Sil: How is minimum depth calculated?

[absolutego]

you could get a message when you're forced down on a staircase, yeah. i thought that existed already but i disable -more- prompts so i don't notice these things.

imo default-gameplay when you start out without previous angband exposure is ironman-like "clear floor, go down", not scumming low depths as implied by the last posts.

i noted that myself! it was an off-hand comment, but surely we can discuss other min-depth-related things?

the notion of diving and scumming 950' is quite incomprehensible for people who run into time problems at 500-600' - although I agree about 950' scumming - how about just not doing it anymore? still, i believe the game should be balanced for the first win, not the tenth.

if you agree that there is a problem, why not try to address it? you can care about returning players, as long as you don't upset the balance for new ones. i actually experimented with this and had what i thought was an elegant solution, but others disagreed and i dropped it... in any case, i don't think "you can just not do optimal things" is a good way to defend an argument.

[taptap]

if you agree that there is a problem, why not try to address it?

what is considered optimal play is mainly due to killing morgoth being encouraged by the current score system. drop that, and nobody would scum at 950 ft., beren did not after all the game would play very different if you would try to win with least xp spent. personally, i believe there are many good reasons for non persistent dungeons, that this at one point or another allows/encourages scumming just can't be helped.

[absolutego]

i'm all for that, but there's also the fact that if you can actually kill morgoth the ascent is much safer.

as for the rest, i still believe there are ways to potentially handle this but i don't want to get into it now. i would like to say this though: how about just removing _treasures, for a start? it reduces the appeal of scumming a lot, and gets rid of the worst interface in the game. _revelations already reveals vaults anyway.

[clouded]

Sil isn't Sil without mindepth and turns-as-a-resource. You could argue that maybe the light-clock and food-clock aren't adding much and they should be removed, though.

I think they should be. It would have no effect on light other than maybe allowing you to use a lantern earlier than you would if you had to wait for oil (lanterns could spawn deeper). The only real effect from hunger is "balancing" some items, but these are problematic in themselves, encouraging you to swap them out when they aren't needed. Oathwraiths can cause havoc for 0 will characters with their hunger melee, but oathwraiths are bad enough in other ways for 0 will characters. Should I also mention the complete "gotcha" of gorged stopping you from drinking potions.

how about just removing _treasures, for a start? it reduces the appeal of scumming a lot, and gets rid of the worst interface in the game. _revelations already reveals vaults anyway.

I would also go along with this. It is not very fun to scour floors, examining every item before you take a move. It is a large part of how characters can get too many consumables and power, which is probably my main point on scumming 950'... Characters in Sil can get much too strong. I can't stand to do it more than a few times, but there really isn't much of a reason not to repeat 950' as much as you can, it is a relatively safe depth.

[taptap]

I only ever repeat 900/950' each once if at all these days, although I play quite slow compared to other people. I have lost quite a few chars that were strong enough to face the throneroom when procrastinating at 950'. Trying to optimize some aspect but falling victim to a lucky D or tricky serpent/rauko trap with too much darkness or or ... Aren't a few of the late game uniques more notable / dangerous now? (Tevildo has cruel blow, hasn't he?)

Removing treasures would also be a nerf to loremaster, which is great. (At least I do lose quite a few recharge charges to understanding staffs playing without.)

[BlueFish]

I appreciate the game design intent of removing those staves but in practice they are some of the most powerful items in the game, and by removing them, the game would be made significantly harder. In case that matters to anybody.

[Scatha]

We could gently discourage the dive-and-repeat-950' strategy without hurting novice players by making the minimum depth increase marginally faster if you're well below your current one. It's a relatively opaque mechanic anyway, but maybe this would feel bad for people.

BlueFish: we do keep an eye on balance, but typically make several changes which make the game both easier and harder. There are some making-it-easier ones we have in mind for future versions.

[absolutego]

We could gently discourage the dive-and-repeat-950' strategy without hurting novice players by making the minimum depth increase marginally faster if you're well below your current one. It's a relatively opaque mechanic anyway, but maybe this would feel bad for people.

that's what i tried. it felt fine to me (a few of my ladder games were played this way) but others didn't like the idea and i forgot about it. let me know if you want to try it out.

[locus]

I like having an incentive to dive to the deepest level I can handle. That's where I'm having the most fun, where the challenges are best tuned to my current capabilities. If you punish players for diving, it becomes like Crawl, where players are rewarded for making the game as boring as possible, taking the easiest option at each step. The interaction between the mindepth clock and diving is the biggest thing that makes Sil great.

Staff of Treasures can be fun, but I'd rather you kill that than kill diving entirely.

[BlueFish]

if you agree that there is a problem, why not try to address it? you can care about returning players, as long as you don't upset the balance for new ones. i actually experimented with this and had what i thought was an elegant solution, but others disagreed and i dropped it... in any case, i don't think "you can just not do optimal things" is a good way to defend an argument.

I consider a few thousand turns available to explore 900-950 to be some of the most enjoyable turns of Sil I ever play.

One way for a player to re-balance the game on their own, if they feel optimal play of one build is boring due to too many turns available at 950, is to just play a different build, maybe weaker but more interesting in some way.

Not to state the obvious, but Sil already has a turncount limit, at least until min depth is 1000. That's pretty fundamental. A player playing around within that turn limit - maybe doing a build that can dive quick due to stealth but who wants lots of turns deep to gather a sufficient kit to win, seems exactly the sort of rich gameplay possibility the flat turncount allows, as it stands.

This idea that too many turns available when the player reaches 950 is a game design deficiency seems utterly bizarre to me, and surely comes exclusively from veteran players who have spent a long time excruciating about the best way to game the game. I suspect most if not all of these players already care about minimizing turncount anyway, for the ladder score.

[debo]

Re: treasures, I'm pretty sure getting rid of these would have a similar effect to no-selling in vanilla: it removes a pressure to minmax. As others have mentioned, you can generally tell where tasty things are by using _rev anyways.

Alternately, you could change "song of trees" to be "song of treas" and have the discovery radius be relative to your song skill

[Nick]

Alternately, you could change "song of trees" to be "song of treas" and have the discovery radius be relative to your song skill

Game development by typo. I like it

[absolutego]

I like having an incentive to dive to the deepest level I can handle. That's where I'm having the most fun, where the challenges are best tuned to my current capabilities. If you punish players for diving, it becomes like Crawl, where players are rewarded for making the game as boring as possible, taking the easiest option at each step. The interaction between the mindepth clock and diving is the biggest thing that makes Sil great.

i'm not sure i follow you, both games score mostly on turncount (leaving aside optional content). in crawl you dive for score and the game is harder. in sil, once you're familiar with the game, diving becomes the best option because it's relatively safe given how xp works, and you get better loot. if you let people grind 950' for ~10k-15k turns (out of 30k) you have what clouded mentioned, characters that are much too strong.

a related problem is that 950' is inherently safer for many characters than 900' because you cannot be shafted. "common sense" aside, you could have shafts drop you on a new 950' level.

anyway this is frosting on the cake, what i really dislike is treasures-scumming. :P

[locus]

i'm not sure i follow you, both games score mostly on turncount (leaving aside optional content). in crawl you dive for score and the game is harder. in sil, once you're familiar with the game, diving becomes the best option because it's relatively safe given how xp works, and you get better loot.

If people played for score, that would make sense. But only like, the top 1% of roguelike players, people who can fairly consistently win, play for score. Most players are just trying to maximize their chances of winning, which in Crawl is done by methodically clearing the easiest level at any given time, and in Sil is done by diving to the lowest level you can handle so you have the best loot when it comes time to face the throne room.

[taptap]

If people played for score, that would make sense. But only like, the top 1% of roguelike players, people who can fairly consistently win, play for score. Most players are just trying to maximize their chances of winning, which in Crawl is done by methodically clearing the easiest level at any given time, and in Sil is done by diving to the lowest level you can handle so you have the best loot when it comes time to face the throne room.

This wouldn't really change without treasures though. It just would make scumming a little less efficient, while ironman / beginners-in-time-trouble play is not affected at all.

Imo, people play for score in Sil, and the single most important score-bonus is kill / not-kill of Morgoth, followed by Sils taken. That is why so many players go for the kill and three Sils all the time (and so many winning chars look so similar) and it is apparently such a no-brainer to do so, that people are under the impression Sil scores "mostly on turncount" - if people were trying to maximize chances of winning, they would go for simple Sil-grabbers more often.