Positional Attack Spells


Although the central place for design discussion is ##crawl-dev on freenode, some may find it helpful to discuss requests and suggestions here first.

Dungeon Master

Posts: 250

Joined: Thursday, 27th November 2014, 19:12

Post Sunday, 20th January 2019, 18:49

Positional Attack Spells

I'm going to break off this thread from the needles thread to discuss tealizard's suggestion in that thread that positional attack spells be considered instead of arbitrary targeting. This is by no means a full endorsement of the idea so do not take this thread as a devteam sign off that we'll definitely do whatever is written, but it's interesting and I'd like to see what comes out of fleshing out specific ideas for positional attack spells. As a working definition a positional attack spell is a non-targeted spell which interacts with the player's position (including in this the player's tactical status, hp, mp, status effects) for determining its effects. Some flexibility is needed for this definition: the spell can be targetable but still require good positioning; for the purposes of this discussion assume the ui makes the right auto-targeting choice when in the correct position.

tealizard took an extremist point of view: remove all targeting. I'm skeptical of this point of view since it's so intertwined with automation and ui improvements, but I'm willing to bet there are some good spell ideas (new, revision, or replacement) in specific examples.

Targeting vs positioning Ranged attacks care less about the position, since many monsters are less dangerous at range, and targeting takes no time, which limits the tactical relevance of positioning when using a ranged attack. Changing position takes time, so using positional spells effectively requires changing the tactical situation which comes at a cost since monsters react.

Regarding automation As pointed out to death in that thread there is already automagic and Lua targeter control. Automation and ui improvements are an orthogonal issue to "what tactical depth can be provided by positional spells", so let's leave it aside in this thread. If you've got an idea for a targeted spell that usually has a "best" target that requires positional set up, just assume the ui figures that out for the purposes of this thread.

Regarding summons You could make the argument that summons are "positional" in that the player has allies to position around and with respect to. With a few exceptions these should be left aside in this thread.

Facing Adding facing to crawl would be bad. Don't suggest that.

With all that said, have at it, pitch some interesting positional attack spells, generate ideas. Two ground rules: one, single target but auto-targeting is not "positional", it's a ui improvement; two, I'd like the focus to be on spell ideas, try not to focus too much automation/ui concerns in an idea unless you're certain they'll screw things up for the spell.

To kick things off, here's my list of crawl's positional attack spells with some notes, to give some examples to keep in mind.

  • Tornado
  • Ozo's Fridge
  • Olgreb's
  • Ignite Poison
  • Discord
  • Shatter
  • Ignition
These are all AOE attack patterns, mostly with LOS radius. Tornado has some terrain constraints. The positional impacts of these spells are "can I manage the drawback (noise, -Potion, being in the open, having a frenzied monster target me) and are the monsters in the correct place to get good effects from my cast".

  • Freeze
  • Vampiric Draining
  • Sticky Flame
  • Static Discharge
  • Irradiate
These are melee attack patterns. Three are targeted melee (room for automation here to make their ui more like melee weapons and reduce keypresses, but we're not focusing on ui), and the other are melee range but untargeted. The positional considerations with these spells are essentially the same as those with melee.

  • Conjure Ball Lightning
  • Chain Lightning
These are autotargeted patterns. The spells don't use the targeter, but don't (necessarily) hit every monster in LOS. CBL requires some open space and distance from the target because of the self-damage. Chain Lightning has its own considerations to make sure the desired target gets the mega damage.

  • Shock
  • Lightning Bolt
These are targeted spells, but if you want to maximize their use then you want to bolt bounce, which means working out relative positions of yourself, monster, and wall. The positioning does lead to tactical choices "do I abandon a better defensive post to get a better killshot on monster X, or take two casts and stay where I am?" but often also needs a lot of manual targeting. The latter aspect is an automation question, so for now assume that an "optimal bounce autotargeter" exists for the automatically selected default.

  • Summon Forest
  • Malign Gateway
  • Summon Lightning Spire
These are terrain altering summons. Unlike other summons, they are (mostly) fixed in place. Malign Gateway currently isn't that tactically deep, since it attacks out of LOS. Making it only attack in LOS would be a nerf, but also make it tactically interesting: the player must contend with the chaos effects on monsters and judge how long to stick around once it turns. Summon Forest has the dryad, which has its ally positioning that I'm leaving aside in this thread, but it also turns some walls into trees, can change los, and place water, which means that taking advantage of it after its cast requires maneuvering. SLS is targeted, but won't move once placed. The placement choice does provide it with some tactical flexibility, and as a situation evolves a player may want to move relative to it, so I think it works as a good example of how a limited targeter can play a role in a positional spell.

Finally, some spells with positional considerations that aren't translocations or attack spells, just to keep in mind.

  • Silence
  • Darkness
  • Ozo's Armour
  • Swiftness

Vestibule Violator

Posts: 1508

Joined: Monday, 21st November 2011, 07:40

Post Sunday, 20th January 2019, 19:45

Re: Positional Attack Spells

A few more examples of positional spells/god abilities:
  • Ring of Flames
  • Cleansing Flame
  • Heavenly Storm
These are very small AoEs, in contrast to the nearly full-screen AoEs you mentioned, and they're relatively rare and somewhat more positionally complex.

Edit: I guess they're functionally quite similar to your second category
Usual account: pblur on kelbi

Vestibule Violator

Posts: 1508

Joined: Monday, 21st November 2011, 07:40

Post Sunday, 20th January 2019, 19:57

Re: Positional Attack Spells

A bomber-man inspired spell; drops at current location, and fires a beam in each of the principal directions after a couple turns.

A lot like Fulminant Prism, but with lower depth and lower UI complexity.
Usual account: pblur on kelbi
User avatar

Zot Zealot

Posts: 982

Joined: Monday, 29th September 2014, 09:04

Post Sunday, 20th January 2019, 20:32

Re: Positional Attack Spells

A homing missile type effect which just hits the closest target.

Ziggurat Zagger

Posts: 8786

Joined: Sunday, 5th May 2013, 08:25

Post Sunday, 20th January 2019, 23:15

Re: Positional Attack Spells

I would start by looking at changing existing spells. byrel suggested a Fulminant Prism-like spell, but why not just use that gimmick on Fulminant Prism itself?

In fact, I think a good starting point for this would be to revamp the book of Conjurations. Except for Magic Dart, the spells in it are pointlessly overcomplicated, ridiculously so for a hints mode background. Here's where I'd go with that:

Searing Ray
If there's one spell crying out for a "homing missile type effect", it's this one. The way it currently works is ridiculous - to use it effectively at angles other than the 8 obvious ones, you have to know how the pathfinding of the monster(s) will alias. It should either pick a monster upon casting and fire each beam at that monster (even if it moves), or pick a new target each turn.
Also, make all 4 beams have the same accuracy/damage/penetration.

Dazzling Spray
I don't think this should even be in the book, it belongs on En not Cj.

Force Lance and IMB
IMB's wacky-shaped explosion was added in an attempt to differentiate it from Throw Icicle, Stone Arrow, etc., but the explosion never does anything but make noise. It's not guaranteed to explode and even when it does the explosion has a 25% of failing to affect any given square anyway.
Force Lance's gimmick is knockback, but the best use of the knockback is to just knock the monster into a wall because that does extra damage.
So the result is that these spells end up being just like Icicle/Stone Arrow/each other.
My suggestion is to remove these spells and introduce a new spell, "Iskenderun's Force Blast". This spell creates a 5x5 explosion of force centered on the caster that damages monsters and knocks them back, significantly further than Force Lance currently does, like 4 or 5 squares (also can knockback stop depending on size please? thanks). This is a lot more different from Icicle/Stone Arrow, and combines well with Searing Ray. It is somewhat similar to Irradiate, though...
Replace Force Lance/IMB in monster books with Icicle or Stone Arrow or Iron Shot or whatever. Also, either remove collision damage or mention collision damage in the spell description.



Also, Conjure Flame
Conjure Flame should create the flame cloud underneath the caster rather than have targeting. Give the caster immunity to the cloud until they step off of it. This solves several issues with the spell (refreshing it, using it to prevent a monster from ever entering LOS) and also makes it less crazy overpowered.

For this message the author duvessa has received thanks: 6
nago, sdynet, Sprucery, tealizard, VeryAngryFelid, Yermak
User avatar

Vaults Vanquisher

Posts: 454

Joined: Thursday, 1st November 2018, 02:33

Post Monday, 21st January 2019, 02:21

Re: Positional Attack Spells

I agree that changing existing spells to have little (e.g. melee range only) or no explicit targeting is the best way forward.

One thing to concentrate on is trying to make certain obvious tactical positions less advantageous, e.g. corridors and diagonal gaps. Certain homing-type mechanics already do a reasonable job of that. OoD does too, though OoD could easily just do homing without targeting. The bomberman-style fulminant prism idea also encourages fighting in open spaces to some extent, particularly if the blast effect reaches far enough that the player can't escape it in time without diagonal moves.

A number of existing spells and suggestions already have LoS with limited range mechanics, but LoS with minimum range has not been used. For example, the AOE force lance suggestion might be improved if it did not affect a monster at melee range.

You might also base targeting on terrain, for example you could have effects that travel along a contiguous formation of wall tiles and require a position on or near a wall to activate. For example, you could imagine a version of LRD that blows up one or more tiles near monsters along position-selected walls (or perhaps all sufficiently close walls). Dividing the effect among multiple walls could discourage use of corridors and would pair well with the weird AC mechanics of some earth spells.

Regarding hexes, making dazzling spray a short-range AOE effect would be a significant improvement. I would also like to see a merge of the confusing touch and confusion spells. Melee-triggered hexes are a pretty good model in general, I think.

edit: To expand a bit philosophically, there are possible targeting modes that would be awful with explicit targeting but work with implicit targeting. I would include cone effects here. For example, consider a narrow cone attack pattern that can be interrupted by walls, in other words you have a breadth-first ordering on the set of tiles the effect can hit and you disqualify all tiles past the first one that's a wall. If you could manually target this thing, like glaciate (which already has bad targeting), you could mostly eliminate the wall-interruption effect through careful fiddling, but if it always targets a particular way, for example say it always centers the susceptible target with most hp or HD or greatest size or nearest to the player or whatever, then you have something because sometimes you'll hit a lot of stuff and sometimes you won't and this is all a matter of position and terrain. This sort of cone effect also penalizes corridor fighting. A spell like this would make a good replacement for bolt of cold, imo, flavored as something involving ice crystals.
This is where mechanical excellence and one-thousand four-hundred horsepower pays off.

Slime Squisher

Posts: 392

Joined: Sunday, 11th September 2016, 17:21

Post Monday, 21st January 2019, 06:25

Re: Positional Attack Spells

Searing Ray would benefit greatly from some sort of homing effect, but thought needs to be given to how it should behave if the initial target dies or shuffles out of sight for a turn, to whether the spell should prioritize the initial target over the chance to hit multiple targets, and to whether it should try to avoid friendly/neutral creatures. My preference would be to maximize the number of hostile creatures hit, then to avoid friendly/neutral creatures unless a friendly or neutral create is the initial target, then to aim for the initial target as a tie breaker, then aim for whichever option has the lowest HP, moving on to 2nd lowest if they are tied for lowest, and finally, choosing randomly if there is still a tie. But, that sounds like a pain to code. Also, it should probably ask for confirmation if there's a friendly/neutral anywhere in LoS when casting the spell.

Dazzling Spray is like a weird hybrid of a Conjurer spell and an Enchanter spell, so it makes sense that it appears in both starter books. If it were redesigned to remove the damage component, I'd say it should only go to Enchanter, and if the blind effect were removed, I'd say it should only go to Conjurer. I don't think it's a bad idea to give Enchanters a means of doing damage at range or Conjurers a magical means of defense, however, so I'm fine with the spell as-is.

The best use of Force Lance isn't always knocking enemies into walls. If you can kill an enemy by knocking it into another enemy, that extra damage to the 2nd enemy is strictly superior to the case where you threw the first enemy into a wall. Also, even if there's just one enemy, it's usually better to knock them back into a cloud and a wall than just into a wall. If you just want to escape from something (or create room to climb stairs without being followed), you don't want anything stopping them from moving the full 2 tiles. Also, if you are next to an ogre and a centaur, for example, you don't want the spell to autotarget the centaur. There's way too much going on with this spell to make it autotarget by default (or worse, as the only option).

IMB is just weird, but has its behavior been altered since its wiki page was updated? That says the chance is between 3/4 and 1/6 depending on distance (it doesn't take Barachi into account). This is actually one of my least favorite Conjurations spells, and I basically only use it if I want to do AoE damage to something right next to me without getting caught in my own spell... even then, I usually prefer Dazzling Spray to reduce my chance of being pummeled, and moreover, I try to avoid situations like that in the first place.

Not being able to target Conjure Flame would be a significant nerf that I don't think is warranted (it's often valuable to drop a flame could behind an enemy, or to litter several in a crowd of stumbling enemies). Besides, there are already several ways to drop a Conjure Flame next to you then walk through it with little or no damage, and the situational appeal that creates for those defenses adds interesting strategic variety to the game.

Not being able to target Fulminant Prism would be an extreme nerf to the spell without drastically altering its behavior. If positioned badly, it's wasted MP and generated noise for little to no benefit, either because an enemy kills it prematurely, or because the enemy isn't in range when it explodes. The best ways to use Fulminant Prism involve careful consideration of the terrain and both the player's and the monsters' capabilities. There's just no realistic way to get the same performance from an auto-targeted version, not to mention it would mean a huge loss of meaningful decisions. Fulminant Prism is a valuable component to a tactical, control-oriented style of magic use that, while more time consuming and therefore not appealing to everyone, I absolutely love (I'm primarily a Johnny in MtG parlance), and I can't imagine how Crawl could benefit from losing the ability to target this spell.

The idea of creating a spell with an effect that traces adjacent walls is quite intriguing, although I don't see how this is supposed to devalue corridors.

Cone AoE would be a nice addition and would encourage fighting in the open, but I would want the ability to target it.

Ranged hexes are a valuable tool for defending against ranged threats, even for non-stabbers, and the removal of that tool would further encourage luring.

I do agree that manual targeting can sometimes be frustrating, for example, trying to find a tile to optimally target spells that can potentially hit multiple, but not all, enemies in range (e.g. bolts). Sometimes it looks like you should be able to line up 2 enemies so that if a projectile misses the first, it might hit the second, but to get the UI to cooperate, you'd have to be able to either target a tile outside the spell's range (even if both enemies are in range), target a tile that's out of sight (e.g. obscured by clouds), or target a point between two tiles. I wouldn't want to lose the ability to choose which targets to prioritize with these spells, but I would like the UI to be a bit smarter about how it presents the options. As for how to implement this, maybe + and - could be reworked to cycle through possible targeting options clockwise or counterclockwise, respectively. Cardinal directions could keep the current behavior, so you could use those to specify the general vicinity, then switch to + and - to fine-tune.

Ziggurat Zagger

Posts: 4432

Joined: Friday, 8th May 2015, 17:51

Post Monday, 21st January 2019, 08:14

Re: Positional Attack Spells

If you want to make positioning more interesting, we can look at divinity:origin sin. It has spells which work like this:
1) Caster jumps 2-4 times from a monster to monster, dealing damage each time. So you have to trade high damage for change of position (potentially landing in the middle of some crowd)
2) You can combine several spells.
a) For instance, you can throw oil (so monsters movement is slowed down) and then set the oil on fire so monsters have a chance to get burning damage over time and also they take damage every turn while they are in the cloud. Also burning monsters get more damage from fire spells and less damage from ice spells. As an extra bonus, the burning oil creates a cloud of smoke so the monsters in the cloud (and those behind it) cannot shoot at you (neither can you shoot at them).
b) Another example is you can create a rain in the area to extinguish fire, to increase damage from ice spells, to decrease damage from fire spells. Ice spells include "freeze" (shot range spell which deals no damage but has a chance to temporarily convert the target into ice statue making it unable to act but increasing its armour also), so rain can act like decreasing MR in crawl but only for some spells.
c) You can teleport a monster and make it land into that burning oil you created so it will get damage from smashing into land, damage from burning and be further away from you
3) Archers can trade accuracy for damage (great when shooting at disabled targets which cannot evade shots), movement speed for accuracy (great when shooting at targets with high EV like spriggan monsters in crawl) so even archers still needs to make some decisions.

As you can see my take on positioning is that you MUST not lure a monster into favorable terrain (it is boring and time-consuming), you can create interesting positioning traps on your own.
Underestimated: cleaving, Deep Elf, Formicid, Vehumet, EV
Overestimated: AC, GDS
Twin account of Sandman25
User avatar

Vaults Vanquisher

Posts: 454

Joined: Thursday, 1st November 2018, 02:33

Post Monday, 21st January 2019, 08:49

Re: Positional Attack Spells

re: Nekoatl, I think you would benefit from taking it as read that people understand things like "ranged hexes are useful." People are very well aware that, for example, the cflame placement duvessa suggested is a nerf to the spell. There's a spectrum of usefulness ranging from useless to overpowered and between those there may exist an appropriate degree of usefulness for a given effect. Conjure flame is overpowered. I would argue that ranged confusion is also overpowered, but that's probably a more controversial opinion. When something is overpowered, it removes strategic and tactical depth from the game because it does too many things too well. The question is always how to make something useful but not too useful.

The reason that dividing an LRD effect into multiple explosions with roughly a set total damage distributed among them is that monster AC would be applied for each explosion. Often this is a small effect, but not always. Of course, there could be additional penalties for dividing the effect.

By the way, the reason my cone thing is in an edit is that I knew someone would say that they want to target it arbitrarily and I didn't want to present that possibility, but against my better judgement I broke down and added it anyway. Targeting cones is terrible and should not be taken seriously as a mechanic. It's even worse than beams, which should also mostly be eliminated (I could see limited use ranged weapons allowing penetration mechanics, for example). You might as well say you want to sequentially target chain lightning.

* * *

I'd like some clarification on this thing in the OP about "UI being orthogonal" and "no single target/autotarget effects." Between showing way too much faith in automation as an optional alternative to full user control and being very vague about where the line is drawn, I'm pretty skeptical about the position taken. For example, which of the following runs afoul of this standard: Autotargeted fireball, autotargeted fire bolt, autotargeted cone as described in my previous post? If someone said let's "have a two-arc lightning bolt replacement," do you entertain the notion of explicit targeting there? How much obfuscation does it take to say "well, obviously you can't do that" like you would with sequentially targeted chain lightning or bitmap editor freezing cloud?

edit: re: duvessa's conjure flame, the reason I suggested reasonably intelligent autotargeting of adjacent tiles in the previous thread is that placing the cloud on the player then letting them walk off allows you to create a gap with many monsters where there was none before. Increases the power of the spell in certain use cases.
Last edited by tealizard on Monday, 21st January 2019, 12:01, edited 1 time in total.
This is where mechanical excellence and one-thousand four-hundred horsepower pays off.

For this message the author tealizard has received thanks:
duvessa

Slime Squisher

Posts: 392

Joined: Sunday, 11th September 2016, 17:21

Post Monday, 21st January 2019, 10:41

Re: Positional Attack Spells

Confuse isn't OP. It has a chance to have no effect even when successfully cast, and has little or no chance to affect some monsters, even at max spell power. If it fails, that's 3 MP and a turn wasted, plus the possibility of attracting attention. These drawbacks are important strategic considerations when deciding whether to invest in being able to effectively cast the spell. In fact, at the moment, I can't think of any ranged confusion effect that doesn't have some significant drawback or limitation.

Conjure Flame might be, but I don't agree that this proposed nerf is a good way to tone it down. Against most enemies, it's not very useful on its own. Zombies and other mindless foes that just walk right into it can be taken out efficiently, but most won't, and you can't cast it in occupied tiles. You need some means of either luring or forcing an enemy into the cloud, or some means to wear down the enemy from a distance, to kill anything with it, and once enemies with more HP start appearing, it's not even that useful as a deterrent. I do think intelligent monsters should be a bit smarter about reacting to clouds of flame, however, for example circling around behind you or stepping out of view to wait for you to emerge if they're not confident enough to brave the flames, and retreating if you drop some other cloud on top of them instead of deciding "Hey, I'm already in a cloud, so why not march through this several-tile-long inferno, when I could just take a step backwards and be safe?".

I think with my responses to both the proposal to eliminate ranged hexes and to modify the behavior of Conjured Flame, I already explained how, specifically, those proposals would reduce strategic depth / meaningful decisions. Improving the monster AI would tone down Conjure Flame's effective strength while adding strategic depth instead of removing it. Alternately, the damage could scale with spell power so it wouldn't be so significant in the early game, and that would at least not sacrifice strategic depth unnecessarily.

I'm not sure what exactly you're requesting clarification on, as the text you're quoting doesn't seem to exist, but I don't think there's a single right answer to how much automation is correct for all spells. Different types of players will appreciate different things, and trying to make every spell appealing to all players would unnecessarily hinder the satisfaction that any player type could derive by picking and choosing the spells that most appeal to them from a broader pool. For example, I would personally prefer complete precise control over every spell, down to the spawn locations and actions of every summon, regardless of how much it slowed me down, but obviously not everyone feels that way, and that's fine... I would never seriously suggest that all spells should be so precisely controlled. I think the approach of having spells which fill similar roles but fit different play styles, and the approach of compromising between different play styles in a spell's design, both have value and compliment each other well. I wouldn't want to use an autotarget fire bolt, fireball, or cone, and if I were to use one, it would either be because I had no viable alternative available, or because the corresponding damage potential was significantly (at least +50%) higher than comparable targeted spells to compensate for the lack of control. But, I wouldn't mind if they existed, so long as I could usually choose alternatives that I could target with. Maybe there could even be a specific school of magic dedicated to being powerful but hard to control (Fire or Air, perhaps, or maybe even a new school altogether, e.g. Chaos [hated by Zin, loved by Xom!]).

re: tealizard, I've found that assuming that everyone else understands something that seems obvious to me can lead to confusion, which is why I usually try to say what I mean instead of trusting that it's understood, and why I appreciate it when others do the same.

For this message the author Nekoatl has received thanks:
VeryAngryFelid
User avatar

Vaults Vanquisher

Posts: 454

Joined: Thursday, 1st November 2018, 02:33

Post Monday, 21st January 2019, 11:07

Re: Positional Attack Spells

To clarify, my questions about the OP were addressed to the OP.
This is where mechanical excellence and one-thousand four-hundred horsepower pays off.

For this message the author tealizard has received thanks:
duvessa

Dungeon Master

Posts: 250

Joined: Thursday, 27th November 2014, 19:12

Post Monday, 21st January 2019, 15:05

Re: Positional Attack Spells

Well the first 5 posts of the thread got off to an ok start before we ran right into a wall. If you can't tell from my other posts, I loathe abstract theory crafting and really hoped this thread would focus on specific ideas.

Regarding the ui disclaimer: in the realm of "positional spells" there is a category of spells that are "take something that effects a single monster and do it to an algorithmically chosen one". If someone does not share your extremist point of view on targeting, tealizard, this will be (as you found out in the other thread) replied to with "just improve the default targeter choice this is a ui-issue" and the thread will be derailed. The concept is much broader than that little slice of design space, so let's get some specific ideas from the rest of the "positional spells" design space.

Another ground rule: nerfs are good, players are too strong. If your only comment on someone else's spell revision is "this is a nerf" and the spell isn't Leda's then don't press the submit button.

For this message the author ebering has received thanks: 2
duvessa, sdynet
User avatar

Vaults Vanquisher

Posts: 454

Joined: Thursday, 1st November 2018, 02:33

Post Tuesday, 22nd January 2019, 00:37

Re: Positional Attack Spells

Now I'm just a little old country video game forumer and I only think about things in terms of specific ideas, so why don't y'all go on and tell me which of them specific ideas I mentioned in yonder question meets y'all's criteria and we can git this thing back on track.
This is where mechanical excellence and one-thousand four-hundred horsepower pays off.

For this message the author tealizard has received thanks:
duvessa
User avatar

Zot Zealot

Posts: 982

Joined: Monday, 29th September 2014, 09:04

Post Tuesday, 22nd January 2019, 02:03

Re: Positional Attack Spells

When you cast this spell, it causes clouds to grow from outwards a random water/lava square in los for several turns. (Partially controllable with clever positioning)

All monsters in los (including you) leave a cloud trail as they move. (Like the company effect) (good for kiting but you can't stop moving so it's way to trap yourself).

Slime Squisher

Posts: 377

Joined: Thursday, 12th June 2014, 06:56

Post Tuesday, 22nd January 2019, 05:39

Re: Positional Attack Spells

Specific idea: A single-target conjuration whose damage is proportional to the number of non-wall squares around you. If that's too gameable, it could be instead based on the number of non-wall squares in a 5x5 or 7x7 square around you. The damage/MP of the fully "open" version of the spell would have to considerably exceed that of any comparable spell in order for it to be worthwhile. The current damage bonus could be communicated with a (toggle-able) status light, like "Open (+125%)" or similar.

A more radical idea would be to do this to all conjurations, eliminating the "people will just use conjurations without that restriction" problem. That would certainly distinguish the school from the others, and also help distinguish it from ranged combat.

Ideas to make people think about positioning don't need to be complicated. Simple rules like "you do more damage out in the open" are easy to communicate, and the rewards are easily adjusted if players still consider the risk of fighting in the open too great.

Dungeon Dilettante

Posts: 4

Joined: Monday, 3rd September 2018, 00:46

Post Tuesday, 22nd January 2019, 07:43

Re: Positional Attack Spells

I'll try my hand at one. Not sure if this counts as autotargeting, but the same effect couldn't be achieved with a UI/targeter improvement, so I believe it isn't.

Iskundrun's Starshower: Casting this spell gives you a status that drops a magic dart-like projectile on the enemy furthest from you, by Manhattan distance, every 10 auts. This encourages getting weaker enemies between you and your target, managing los, and gives the choice of whether improving your overall damage by splitting it between monsters in a a pack since the furthest monster will be hardest to hit with your other conjurations, and as things move the target of this spell might change. It, I hope, gets the feel of battlesphere without using clunky ally mechanics or making you continue to cast at the enemy - I'd like casting this as a precombat buff and then wading into melee to be an option. Perhaps it could serve as a midlevel conjuration/charms for your skald-types? Levels 4 and 5 are kind of lacking in general-purpose charms.

One issue I can see is the nuance of the spell gets lost if you are engaging one enemy at a time, as you should. To counter this, we could make the spell hit every target if multiple are at the furthest distance from you, rather than picking one at random as you might expect. This could enable players to look for opportunities to get enemies equidistant and get an effective multiplier to their damage. Maybe the fact that most monsters eventually end up in melee range and the spell would hit all of them then is a problem, but since it would only take one monster further away to ruin it, I don't think it would be.

The spell also lightly disincentives corridors, since the target you are hitting in a chokepoint situation is the one you are furthest from actually engaging. Not near enough to make corridors not good, but it's nice to not further encourage them.

EDIT: Upon reconsidering for a couple minutes, manhattan distance is overcomplicated and probably leads to too much tile counting. It should probably just try to hit everything at range 7 (or whatever your los is), and if nothing is there try range 6, etc. Another facet I forgot to mention is that the spell ends immediately if you can't see any monsters, to prevent just walking around with it up all the time.

Vestibule Violator

Posts: 1508

Joined: Monday, 21st November 2011, 07:40

Post Tuesday, 22nd January 2019, 17:30

Re: Positional Attack Spells

I'm not sure which side of the line between UI and New spell this is, but... I really like the idea of a few spells with cones/AoEs that always auto target to hit the maximum number of enemies (ignoring things like HD completely).

For cones, this eliminates the fiddlyness that tealizard described.

For both cones and AoE, it leads to a slight twist on optimal luring to get the thing you WANT to hit into a pack so it gets hit, which sounds more engaging than tedious to me.
Usual account: pblur on kelbi

Ziggurat Zagger

Posts: 6454

Joined: Tuesday, 30th October 2012, 19:06

Post Tuesday, 22nd January 2019, 18:19

Re: Positional Attack Spells

Pretty much any attack pattern relative to the player would be considered positional, you could do a cross, a knight-attack pattern, a ring (at range > 1) diagonals etc. etc. etc. doing a "Cone that faces in some pretargeted direction" is probably the worst option, as you have to be able to somehow predict what it's going to attack before you do it, which means either a targeting cursor (defeating the purpose), or some kind of feedback that lets the player know in advance of casting what they would be targeting if they were to use this spell.

Having a spell that attacks "something that you can't reliably predict" is a terrible interface and really defeats any possibility of tactical thinking.

Any area effect is going to naturally be more powerful than a single-target attack (which for non-directional attacks is pretty much limited to some form of auto targeting, which IMHO is just a worse form of manual targeting, but if it's something you can reliably predict it at least wouldn't be terrible) so that needs to be accounted for in spell design, but that's mostly just a matter of making it the right level for how much damage it does.

A cross-type or ring-type area effect would probably be the most interesting (being able to hit things *only at a specific range* or *only in a specific subset of directions* is at least different from the mechanics we have now)
Spoiler: show
This high quality signature has been hidden for your protection. To unlock it's secret, send 3 easy payments of $9.99 to me, by way of your nearest theta band or ley line. Complete your transmission by midnight tonight for a special free gift!

For this message the author Siegurt has received thanks: 2
duvessa, ebering

Ziggurat Zagger

Posts: 8786

Joined: Sunday, 5th May 2013, 08:25

Post Tuesday, 22nd January 2019, 19:12

Re: Positional Attack Spells

Siegurt wrote:Any area effect is going to naturally be more powerful than a single-target attack (which for non-directional attacks is pretty much limited to some form of auto targeting, which IMHO is just a worse form of manual targeting, but if it's something you can reliably predict it at least wouldn't be terrible) so that needs to be accounted for in spell design, but that's mostly just a matter of making it the right level for how much damage it does.
If you want fighting one monster at a time to be less of a dominant strategy, then you should probably not be afraid of area attacks being good.

For this message the author duvessa has received thanks:
Rast

Ziggurat Zagger

Posts: 6454

Joined: Tuesday, 30th October 2012, 19:06

Post Tuesday, 22nd January 2019, 19:16

Re: Positional Attack Spells

duvessa wrote:
Siegurt wrote:Any area effect is going to naturally be more powerful than a single-target attack (which for non-directional attacks is pretty much limited to some form of auto targeting, which IMHO is just a worse form of manual targeting, but if it's something you can reliably predict it at least wouldn't be terrible) so that needs to be accounted for in spell design, but that's mostly just a matter of making it the right level for how much damage it does.
If you want fighting one monster at a time to be less of a dominant strategy, then you should probably not be afraid of area attacks being good.

I don't particularly want fighting one monster at a time to be less dominant, and I'm not afraid of area attacks being good, I'm just pointing out that they are naturally better than non-area-attack spells, as a thing we should keep in mind.
Spoiler: show
This high quality signature has been hidden for your protection. To unlock it's secret, send 3 easy payments of $9.99 to me, by way of your nearest theta band or ley line. Complete your transmission by midnight tonight for a special free gift!

Ziggurat Zagger

Posts: 4432

Joined: Friday, 8th May 2015, 17:51

Post Tuesday, 22nd January 2019, 19:55

Re: Positional Attack Spells

Can we have AoE spells that deal more damage per monster when they affect multiple monsters? Ideally the more monsters the more damage per monster. That might lead to fair deaths when player got too greedy for damage and gathered too many monsters in a crowd.

Edit. Something like that fire level 8 spell but without requiring monsters be adjacent to each other.
Underestimated: cleaving, Deep Elf, Formicid, Vehumet, EV
Overestimated: AC, GDS
Twin account of Sandman25

Vestibule Violator

Posts: 1508

Joined: Monday, 21st November 2011, 07:40

Post Tuesday, 22nd January 2019, 20:25

Re: Positional Attack Spells

A bit like the singing sword maybe. I mean that's technically tension not number of enemies, but it has the same over-extending incentives. Qazlal has a similar feel, and IMO is one of the more fun gods to play (if somewhat suicidal).

Lightning Storm
Every monster in LOS is hit with an arcing electrical charge with a significant arc range (say, 3 tiles). It preferentially jumps to closer targets, so it will absolutely crush clumps of adjacent enemies, but will still scale up with multiple enemies that aren't touching as well.

Earthquake (Earth/Hexes)
Every monster in LOS radiates a shock 2-range AOE that slows and does minor irresistable damage. Slow duration for a single shockwave is short (10 auts maybe), but durations stack with multiple shockwave hits.

Edit:
Agoraphobia
Cause Fear, but with power increased/decreased based on number of creatures in each enemy's line of sight.
Usual account: pblur on kelbi

For this message the author byrel has received thanks:
VeryAngryFelid

Slime Squisher

Posts: 406

Joined: Thursday, 16th June 2011, 18:36

Post Tuesday, 22nd January 2019, 21:11

Re: Positional Attack Spells

Effect: Every monster in LOS performs N attack(s) on a neighbor, if it has one. More fit for Skald theme/gameplay, not efficient for a nuker. Allow spell attacks on the player if in range?
Won with: KeAE^Sif, NaWz^Sif, NaTm^Chei, SpEn^Nmlx, GrEE^Qaz, HOFE^Veh, MiBe^Trog, DrFE^Hep, FoFi^Zin, CeHu^Oka, DjFE^Ash, DrIE^Ru, FeSu^Jiy, GnCA^Usk.
In Progress:
Long-term goal: complete the pantheon.

Ziggurat Zagger

Posts: 8786

Joined: Sunday, 5th May 2013, 08:25

Post Tuesday, 22nd January 2019, 21:44

Re: Positional Attack Spells

VeryAngryFelid wrote:Can we have AoE spells that deal more damage per monster when they affect multiple monsters? Ideally the more monsters the more damage per monster. That might lead to fair deaths when player got too greedy for damage and gathered too many monsters in a crowd.

Edit. Something like that fire level 8 spell but without requiring monsters be adjacent to each other.
For the record, this isn't quite exclusive to Ignition: Static Discharge also offers more damage-per-monster when used against multiple monsters, though the requirements are pretty stringent - the monsters have to be adjacent to each other and to you, and it starts getting worse after 2 to 4 monsters, depending on your spell power.

Blades Runner

Posts: 593

Joined: Tuesday, 11th December 2018, 19:14

Post Tuesday, 22nd January 2019, 22:16

Re: Positional Attack Spells

ebering wrote:Well the first 5 posts of the thread got off to an ok start before we ran right into a wall. If you can't tell from my other posts, I loathe abstract theory crafting and really hoped this thread would focus on specific ideas.

Regarding the ui disclaimer: in the realm of "positional spells" there is a category of spells that are "take something that effects a single monster and do it to an algorithmically chosen one". If someone does not share your extremist point of view on targeting, tealizard, this will be (as you found out in the other thread) replied to with "just improve the default targeter choice this is a ui-issue" and the thread will be derailed. The concept is much broader than that little slice of design space, so let's get some specific ideas from the rest of the "positional spells" design space.

Another ground rule: nerfs are good, players are too strong. If your only comment on someone else's spell revision is "this is a nerf" and the spell isn't Leda's then don't press the submit button.


How strong is too strong per design vision, or rather how weak is sufficient?

It matters when it comes to suggesting spells with positional consideration. For example searing ray is a positional spell. Its current theme is "searing", aka a very hot ray that does increased damage when aimed at the same thing for longer duration. The current mechanic of "you must stay in place and fire this in the correct direction to get maximum damage" is where the positioning comes into play; crawl presents scenarios where you do not want to stand in place.

If we just want to nerf the player, you could literally constrain it to the 8 basic directions outright, which would be a small, positional-forcing nerf while keeping the theme and stripping the dependency on knowing monster pathing algorithms to the extent duvessa implies.

However, it could also be altered into a single target, stick-on spell that follows an individual monster to keep burning them harder, still restricting the player from moving for the duration. This would be easier to use and thus presumably nerf the spell more in consequence, but maybe not?

New spell idea:

Force blast - like an inverted gell's gravitas, it shoves monsters outward from the center and does damage if they collide with a wall or each other. If misplaced (or maybe not misplaced) could therefore shove monsters toward the player, and tends to scatter monsters. Fan of gales is kind of similar, but only does this around the player as a one-time evoke. To differentiate that, make it only effect an area like a fireball, but have the blast push monsters further.

Maybe this makes gravitas useful, but probably not :p.

Slime Squisher

Posts: 344

Joined: Tuesday, 14th April 2015, 19:56

Location: France

Post Tuesday, 22nd January 2019, 22:27

Re: Positional Attack Spells

Idea : some kind of slow-moving omnidirectional shockwave originating from the player. Like shatter (hits tiles next to the player, then two tiles away, then three, etc) but with a 10 aut delay between each.
3 runes : MiMo^Ru, HOFi^Beogh, TrMo^Yredelemnul, GrFi^Ru, FoFi^Gozag, MiGl^Okawaru
4 runes : DDFi^Makhleb
5 runes : GrEE^Vehumet
15 runes : MiFi^Ru, NaWz^Sif Muna, GrWz^Sif Muna
I mostly play offline or online on CXC

For this message the author Fingolfin has received thanks: 2
byrel, ebering

Vestibule Violator

Posts: 1508

Joined: Monday, 21st November 2011, 07:40

Post Wednesday, 23rd January 2019, 03:47

Re: Positional Attack Spells

TheMeInTeam wrote:
ebering wrote:Another ground rule: nerfs are good, players are too strong. If your only comment on someone else's spell revision is "this is a nerf" and the spell isn't Leda's then don't press the submit button.


How strong is too strong per design vision, or rather how weak is sufficient?

It matters when it comes to suggesting spells with positional consideration. For example searing ray is a positional spell. Its current theme is "searing", aka a very hot ray that does increased damage when aimed at the same thing for longer duration. The current mechanic of "you must stay in place and fire this in the correct direction to get maximum damage" is where the positioning comes into play; crawl presents scenarios where you do not want to stand in place.


I think balance considerations are out of scope entirely here. Even if a proposal is a targetting modification to an existing spell, I think we should simply assume that damage, spell level or whatever would be tweaked to make the new version balanced if it was made.
Usual account: pblur on kelbi

For this message the author byrel has received thanks: 3
duvessa, ebering, mattlistener

Slime Squisher

Posts: 406

Joined: Thursday, 16th June 2011, 18:36

Post Wednesday, 23rd January 2019, 04:29

Re: Positional Attack Spells

Change Airstrike to scale with the amount of air (unoccupied tiles) next to the target. Favors an in-the-open playstyle rather than the same old retreat-to-killhole.

Monster Airstrike could be changed in the same way -- makes you want to get into cover rather than be caught out in the clear, very Air-schoolish.
Last edited by mattlistener on Friday, 25th January 2019, 16:22, edited 1 time in total.
Won with: KeAE^Sif, NaWz^Sif, NaTm^Chei, SpEn^Nmlx, GrEE^Qaz, HOFE^Veh, MiBe^Trog, DrFE^Hep, FoFi^Zin, CeHu^Oka, DjFE^Ash, DrIE^Ru, FeSu^Jiy, GnCA^Usk.
In Progress:
Long-term goal: complete the pantheon.

Dungeon Master

Posts: 250

Joined: Thursday, 27th November 2014, 19:12

Post Wednesday, 23rd January 2019, 04:43

Re: Positional Attack Spells

Another one in the game (new!) is the new storm card's effect. First it does a wind blast. Then It picks a random number 1-9 (via a complicated process that is a legacy of old Nem mechanics) and draws that many cells at a distance more than three from the player randomly. It then detonates an orb of electricity (think cbl explosion) on each of the chosen cells, but with the limit of at most one explosion per cell.

To maximize the hit on your target you want to see the target and many cells around the target (within the explosion radius) and not much else. Too narrow and power is wasted: you cut down the number of explosions because you only get one per cell and the same cell could get randomly picked twice. Too wide and they go everywhere. At high enough power if you get a good roll you can AOE the screen, if that's what you want, but you might also just get one quadrant and make a ton of noise.

For this message the author ebering has received thanks: 2
byrel, tealizard
User avatar

Vaults Vanquisher

Posts: 454

Joined: Thursday, 1st November 2018, 02:33

Post Wednesday, 23rd January 2019, 05:07

Re: Positional Attack Spells

re: Siegurt's theory that any degree of unpredictability destroys the possibility of tactics, I just have to make the usual point that I've never heard this point of view deployed against existing features (say, chain lightning, freezing cloud, disaster area, to name a few). Of course, in the face of actual examples, the theory is just obviously wrong. The fact that a player would have to consider more than one possible outcome, think about average and outlier cases, would at least potentially increase tactical depth, subject to the skill of the designer of course. That's why we randomize damage in this game, among other things.

re: Cones, my read is that anything that could potentially be targeted with a single "control point" is likely to be coopted into a fully targeted effect if it's acted on at all. This is too bad, because there is an obvious tactical difference between the dcss fireball and a fireball that automatically centers the closest monster -- for example, the latter will always hit the player character at point blank range. On the other hand, you could imagine a version of chain lightning that jumps from target to target but where the jumping behavior is biased so that it tries to continue in roughly the direction it's already moved from the player character, in effect confining it to a cone which is randomly chosen. I hope the incoherence of the strong version of single target/autotarget prohibition comes through in this example.

Many, perhaps most up to obvious equivalence, possible attack patterns that would work in crawl combat are going to have some sort of directional or targetable aspect to them. If you insist on explicitly targeting them whenever it's possible, you'll get obvious tactics. Making the player play the numbers and judge how terrain and/or monster placement is likely to affect outcomes of somewhat randomly determined patterns is a good in itself. It also puts more power in the hands of the designer, preventing obvious ways of avoiding potential downsides to spells, e.g. the caddy-corner fireball.
This is where mechanical excellence and one-thousand four-hundred horsepower pays off.

For this message the author tealizard has received thanks:
duvessa

Ziggurat Zagger

Posts: 6454

Joined: Tuesday, 30th October 2012, 19:06

Post Wednesday, 23rd January 2019, 08:32

Re: Positional Attack Spells

What I said was
Siegurt wrote:Having a spell that attacks "something that you can't reliably predict" is a terrible interface and really defeats any possibility of tactical thinking.

Which you interpreted as:
tealizard wrote:re: Siegurt's theory that any degree of unpredictability destroys the possibility of tactics,

So I didn't say that *any and all* randomization destroys tactics, only that *you can't think tactically about an action that you can't predict the results of* (I also said that it was a terrible UI, it's terrible because people like to feel in control, and like their actions have a direct impact, it's terribleness has nothing directly to do with its tacticlessness, those are two separate things)
tealizard wrote: just have to make the usual point that I've never heard this point of view deployed against existing features (say, chain lightning, freezing cloud, disaster area, to name a few). Of course, in the face of actual examples, the theory is just obviously wrong, which solidifies my belief that this is standard bias against change, even in point of view. The fact that a player would have to consider more than one possible outcome, think about average and outlier cases, would at least potentially increase tactical depth, subject to the skill of the designer of course. That's why we randomize damage in this game, among other things.

Damage has a range and an average, and a reasonably predictable result (you can estimate that you will kill things with damage spells, and even have a rough idea of how many it will take)

If there's a predictable outcome (even if it's somewhat randomized predictability) you can base decisions on it and act tactically, if you can't estimate *at all* what will happen, then you can't make any plans based on the outcome, or make an informed choice. (On the other side of the coin if the outcome is *completely* predictable, it's repetitive, *that* is why we randomize damage, it has nothing to do with increasing tactical depth, which it doesn't do, variance and tactics are actually natural enemies)

Of the things you mention (chain lightning, freezing cloud, and disaster area) all of them pretty reliably hit targets and the effects can be predicted within a reasonable range, and can even be to some extent be controlled (freezing cloud more than the other two) you get to make choices about what you're going to hit. If you have no decision about what you're going to hit (or if you're going to even hit anything) you can't act on it tactically (see Qaz's passive cloud effect for an actual in-game example of an in-game unpredictable attack, since it's unpredictable what square it will even effect, using it as your primary attack, while theoretically possible, would be absolutely horrible.)

That doesn't automatically preclude those sorts things from being a *good thing to have in game* it just means it's not a tactical thing, it's a non-tactical thing. I simply contend that having your primary attack tactic be non-tactical is inherently terrible. (Although even if your attack was completely non-tactical, you could still of course position tactically and retreat tactically)

Presently, all of the things in the game you could conceivably use to attack in a tactical fashion have *some* kind of predictability (even if it's by controlling the position of things as they relate to you by moving around) That's what tactics *are*: Choosing the (presumed) best outcome by selecting among predicted results of future action. No predictability equates *by definition* to no tactics (Also, as an aside, lack of the ability to choose among options also precludes tactics by definition, as both choice and prediction are required elements).

Imagine for a moment you were suggesting that not just spells, but all attacks (including melee) functioned this way (both for monsters and players), where all attacks would hit "something" and you had no control over what they would even attack (or indeed if it was even a monster at all), all ability to prioritize important (or just plain wounded) targets was completely eliminated, and your position relative to monsters was immaterial as to whether they would hit you, would you consider that to be an improvement? Would you want to play 'permanently confused, the game'?
Spoiler: show
This high quality signature has been hidden for your protection. To unlock it's secret, send 3 easy payments of $9.99 to me, by way of your nearest theta band or ley line. Complete your transmission by midnight tonight for a special free gift!
User avatar

Vaults Vanquisher

Posts: 454

Joined: Thursday, 1st November 2018, 02:33

Post Wednesday, 23rd January 2019, 09:11

Re: Positional Attack Spells

So first of all, it's literally true of melee combat that there is a nonzero chance that absolutely nothing happens when you attack. In fact when you melee a single target, of all the possible amounts of damage, zero is almost always the most probable by a wide margin. Secondly, if you look at cleaving melee, it's totally possible to have numerous eligible targets and hit none of them or any combination of them. The player has absolutely no control over this, so maybe wrap your brain more thoroughly around that situation.

What these situations share is that the player can think through the average case, the best and worse cases, and likely outcomes of repeated attacks, as you and I clearly agree. But this is no different from a spell that has a somewhat randomized attack pattern that may hit different monsters in different situations or just from one cast to the next in the same situation. This sort of judgement is what tactics are made of. (By the way all of the specific examples you mention can and do randomly fail to hit things the player may want them to.)

I don't see how my summary of your post differs substantially from what you posted either.

[edit: as an aside, I used to have views about randomization and "variance" similar to those you profess in the previous comment. In a game format as naturally static as crawl's, though, you need surprising outcomes (among other things, obviously) to create action. That's a consequence of lacking the randomization of human reflexes in real time games and allowing unlimited consideration as opposed to time-limited turn-based play, which would bring in quickness of wit.]
This is where mechanical excellence and one-thousand four-hundred horsepower pays off.

For this message the author tealizard has received thanks:
duvessa

Swamp Slogger

Posts: 182

Joined: Monday, 2nd July 2018, 16:47

Location: United States

Post Wednesday, 23rd January 2019, 17:38

Re: Positional Attack Spells

Siegurt wrote:Any area effect is going to naturally be more powerful than a single-target attack...so that needs to be accounted for in spell design, but that's mostly just a matter of making it the right level for how much damage it does.

I think it would be interesting to see some "negative" AoE spell designs that encourage you to hit only one enemy in the AoE (and are naturally weaker than their single-target analogs). I'll start with a concrete example: change Freeze to affect all enemies adjacent to the caster, but the effect is split between all enemies adjacent, so that as the number of targets increases the total damage stays the same and the damage-per-target decreases.

Ziggurat Zagger

Posts: 6454

Joined: Tuesday, 30th October 2012, 19:06

Post Wednesday, 23rd January 2019, 18:16

Re: Positional Attack Spells

tealizard wrote:But this is no different from a spell that has a somewhat randomized attack pattern that may hit different monsters in different situations or just from one cast to the next in the same situation.

I guess it depends on the degree of "somewhat randomized" you're talking about, while melee can miss, or do zero damage, you are at least guaranteed to hit (or not) a specific subset of creatures, notably those next to you (and for non-cleave non-riposte, the specific creature in the direction you specify, even if you're using tab, it will always select the same creature when given a subset of creatures it could attack, it's not "random" at all) So in melee, presently you generally have a choice about *what* to attack, even if you don't have any choice about how effective that attack will be. It's not like you're going to hit 'l' to attack right, and your mace is going to go wild and whack that thing that's across the room on the opposite side, not being able to explicitly attack the things you're actually next to in melee would suck.

*algorithmic* (for example "hit closest", "hit furthest", "hit highest HD" etc.) methods of choosing spell attack targets are *also* predictable (although using out of game information like HD for choosing would suck for information-conveying purposes, and is a terrible metric for doing anything in game anyway) Even the somewhat-arbitrary method of selecting targets that tab uses isn't *random* and is therefore predictable.

So in a very real sense, an attack that hits an *actually attacks a completely random target* is substantially different than one that attacks a known (predictable) target and has random effectiveness. Particularly since i don't see it as likely that a "random target attack" would the additional randomness that comes along with most attacks (the to hit and damage rolls), if you propose that spells attack random targets and also don't miss and always do a set amount of damage, then at least you preserve somewhat a similar *degree* of randomness.

In a practical sense, when targeting matters, it matters a *lot* that's why smiting and piercing attacks are useful, being able to take out that summoner, convoker, or orc priest with prejudice is one of the things that makes a better player better. Knowing what's likely to cause problems if left alone and taking actions that are likely to resolve them quickly is good tactics, knowing something will cause problems and not being able to take any action that will improve the situation is just annoying. If I have to flee every time I see a kobold demonologist, because he happens to come with a pack of wimpy kobolds, and I therefore can't attack him explicitly or with priority, I'm going to rapidly get annoyed with the situation.

Note that *random* targeting is totally off topic from *positional* targeting, which is the subject of this thread, your position is under your control, and therefore spells which attack relative to your position are not random, and are predictable (and may or may not constitute good spells, as per their individual design) *random* targeting is a different discussion, and probably should get it's own thread, if you want to advocate it.
Spoiler: show
This high quality signature has been hidden for your protection. To unlock it's secret, send 3 easy payments of $9.99 to me, by way of your nearest theta band or ley line. Complete your transmission by midnight tonight for a special free gift!
User avatar

Vaults Vanquisher

Posts: 454

Joined: Thursday, 1st November 2018, 02:33

Post Thursday, 24th January 2019, 02:10

Re: Positional Attack Spells

The chain lightning example clearly puts effects that hit a subset of monsters potentially different numbers of times all subject to position and chance on the table in this thread.

Also, "you are at least guaranteed to hit (or not) a specific subset of creatures"? Really?
This is where mechanical excellence and one-thousand four-hundred horsepower pays off.

Spider Stomper

Posts: 200

Joined: Sunday, 11th May 2014, 11:26

Post Thursday, 24th January 2019, 05:51

Re: Positional Attack Spells

Dead line(Necromancy/Translocation)
Channeling magic. Create a red cloud on the 'edge of the LOS(Square)'. When an enemy touches this cloud, a vampire effect occurs. Instead, while maintaining this magic, the player gets a constant drain.

Shroud of Force(Conjuration/Charm/Translocation)
Channeling magic. If you attack the enemy of next tile, or enemy attack the you, fire Force Lance at the that tile.

Blizzard(Ice/Hex)
Ice blast in player 5x5. The enemies who are hit by this attack are slowed down.

Guardian dragon(Air/Transmutation)
the head of a dragon on one's chest. As a result, armor is absorbed by the body. If you attack the enemy, or move to the next tile, fire cone of electrical explosion at the target.

For this message the author sdynet has received thanks:
Vanguardan
User avatar

Halls Hopper

Posts: 77

Joined: Thursday, 17th August 2017, 18:47

Post Thursday, 24th January 2019, 22:54

Re: Positional Attack Spells

sdynet wrote:Shroud of Force(Conjuration/Charm/Translocation)
Channeling magic. If you attack the enemy of next tile, or enemy attack the you, fire Force Lance at the that tile.

Blizzard(Ice/Hex)
Ice blast in player 5x5. The enemies who are hit by this attack are slowed down.


I like these two, particularly the first one. Like DMsl but for melee.
Current Victories:
Spoiler: show
3 Runes: GrEE^Ru, CeBe^Trog, SpEn^Gozag, KoAr^Dith, TeAE^Kiku, TrCK^Xom.
4 Runes: VSFi^Qaz.
5 Runes: DsGl^Oka, MiGl^Hep, GnWz^Sif.
9 Runes: DDAr^Makh.
15 Runes: HOFi^TSO->Zin, DEFE^Veh
For now...
User avatar

Halls Hopper

Posts: 77

Joined: Thursday, 17th August 2017, 18:47

Post Friday, 25th January 2019, 03:44

Re: Positional Attack Spells

As for my own idea(s):

1. Steps of Chaos, Level 5(6?) Charms/Conjuration. For a short duration, summons clouds of seething chaos behind you as you walk (i.e. on the tile you step off of). Probably lasts 2-3 turns at lower power, scaling up to 6-8 at max power.
Xom would appreciate the use of this spell, and naturally would incur penance under Zin.
Current Victories:
Spoiler: show
3 Runes: GrEE^Ru, CeBe^Trog, SpEn^Gozag, KoAr^Dith, TeAE^Kiku, TrCK^Xom.
4 Runes: VSFi^Qaz.
5 Runes: DsGl^Oka, MiGl^Hep, GnWz^Sif.
9 Runes: DDAr^Makh.
15 Runes: HOFi^TSO->Zin, DEFE^Veh
For now...

Vestibule Violator

Posts: 1508

Joined: Monday, 21st November 2011, 07:40

Post Friday, 25th January 2019, 18:23

Re: Positional Attack Spells

Vanguardan wrote:As for my own idea(s):

1. Steps of Chaos, Level 5(6?) Charms/Conjuration. For a short duration, summons clouds of seething chaos behind you as you walk (i.e. on the tile you step off of). Probably lasts 2-3 turns at lower power, scaling up to 6-8 at max power.
Xom would appreciate the use of this spell, and naturally would incur penance under Zin.

If it summoned clouds of something besides chaos, this might be playable. Chaos will do nasty things like berserk/haste the stuff you're running from.
Usual account: pblur on kelbi
User avatar

Halls Hopper

Posts: 77

Joined: Thursday, 17th August 2017, 18:47

Post Saturday, 26th January 2019, 02:45

Re: Positional Attack Spells

byrel wrote:
Vanguardan wrote:As for my own idea(s):

1. Steps of Chaos, Level 5(6?) Charms/Conjuration. For a short duration, summons clouds of seething chaos behind you as you walk (i.e. on the tile you step off of). Probably lasts 2-3 turns at lower power, scaling up to 6-8 at max power.
Xom would appreciate the use of this spell, and naturally would incur penance under Zin.

If it summoned clouds of something besides chaos, this might be playable. Chaos will do nasty things like berserk/haste the stuff you're running from.


I disagree. I don't know exactly how chaos cloud effect and chaos weapon effects differ, but if the probabilities are similar there's only about a 5% chance for a negative effect like haste/berserk/invis/healing.

That's why I picked chaos clouds, actually, because the chance for you to have to rethink your strategy gives you pause and makes you consider what situations you want to use this spell in (i.e., it's not always for a better escape route).
Current Victories:
Spoiler: show
3 Runes: GrEE^Ru, CeBe^Trog, SpEn^Gozag, KoAr^Dith, TeAE^Kiku, TrCK^Xom.
4 Runes: VSFi^Qaz.
5 Runes: DsGl^Oka, MiGl^Hep, GnWz^Sif.
9 Runes: DDAr^Makh.
15 Runes: HOFi^TSO->Zin, DEFE^Veh
For now...

Vestibule Violator

Posts: 1508

Joined: Monday, 21st November 2011, 07:40

Post Saturday, 26th January 2019, 20:24

Re: Positional Attack Spells

Vanguardan wrote:I disagree. I don't know exactly how chaos cloud effect and chaos weapon effects differ, but if the probabilities are similar there's only about a 5% chance for a negative effect like haste/berserk/invis/healing.

That's why I picked chaos clouds, actually, because the chance for you to have to rethink your strategy gives you pause and makes you consider what situations you want to use this spell in (i.e., it's not always for a better escape route).


I'm not going to train and learn a level 5 spell that has a chance of ending my game any time I use it on anything scary... But really, I think discussion of which clouds this produces is off topic. 'leaving a cloud trail wherever you walk' is an interesting positional design.
Usual account: pblur on kelbi

For this message the author byrel has received thanks:
Vanguardan

Ziggurat Zagger

Posts: 5382

Joined: Friday, 25th November 2011, 07:36

Post Tuesday, 29th January 2019, 01:10

Re: Positional Attack Spells

tealizard wrote:edit: re: duvessa's conjure flame, the reason I suggested reasonably intelligent autotargeting of adjacent tiles in the previous thread is that placing the cloud on the player then letting them walk off allows you to create a gap with many monsters where there was none before. Increases the power of the spell in certain use cases.

I realize not being able to place it at a distance is a nerf, but imho the ability to create space between you and a monster which is already next to you is so powerful that this would be an overall power increase. YMMV.

mattlistener wrote:Change Airstrike to scale with the amount of air (unoccupied tiles) next to the target. Favors an in-the-open playstyle rather than the same old retreat-to-killhole.

Monster Airstrike could be changed in the same way -- makes you want to get into cover rather than be caught out in the clear, very Air-schoolish.

Really like the idea for the player version, but I'm a bit worried about it for a monster version. Could definitely see this making auto explore very dangerous as you just wander around a level when suddenly an airstriker hits you with 8 open tiles around you.

I think there's room for an interesting imprison type of spell in the earth school; something where you can tomb a monster, rather than yourself. Create temporary walls around any monster with duration based on spellpower. Cannot create walls over other monsters, dungeon features, etc. Ideally I'd like it to not be able to cut off a path entirely, ie, you can't cast it in a corridor and completely seal the passage, although I'm not sure how complicated checking that would be. Maybe make it so it requires an open 5x5 area in order to be cast:

  Code:
x x x x x
x w w w x
x w m w x
x w w w x
x x x x x


The monster trapped is m, the w are new temporary walls, and all x's need to be walkable terrain - they would be allowed to have monsters. Possibly allow for monsters to be in the W tiles, and push them to the X tiles, if enough space is available.

Assuming the duration can scale up to 10+ turns, it's a pretty powerful spell and should probably be level 5-7, if it just lasts 2-4 turns it could be a bit lower. Most of the suggestions in the thread have been offensive positional spells, I think having a defensive positional spell is interesting. This essentially allows for pack splitting in open terrain without luring, but at a mana cost and skill training cost.
User avatar

Halls Hopper

Posts: 77

Joined: Thursday, 17th August 2017, 18:47

Post Tuesday, 29th January 2019, 04:13

Re: Positional Attack Spells

byrel wrote:
Vanguardan wrote:I disagree. I don't know exactly how chaos cloud effect and chaos weapon effects differ, but if the probabilities are similar there's only about a 5% chance for a negative effect like haste/berserk/invis/healing.

That's why I picked chaos clouds, actually, because the chance for you to have to rethink your strategy gives you pause and makes you consider what situations you want to use this spell in (i.e., it's not always for a better escape route).


I'm not going to train and learn a level 5 spell that has a chance of ending my game any time I use it on anything scary... But really, I think discussion of which clouds this produces is off topic. 'leaving a cloud trail wherever you walk' is an interesting positional design.


Fair enough. So that leaves a few options:
Spoiler: show
-Assuming we don't want overlap, that means no fire, freezing, poison, or meph clouds.
-Miasma/Negative Energy/Spectral Flames might work, but Corpse Rot already exists and Necromancy doesn't really need another spell with that much utility.
-Acid fog is definitely too strong.
-Mutagenic Fog has the same issues as Seething Chaos, plus it's basically Polymorph Other which was removed.
-Raging Winds or Thunder Clouds might work, and would fit thematically with air. If it was a very high level spell, a speed boost might be appropriate.
-Calcifying Dust has potential, but it doesn't fit with the Earth school very well, and might be too strong.
-That's it for existing clouds, but if a 'violent distortion energy' cloud were added inflicting damage/distortion effects, this concept would fit as a Translocation/Charms spell.
Current Victories:
Spoiler: show
3 Runes: GrEE^Ru, CeBe^Trog, SpEn^Gozag, KoAr^Dith, TeAE^Kiku, TrCK^Xom.
4 Runes: VSFi^Qaz.
5 Runes: DsGl^Oka, MiGl^Hep, GnWz^Sif.
9 Runes: DDAr^Makh.
15 Runes: HOFi^TSO->Zin, DEFE^Veh
For now...

Slime Squisher

Posts: 344

Joined: Tuesday, 14th April 2015, 19:56

Location: France

Post Tuesday, 29th January 2019, 06:49

Re: Positional Attack Spells

If you're willing to relax the no-targeting requirements a little, then you could have the following spells that can only target an adjacent tile (à la Vampiric Draining) : Raise Walls and Wall of Flame, puts temporary walls / clouds of flame on the targeted tile and both its neighbors (still adjacent to you).
3 runes : MiMo^Ru, HOFi^Beogh, TrMo^Yredelemnul, GrFi^Ru, FoFi^Gozag, MiGl^Okawaru
4 runes : DDFi^Makhleb
5 runes : GrEE^Vehumet
15 runes : MiFi^Ru, NaWz^Sif Muna, GrWz^Sif Muna
I mostly play offline or online on CXC

Zot Zealot

Posts: 1004

Joined: Thursday, 16th August 2018, 21:19

Post Tuesday, 29th January 2019, 19:15

Re: Positional Attack Spells

Vanguardan wrote:-Acid fog is definitely too strong.
-Calcifying Dust has potential, but it doesn't fit with the Earth school very well, and might be too strong.


If it's too strong, you can balance it by changing the spell level. If it's only Acid fog following you, I think it'd still be less useful than Freezing Cloud. It's definitely not too strong for a level 9 spel XD

Calcifying dust would be interesting in the earth school to petrify things with high MR, relating to petrify the same way 'meph cloud' relates to 'confuse'. I think it'd be nice to have an air/earth spell ;)

Blades Runner

Posts: 593

Joined: Tuesday, 11th December 2018, 19:14

Post Tuesday, 29th January 2019, 20:36

Re: Positional Attack Spells

Apocalypse:

Creates explosions similar to an immolation explosion, possibly stronger. Can be pure conjuration or elemental. Picks several locations in a random area around the caster.

This spell could be stronger than comparable spells of its level since its is more situational/random. How many apoc procs + how much damage is a matter of balance for spell level, though given the name there's some implication it would be a reasonably high level spell.
User avatar

Halls Hopper

Posts: 77

Joined: Thursday, 17th August 2017, 18:47

Post Wednesday, 30th January 2019, 22:29

Re: Positional Attack Spells

petercordia wrote:
Vanguardan wrote:-Acid fog is definitely too strong.
-Calcifying Dust has potential, but it doesn't fit with the Earth school very well, and might be too strong.


If it's too strong, you can balance it by changing the spell level. If it's only Acid fog following you, I think it'd still be less useful than Freezing Cloud. It's definitely not too strong for a level 9 spel XD

Calcifying dust would be interesting in the earth school to petrify things with high MR, relating to petrify the same way 'meph cloud' relates to 'confuse'. I think it'd be nice to have an air/earth spell ;)


Good point. We don’t have a level 7 earth spell either so that might be a good balance point. Plus high-MR targets often have enough intelligence to not walk right into clouds so I dunno how this will play out without testing.
Current Victories:
Spoiler: show
3 Runes: GrEE^Ru, CeBe^Trog, SpEn^Gozag, KoAr^Dith, TeAE^Kiku, TrCK^Xom.
4 Runes: VSFi^Qaz.
5 Runes: DsGl^Oka, MiGl^Hep, GnWz^Sif.
9 Runes: DDAr^Makh.
15 Runes: HOFi^TSO->Zin, DEFE^Veh
For now...

Dungeon Master

Posts: 250

Joined: Thursday, 27th November 2014, 19:12

Post Monday, 14th October 2019, 22:07

Re: Positional Attack Spells

I’m going to necro this thread. I’ve let these ideas summer for a while and plan on implementing an overhaul of magic in 0.25. The current state of my thinking can be viewed here https://github.com/crawl/crawl/wiki/Pos ... tack-Magic

For this message the author ebering has received thanks: 3
Implojin, petercordia, sdynet

Slime Squisher

Posts: 368

Joined: Thursday, 11th April 2013, 21:07

Post Tuesday, 15th October 2019, 00:48

Re: Positional Attack Spells

It's great that someone is making the time to breakdown and thoroughly address ranged damage differentiation in DCSS. I wish this had happened years ago! The changes in Hellcrawl and other forks were nice, but Stone Soup has languished.

Have you already implemented the new spells described on the wiki page?
Last edited by Implojin on Tuesday, 15th October 2019, 02:22, edited 1 time in total.

Dungeon Master

Posts: 250

Joined: Thursday, 27th November 2014, 19:12

Post Tuesday, 15th October 2019, 02:21

Re: Positional Attack Spells

Implojin wrote:Have you already implemented the new spells described on the wiki page?


Implementation is a work in progress (the branch linked on the wiki page is not yet in the main crawl repo, it's my draft work). Currently I've done Air and Fire, and am working on Ice. There's a lot going on here, so it will be rolled out slowly over 0.25 (a school at a time or so). Watch this space for when it reaches a play testing state.

Regarding your edited away UI tangent: necro the thread and I'll reply there!

For this message the author ebering has received thanks:
Shtopit
Next

Return to Game Design Discussion

Who is online

Users browsing this forum: No registered users and 99 guests

Powered by phpBB © 2000, 2002, 2005, 2007 phpBB Group.
Designed by ST Software for PTF.