, this might be workable. The main problem is crawl has a lot of strategic items, and it sort of assumes you will never want to carry them around.

[Sprucery]

Even if you do something like uniformly replace "orc" with "orc warrior" in Orc (surely no one is complaining about priests, though I could go either way with wizards) you don't actually end up fixing very much. The proper player response would be to instead just do Orc even later, when orc warriors are completely unthreatening.

I still think that just replacing (most, if not all) orcs with warriors would be a very good first step, at least. Orc has lots of gold and shops. So there's definitely a motivation to go there rather sooner than later, unlike pre-runelock rune branch endings. Perhaps you would do a couple more Dungeon levels then before Orc if you were scared of warrior packs.

[byrel]

Everyone goes to Orc after lair, of course, so probably that would be the place to aim. Even if you do something like uniformly replace "orc" with "orc warrior" in Orc (surely no one is complaining about priests, though I could go either way with wizards) you don't actually end up fixing very much.

I don't agree with this; my venom mages uniformly go to orc first, because fast poison-resistant evasive monsters (spiny frogs, black mambas) are more dangerous to them than anything in orc, short of a stone giant. Basically if I have great ranged attacks, but will have trouble dealing with mambas, orc is easier.

You might make the threshold based on how much XP that monster will grant. A ballpark guess would be if the monster's worth less than 2.5% of the next XP level, it's popcorn.

This particular metric is bad because it punishes races with a good xp apt. Total xp gained is better to look at instead.

The monsters you give are probably actually a good argument against implementing your particular numbers; from Sequell I get that average lair entry (using playable recent chars) is at xl 11 (so Yaks are gone before the end of lair). Yaktaurs and draconians vanish from the game entirely, on average (V entry is at xl 19, Depths at xl 20, Zot at xl 25). But if you change the numbers this might be okay.

With your particular example of draconians, I personally think that draconians are one of the best encounters in crawl (you can tweak their numbers upward if you think they are too easy; their general behaviour is very good though) while tentmons and orbs of fire are bad (tentmons are speed-10 melee-only monsters, orbs of fire are just a floating gear check). One problem is that you're looking at individual monster xp, though pack monsters (like draconians) should be handled differently from solo monsters (like most things in crawl). Maybe use the entire draconian pack xp as its xp value for your method?

[/quote]
I agree this shows this metric is bad, so the numbers need tweaked down a bit. And yeah, based off of XP gained instead of incorporating XP aptitude.

Specifically though, I'm entirely OK with yaks disappearing by the end of lair. They should generally be there at the beginning, but start being replace by death yaks, hydras, etc. by the end. Yaks tend to be popcorn at lair end.

In zot, I disagree. I almost always find draconians trivial and OoF terrifying. Even if I have rF++ (which I don't always). Draconians are trivialized by invis, so I just pop a potion or wand charge if the fight looks at all hard, and they die.

[Sprucery]

my venom mages uniformly go to orc first, because fast poison-resistant evasive monsters (spiny frogs, black mambas) are more dangerous to them than anything in orc, short of a stone giant.

...and I think that even if plain orcs were replaced with warriors you could still go to Orc first. Currently OTR makes orcs drop like flies. Mephitic Cloud destroys orc warriors (and all orcs easily except warlords) anyway.

[Siegurt]

my venom mages uniformly go to orc first, because fast poison-resistant evasive monsters (spiny frogs, black mambas) are more dangerous to them than anything in orc, short of a stone giant.

...and I think that even if plain orcs were replaced with warriors you could still go to Orc first. Currently OTR makes orcs drop like flies. Mephitic Cloud destroys orc warriors (and all orcs easily except warlords) anyway.

No, you could not. I have tried this when a pair of orc warrior happened to be at the entrance to orc, it doesn't work, orc warriors take far too long to die to do this before you have lair exp when there are only two of them much less when there are 6-8 and you can cast otr probably twice and meph twice.

[Sprucery]

No, you could not. I have tried this when a pair of orc warrior happened to be at the entrance to orc, it doesn't work, orc warriors take far too long to die to do this before you have lair exp when there are only two of them much less when there are 6-8 and you can cast otr probably twice and meph twice.

Hmm, maybe a whole pack of warriors would be too much at that point indeed. Then I guess VMs would have to do as much Dungeon as possible and then decide whether Lair or Orc is the better option.

[byrel]

my venom mages uniformly go to orc first, because fast poison-resistant evasive monsters (spiny frogs, black mambas) are more dangerous to them than anything in orc, short of a stone giant.

...and I think that even if plain orcs were replaced with warriors you could still go to Orc first. Currently OTR makes orcs drop like flies. Mephitic Cloud destroys orc warriors (and all orcs easily except warlords) anyway.

Some venom mages could, some couldn't. Depends on what else you have going for you. You can usually handle 1-2 warriors, by kiting them in melee range with OTR and sting. If you have enough defenses (say, you stopped training magic when OTR was castable and went pure dodging+fighting, which I do sometimes) you could concievably handle having all the orcs replaced with warriors, with a ton of stair dancing. Vehumet VMs would be screwed; they rely on the extra MP from the free popcorn kills to allow them to keep training magic skills.

But the suggestion was never they all are replaced. piety/20 maxes out at a 10% chance. So if you removed all the orcs, and replaced 10% of them with warriors, yes, a VM could still handle it.

[Sprucery]

Vehumet VMs would be screwed

Removing popcorn monsters would be a nerf to at least Vehumet, Makhleb and Fedhas... Gozag too I guess.

Edit: sorry for derailing the thread, this isn't really discussing the proposed gods.

[archaeo]

crate is basically right, but a couple of responses (even if he doesn't see them, thanks for stopping by crate):

Making crawl floors very empty is not exactly a good solution, since you're just exchanging one problem for another. We don't want to make crawl into a collection of fifty floors of labyrinths where you have to go search for the one or two packs of monsters on the floor and then get the loot. Well, I certainly don't, and I think most people would agree with me there.

I think the problem of "few monsters in big levels" would end up mostly disappearing as players hit o. Otherwise, I don't really see why removing a healthy chunk of popcorn monsters would be that much more annoying than Chei's slowness or Qaz's noise. I think lots of people agree that those gods aren't much fun, but they have their fans too, and I see enough players complaining about popcorn encounters that it seems like a worthwhile idea to try out.

That said, it's totally clear that HD and XP are both probably not great metrics, and it's not obvious to me what would end up working. Nor do I have particularly good ideas for what kinds of monster encounters should be culled as popcorn. It's a hard problem.

Unfortunately crawl inventory management is a huge pain in the ass with these limitations and I don't see a way to fix that. This was the biggest problem with "old" W, based on my experience. If you can fix inventory management and items, this might be workable.

A powerful (but dangerous to overuse) branchport ability and a sensible landing zone when you've picked up runes would probably be enough to alleviate these problems, imo. Some inventory issues remain, but "don't stash" is a conduct players play informally all the time. Goldified ammo and reduced permafood types (or food removal) would help a lot too.