Exploration-based piety is bad


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

bel

Cocytus Succeeder

Posts: 2184

Joined: Tuesday, 3rd February 2015, 22:05

Post Saturday, 30th June 2018, 06:04

Exploration-based piety is bad

There are, broadly speaking, two main ways in which gods give piety: Through kills (Trog, Makhleb, for instance), and through exploration (Elyvilon, Nemelex, for instance). I will argue that the second mechanism is bad. I'll be ignoring cases which don't fit into either category (like Ru). Zin's piety gain is a variant of exploration which is even worse, but I'll ignore that here too.

------------------------

Firstly, why should there be two piety mechanisms? Is anything gained from this? From my understanding, Ely and Nemelex give exploration piety as a sort of "placeholder" when their old piety mechanisms were changed, and nobody has bothered to change it yet.

Secondly, what is piety gain meant to accomplish? There is no explicit design goal (which I can find), but the most coherent explanation is that you are meant to get piety as you progress through the game, which you can spend to make your life easier. It's a sort of consumable. Generally speaking, you should get more piety as you progress deeper into the game and venture into more difficult areas.

Piety through kills approximates this intuition quite well. Exploration-based piety does not.

Exploration-based piety does not adequately capture the density of monsters, nor the hardness of the floor. From what I remember when I read the code, exploration-based piety is based on floor depth. So Vaults:5 will give more piety than Vaults:4. But the difference between them is negligible, in practice. (Try it out in wizmode if you like). Also, stuff like Abyss existing screws up the mechanism even more.

There is an obvious and decent proxy for how difficult the floor on which you are right now: the XP which you get from the monsters you kill, together with how much you need to rest in between. The obvious way to handle this situation is to have kills-based piety, together with piety decay.

For this message the author bel has received thanks: 3
duvessa, nago, Pekkekke

Spider Stomper

Posts: 248

Joined: Monday, 4th September 2017, 10:53

Post Saturday, 30th June 2018, 08:32

Re: Exploration-based piety is bad

Why not both, speaking from a purely mechanistic perspective?

Revealing tiles is a risk. Killing monsters is a risk. All gods (except outliers like Ru) should reward the charcter surviving and "growing more pieous" in the same manner here, rewarding for both kills and exploration, as it's easier to balance and more easily understood for the player, than if each god has there own formulas. Flavour differentiations tied into mechanism should be on piety decay; i.e. you will lose piety if you kill good creatures with zin, etc.

Another benefit of this is that multiple play styles are supported, do you dive or speedrun, or do you always clear the whole floor, or are you somewhere in the middle? As long as you survive and you don't camp, you are going to see some piety gain. That's good.

As far as scaling benefits to risk, I think having a modifier that increases gains dependent on number of new tiles revealed in a step, or number of hostile enemies on screen during a kill, is probably reasonable, and would be easy to describe to players in the various help files. So here the baseline is you get piety for taking risks, but you get more piety for taking more risks. Gods are powerful and ambitious. The mechanisms and flavour tied to them should be as well.

As far as some characters essentially gaining "free piety" because they are too powerful for a given floor, well this is a larger problem (and probably unsolvable in a continually evolving game -see power creep) so trying to address it from subordinate architectures is both misplaced and ineffective. It's also pointless in that those characters are already OP so it's not like having easy access to high piety or not is going to make a big difference to them.

For this message the author crawlnoob has received thanks:
PlatinumSpider

bel

Cocytus Succeeder

Posts: 2184

Joined: Tuesday, 3rd February 2015, 22:05

Post Saturday, 30th June 2018, 08:49

Re: Exploration-based piety is bad

I don't understand most of your post, but I can point out that speedrunning is irrelevant to this issue. Speedrunners dive, which means they explore deeper dungeon floors in a given time period; which in turn means that they get more exploration-based piety. But they also kill harder monsters in the same time period, so they get more kills-based piety too.

In any case, Crawl design is not driven by speedrunning concerns. Speedrunners exploit game mechanics, not the other way around.

Spider Stomper

Posts: 248

Joined: Monday, 4th September 2017, 10:53

Post Saturday, 30th June 2018, 18:48

Re: Exploration-based piety is bad

bel wrote:I don't understand most of your post, but I can point out that speedrunning is irrelevant to this issue. Speedrunners dive, which means they explore deeper dungeon floors in a given time period; which in turn means that they get more exploration-based piety. But they also kill harder monsters in the same time period, so they get more kills-based piety too.

In any case, Crawl design is not driven by speedrunning concerns. Speedrunners exploit game mechanics, not the other way around.


Whats so hard to understand about it? You argue that exploration based piety gain is bad in favor of kill based piety gain, and I argue that both should be used for most gods, and give reasons. Maybe read it again, especially as you clearly missed the part where I point out that an entire spectrum of playstyles, from speedrunning (one extreme) to totally clearing the floor (more common but still an extreme) are supported by this. I certainly don't think speedrunning alone should drive game development, but nor did I allude to such a position in my post.

Temple Termagant

Posts: 10

Joined: Sunday, 1st July 2018, 19:32

Post Sunday, 1st July 2018, 20:07

Re: Exploration-based piety is bad

Kill-based piety is bad.

Killing enemies is already one of the main things you do in the game and you already receive exp and sometimes loot and food for doing so, why should they give piety other than flavor reasons?

bel wrote:Firstly, why should there be two piety mechanisms? Is anything gained from this?

Yes, for instance, if i am playing a weak race or having a hard time early on, or i am a spriggan, i can base my god choice around one that allows me to gain piety without facing enemies so i can use that piety to defeat enemies and gain exp.
bel wrote:stuff like Abyss existing screws up the mechanism even more.

The existence of abyss screws up kill-based piety just as much. Also you can kill-based piety farm hell by standing still in a staircase.
bel wrote:Secondly, what is piety gain meant to accomplish?

Piety is the middle ground between magic and consumables, but not always. If you worship gozag/Zin piety is your money, is you worship Ru/Xom there is no piety, if you worship Usk piety is "bizarro" magic, that fills up to be used instead of drained as you use it to defeat foes. Different piety systems leads to more dynamic gods.
bel wrote:Piety through kills approximates this intuition quite well. Exploration-based piety does not.

Its explained within each god how one obtains piety. There is nothing unintuitive about it.
bel wrote:Exploration-based piety does not adequately capture the density of monsters, nor the hardness of the floor.

Kill-based piety SPECIALLY doesnt do that! How much piety is a d:3 two-headed ogre worth? How much piety is a d:6 two-headed ogre worth if i fell down a shaft from d:3? 10 goblins = 5 orcs = 1 ogre? How much piety is each of those things? If its too weak i dont get any piety even if i get the message "accepts your kill"?
bel wrote:There is an obvious and decent proxy for how difficult the floor on which you are right now: the XP which you get from the monsters you kill, together with how much you need to rest in between. The obvious way to handle this situation is to have kills-based piety, together with piety decay.

There never was and its even worse now that monster spawns over time have been removed, you can litteraly explore a empty level where no enemies have spawned(or the few who did fell down a shaft), so not only you dont get loot, you dont get experience, your piety decays and you are less prepared for the next floor, who will have stronger enemies.

Even though kill-based piety is bad, i can aknowledge is does bring something to the game, so let me ask you:
How does removing exploration-based piety makes the game better? All your reasonings so far are arbitrary, wrong or they also apply to kill-based piety.

For this message the author sundaye has received thanks:
petercordia

Ziggurat Zagger

Posts: 8786

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

Post Sunday, 1st July 2018, 20:27

Re: Exploration-based piety is bad

sundaye wrote:you can litteraly explore a empty level where no enemies have spawned
Practically speaking, only if that level is the Ecumenical Temple. Otherwise, every level generates 3d12 monsters, 3d8 monsters (Pan), 3d10 (Crypt), or 3d23 capped at 60 (Hell), plus uniques and vaults and ghosts.

While it is theoretically possible for a level to pick MONS_NO_MONSTER for all its monsters, place no uniques, and place no vaults containing monsters, the probability of this happening is roughly 1 in 10000 and that's being generous.

I would also consider such a level giving no piety to be a good thing. Why should you get piety for exploring a level that has no danger?

Temple Termagant

Posts: 10

Joined: Sunday, 1st July 2018, 19:32

Post Sunday, 1st July 2018, 21:03

Re: Exploration-based piety is bad

duvessa wrote:Why should you get piety for exploring a level that has no danger?

Losing piety because the level generator decided so is not good design. Basic dungeon crawler design is to make each floor below harder than the previous one and you are supposed to explore that floor to get exp/piety/consumables to deal with the next one. When the floor decides to not that and instead be just a time waster, how is that good?

Ziggurat Zagger

Posts: 8786

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

Post Sunday, 1st July 2018, 21:18

Re: Exploration-based piety is bad

Yes, perhaps the game shouldn't be able generate empty floors. But you didn't suggest that, you just suggested that exploration piety would work better than kill piety on those empty floors. I submit that the exact opposite is true, and giving 0 piety for those floors is an advantage of kill-based piety.

Temple Termagant

Posts: 10

Joined: Sunday, 1st July 2018, 19:32

Post Sunday, 1st July 2018, 22:08

Re: Exploration-based piety is bad

Offtopic:
duvessa wrote:Yes, perhaps the game shouldn't be able generate empty floors. But you didn't suggest that

Why would i suggest that when the game was changed to be the opposite of that? Its a wasted effort to try and debate the importance of not sending a character from d:1 to d:2 at xl1 because the game didnt generate enough experience on the first floor, amongst other things. This game has the need for absurdly random scenarios to be a possibility, thats why the 1 in a 10000 chance of something happening shouldnt be a valid argument when the game itself considers that chance a reality.

Ontopic:
duvessa wrote:you just suggested that exploration piety would work better than kill piety on those empty floors. I submit that the exact opposite is true, and giving 0 piety for those floors is an advantage of kill-based piety.

You are expecting other things that makes it bad to be eventually changed to accomodate your reasoning, which i will agree with when they happen.

bel

Cocytus Succeeder

Posts: 2184

Joined: Tuesday, 3rd February 2015, 22:05

Post Monday, 2nd July 2018, 02:41

Re: Exploration-based piety is bad

sundaye wrote:
Spoiler: show
Kill-based piety is bad.

Killing enemies is already one of the main things you do in the game and you already receive exp and sometimes loot and food for doing so, why should they give piety other than flavor reasons?

bel wrote:Firstly, why should there be two piety mechanisms? Is anything gained from this?

Yes, for instance, if i am playing a weak race or having a hard time early on, or i am a spriggan, i can base my god choice around one that allows me to gain piety without facing enemies so i can use that piety to defeat enemies and gain exp.
bel wrote:stuff like Abyss existing screws up the mechanism even more.

The existence of abyss screws up kill-based piety just as much. Also you can kill-based piety farm hell by standing still in a staircase.
bel wrote:Secondly, what is piety gain meant to accomplish?

Piety is the middle ground between magic and consumables, but not always. If you worship gozag/Zin piety is your money, is you worship Ru/Xom there is no piety, if you worship Usk piety is "bizarro" magic, that fills up to be used instead of drained as you use it to defeat foes. Different piety systems leads to more dynamic gods.
bel wrote:Piety through kills approximates this intuition quite well. Exploration-based piety does not.

Its explained within each god how one obtains piety. There is nothing unintuitive about it.
bel wrote:Exploration-based piety does not adequately capture the density of monsters, nor the hardness of the floor.

Kill-based piety SPECIALLY doesnt do that! How much piety is a d:3 two-headed ogre worth? How much piety is a d:6 two-headed ogre worth if i fell down a shaft from d:3? 10 goblins = 5 orcs = 1 ogre? How much piety is each of those things? If its too weak i dont get any piety even if i get the message "accepts your kill"?
bel wrote:There is an obvious and decent proxy for how difficult the floor on which you are right now: the XP which you get from the monsters you kill, together with how much you need to rest in between. The obvious way to handle this situation is to have kills-based piety, together with piety decay.

There never was and its even worse now that monster spawns over time have been removed, you can litteraly explore a empty level where no enemies have spawned(or the few who did fell down a shaft), so not only you dont get loot, you dont get experience, your piety decays and you are less prepared for the next floor, who will have stronger enemies.

Even though kill-based piety is bad, i can aknowledge is does bring something to the game, so let me ask you:
How does removing exploration-based piety makes the game better? All your reasonings so far are arbitrary, wrong or they also apply to kill-based piety.

I find fisking annoying and hard to read, so I'll respond in one piece. Hope that's ok.

First, I agree with you on one point: farming XP by standing on an upstairs in Hell is an exploit. If it were up to me, I'd make the spawns in Abyss and Hell durable summons, but that's not on the horizon, so I don't talk about that.

Second, to remove an irrelevance: you mention gods like Uskayaw, Ru and Xom, gods which I specifically excluded from my post. I have no problem with different piety mechanisms if they make sense. I am strictly talking about two basic systems: kills-based and exploration-based piety.

-----------------------

Now to the main point of disagreement. I suggest that you're starting from the wrong place. The correct place to start is the following: suppose there was no piety gain mechanism, and we are interested in designing it from scratch.

The first question to ask would be: what is the goal of the piety gain mechanism? As the saying goes: "if you don't know where you want to go, any direction is fine". The game does not outline any goal. So, I suggested a plausible goal which one should follow: piety is a consumable which you get from following a god, to help you on your way. This rough goal is true for both exploration and kills-based piety, so I started here. From this starting point, I went on to a simple extension: as you progress through the game, the piety you get should increase. Again, this extension is true for both kills- and exploration-based piety.

In contrast, I don't see an answer to this basic question in your post, except "flavour". Flavour is mostly meaningless, because anything can be justified as "flavour". Besides, what is flavourful about giving Elyvilon or Nemelex exploration-based piety? The answer is: nothing. They are the way they are because their old piety gain systems were bad, and someone changed it to the current system as a placeholder.

Once we are clear about the goal, we can look at proposals to achieve that goal. Two of them which exist in the game are: kills-based piety, and exploration-based piety. Which is better? Is one unambiguously better than the other? Is anything gained by having two mechanisms? Note that nobody claimed that kills-based piety is perfect, just that it is much better than exploration-based piety.

-----------------------

Once we use this starting point, the answers to a few questions of yours become clear. For instance, you ask: why should kills give both XP and piety? The simple answer: this is the wrong way to think about it. Kills give XP either way, so it's completely irrelevant. The question is: "how does one get piety"? Should one get piety from killing monsters, or revealing new tiles? I say: the first is better -- much better.

You claim that kill-based piety doesn't capture the hardness of the floors or density of monsters. Yet, you fail to say how exploration-based piety does it either. The case of exploring the empty Temple makes this clear. Why should you get piety for exploring Temple? This bizarre outcome is because of a simple reason: simply revealing tiles does not entail any risk; it is only a risk if the tile contains a monster on it. But if the risk is tied to the monster anyway, then why bother going this roundabout way? Again: I am not claiming that kills-based piety is perfect, but that exploration-based piety is worse -- much worse.

-------------------

Lastly, you (and crawlnoob above), raise the concern that having exploration-based piety supports a different playstyle, so cutting it out could cut out something worthwhile. Here, one has to actually look at the evidence -- one can't simply make things up. I have played Crawl for years, and I have yet to hear of anyone who picks a god based on whether they give exploration- or kills-based piety. If you look at the top speedrunning gods (Chei, Trog, Makhleb, Vehumet), they all give kills-based piety. I do not observe many people playing Spriggan stabbers of Elyvilon simply because Elyvilon gives exploration-based piety. So what is the evidence for the claim that exploration-based piety supports a unique playstyle?

For this message the author bel has received thanks:
duvessa

Ziggurat Zagger

Posts: 4432

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

Post Monday, 2nd July 2018, 06:58

Re: Exploration-based piety is bad

My main complaint about piety decay is that it punishes using autoexpore instead of manual exploration and in that regard exploration piety is better than piety for kills.
If exploring Abyss is a problem (I don't think Pan is easy to explore even as high level character), then there are several solutions:
1) High level character cannot enter low levels of Abyss (if going for Abyssal rune, characters already can unwield distortion weapon to get into late Abyss instantly)
2) Piety gain for explore is decreased in low levels of Abyss for high level character
Underestimated: cleaving, Deep Elf, Formicid, Vehumet, EV
Overestimated: AC, GDS
Twin account of Sandman25

Tartarus Sorceror

Posts: 1774

Joined: Tuesday, 23rd December 2014, 23:39

Post Wednesday, 4th July 2018, 00:54

Re: Exploration-based piety is bad

I think it's just a matter of flavor. Elyvilon is a healing and peaceful god, it should not reward killing. Ash is a god about knowledge, so exploration piety makes sense. In the case of Nemelex the exploration piety does seem arbitrary.
streaks: 5 fifteen rune octopodes. 15 diverse chars. 13 random chars. 24 NaWn^gozag.
251 total wins Berder hyperborean + misc
83/108 recent wins (76%)
guides: safe tactics value of ac/ev/sh forum toxicity

For this message the author Berder has received thanks: 3
byrel, Rast, Wahaha

Spider Stomper

Posts: 248

Joined: Monday, 4th September 2017, 10:53

Post Tuesday, 24th July 2018, 18:29

Re: Exploration-based piety is bad

Bel, whether or not people actually pick a god based on the type of piety-mechanism it supports does not necessarily indicate whether or not that mechanism supports a specific play style or not. People can and do decide to pick gods for other reasons entirely, some of which may or may not be more important than the piety gain mechanism.

Furthermore, you have no idea how all the players pick their gods. You can argue that you've never seen (or remember,actually) anybody talking about it in the percentage of total threads in the tavern which you have read, but thats a sample of a sample of a sample, so hardly a solid argument for your claim.

I myself specifically avoided certain gods for a long time (and still do) based on their, as I saw them, quirky mechanisms.

Vestibule Violator

Posts: 1508

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

Post Wednesday, 3rd July 2019, 13:49

Re: Exploration-based piety is bad

First, I agree with the design goal for piety that you proposed; piety is a tool you gain 'as you progress through the game, which you spend to make your life easier' and 'generally speaking, you should get more piety as you progress deeper into the game and venture into more difficult areas'.

I think that your objections to exploration-based piety can be summarized as:

  1. It doesn't model the difficulty of the challenges you're facing (monster density, etc.)
  2. It is superfluous, and provides no benefit to offset the complexity of an additional piety-gain mechanic

Point 1 seems to take a leap away from your initial premise that piety is a progression reward, that should scale with progression. Crawl is not a game of 'kill all the monsters'. Killing more monsters is at least as tenuously related to progression as exploration. Neither are generated after level gen (ignoring Pan/Abyss/Hell), and level gen has high variance for both for a given amount of progression. I don't see an argument for kill-piety here. If Shoals:4 has higher monster density, does the barnacled rune contribute to your progression more? If it has fewer monsters, does it somehow progress less? At the most fundamental level, crawl's a game of macguffin retrieval, not killing. There's a mild correlation between level XP and 'being a rune ending', so kill piety could be a better fit... but that that assumes that characters actually clear the entire rune end floor. That's almost certainly not optimal for some of them (like Shoals) for some characters, but is optimal for others (Vaults) and other characters.

And that's where point 2 is wrong as well. Different characters progress through the game with different percentages of the floor killed. A melee spriggan spends a lot of time putting exclusions on all the hard sleeping monsters, and never waking them. Does he somehow not progress as much because he's avoiding threats instead of killing them? IMO, that's one of the core engagements of crawl; you don't have to kill something to progress. And exploration piety does a much better job measuring her progress; you get almost full explore piety for the floor regardless of the number of hard monsters you had to exclude.

Given that each models our desired standard (game progression) better in some cases and for some characters, it makes sense for there to be some gods with each to cater to characters on each end of the spectrum.
Usual account: pblur on kelbi

bel

Cocytus Succeeder

Posts: 2184

Joined: Tuesday, 3rd February 2015, 22:05

Post Wednesday, 3rd July 2019, 14:26

Re: Exploration-based piety is bad

In general, the point I'm making is not that kills-based piety is perfect, but that it captures the goal better then exploration-based piety.

Currently, exploring Shoals:4 gives almost as much piety as exploring Shoals:3, even though the former is tougher. Why is it tougher? Because it has tougher and more monsters. It is deliberately made tougher because the game is trying to communicate that there is a reward for you if you overcome the challenges. If piety is supposed to help you overcome challenges as you progress through the game, then it should depend on the challenges. The XP of the monsters you kill provides a good proxy for the challenges you encounter; exploration does not.

It is true that you do not have to kill monsters to win. However, that does not change the basic argument. Even if you kill half of the monsters on the floor, the reward will still be proportional to the challenge. Leaving monsters alive is completely fine, because you have successfully avoided the challenge of fighting them (using stealth or whatever). I am not opposed to using a Sil-style mechanic where you get some XP for just seeing the monster and some more XP for actually killing it. I think bcrawl or X-crawl implements this mechanic of "get XP for seeing the monster" mechanic for Felids. However, it's not really necessary for my argument.

Moreover, Crawl is not really designed as a pacifist game; you generally kill most monsters you encounter (except if you're in the Abyss or something). Only some monsters which are hard to kill because of your build (say, thorn hunters as an Enchanter), or few uniques which are not worth killing (because they don't give enough XP, hint, hint) are typically left alone. If Mara gave 10 times the XP it gives currently, I would definitely try to kill it most of the time.

As a specific point, I do not think point 1 is contradictory to my premise. Recall that my premise says: "...you progress deeper into the game and venture into more difficult areas." (emphasis added). So it is not just progression, but also how tough the challenges you encounter as you progress. Point 1 is just the second part of my premise.

Return to Game Design Discussion

Who is online

Users browsing this forum: No registered users and 21 guests

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