Shafts - Too common now?


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

Blades Runner

Posts: 586

Joined: Wednesday, 19th June 2013, 09:31

Post Sunday, 20th January 2019, 02:09

Shafts - Too common now?

This has been discussed quite a few times, and i believe that each time the conclusion has been that its fine as it is.

However, the fact that its getting discussed a lot recently leads me to believe something must have changed. I regularly get shafted 2-3 floors in the early game. I think other players are experiencing this too from the posts on the forum going on.

The answer seems to have been that being shafted provides an interesting challenge rather than an unavoidable death; and so its fine. This is true in some cases, but not all. Up until around D:7 or 8, getting shafted 3 floors generates a huge spike in the difficulty of the game, and going down 3 floors from D:2 can turn a strong start, into a game of pot luck quaffing un ID'd potions and scrolls.

I certainly have found myself getting shafted more recently and dying as a result of it. Whether its avoidable or not is debateable, but it's causing me - and it would seem others problems. It makes me re-do the most boring part of the game (early D) more often.

How does everyone else feel? I wonder if it would be helpful to:

a) reduce the number of shaft traps. Potentially implement a formula that reduces the number of shaft traps by a percentage depending on floor... E.g D:1 has 50% less shaft traps than normal. D:2 has 40% less. By the time you get to D:6 you are at regular numbers of shaft traps.
b)Increase the number of traps in line with level. Something similar to the above but scaling up with level rather than down with depth of D:x

Ziggurat Zagger

Posts: 8544

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

Post Sunday, 20th January 2019, 04:09

Re: Shafts - Too common now?

Why is increasing the number of shafts a bad thing?

Blades Runner

Posts: 586

Joined: Wednesday, 19th June 2013, 09:31

Post Sunday, 20th January 2019, 05:00

Re: Shafts - Too common now?

All IMO of course... The game is pretty hard as it is anyway. Shafts add difficulty that doesn't need to be there, and introduce tedium by way of having to replay the early game if you don't always deal with shaftings in an optimum way. Even then sometimes you cant get out of it.

Shafting 3 floors on D:10 is pretty OK to add challenge. Shafting 3 floors on D:3 is a very bad situation.

Dungeon Master

Posts: 151

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

Post Sunday, 20th January 2019, 05:43

Re: Shafts - Too common now?

The shaft rate in 0.22 (trunk and stable) and 0.23 (trunk) games from before the trap change for the first six dungeon levels:
  Code:
<ebering> .echo $(/ (!lm * vlong>=0.22 vlong<0.23-a0-495 shaft br=D lvl<7) (!lm * vlong>=0.22 vlong<0.23-a0-495 shaft br=D lvl<7 x=cdist(gid) fmt:"${x}"))
 <Sequell> 1.0524399289712063

is slightly smaller than post trap change (excluding games when those changes were in flux)
  Code:
<ebering> .echo $(/ (!lm * vlong>=0.23-a0-517 shaft br=D lvl<7) (!lm * vlong>=0.23-a0-517 shaft br=D lvl<7 x=cdist(gid) fmt:"${x}"))
<Sequell> 1.0744844042952104


On average, in the first 6 dungeon levels, you'll run into an extra 0.022 shafts. There's not much here to reduce.

The new trap change did increase the later game trap rate by a bit more, using shafts as a proxy (and since this is a thread about shafts)
  Code:
<ebering> .echo $(/ (!lm * vlong>=0.22 vlong<0.23-a0-495 shaft) (!lm * vlong>=0.22 vlong<0.23-a0-495 shaft x=cdist(gid) fmt:"${x}"))
<Sequell> 1.156431054461182
<ebering> .echo $(/ (!lm * vlong>=0.23-a0-517 shaft) (!lm * vlong>=0.23-a0-517 shaft x=cdist(gid) fmt:"${x}"))
<Sequell> 1.4142197125256672

A whopping 0.36 new shafts on average.

Why is it more of an increase later in the game? The old traps and doors skill, when it was removed, survived in an xl chance to reveal traps. In the new trap system, the trap rate is determined entirely by dungeon level, so late game the player will hit more trap effects than they used to. Yes, this is a player nerf. No, that's not a bad thing.

If you want to read more about why the devteam is in agreement that shafts are good dpeg wrote a great post years ago that I agree with. The short of it is: shafts increase the probability of a tense memorable random encounter in which skill makes a real difference; such encounters are a goal of crawl development.

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

Blades Runner

Posts: 586

Joined: Wednesday, 19th June 2013, 09:31

Post Sunday, 20th January 2019, 06:03

Re: Shafts - Too common now?

Thats interesting that the change is in fact so small. Can't argue with figures! Funny that lots of people seem to be noticing it...but I guess they might just be having a placebo because they see other people mentioning it too....

Spider Stomper

Posts: 187

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

Post Sunday, 20th January 2019, 10:58

Re: Shafts - Too common now?

I like shafts, or more accurately, I enjoy trying to climb back up to the level I got shafted from... but not in conjunction with portal timers, because the completionist in me can't abide continuing a run if I miss a timed portal. This wasn't as big a problem before the traps change, but now I find myself occasionally getting shafted a few floors, then before I find an up staircase, getting shafted a few more, and then encountering a portal timer prematurely, or (more rarely) getting shafted from a level with a portal timer multiple times and thus being unable to climb back up in time to reach it. Maybe it's just a coincidental run of bad luck, but if the chance of getting shafted increases with depth, that also means that one of the consequences of getting shafted is an increased risk of getting shafted.

I'd like to think there's some way that portal timers could be reformed to not ruin the fun of getting shafted... perhaps by pausing the timer if a player leaves a portal level other than by voluntarily descending? That way, getting shafted from a level with a timed portal or climbing back up after being shafted and encountering one wouldn't cause a player to lose the chance to enter the portal, but choosing to skip the portal to farm beyond it would. It's not a perfect solution, because if a timed portal generates past a branch, the player could climb back up to the branch and explore it while the timer was paused, but then again, if that causes players to consider exploring branches sooner than they otherwise would, that could be kind of interesting too.

(Btw, I noticed an increase in my shafting experience before I saw anyone mention it, so at least in my case, the placebo theory doesn't hold up.)

For this message the author Nekoatl has received thanks:
VeryAngryFelid

Ziggurat Zagger

Posts: 5964

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

Post Monday, 21st January 2019, 00:57

Re: Shafts - Too common now?

Something about the math is fishy, I definitely have experienced more traps by a large margin above what I previously had roughly on the order of twice as many shafts. In each of my previous four games I was shafted 4 times before the lair, something I've never experienced in all the prior years of my playing once, much less four games in a row.

I suspect there's a factor at play that hasn't been explained, and if the intent is to double the number of shafts encountered, then that's fine, but if it's intended to be roughly equal, it's not.
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: 5964

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

Post Monday, 21st January 2019, 01:24

Re: Shafts - Too common now?

Siegurt wrote:Something about the math is fishy, I definitely have experienced more traps by a large margin above what I previously had roughly on the order of twice as many shafts. In each of my previous four games I was shafted 4 times before the lair, something I've never experienced in all the prior years of my playing once, much less four games in a row.

I suspect there's a factor at play that hasn't been explained, and if the intent is to double the number of shafts encountered, then that's fine, but if it's intended to be roughly equal, it's not.

Looking back at my actual morgues, I misremembered, there was only 4 shafts in one game, 3 in two and 2 in another, and there was a short game where I died before the lair in the middle when I didn't get any shafts, so my statement was misreperesentative (however that's still an unusually high number of shafts, it's possible that my games were just a outliers though)
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!

Dungeon Master

Posts: 151

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

Post Monday, 21st January 2019, 01:48

Re: Shafts - Too common now?

You're right, though for the wrong reason (as I was typing this I see you looked at your morgue and see that the post was an exaggeration). Your sample size is too small to make definitive judgements about the shape of the distribution, and you might be including a game played in the interval where the rate was being tinkered with post-merge.

However, the denominator in my previous number includes the large number of early deaths, which skews it. Looking at won games over the whole game (which includes some of extended) the shaft rate has increased, on average, by one whole shaft, but the experience on early D is approximately the same (increase of .004 shafts on average). As I said above, the increase in later traps is intentional.

The calculations for winning games (old followed by new).

Early game:
  Code:
<ebering> .echo $(/ (!lm * vlong>=0.22 vlong<0.23-a0-495 shaft br=D won lvl<7) (!lm * vlong>=0.22 vlong<0.23-a0-495 shaft br=D lvl<7 won x=cdist(gid) fmt:"${x}"))
<Sequell> 1.0765407554671969
<ebering> .echo $(/ (!lm * vlong>=0.23-a0-517 shaft br=D lvl<7 won) (!lm * vlong>=0.23-a0-517 shaft br=D lvl<7 x=cdist(gid) won fmt:"${x}"))
<Sequell> 1.0807174887892377


Entire game:
  Code:
<ebering> .echo $(/ (!lm * vlong>=0.22 vlong<0.23-a0-495 shaft won) (!lm * vlong>=0.22 vlong<0.23-a0-495 shaft x=cdist(gid) won fmt:"${x}"))
<Sequell> 1.4518862766539093
<ebering> .echo $(/ (!lm * vlong>=0.23-a0-517 shaft won) (!lm * vlong>=0.23-a0-517 shaft x=cdist(gid) won fmt:"${x}"))
<Sequell> 2.5343642611683848

For this message the author ebering has received thanks:
duvessa

Spider Stomper

Posts: 187

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

Post Monday, 21st January 2019, 04:34

Re: Shafts - Too common now?

If getting shafted makes winning more difficult, then I would expect that selectively limiting the sample to games that were won would also skew the shaft rate down. I wonder how the data would look if average number of shafts encountered for characters that died at each XP level 1-26 were compared between pre- and post-rework.

Ziggurat Zagger

Posts: 8544

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

Post Monday, 21st January 2019, 05:39

Re: Shafts - Too common now?

Comparing the numbers of shafts that people fell down doesn't really say much of anything, because most people use autoexplore, and even those who don't aren't exploring optimally. If you explored optimally in old versions then you would practically never fall down a shaft.

Spider Stomper

Posts: 187

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

Post Monday, 21st January 2019, 06:33

Re: Shafts - Too common now?

Pre-rework, if you were optimizing to minimize occurrences of traps, then you would spend most of your turns standing in place to generate chances for traps to be discovered, so I agree with you about that. I think it makes more sense to compare to average player behavior than a theoretical optimum, however, which would seem to be achieved by averaging the data.

In any case, I think you've hit on a point in that the way traps used to work isn't really important compared to whether or not the current behavior is healthy, and aside from the portal timers issue, I think it is.

Dis Charger

Posts: 2031

Joined: Saturday, 2nd February 2013, 09:52

Post Monday, 21st January 2019, 07:27

Re: Shafts - Too common now?

duvessa wrote: If you explored optimally in old versions then you would practically never fall down a shaft.

Am I the only one who do not understand this?
User avatar

Ziggurat Zagger

Posts: 4168

Joined: Wednesday, 23rd October 2013, 07:56

Post Monday, 21st January 2019, 07:54

Re: Shafts - Too common now?

Magipi wrote:
duvessa wrote: If you explored optimally in old versions then you would practically never fall down a shaft.

Am I the only one who do not understand this?

Enter level, make noise while standing on upstairs, note which tiles the monsters step on, only use those tiles to explore etc.?
...DDFiHOWzVpVMHuEE{HEMoDEHuDDAs}{HaBeKoAK}CeVM{MfWnMiAK}TeAMDrIE{FoVMVSFi}{MuVMGhGlVpMo}
HaWrSpWz{OgGlTrMo}{CeWnMfBeMiSk}DrEE{GrFiFoGl}DgEnFeNe{OpGlHuSu}DDArHaCKSpAEGrTmDgFEDsCj
GhMoHuVM{HaAMBaEn}{HuMoHOWn}DsWzDDHu{DgWnGnBe}FeIE{MiEnMfCj}SpNeBaEEGrFEHaAK

Ziggurat Zagger

Posts: 5964

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

Post Monday, 21st January 2019, 16:33

Re: Shafts - Too common now?

Sprucery wrote:Enter level, make noise while standing on upstairs, note which tiles the monsters step on, only use those tiles to explore etc.?

Additionally, Just regular old attention-paying manual exploration reduces your trap likelihood, even if you don't go to extremes about optimizing it. Because when you explore manually you generally: 1. do less backtracking than autoexplore does, because you have a better ability to anticipate where unexplored parts of the level will be than autoexplore uses, and 2. When you do backtrack, you'll more frequently reuse the same paths, because people are habitual and what looks like the 'right' way to get from point a to point b will generally still look like the right way when coming back.

Over the course of the game, that means that a manual explorer will generally step on less *distinct* tiles than an an autoexplorer will for the same amount of tiles revealed, even if you don't optimize for it (optimizing for it could mean tracking where monsters and you stepped and the aforementioned shouting thing)

It may be that part of my personal experience with new traps is that my general habit is to manually explore most levels out somewhat to start with, and only use autoexplore at the end, if at all (for wholly un-trap-related reasons), which may have previously reduced my trap experience below that of the average person.
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:
byrel

Return to Game Design Discussion

Who is online

Users browsing this forum: No registered users and 7 guests

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