Understanding spell power


Ask fellow adventurers how to stay alive in the deep, dark, dangerous dungeon below, or share your own accumulated wisdom.

User avatar

Dis Charger

Posts: 1949

Joined: Monday, 18th February 2013, 07:59

Location: France

Post Saturday, 10th May 2014, 19:08

Understanding spell power

Quoted from wiki :
  Code:
Value          Description
0-9          #.........
10-14        ##........
15-24        ###.......
25-34        ####......
35-49        #####.....
50-74        ######....
75-99        #######...
100-149      ########..
150-199      #########.
200          ##########

If I get it well, having for example Firestorm at 100 spell powe will display ########..
But it's half as good as ########## ?

Isn't it a bit misleading ?
Online stats
Fastest Hell runes (enter Hell branch -> get the rune)
Icy : 56 / Iron : 126 / Obsidian : 215 / Bone : 125

Ziggurat Zagger

Posts: 11111

Joined: Friday, 8th February 2013, 12:00

Post Saturday, 10th May 2014, 19:13

Re: Understanding spell power

Yes, it is misleading. Primarily because displayed spell power does not help to understand expected damage. For some spells doubled spell power increases damage by roughly 15%.

Shoals Surfer

Posts: 257

Joined: Tuesday, 24th September 2013, 17:52

Post Saturday, 10th May 2014, 19:25

Re: Understanding spell power

What's the purpose of abstracting it to such a misleading format then? Why not list effective spellpower as a number (including the currently hidden boosts that Sandman says exist)?
Current foes list: duvessa, TheDefiniteArticle

Ziggurat Zagger

Posts: 11111

Joined: Friday, 8th February 2013, 12:00

Post Saturday, 10th May 2014, 19:30

Re: Understanding spell power

"Hiding numbers which don't matter" I guess.

Shoals Surfer

Posts: 257

Joined: Tuesday, 24th September 2013, 17:52

Post Saturday, 10th May 2014, 19:32

Re: Understanding spell power

*faceplam* Protecting us from ourselves. I get enough of that from my government. I don't need more in my video games.
Current foes list: duvessa, TheDefiniteArticle

For this message the author skjarl has received thanks:
mopl

Slime Squisher

Posts: 354

Joined: Tuesday, 14th January 2014, 23:33

Post Saturday, 10th May 2014, 19:55

Re: Understanding spell power

Suppose we use numbers for spellpower. So you have 50 spellpower for lightning bolt, and 60 spellpower for Fireball. In what way does that information help you win the game? Unless you also know the accuracy and damage formulas for both of those spells, precise spellpower numbers are useless to you. You might say "it's useful to know which spell has more power" but actually Lightning Bolt uses 3/5s of your spellpower (or 30 in this example), and Fireball only uses 1/2 (also 30 in this example). Does that mean they do the same damage? Nope.

But forget reasonable stuff. Let's look at a fun example! You have 80 power for Airstrike. What does that mean? Do you have any idea? I don't, and I've won the game a couple dozen times. Here, I'll check the damage formula:
  Code:
7 + 1d(1d4 - 1 + (1d(Power) - 1)/6) + (1d(Power) - 1)/7

Nope, still no fucking clue what 80 power means for Airstrike.

mopl wrote:Isn't it a bit misleading ?

Not any more misleading than the fact that raising a skill from 25 to 26 costs like a hundred times as much XP as raising the same skill from 5 to 6.

For this message the author TheDefiniteArticle has received thanks: 2
nago, Sar

Shoals Surfer

Posts: 257

Joined: Tuesday, 24th September 2013, 17:52

Post Saturday, 10th May 2014, 20:05

Re: Understanding spell power

Ok, I get it that spell damage formulas are complex. So present players with what they REALLY care about: damage ranges (heck, you could even calculate their average damage for them so they can eyeball it). That's not useless information. Right now, the graph in the OP's post is just as useless as a spell power number is for the exact same reason you just said.
Current foes list: duvessa, TheDefiniteArticle

For this message the author skjarl has received thanks: 2
Hurkyl, zxc23

Ziggurat Zagger

Posts: 8786

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

Post Saturday, 10th May 2014, 20:08

Re: Understanding spell power

TheDefiniteArticle wrote:But forget reasonable stuff. Let's look at a fun example! You have 80 power for Airstrike. What does that mean? Do you have any idea? I don't, and I've won the game a couple dozen times. Here, I'll check the damage formula:
  Code:
7 + 1d(1d4 - 1 + (1d(Power) - 1)/6) + (1d(Power) - 1)/7

Nope, still no fucking clue what 80 power means for Airstrike.
It means 7 + 1d(1d4 - 1 + 1d(80) - 1)/6) + (1d(80) - 1)/7

For this message the author duvessa has received thanks: 5
Arrhythmia, TheDefiniteArticle, vengefulcarrot, Weeksy, zxc23

Ziggurat Zagger

Posts: 11111

Joined: Friday, 8th February 2013, 12:00

Post Saturday, 10th May 2014, 20:28

Re: Understanding spell power

skjarl wrote:Ok, I get it that spell damage formulas are complex. So present players with what they REALLY care about: damage ranges (heck, you could even calculate their average damage for them so they can eyeball it). That's not useless information. Right now, the graph in the OP's post is just as useless as a spell power number is for the exact same reason you just said.


Unfortunately it's quite hard to calculate in advance.
My mod displayed possible range for everything, something like "Rolled 17 out of 3-25. Black bear decreases damage by 2, loses 15 HP, 6 HP remaining" and I stopped paying attention to it after some time.

Shoals Surfer

Posts: 257

Joined: Tuesday, 24th September 2013, 17:52

Post Saturday, 10th May 2014, 21:07

Re: Understanding spell power

Sandman25 wrote:Unfortunately it's quite hard to calculate in advance.
My mod displayed possible range for everything, something like "Rolled 17 out of 3-25. Black bear decreases damage by 2, loses 15 HP, 6 HP remaining" and I stopped paying attention to it after some time.


I'm not so much concerned about getting an actual damage dealt value from the game as being able to determine how much stronger iron shot is than stone shot in a rough fashion without source diving the damage formulas. It would be nice to get a resultant damage number, but one step a time I guess. You may recall a similar thread I started a while back about determining the relative strength of weapons. There are some clever folks working on this game and players should not be having trouble figuring out which weapon or spell is most effective at any given point. More information is always better than less when you're trying to make informed decisions.

Anyway, I'm sure this has been argued to death already so I'll leave it alone.
Current foes list: duvessa, TheDefiniteArticle

Ziggurat Zagger

Posts: 11111

Joined: Friday, 8th February 2013, 12:00

Post Saturday, 10th May 2014, 21:37

Re: Understanding spell power

I mean I used those "rolled ..." lines to compare spells, weapons, god abilities, large rocks etc. There is no other way IMHO as it depends on target (AC/EV/resistances).

Ziggurat Zagger

Posts: 6454

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

Post Saturday, 10th May 2014, 21:56

Re: Understanding spell power

skjarl wrote:
Sandman25 wrote:Unfortunately it's quite hard to calculate in advance.
My mod displayed possible range for everything, something like "Rolled 17 out of 3-25. Black bear decreases damage by 2, loses 15 HP, 6 HP remaining" and I stopped paying attention to it after some time.


I'm not so much concerned about getting an actual damage dealt value from the game as being able to determine how much stronger iron shot is than stone shot in a rough fashion without source diving the damage formulas. It would be nice to get a resultant damage number, but one step a time I guess. You may recall a similar thread I started a while back about determining the relative strength of weapons. There are some clever folks working on this game and players should not be having trouble figuring out which weapon or spell is most effective at any given point. More information is always better than less when you're trying to make informed decisions.

Anyway, I'm sure this has been argued to death already so I'll leave it alone.


Indeed, part of the problem is that "Average damage" changes depending on the target, so one spell may roll 4d2, one 2d4 and one may roll 1d8 for damage, and while they may technically do the same damage on average *to a creature with 0 AC* (Note that these rolls do not, this is simplified for the purposes of clarity only.) the average damage will change as the AC of the critter you're facing does (In general, if the "before AC" average is above the AC of the critter, it's better to have more dice and more of your damage close to middle of the distribution, and vice versa) So even just presenting an "average damage" is misleading when comparing different spells to each other, as the distribution of pre-ac damage also changes the net post-ac average damage, even indicating a range would be insufficent to convey this information.

Really the power level of the spell is less important than the fact that you can cast it at all. For most spells it's important that you get your failure rate down low, it's not important whether you have one more # of spellpower (And it takes exponentially more training to get each # anyway, there's no way you'll get to 200 spellpower in a normal game until long after you could have won.)

The one exception to this are hexes (where there's a pretty rapid fall-off in your odds of success as your spellpower is contested against their MR) However (now) you see a message which indicates your odds of success with a hex, so if you're failing you know whether it's because you need more spellpower, or because you just have bad luck.
Last edited by Siegurt on Saturday, 10th May 2014, 21:56, edited 1 time in total.
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:
Lasty

Slime Squisher

Posts: 330

Joined: Thursday, 10th May 2012, 03:29

Post Saturday, 10th May 2014, 21:56

Re: Understanding spell power

skjarl wrote:Ok, I get it that spell damage formulas are complex. So present players with what they REALLY care about: damage ranges (heck, you could even calculate their average damage for them so they can eyeball it). That's not useless information. Right now, the graph in the OP's post is just as useless as a spell power number is for the exact same reason you just said.


Except you don't know how much hp a monster has. Well I don't at least. when I get more ## it feels good because it means my spell got more powerful somehow. Mostly I just go by feeling though, but it's nice to be able to compare between characters sometimes: "I had about half the bar filled up with # by this point last time and now I have a bit more, nice".

One thing I don't want in crawl is more numbers since they tempt you do calculations instead of playing and just going by feel (which is more fun). Also the specific numbers are really not what you need to be thinking about when you're trying to win.

Basically you use your powerful spells on dangerous enemies when it's safe to do so.

Ziggurat Zagger

Posts: 11111

Joined: Friday, 8th February 2013, 12:00

Post Saturday, 10th May 2014, 22:24

Re: Understanding spell power

Technically it's very easy to display
"You hit the sheep (3 damage)"
when
show_damage=true (false by default) in your RC/init.txt and I am not sure why it is not implemented provided we have "!!" hints for damage anyway.
It would greatly ease weapon/spells comparison

Ziggurat Zagger

Posts: 4055

Joined: Tuesday, 10th January 2012, 19:49

Post Sunday, 11th May 2014, 01:20

Re: Understanding spell power

It is a design decision for crawl to not give you damage numbers and I think it is very unlikely to be changed. Posting here is not even talking to the correct people if you are trying to get that changed.

Ziggurat Zagger

Posts: 11111

Joined: Friday, 8th February 2013, 12:00

Post Sunday, 11th May 2014, 01:26

Re: Understanding spell power

No, I am not trying to change it. Just whining a little as usual...

Vestibule Violator

Posts: 1601

Joined: Sunday, 14th July 2013, 16:36

Post Sunday, 11th May 2014, 01:42

Re: Understanding spell power

Siegurt wrote:the average damage will change as the AC of the critter you're facing does (In general, if the "before AC" average is above the AC of the critter, it's better to have more dice and more of your damage close to middle of the distribution, and vice versa) So even just presenting an "average damage" is misleading when comparing different spells to each other, as the distribution of pre-ac damage also changes the net post-ac average damage, even indicating a range would be insufficent to convey this information.

Actually, I'm pretty sure that's exactly backwards: you want to have high variance, not low variance. If your damage roll tends strongly to the mean, you will be consistently doing a small amount of damage. But if your damage roll has high variance, your low damage rolls lose very little but your high damage rolls punch through the AC better.

More mathematically, the difference that AC makes is that if the damage roll is lower than the AC check, the damage done is increased to zero. The average damage you do is

  Code:
E[damage roll] - E[AC roll] + E[bonus due to raising negative damage to zero] - E[overkill]


Thus, 1d15 should do a lot better than 4d3 against monsters that have AC, because it gets more of that bonus. Of course, high variance makes the overkill penalty worse, but that's separate from the effect you were discussing.
User avatar

Dis Charger

Posts: 1949

Joined: Monday, 18th February 2013, 07:59

Location: France

Post Sunday, 11th May 2014, 08:34

Re: Understanding spell power

Thanks for all answers, what I needed to know is that spell damage isn't easily related to spell power (200 spell power isn't twice the damage for 100 spell power)
Online stats
Fastest Hell runes (enter Hell branch -> get the rune)
Icy : 56 / Iron : 126 / Obsidian : 215 / Bone : 125

Ziggurat Zagger

Posts: 5382

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

Post Monday, 12th May 2014, 02:31

Re: Understanding spell power

Sandman25 wrote:Technically it's very easy to display
"You hit the sheep (3 damage)"
when
show_damage=true (false by default) in your RC/init.txt and I am not sure why it is not implemented provided we have "!!" hints for damage anyway.
It would greatly ease weapon/spells comparison

In my dreams I make a fork of DCSS called "transparent crawl" and I show this damage, your actual piety, your actual spell power, your regeneration rate, make a new "statistics" screen which shows things like the most damage you've taken in one hit, most in one turn, the most damage you've done in a single hit, in a single action, and other sorts of things. I then make windows/OSX/linux builds, post them on sourceforge somewhere, and am everyone's hero. This wouldn't even really be that hard to do, if I was a slightly more competent programmer. Probably like a week's worth of work for someone with good skills. I, however, am lazy and stupid, so I just go back to playing DCSS instead. Making a fork is the only way you'll get this - it's been turned down hundreds of times.

Dungeon Master

Posts: 1613

Joined: Thursday, 16th December 2010, 21:54

Post Monday, 12th May 2014, 02:54

Re: Understanding spell power

Luckily that already exists under the name of "make debug", it's ever so transparent.

For this message the author Kate has received thanks:
sgrunt
User avatar

Blades Runner

Posts: 538

Joined: Saturday, 15th February 2014, 03:22

Location: NYC

Post Monday, 12th May 2014, 07:17

Re: Understanding spell power

MarvinPA wrote:Luckily that already exists under the name of "make debug", it's ever so transparent.

Doesn't that negate your score?

Ziggurat Zagger

Posts: 6454

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

Post Monday, 12th May 2014, 17:18

Re: Understanding spell power

Hurkyl wrote:Actually, I'm pretty sure that's exactly backwards: you want to have high variance, not low variance. If your damage roll tends strongly to the mean, you will be consistently doing a small amount of damage. But if your damage roll has high variance, your low damage rolls lose very little but your high damage rolls punch through the AC better.


Yes, I that's actually what I meant to say (it's there, but confusingly worded)

High variance is better against high AC, low variance is better against low AC.
  Code:
 | /\                    _|____
_|/  \_ is better than    |   

but
  /\|                  ___|__
_/  |\_ is worse than     | 

(Where the vertical line indicates the AC reduction and the horizontal/diagonal lines are the "before AC damage distribution")
Oh and yay for ascii graphs :)
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!

Abyss Ambulator

Posts: 1205

Joined: Friday, 8th November 2013, 17:02

Post Monday, 12th May 2014, 17:56

Re: Understanding spell power

I know people keep saying it's complicated to show current spell damage, but at the same time, it seems like a lot of spells do something like "XdieY damage", where X is spellpower, and Y is some constant.

Couldn't the game just show us that? So I'd see my magic dart does 2D3, and my mystic blast does 5D9, and iron shot does 7D8. Now I at least have a basic idea of how much damage to expect from each spell. Maybe some spells like airstrike could get a note saying "This spell does considerably lower damage to targets with high AC" or something along those lines.

Then there's some way of comparing spells without spending an hour in wizmode, or reading through outdated spreadsheets.

Barkeep

Posts: 3890

Joined: Wednesday, 14th August 2013, 23:25

Location: USA

Post Monday, 12th May 2014, 19:02

Re: Understanding spell power

A lot of spells have much more complicated damage formulae than "XdieY"

Also some spells check AC multiple times, some ignore EV, some ignore part of AC, some ignore all of AC. Etc. This helps differentiate a bit amongst the many spells that deal damage, but it makes it very difficult to give the player information that is simple, meaningful, and not misleading.

I think it is more important to avoid giving misleading information and to keep information simple than it is to help a player foist some kind of min-maxing strategy upon a game* in which such approaches don't work well for lots of other reasons, anyway. The spell power display let's you know when you have hit max or near-max power, that's the important thing.

* I don't mean this dismissively. But it is simply a fact that you do not win Crawl, or even notably improve your odds of winning, by doing some sort of uber-OCD optimization of skill allocation. When you get good success and solid spell power going on mid- or high-level spells, training further just to improve those spells is low priority. Low level spells will usually be maxed out in the process of getting higher level spells. When you hit min-delay with a weapon, training further is extremely low priority. Hexes are a bit different and there are some exceptions but for the most part, if you get this, you know what you need to.

Temple Termagant

Posts: 6

Joined: Friday, 2nd May 2014, 21:53

Post Monday, 12th May 2014, 20:40

Re: Understanding spell power

crate wrote:It is a design decision for crawl to not give you damage numbers and I think it is very unlikely to be changed. Posting here is not even talking to the correct people if you are trying to get that changed.


I don't want to argue about it but I am curious about the justification for it because I can't think of any.

Ziggurat Zagger

Posts: 4055

Joined: Tuesday, 10th January 2012, 19:49

Post Monday, 12th May 2014, 21:51

Re: Understanding spell power

under_bridge wrote:
crate wrote:It is a design decision for crawl to not give you damage numbers and I think it is very unlikely to be changed. Posting here is not even talking to the correct people if you are trying to get that changed.


I don't want to argue about it but I am curious about the justification for it because I can't think of any.

As has been pointed out in this topic, for actual damage numbers to mean anything you also need to know several other numbers. This is a lot of numbers to show, and crawl wants (you can argue about how much it succeeds, this is not the place for that) to show only numbers which are important and, ideally, also easy-to-figure-out. This is useful because it means if crawl is showing you a number you know it matters; knowing that flame tongue deals up to 14 damage and throw flame up to 18 (I made these numbers up) is not something the devs think is that important.

Broadly, figuring out which spells deal more damage than other spells is something you can learn through experience. You are unlikely to be able to tell the difference between spells that are similar, but seeing that Poison Arrow outdamages Iron Shot is not terribly hard.

Certainly you can debate about how well crawl succeeds in this goal of trying to display important information and hide unimportant information (and whether it is desirable in the first place) but that is the idea.

Snake Sneak

Posts: 128

Joined: Friday, 26th April 2013, 05:09

Post Monday, 12th May 2014, 22:05

Re: Understanding spell power

under_bridge wrote:
crate wrote:It is a design decision for crawl to not give you damage numbers and I think it is very unlikely to be changed. Posting here is not even talking to the correct people if you are trying to get that changed.


I don't want to argue about it but I am curious about the justification for it because I can't think of any.


Numbers make you focus on the details and forget the big picture and they introduce tedium.

For me, 2d6 doesn't tell me that much about how much damage I'm going to do. On the other hand, firing off a spell at an enemy a few times tells me exactly what I need to know. I would rather that players were encouraged to take this experimental method than forced to get their calculators out and crunch the numbers in order to play optimally.

I've been playing a bit of Battle for Wesnoth lately and in BfW you get to see all the numbers. It even gives you a histogram for how much HP your unit and the enemy unit might have after an attack. BfW is a great game (you should try it if you haven't before, it's opensource :) ) but in my opinion these numbers introduce tedium and I would rather have it somehow done similarly to crawl.

I believe this is the primary reason for avoiding numbers in crawl and in fact it is one of the things I love about the game.

Ziggurat Zagger

Posts: 11111

Joined: Friday, 8th February 2013, 12:00

Post Monday, 12th May 2014, 22:10

Re: Understanding spell power

Kramin42 wrote:I've been playing a bit of Battle for Wesnoth lately and in BfW you get to see all the numbers. It even gives you a histogram for how much HP your unit and the enemy unit might have after an attack. BfW is a great game (you should try it if you haven't before, it's opensource :) ) but in my opinion these numbers introduce tedium and I would rather have it somehow done similarly to crawl.

I believe this is the primary reason for avoiding numbers in crawl and in fact it is one of the things I love about the game.


Damage is deterministic in BfW, that's why it is very important. I often look at those histograms to see if I really want to risk or better use two units playing safe.

Vestibule Violator

Posts: 1601

Joined: Sunday, 14th July 2013, 16:36

Post Tuesday, 13th May 2014, 03:19

Re: Understanding spell power

Kramin42 wrote:For me, 2d6 doesn't tell me that much about how much damage I'm going to do. On the other hand, firing off a spell at an enemy a few times tells me exactly what I need to know. I would rather that players were encouraged to take this experimental method than forced to get their calculators out and crunch the numbers in order to play optimally.

If I see 2d6 and 2d5, I immediately know how those values relate. If I'm firing spells at enemies, it will take hundreds of attempts to have anything mildly statistically significant, and I'm quite liable to infer the wrong thing long before I get to that point and get stuck on it due to confirmation bias.

Swamp Slogger

Posts: 137

Joined: Wednesday, 14th December 2011, 16:11

Location: Australia

Post Tuesday, 13th May 2014, 09:27

Re: Understanding spell power

Speaking of numbers, I think showing piety count would be very helpful (and would also tell you if you're sustaining piety or about to lose a particular ability - important for gods like zin).

Ziggurat Zagger

Posts: 4055

Joined: Tuesday, 10th January 2012, 19:49

Post Tuesday, 13th May 2014, 09:50

Re: Understanding spell power

Well piety is hidden from the player because you are not supposed to know your exact piety value (and in fact you cannot, since piety gain, piety decay, and piety costs are all randomised).

Abyss Ambulator

Posts: 1205

Joined: Friday, 8th November 2013, 17:02

Post Tuesday, 13th May 2014, 14:26

Re: Understanding spell power

Hurkyl wrote:
Kramin42 wrote:For me, 2d6 doesn't tell me that much about how much damage I'm going to do. On the other hand, firing off a spell at an enemy a few times tells me exactly what I need to know. I would rather that players were encouraged to take this experimental method than forced to get their calculators out and crunch the numbers in order to play optimally.

If I see 2d6 and 2d5, I immediately know how those values relate. If I'm firing spells at enemies, it will take hundreds of attempts to have anything mildly statistically significant, and I'm quite liable to infer the wrong thing long before I get to that point and get stuck on it due to confirmation bias.


Yes, exactly this, because crawl is played by humans, who are AWFUL at statistical analysis. The idea that you should be able to figure out which of two weapons or spells is better through experience is based on the idea that humans will percieve slight differences in randomized patterns over time. In fact, people cannot do that, and instead will inject their own incorrect assumptions based on their incomplete knowledge, then various biases will only entrench the player further into their incorrect assumptions.

Temple Termagant

Posts: 6

Joined: Friday, 2nd May 2014, 21:53

Post Tuesday, 13th May 2014, 18:42

Re: Understanding spell power

Even games that tell you almost nothing else in the way of statistics provide numbers for damage dealt after a hit lands. It's by no means a meaningless, abstract, or incomprehensible value. Hiding it is as weird as hiding the player's HP, in my view.

(Wait, I said I *didn't* want to argue this, my bad)

For this message the author under_bridge has received thanks:
Sandman25

Ziggurat Zagger

Posts: 5382

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

Post Tuesday, 13th May 2014, 18:51

Re: Understanding spell power

crate wrote:Well piety is hidden from the player because you are not supposed to know your exact piety value (and in fact you cannot, since piety gain, piety decay, and piety costs are all randomised).

If you haven't used any chei powers in the last few floors, after being at 6*, it is extremely likely your piety has reached 200 :)

Temple Termagant

Posts: 6

Joined: Friday, 2nd May 2014, 21:53

Post Tuesday, 13th May 2014, 18:54

Re: Understanding spell power

Even games that tell you almost nothing else in the way of statistics provide numbers for damage dealt after a hit lands. It's by no means a meaningless, abstract, or incomprehensible value. Hiding it is as weird as hiding the player's HP, in my view.

(Wait, I said I *didn't* want to argue this, my bad)

Ziggurat Zagger

Posts: 11111

Joined: Friday, 8th February 2013, 12:00

Post Tuesday, 13th May 2014, 19:46

Re: Understanding spell power

I really enjoyed watching my piety when playing with my mode. Piety reached 200? Time to use some god abilities just for fun.

Lair Larrikin

Posts: 17

Joined: Sunday, 15th December 2013, 05:46

Post Wednesday, 14th May 2014, 18:31

Re: Understanding spell power

As a complete aside, this thread has made me want to see how BFW is doing since I stopped playing back in 2008 or so.

Barkeep

Posts: 3890

Joined: Wednesday, 14th August 2013, 23:25

Location: USA

Post Wednesday, 14th May 2014, 19:23

Re: Understanding spell power

damiac wrote:
Hurkyl wrote:
Kramin42 wrote:For me, 2d6 doesn't tell me that much about how much damage I'm going to do. On the other hand, firing off a spell at an enemy a few times tells me exactly what I need to know. I would rather that players were encouraged to take this experimental method than forced to get their calculators out and crunch the numbers in order to play optimally.

If I see 2d6 and 2d5, I immediately know how those values relate. If I'm firing spells at enemies, it will take hundreds of attempts to have anything mildly statistically significant, and I'm quite liable to infer the wrong thing long before I get to that point and get stuck on it due to confirmation bias.


Yes, exactly this, because crawl is played by humans, who are AWFUL at statistical analysis. The idea that you should be able to figure out which of two weapons or spells is better through experience is based on the idea that humans will percieve slight differences in randomized patterns over time. In fact, people cannot do that, and instead will inject their own incorrect assumptions based on their incomplete knowledge, then various biases will only entrench the player further into their incorrect assumptions.


If the patterns are so slight they cannot be noticed, I would posit that the player should not concern him or herself with them. I would also posit that the very fact many people are not very intuitively good at processing probabilities also greatly limits the usefulness of providing people with information like, "This spell does 3d6 damage," "This other spell does 5+2d4 damage," "This spell does 1d20 damage." Especially if one of those spells ignores EV while another one checks AC twice and takes highest value.

There are unintuitive things about Crawl that are problematic, and some things which I think could/should be made transparent. But exact damage values and the like is not one of them.

Nearly every death is directly and primarily attributable to a tactical mistake. This is what it means to say that a death (even with a badly skilled dude) is avoidable. Of those relatively few instances in which bad skilling is a significant contributing factor to a death, the bad skill management and/or gear choices were bad because they left a character too vulnerable for whatever point in the game they were at.

Maybe you disagree with the above assessment. But if not, then I think it is fair to ask what value would actually be gained by foisting this information upon players, and especially upon players who already have enough on their plate trying to really master/internalize the interface and commands and basic tactics. People who really want to know can just look the stats up online.

The actually useful information for a player is stuff like, "Is this sufficient to kill an ogre before it kills me?", and realistically, people are only going to develop a sense of that through experience, anyway.

For this message the author and into has received thanks: 3
Kramin42, Lasty, zardo
User avatar

Blades Runner

Posts: 538

Joined: Saturday, 15th February 2014, 03:22

Location: NYC

Post Wednesday, 14th May 2014, 23:09

Re: Understanding spell power

and into wrote:[snipped quotes]

If the patterns are so slight they cannot be noticed, I would posit that the player should not concern him or herself with them. I would also posit that the very fact many people are not very intuitively good at processing probabilities also greatly limits the usefulness of providing people with information like, "This spell does 3d6 damage," "This other spell does 5+2d4 damage," "This spell does 1d20 damage." Especially if one of those spells ignores EV while another one checks AC twice and takes highest value.

There are unintuitive things about Crawl that are problematic, and some things which I think could/should be made transparent. But exact damage values and the like is not one of them.

Nearly every death is directly and primarily attributable to a tactical mistake. This is what it means to say that a death (even with a badly skilled dude) is avoidable. Of those relatively few instances in which bad skilling is a significant contributing factor to a death, the bad skill management and/or gear choices were bad because they left a character too vulnerable for whatever point in the game they were at.

Maybe you disagree with the above assessment. But if not, then I think it is fair to ask what value would actually be gained by foisting this information upon players, and especially upon players who already have enough on their plate trying to really master/internalize the interface and commands and basic tactics. People who really want to know can just look the stats up online.

The actually useful information for a player is stuff like, "Is this sufficient to kill an ogre before it kills me?", and realistically, people are only going to develop a sense of that through experience, anyway.

I entirely agree with what you said here, YET I still think you and others against this feature may be missing a primary point. Fun. Some people (I am not saying all or even most by any means, but definitely some) enjoy seeing D&D like stats such as a fireball that does 8d6+12. Sure the numbers don't really mean anything to most people but some people will find them meaningful in a ~ way and others will just enjoy thinking that they do. I am against hiding information in order to "protect" the user. I think that should be something that is optional. Maybe the option starts out as off (the same way that manual skill selection does) but is set up to allow people to turn it on at whim.

Vestibule Violator

Posts: 1601

Joined: Sunday, 14th July 2013, 16:36

Post Thursday, 15th May 2014, 00:32

Re: Understanding spell power

and into wrote:If the patterns are so slight they cannot be noticed, I would posit that the player should not concern him or herself with them.

"So slight a human doesn't notice them" is still rather big. Correlating the effect to the cause is an even huger problem.

The actually useful information for a player is stuff like, "Is this sufficient to kill an ogre before it kills me?", and realistically, people are only going to develop a sense of that through experience, anyway.

Actual game experience told me that a +9 quick blade of freezing will kill that Skeletal Warrior before a +6 long sword of holy wrath.

That's actually very wrong.

Slime Squisher

Posts: 354

Joined: Tuesday, 14th January 2014, 23:33

Post Thursday, 15th May 2014, 01:13

Re: Understanding spell power

Correlating the effect to the cause is an even huger problem.

I don't understand what you mean. The cause is "I used attack A on target X" or "I used attack B on target X". The effect is damage. It couldn't be more straightforward.

Actual game experience told me that a +9 quick blade of freezing will kill that Skeletal Warrior before a +6 long sword of holy wrath.

That's actually very wrong.

There's only one possible way that "actual game experience" told you that weapon A could kill a skeletal warrior faster than weapon B. Which is that in an actual game you experienced weapon A killing a skeletal warrior faster than weapon B killing another skeletal warrior. And the only way that could be wrong is if you experienced it in an older version of Crawl (i.e. a game with different rules)

I still think you and others against this feature may be missing a primary point. Fun. Some people (I am not saying all or even most by any means, but definitely some) enjoy seeing D&D like stats

Fun is not an argument. The word means everything, and therefore also means nothing. It's perfectly possible - and reasonable - for someone to enjoy a bad mechanic.

Vestibule Violator

Posts: 1601

Joined: Sunday, 14th July 2013, 16:36

Post Thursday, 15th May 2014, 02:31

Re: Understanding spell power

TheDefiniteArticle wrote:
Correlating the effect to the cause is an even huger problem.

I don't understand what you mean. The cause is "I used attack A on target X" or "I used attack B on target X". The effect is damage. It couldn't be more straightforward.

Crawl is deliberately designed so you don't see that effect.

The effect "I'm killing dudes faster" is impossible to see by eye except in very drastic circumstances: this is ensured by randomness, the heterogeneous distribution of monsters, inconsistencies in which options you actually have to consider, and the amount of time between comparing different alternatives. (all of these factors would apply to seeing the actual damage rolls too)

The effect "I do better in some games than others" could be caused by a bajillion things, including sheer dumb luck.

So no, it really isn't straightforward at all.

There's only one possible way that "actual game experience" told you that weapon A could kill a skeletal warrior faster than weapon B. Which is that in an actual game you experienced weapon A killing a skeletal warrior faster than weapon B killing another skeletal warrior. And the only way that could be wrong is if you experienced it in an older version of Crawl (i.e. a game with different rules)

Or, y'know, it could be all of those other things.

Fun is not an argument. The word means everything, and therefore also means nothing. It's perfectly possible - and reasonable - for someone to enjoy a bad mechanic.

To wit, many people seem to enjoy ignorance and superstitions to actually knowing how effective their choices are.

For this message the author Hurkyl has received thanks: 2
Sandman25, zardo

Swamp Slogger

Posts: 143

Joined: Friday, 15th March 2013, 23:33

Post Thursday, 15th May 2014, 03:22

Re: Understanding spell power

(i'm not in favor of showing exact numbers for several reasons but you are at least right about the murkiness of observations were one to want to know exactly what's going on)

For this message the author zardo has received thanks:
duvessa

Slime Squisher

Posts: 354

Joined: Tuesday, 14th January 2014, 23:33

Post Thursday, 15th May 2014, 03:58

Re: Understanding spell power

The effect "I'm killing dudes faster" is impossible to see by eye except in very drastic circumstances

Fortunately minor differences don't matter in Crawl :)

Ziggurat Zagger

Posts: 11111

Joined: Friday, 8th February 2013, 12:00

Post Thursday, 15th May 2014, 04:06

Re: Understanding spell power

TheDefiniteArticle wrote:
The effect "I'm killing dudes faster" is impossible to see by eye except in very drastic circumstances

Fortunately minor differences don't matter in Crawl :)


I still remember why I created my mod which displayed rolls, damage, to-hit etc.: I encountered an extremely dangerous monster and I didn't know which weapon/spell was going to deal the most damage. The character died while I was trying different weapons/spells and my later tests showed that one of those tools was strictly superior and I would very likely survive if I knew it in advance.
My point is that some players hate to play hundreds of games to gather statistics for future games. Hmm, even after hundreds of games I still run fsim and am surprised by its results sometimes.

Barkeep

Posts: 3890

Joined: Wednesday, 14th August 2013, 23:25

Location: USA

Post Thursday, 15th May 2014, 19:59

Re: Understanding spell power

So your first thought when you encountered an enemy that you knew to be dangerous was, "Time to experiment with a bunch of weapons and spells"? There are a few ironies here.

The first irony being that choosing a good option and sticking with it and using it well—even if it wasn't the *best* option—probably would have given you a better chance of winning the fight than testing out a bunch of different random stuff against a dangerous opponent.

In general you have no problem winning Crawl with reasonable play. This means choosing good weapons, good spells, and using them well. "Optimal play" is a thing some people argue about on the Tavern. For practical purposes, becoming a consistently reasonable player is way more important than knowing what's "optimal" against a particular dude.

But that's minor. The real issue is that if you are fighting a monster that is really dangerous, you shouldn't be trying a bunch of different stuff against him. You should be making your exit/escape plan, and either execute it immediately, or carefully engage in a fight but be ready to flee the moment the fight *begins* to go awry.

Not to put too fine a point on it, but you are basically saying, "If I had known in advance that this weapon/spell was significantly better against this particular dangerous monster, and if I had used that from the get go, then I might have avoided a just and deserved punishment for the bad tactics I was using." It seems really strange to me that the lesson you drew from this was, "I need to know everything so I can 100% optimize my weapon choice," rather than "I should be more willing to run away from dangerous monsters."

Bad tactics leading to death is good design in Crawl. Again, I'm failing to see the problem. Putting in lots of detailed damage read outs just encourages people to think that you win Crawl by doing 1.0 effective damage more with the "optimal" weapon against a particular monster. This isn't true, and the fact that Crawl intentionally strives to exactly *not* be that type of game means that it is unlikely to become true.

For this message the author and into has received thanks: 2
archaeo, Lasty

Ziggurat Zagger

Posts: 11111

Joined: Friday, 8th February 2013, 12:00

Post Thursday, 15th May 2014, 20:13

Re: Understanding spell power

and into wrote:So your first thought when you encountered an enemy that you knew to be dangerous was, "Time to experiment with a bunch of weapons and spells"? There are a few ironies here.


Isn't it expected way of playing crawl?
You meet a dangerous monster for the first time with a character, you know nothing about its spells, AC, EV, HP, its max damage so you try to damage it with different weapons/spells?


The first irony being that choosing a good option and sticking with it and using it well—even if it wasn't the *best* option—probably would have given you a better chance of winning the fight than testing out a bunch of different random stuff against a dangerous opponent.


It's possible to use different tools without wasting time. I cast a spell, it dealt some damage (I didn't know how much damage), I casted another spell, again it deals some damage (again I didn't know how much damage), I hit the monster with my weapon, again it dealt some damage and again I didn't know how much damage. Eventually the character died but one of those spells was much better than the weapon and the other spell.

Edit.
Not to put too fine a point on it, but you are basically saying, "If I had known in advance that this weapon/spell was significantly better against this particular dangerous monster, and if I had used that from the get go, then I might have avoided a just and deserved punishment for the bad tactics I was using." It seems really strange to me that the lesson you drew from this was, "I need to know everything so I can 100% optimize my weapon choice," rather than "I should be more willing to run away from dangerous monsters."


This is part of the problem - crawl does not show when you should run and running from everything remotely dangerous leads to later deaths because of running out of escape tools. You lost 5 HP, 10 HP, 7 HP and then you suddenly lose 80 HP and die. "Sun Demon rolls 5 out of 1-49" could be very useful, I know it from experience, I played with such messages

Sar

User avatar

Ziggurat Zagger

Posts: 6418

Joined: Friday, 6th July 2012, 12:48

Post Thursday, 15th May 2014, 20:23

Re: Understanding spell power

Sandman25 wrote:you know nothing about its spells

Well, not anymore.

Barkeep

Posts: 3890

Joined: Wednesday, 14th August 2013, 23:25

Location: USA

Post Thursday, 15th May 2014, 20:28

Re: Understanding spell power

Uh well let's say I come upon a new thing and I xv to see something about it, and let's say it says "extremely dangerous," and its tile does make it look pretty nasty. Okay so I check description again. Is it faster than me? Does it resist poison? Does it use weapons/items? Does it have low-ish MR? Is there an easy way to make it a lot less dangerous with curare or the like?

If you can safely back up, that's always a good idea. Give it a few turns while you tactically retreat and see if it does something tricky, such as (if it can use items) switch to a crossbow and start shooting you, or maybe turns out it was carrying a wand. This also has the advantage that you are moving the fight to already explored terrain which is a good idea usually, especially if you are casting spells with some noise on them.

If it hasn't done anything tricky while I was repositioning and I've gotten into a safe area to fight, then I'd just assume it is an "ogre-type" monster: beefy, with a very nasty melee attack. Maybe worse case scenario, it has some melee brand or af_ effect. In that case softening it up with wands or spells and buffing up with spells or consumables means I can most likely take it out, though I'd be ready to flee if I started taking too much damage.

In any case, if something I'm totally unfamiliar with kills me in Crawl I don't beat myself up much about it, particularly if I learned something in the process. It is when I make the same mistakes against stuff I'm familiar with already, that I get frustrated.

Ziggurat Zagger

Posts: 11111

Joined: Friday, 8th February 2013, 12:00

Post Thursday, 15th May 2014, 21:13

Re: Understanding spell power

Compare it with this:

  Code:
You cast Throw Icicle. To-hit rolled 12 out of 1-26, monster EV rolled 10 out of 1-12.
Shard of Ice hits Ettin.
Rolled damage 40 out of 1-46.
Ettin decreases damage by 5 and loses 35, current HP 48.
To-hit rolled 8 out of 1-14, your EV rolled 9 out of 1-20.
Ettin closely misses you.
To-hit rolled 7 out of 1-14, your EV rolled 6 out of 1-20.
Ettin hits you.
Rolled damage 19 out of 1-63.
You decrease damage by 13 and lose 6 HP, current HP 120.


Wouldn't you enjoy making informed decisions based on this log? The battle started lucky but Ettin can one-shot the character so it's time to think about it.
Next

Return to Dungeon Crawling Advice

Who is online

Users browsing this forum: No registered users and 5 guests

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