Dungeon Crawl Reference

This FR attracted 36 comments in 9 days :)
The mutation removal is in. We are planning to modify Recite and to add
Imprison (thanks for the FR and the comments!). With some luck, perhaps
already 0.6 but no promises.

Hmm. Thanks jpeg. All around, doesn't seem too important whether a
creature is normal or high, so it might be ok to move some of the uniques
up (notably, the ones that are casters..) if it would aid the idea of
recite working great on normal intelligence creatures(by keeping it from
making almost all uniques a piece of cake, so it wouldn't be ridiculous).
Would be nice to hear opinions from people on that though.

Megane

Monster intelligence isn't terribly important right now. Off my head I can
think of three situations where it has different effects:

1. Dungeon knowledge:
Highly intelligent monsters will try to step around known traps. Stupid
monsters won't recognize secret doors and will walk through harmful clouds.
The more intelligent a monster the greater its pathfinding range, with
monsters of normal intelligence having a range of 8 (los) whereas highly
intelligent monsters pathfind through the whole level. Also, patrolling
monsters will always be capable of returning to their starting point
(assuming them to be familiar with the level layout).

2. Good god restrictions
All good gods handle monsters differently according to their intelligence,
both in abilities (Recite, pacification) and conducts (Zin: eating corpses,
TSO: attacking fleeing monsters).

3. Group tactics
Monsters of animal/normal intelligence will step aside for monsters of the
same type for combined attack power. Stupid monsters will never step aside,
and highly intelligent monsters will even make space for monsters of
another type and will step aside to make space for an attacker further back
in the corridor that is capable of ranged attacks.

I made a list of natural/normal and natural/high creatures. Also, most
uniques fall under natural/normal, and it might be good for a number of
them to be high instead. Question is just how else does intelligence
factor in, in crawl?

What this list doesn't show, is how many other creatures there are that
aren't affected. It would be good if recite worked extremely well on the
normal creatures, but depending on balance, it could maybe also work
decently, or even well, on high eventually? What's important, in my
opinion, is for there to be some really clear cases where using the ability
is good.

Also, I may have missed a few, not sure..

natural/normal int:
orc wizard/priest/warrior/knight/warlord
hill/stone/fire/frost giant
(deep/iron/rock) troll
kobold (demonologist) (** maybe demonologist->high)
(hob)goblin
orc
gnoll
boggart(** -> high maybe?)
centaur
big kobold
naga
ogre
ettin
harpy
naga mage
yaktaur
siren
mermaid
cyclops
merfolk
vault guard
minotaur
orb guardian

natural/high int:
all deep elves
all draconian
centaur warrior
yaktaur captain
shapeshifter
elf
guardian/greater naga
orc sorcerer/high priest
wizard
titan
hell knight
necromancer
great orb of eyes (?)
ogre mage

Megane

Recite already only affects monsters of above animal intelligence, but
intelligence (normal vs. high) currently plays no role in the calculation
of chances or results.

You're right that intelligence isn't very fine grained. But, there is
enough to differentiate ogres and elves. (Ogres are normal ; all elves are
high ; Ogre-Mages are high). I had forgotten that monster intelligence is
this restricted.

But.. there is:

Normal, High, Plant, Animal, Insect

If you limited it to only working on Normal and High (but much better on
normal - maybe no chance of angering normal, significant on high), and also
restricted it to natural creatures (then avoiding stuff like normal
intelligence demons, undead, etc)..

Then it might work pretty good. Would probably require a word about in the
description of the power, and a line in view monster screen about their
intelligence.

Would result in a tool that's great against just some subsets of creatures,
which is basically standard design in crawl. I'll try to pull out a list
of notable creatures of normal/natural intelligence soon - my glancing at
it makes me think this could be reasonable. Stone giants are normal, all
orcs but for sorcerers and high priests are normal too...Fire/Frost/Stone
giants are normal (titans are high)...

Megane

Megane: that sounds interesting but I am not sure if monster intelligence
is fine-grained enough. (Enough, to differentiate between an ogre and an
elf, say.)

I don't know how it works - I agree that checking MR is odd.

Kotk summarized very well why I don't like recite. It's useful against
speed bumps, and useful for getting you killed against things that pose any
challenge. It's a grind tool for invocations, and a rather annoying one to
use at that.

To second others: non magical. Forget hit dice ( I don't know if this is
currently the case or not). Make it just a check of power vs int, and make
it work very reliably in certain ranges. And make it obvious that it works
this way. So then, people will think ' ah, look, this creature is very
dangerous, but very dumb, so he's a perfect target for recite!'. It might
be all but useless in the Elven halls, but on the other hand, it may be an
awesome tool against orc warlords, stone giants, and ogres. I think that
would be totally reasonable, and gameplay-wise, actually interesting.

Megane

Book of Tax Law - monsters sleep or drop HDx5 gold. Limited uses.

Recitation checking against MR seems a little unusual... I imagine the base
form of recite as an entirely non-magical effect. I imagine it's like a
controversial speaker at a college campus. Some people will get angry, some
convert, some leave, some go to sleep, and some just get confused.

I did mean range of MR of monsters that recite affects at current power.
For example when MR of goblin is less than lower power limit of recite then
goblin becomes indifferent like animal to that speech.

Similarly audience check that is done before and during recite. That may
be made more sophisticated. Currently it checks only lower limit
(refuses/stops when present are only animals and already affected normals).
It can check also upper limit (if listening are only way too resistant
intelligents). That gives better feedback to novice priest and lowers the
need for spoilers.

kotk: I agree that the power is too low.

When you "range", do you mean distance to the player, or range on a
power/MR scale?

dpeg

dpeg: Recite is narrow because of its low power. It works kind of reliably
... in sense that unaffected is maybe one from 8. It has desirable effects
with low hit dice humanoids with low magic resistance. With others it has
undesirable effects. That turns it kind of useless. It is not ogre who i
want to risk hasting early so i recite to goblins and plain orcs who i do
not fear anyway to train invocations and that is grind. When i am
experienced enough to recite ogre to sleep then ogre is also among easiest
monsters that i face.
Two ideas (may be even combined) to theme that Zin does not care about
stupid:
a) Make recite to affect highly intelligent monsters better.
b) Make recites power higher, but also more picky about audience it
affects. In way that lecture has range. Creatures in middle are most
affected below range are unaffected and above range get "angry". That way
you get more unaffected monsters and less angries.

?? animal skin +0 is too much reliability?

kotk: Too much reliability is bad (see Okawaru); some unreliability is good
(Makhleb). The point is find a good balance. As you can see, berserk
happens very rarely. Part of the problem is that you cannot react with god
powers when something goes wrong (this is different to Makhleb or Lugonu,
for example). The prison would help here. Additional Recite effects (I
outlined a few) would also help, as would Recite training Inv a bit faster.

One comment on the prison: ToD style may be too much and would also lead to
abuse (players imprisoning hapless monsters in order to create barriers
etc.) However, we could use Sanctuary-style colouring to depict prisons:
.....
.---.
.-X-.
.---.
.....
where . is floor/feature (normal colour), - is also floor/feture (prison
colour, e.g. darkgrey) and X is the monster.

I like daftfad's book idea a lot because it gives a nice way to link Recite
and Prison. Here is how I see it:
* Zin gifts books of law from time to time. (Priests should start with one.
Once we have presents upon taking on gods, new followers of Zin should also
get one.)
* Reading from such a book is actually Recite [Law], i.e. a special case of
Recite. And recite without a book is just the generic version.
* Each use of a book has a chance to destroy it (as with manuals).
* Reciting from a book increases power (you have more legal force),
restricts the player's sermon to the topic of the book and restrict Recite
outcomes.
* Monsters not obeying can be easier (perhaps guaranteed) and possibly
cheaper imprisoned.
* Usage has no drawbacks, but uses are limited.

Typical laws could be No Mutation, No Cannibalism, Don't Be Greedy etc.
I am against throwing books. (In fact, I want to make all books to be
thrown awkwardly, for Trog purposes.)

cloture: I agree that Zin needs more early game appeal. But I don't see
what's wrong with increasing late game appeal as well. A one off power with
double-edged effect seems interesting to me. (And so fitting.) Note that we
want a new late power as a replacement for Vigour which has been stolen
from Zin.

dpeg

As much as I like the book of law idea, I can't come up with any
particularly satisfactory implementations. About the best I can come up
with is a "book of life" which gives every creature & the player on the
screen regeneration, while damaging undead.

Perhaps a book of considered thought - every creature aside from @ gets -2
speed.

I just can't come up with an specific ideas I like.

In regards to giving a one-time mutation removal gift... I'm not
particularly excited about weighting Zin towards the end-game even more,
that's why I propose starting in on the lesser, external mutations early
on. It would be great to get something to deal with a few of those Ijyb
with poly other wand mutations. Zin's play is already weighted way to
heavily towards full piety, as is his rMut - and if whatever NEW power we
get is used often enough it will keep piety from building up to necessary
levels even more. On the other hand, if it gets used that often, the
starting game is probably more interesting.

@Daftfad- I actually really like the first half of the books of law, the
conducts rules, with perhaps an auto-imprisonment or auto-sleep for
failures to obey.

-tapicell

One comment on original FR:

You should not be able to harm enemies that you have imprisoned. That seems
a bit un-Zinlike (imprisoning someone then brutally murdering them while
they are helpless). Massive piety penalty for this. Also, you should not be
able to imprison anyone who is not evil or who has not yet attacked you.
The idea is that as a follower of Zin you are an agent of law enforcement,
and attacking you should be considered a crime. At any rate, instead of
creating a physical prison, just permanently paralyze the monster.

Other Zin ideas:

Here's a silly idea: Zin occasionally gifts you 'books of law'. Reading
them imposes temporary conducts. Violating these conducts causes penance.
Monsters which break these conducts can be imprisoned (a useful example: a
law that prohibits spellcasting). Reading a book trains invocations
significantly and (increases piety?). Also, you can hurl the books for
significant damage.

Reliability is fine i think. If audience got hasted, went battle frenzy or
berserk during lecture, then that is quite likely hopeless audience at the
moment.

If goal is to have it useful to end game then either its power has to be
boosted (it is up to 50 now), or opponents magic resistance affect lowered.


Other thing (that is probably bug) Options.hp_warning is involved in
decision if lecturer stops recite in middle of it or not and in a very odd
way:
(you.hp*Options.hp_warning <= you.hp_max && delay.parm2*Options.hp_warning
> you.hp_max)
I read it like it is advisable to set hp warning to 2, then if you start
above half health and drop below half health during reciting it stops.
Odd i said because that hp_warning is commonly used as percentage of health
when one gets OUCH messages and can be set in init.txt.

Megane: I think that it's ally play is not fully out of the question with
Elyvilon -- think of wands of enslavement or summon ice beast etc.

On Recite: I've ran a MiPr today. Here are some numbers:
1479 | D:2 | goblin confused
1537 | D:2 | kobold neutral
1694 | D:2 | hobgoblin confused
1933 | D:2 | gnoll berserk
1966 | D:3 | kobold confused
2404 | D:3 | hobgoblin unaffected
2445 | D:3 | hobgoblin confused
2576 | D:2 | goblin asleep
2624 | D:2 | gnoll unaffected
2663 | D:2 | gnoll confused
2697 | D:2 | gnoll confused
2803 | D:2 | hobgoblin asleep
3010 | D:3 | hobgoblin confused
3042 | D:3 | same hobgoblin asleep
3062 | D:3 | orc confused
3119 | D:3 | hobgoblin asleep
3128 | D:3 | hobgoblin now confused, orc asleep, orc priest
unaffected
3140 | D:3 | two orcs asleep, one orc confused
3331 | D:3 | kobold asleep
4049 | D:4 | two gnolls confused
4066 | D:4 | one gnoll hasted
4073 | D:4 | another gnoll unaffected
4231 | D:4 | kobold asleep
4244 | D:4 | another kobold unaffected
4263 | D:4 | orc confused
4345 | D:4 | Sigmund hasted
4388 | D:4 | goblin neutral
4429 | D:4 | goblin unaffected
4469 | D:4 | goblin and Sigmund unaffected
4642 | D:4 | Jessica confused
4916 | D:5 | kobold (with poison needles) neutral
5241 | D:6 | orc band. Two orcs confused, one neutral. Orc wizard not
affected (but half dead)
5415 | D:6 | hobgoblin unaffected (only one turn), centaur hasted
5618 | D:5 | kobold asleep
5661 | D:6 | ogre hasted
6411 | D:5 | hobgoblin confused
6502 | D:5 | hobgoblin neutral
So it works very well on the tiny critters. When I met the first gnolls, I
used Recite in a good position: @hg so the gnoll couldn't reach me, berserk
or not. In other cases, it's good to Recite on the stairs.

Ultimately, Recite faces the same problem as Glamour did: it's unreliable.
That's not the end, we could do something about it:
a) make the bad effects more rare
b) no bad effects, lesser good effects, can Recite each monsters only once
[1]
c) better positive effects [2]

I think the goal should be that Recite has uses until the end of the games.
It should work best against groups (where affected monsters can get in the
way of non-affected monsters). Perhaps the three turn duration is too
long.

[1] This may already be the case.
[2] I'd really like: "Moved by your words, the [foo] kills itself with
[weapon]." (You get credits and xp.) Or also neutral + berserk ("[foo]
breaks free from a live of oppression."). Or "[foo] drops all weapons.". Or
"[foo] prays." (does not move, is awake, becomes strictly neutral).

I've been wondering why recitation holds you in place. (I'm not doing very
well in finding a FR that explains Zin - I might have to check CRD).
Certainly there would be balance issues just having it go on while you go
about your business (running away, casting, paralyzing berserk monsters),
it might be useful to let you be effectively slowed while reciting, or
moving at normal speed while disabling all other actions.

Gods don't need to be better than other gods to have a purpose, so long as
they are reasonably close. Variety is meaningful.

Yeah ; always been a little annoying that having allies is not doable for a
healer. I've still only played ely on a deep dwarf, but I think that
overall ely would be effective (partic. swapping out for shining one for
say elf:7, vault:8, is something not available to other gods).

Back on Zin though;
Imprison would help a lot. making recite not so suicidal would too. Not
sure what to do with recite though - can't completely remove the negative
effects, and the positive effects just aren't very good. Basically , don't
know how to make this into an interesting ability gameplay-wise.

Megane

nobody: I you keep being unable to sign your (close to worthless) comments,
I assume you are trolling by intention.

dpeg

Healers have no way of getting allies, so the concept does not well work in
crawl. Where are Ely and Zin's powers suppose to be useful anyway? There
are better gods for every situation.

The idea of a Healer class is certainly not unique. In Crawl, the idea is
that you can heal yourself, your allies and (now) also your enemies.
Pacification is an interesting mechanic that works surprisingly well, if I
may say so.
The strategcial idea is to have a number of good gods, allowing players to
adapt (this is unique for the good gods). Flavour-wise, the classical ideas
of being "good" have been split up in three: fighting evil (TSO), being
pure (Zin), healing (Ely). More good gods will come but that basic division
will stay (for quite some time, at least).
Enough said. dpeg

How does it even make sense that the god of purity and god of healing are
separate?

I'd be happy to discuss Elyvilon - after playing a deep dwarf healer I'm of
the impression that pacification doesn't cost you any significant
experience - but I won't go in any detail in a Zin FR. Something awful or
IRC pehaps? Sadly, I don't subscribe to CRD _yet_ .... I will just say that
plenty of other gods and play styles deprive you of experience. Yred's
zombie hordes, for instance?

~Lord Sloth

There's nobody who needs healing in the dungeon, since monsters don't fight
each other. Pacification costs experience, which no other god deprives you
of.

borsuk: Same here. This is why I like Prison better than Pestilence.

nobody: Elyvilon has been changed a lot recently. The healer and
pacification themes have been stressed. I suggest you try Elyvilon in 0.5
before (unspecifically) suggesting that "she probably needed some changes
anyway."

I have some trouble comprehending how does spreading filth and plague fit
into Zin (god of purity).

b0rsuk

What if holy was added as a magic school and Zin was the god of holy magic?
There are currently no good gods of magic and this would allow him to give
to grant TSO and Ely abilities without crossing any lines. Although, this
would mean overhauling Ely to be the new priest goddess, but she probably
needed some changes anyway.

I don't think we want classical summons from Zin. A number of gods do that
already (Trog, Makhleb, Yredelmnul, TSO, to some extent Lugonu, Beogh,
Kikubaaqudgha, Elyvilon) and one of the reason for overhaul Zin was the
parallel design of Zin and TSO. This is also a reason why I am not too
happy about reinstating Pestilence.
That said, the golem ideas sound interesting. Will look into them -- this
could definitely be used somewhere.

Megane: Yes, I know that Recite is too weak. We fell into the Glamour trap
:( I like the basic idea of Recite -- you're preaching to the unwashed
masses. Berserk seems to occur too often. My hope is that the power can be
saved with finetuning (instead of scrapping it completely).

b0rsuk: Divine Leash sounds very interesting, I don't think it conflicts
too badly with the Prison, especially if we apply a ToD type condition.
(Because then the prison would work better in the open, and the imprisoned
monster would be safe from you and safe to you -- at least in most cases.)
I like the symmetry between Sanctuary and Prison.

dploog:

Imprisonment fits much better to a hunter god. In fact, it's one of more
prominent ideas I have for him.
I called it 'Divine Leash'.
It's roughly equivalent of Banish. No MR check, and possibly works on all
creatures (It's divine, after all !). Monster can still act as usual, but
it can't move further than X squares from one it was when it was leashed.
Monster can still move within the radius, attack and use all abilities
except perhaps Blink. I think I can get away with such powerful ability
working on most or all creatures - it doesn't prevent monster from
attacking, shooting, casting spells. It's great for a playstyle-agnostic
hunter god. It's about imprisonment, but the creature can still be attacked
using whatever cheesy tactic you choose. Ranged attacks, poison, etc are at
advantage, but they're not *required*, as with most hunter=ranged god
proposals. The duration of Divine Leash should be quite long, but not
infinite. To encourage risk taking - leash one orb of fire and try the
other route... only to be attacked by it a couple of hundred turns later
when you already forgot about it.

b0rsuk

Zin is the god of order right? Maybe he should have control everything
powers then. Although, this would turn Zin into the god of making
everything fight everything, which is extremely chaotic.

I find Zin unsatisfying. Recite burns a lot of time and has so much more
potential to get yourself killed than it has potential to do anything (in
my experience). Also a problem because this is the way to train
invocations for Zin characters. It leads one to think that it must give
piety, but it doesn't, and if it did, it would clash with Elyvilon.
Vitalisation clashes with Elyvilon as well, but also is usually pretty
minor. Similarly, mutation resistance is pretty minor. Sanctuary is cool,
but the piety model seems to limit it (and any other powers) a lot.

Imprisonment would be very nice - reliable crowd control always is. I
think Zin needs at least something like this, but probably more (I like
pestilence more than other pet ideas). Maybe I missed something in looking
over this god, but that's my 2c. Old zin was great, although overpowered.

Megane

Not that I think Zin should get permanent followers - why? That would make
him more useful, sure, but less unique and maybe a bit blander. That said,
if you could find a way to put a unique, or at least different twist on it,
that could be interesting.

If Zin did give pet(s), the most suitable one would be golem(s).

Less exciting route: start off Zin summoning wood golems. They weren't
alive in the first place, so it's not a big deal when they die. Later,
increase the capabilities of your creations.

More inspired route: Crib from various myths, legends, and beliefs
concerning golems, Jewish in particular. The wikipedia entry alone offers
quite a few interesting leads, flavour, and possible behavior or drawbacks.
I particularily like the idea of having one golem (possibly with
regeneration) that grows stronger over time (tied only to time, not kills),
but also less mentally stable. You'd gain the ability to deactivate your
golem - and it would be necessary. This ability would cost current health
proportional to just how powerful AND unstable it has become (if they're
tracked separately). You could easily keep up a steady stream of minor
golem companions, but letting them grow strong is more risky and rewarding.
Also, maybe just as dangerous as recite.

I think he should have permanent followers like the other priests, but then
he would need Ely's healing abilities to maintain them, and that's no good.
Combining Ely and Zin would solve this, but then there would be only two
good gods.

Thanks for the FR!

I suggested Zin mutation removal as well. Recently, I have changed my mind
to a simpler model: at maximal piety (******), you get the one-off ability
to clear all mutations. (This includes good mutations although pure stat
mutations, e.g. Dex-3 Int+2 will stay.) This is double-edged (as are Recite
and Sanctuary) but it is a very good power. Again increases the lategame
appeal of Zin, of course.

There was the idea to resurrect Pestilence from old Zin but I like the
Prison better. Some brainstorming:
Prison could always work but as a Tomb of Dorokhloe type of effect: you ask
for it and the target has to have all adjacent squares wall or empty floor
(no monsters, no water/lava). The prison is temporary (as is the Sanctuary)
and its walls consist of (some odd colour) glass walls. This means that you
can recite at the monster or the monster could still smite at you.
Attacking a monster in a prison is of course a no-no. This power would be
interesting up until Zot:5.

Recite needs some tweaks, I think. After my early game testing I'll turn to
Zinners. Feedback on Recite is welcome already now.

dpeg