More end game stuff?

[Bim]

Hey!
I've had a couple of wins with crawl, and done the Ziggurat run once. Now, obviously I haven't completed all the classes or done some challenge classes, but I was wondering what the thoughts are on adding more end game material. I find personally that once you reach a certain level with enough good gear, most things become easy enough (as long as you don't become careless) and sometimes the luck of not wandering down the stairs into a group of iron trolls.

Any plans or ideas for more end game stuff? I've always thought that it would be nice to have a "so you've got the orb, wanna get the xyz from the 54rd floor?" which opens a portal from where the orb was, to another run with more enemies, maybe a branch or two and a few different enemies (I know, easier said than done!) for if you want to go reallllly hardcore.

Just wondering what other people though really!?

[Curio]

Hells and Pandemonium not quite hardcore for you?

[dpeg]

From the philosophy section of the manual: "By the way, we strongly believe that games are pointless if you can reach the invincible state."

[dpeg]

Exactly. This is why the endgame has to rely on crutches like torment and hellfire. As I see it, the current situation indicates to decrease rather than increase the game's size.

[Bim]

I havn't had many characters that got to the invisible state (only one that was ridiculous (due to unreasonable amounts of mutation bonuses) but I'm sure more experienced players have had a lot more. I agree with Dpeg about the crutches, some of the end game parts feel a bit like the only thing that monsters rely on is spamming.

I suppose its partly a balance issue, that a player shouldn't be able to get to that stage. However, I have to say I really enjoy it when my DDNE gets an axe of pain and can just destroy stuff for a while, I just wish that there was a harder challenge without just wandering around aimlessly once you get really powerful. That being said, this Merfolk IE is being rubbish! FIND A BETTER ROBE!!

[MrMisterMonkey]

Written before minmay:

I think extended endgame material should scale in difficulty over time, particularly Pandemonium, Abyss, and Ziggurats (such as not to fall behind Pan). Hopefully this would make these infinite areas more interesting for those wanting to play further with their current characters, and eventually get to the point where it is impossible to sustain presence, eliminating the invincible state.

As for fun with the current Crawl, you may be interested in getting runes on the orb run. Firstly, avoid any rune branches you want to dive later, to keep them difficult; for the best game, get only Pandemonium/Abyss runes (I think you can get Abyssal on the orb run but the orb run spawns don't happen then so it's boring as usual). After grabbing the orb, you can get any regular dungeon runes you want with orb run spawns chasing you. Oh and if you're feeling great, there's a fun bug you can exploit: dive into the vestibule of hell (take your time clearing it out, as orb run spawns don't happen there) and blow the horn; if it's your first blowing, the hells will open for one attempt at getting one of their runes (with orb run spawns in addition to hell effects). It's sad that you only get to do one hell, but the single-try thing is kind of nice and I imagine hell with panlords would be fun too.
While this isn't really possible on a character you've already developed, as long as your desire for hardcore doesn't also require you to already have a strong character, I think it should be fine. Further on the note of challenge games, you may want to consider speedrunning.


Written after Bim's second post:

Oh and I forgot to mention: useless popcorn, particularly that in the extended endgame, should die painfully.
Pan levels currently feel a bit sparse, too (the worst thing is having to explore into useless ynoxinuls (or even worse, 4s and 5s) and mash autofight to get them out of the way before resuming exploration), so maybe make them smaller, but expand them as they get more difficult.

[Bim]

@MrMonkey,
Love the idea of scaling pan and ziggs, dunno about abyss, as I've sometimes been banished their quite a few times and wouldn't like for it to scale rapidly, although a gradual gradient would be really interesting. I can't really think of a suitable challenge or way to extend the end game without just adding some harder branches, or another ridiculously difficult infinite orbwin level. These both seem like tacky ideas, and I'm sure there would be a more entertaining alternative.

[KoboldLord]

I'm not sure how you'd go about scaling infinite levels in an interesting manner, though. Just jacking the numbers up higher and higher is, in the end, pretty boring. A numerically impossible challenge is less interesting than a possible challenge regardless of whether the possible one is easy or hard. Worse, a stealthy character with lots of escape options can just avoid having those higher and higher numbers aimed in his or her direction.

I think it's perfectly acceptable to say, 'Okay, you win!' Once you've got your 15 Runes and have cleared a Ziggurat with your character, there's no obligation for the designers to code in yet another raid dungeon. Just take your winnings and start a new character. If you want to grind until you have aesthetically pleasing 27s in every skill, go nuts, but don't expect the devteam to care.

Now, I would like to see some Pan or Hell alternatives, and I think there's some pipe dreams already up on the wiki. It would be nice to get a "Primal Heart of Mount Zot" instead of Pandemonium in some games, from which springs the mutagenic power that warps animals into the monsters in the main dungeon. Infested with twelve-headed acid-spitting blink hydrae. Or maybe the classical elemental planes, where the air itself tries to kill you by randomly turning into fire. Or the Celestial Bureaucracy, in which you are bound by red tape and must follow arbitrary and rapidly changing conducts if you want to avoid being punished as a law-breaker. Torment and hellfire are fun, but they don't need to inevitably show up every time.

[Bim]

@Koboldlord,
Number pushing would be very boring, especially as it would eventually just become impossible. I like the idea of those alternatives though! They could also make for good 'end game' challenges. Especially if the Primal Heart randomly mutated you (like a MUT weapon).
Contrary to what dpeg said, I think that the game just improves as it grows. My favourite part about crawl is the wide selection of branches and vaults. I think the main reason for my thread was mainly just that I'd love to see more high level branches, especially ones that are challenging in more unconventional ways.

[galehar]

@Koboldlord,
Number pushing would be very boring, especially as it would eventually just become impossible.

That's the whole point of the idea.

[MrMisterMonkey]

a gradual shift to impossibility is exciting, as it promises constant challenge

[seth]

I can see the appeal of a gradual shift to impossible--the goal would just be to see how far you can go before you die, kinda like an arcade game. Not that I want to see this, particularly, I'm still working on winning

[Bim]

I think if it was really gradual, then it could work out. As you say, it would be kinda fun for an arcade game like highscore thing. I was thinking that you could have an infinite (increasingly difficult) level dungeon entrance where the orb was, and that you had to clear all of the monsters on every level before you could move on (again arcade like) which would stop people just rushing through, and would be a good final indicator (15 runes, orb, 32 levels). Especially if combined with random mutations/removal of mutations after a while for more of a test of skill.

However, I still think this gives it a bit of a 'tacked on' feel. I'm sure there would be better end game challenges. I don't necessarily think it needs to be infinite (although, that could be good), just another big challenge after the zigs and orb.

[dpeg]

The ziggurats were invented to provide such a challenge. Why not make them harder? Likewise with Pan.

[galehar]

If Pan and Abyss get gradually harder, then you can't scum them for infinite XP and loot. Then doing several Zigs become much harder. There's also the idea of scaling the Orb run with the number of runes.

[MrMisterMonkey]

Bim, what's so bad about "rushing"? Preventing it does nothing more than make your challenge zone unplayable for many builds, for which it would be quite fun.

dpeg, the problem with using ziggurats for this is that it's common to have to scum for them, but fixing that would be nice. Scaling pan over time would indeed make this less grindy, as it would ideally actually be challenging, but I doubt it would prove any less annoying.

[ryak]

What about scaling pan based on the number of pan runes you have collected. This allows you to actually have a bit of a choice and follows closely in the high score/arcade suggestions made earlier.

[Bim]

I love the idea of scaling up with pan (or new place) Runes, as that would be an actual choice for players that wanted to do it without forcing it on 'casual' players.

[MrMisterMonkey]

'casual' players would presumably bail out after getting the fiery, glowing, dark, and magical runes, in addition to a single demonic rune
as long as difficulty increases take their time to become noticeable, 'casual' players needn't even worry about it; they're forcing it upon themselves if they hang around in pan for too long

for a compromise approach, perhaps base it on potential runes; scale up the difficulty after passing up levels in which a rune generated (this may be too complicated)

[KoboldLord]

There would be no point in grinding Pan if there were no Ziggurats. Pan loot is mostly floor trash of no special quality. If you bother grinding here, it isn't because you're looking for awesome randarts. You're grinding up the last levels of transmutations so you can survive chain-Torments in a bad Zig level.

I believe we've stretched the definition of 'casual' to the limit here, if we're using it to describe players who are going for 100% completion. There is no game left after you've snagged the Fiery, Glowing, Dark, and Magical Runes.

[MrMisterMonkey]

on your first point:
on loot: well that and anything else that didn't generate earlier, like magical staves, potions of cure mutation, piety, etc. I'm not sure exactly what pan loot is like, including that in the pan vaults, but this floor trash, at the very least, can be quite valuable.
on grinding in general: but that's still grinding, no? (grinding is bad)

on your second point: hell and tomb are both generally harder than pan

[KoboldLord]

Pan loot is, for the most part, equivalent to floor trash in the main Dungeon. On a typical level, you will find a few food items and several consumables, which appear to be distributed roughly the same as they are anywhere else. Cure mutation potions and acquirement scrolls are still crazy-rare. Unlike the main Dungeon, however, you get almost no equipment drops since most of the residents don't require equipment, and what little can drop will be inferior to what you already have. If a random Pan Lord vault generates, you will get one to four slightly better items, usually a staff, spellbook, weapon, armor, or miscellaneous item. Or a Demonic Rune. Sadly, you usually already have a full set of staves and miscellaneous items, and the weapon and armor you already have will always be better otherwise you wouldn't have made it this far.

The real treasure is the fact that you're facing an unlimited supply of predictable punching bags. Once you've gotten used to demon tricks and can deal with 1s without grave risk, you can effortlessly fill your xp pool and spend it as you like. Beyond basic competence at killing things, the only thing required here is the ability to escape from a bad starting position when you arrive at a new level. I've been air-dropped into the Hellion Island vault, for instance, and just last week I got dropped into Lom Lobon's room no more than four tiles away from him. I didn't actually even know that was possible.

Hell and Tomb have higher average difficulty, but they don't have the difficulty spikes involved with getting dropped onto Hellion Island. Additionally, they suffer a bit from Poor Predictable Rock and you can usually clear a couple of them much more easily than the others based on your resistances and available attacks. Rather than having Pan ramp up in difficulty to defeat the badwrongfun of some hypothetical poopsocker, I'd rather Pan start with a difficulty commensurate with the other post-endgame branches and stay there.

I did notice new layouts for some of the unique Pan Lords' levels last time I went through, and I liked what I saw. It might have been just the unfamiliarity throwing me off, but it seemed harder to steal the goodies without coming under fire.

[MrMisterMonkey]

on your first: the biggest difference here is that you can scum pan forever (so you can eventually get what you want), and no, you don't need to be crazy powerful to do it.

on your second: things like hellion island will only generate once, so they have nothing to do with scumming (as a consequence, Pan actually gets more predictable over time (difficulty scaling would help fix this))

on your third:
firstly: note that this is not just to defeat the badwrongfun of some hypothetical poopsocker; it is additionally to provide continued challenge to those seeking it. It is currently even possible to trivialize ziggurats, meaning there is indeed a point at which characters can no longer experience challenge. While scaling ziggurats alone and providing infinite ziggurats for free (such as not to require scumming pan for them) may fix this, there's still the poopsocker thing.
secondly: I'm for pan being generally harder (and hell/tomb being less predictable), and especially more interesting, both tactically and thematically, but I'd still hate for it to stay scummable, and hate even more for characters to crawl to stop providing challenge


also, I forgot to bring abyss back up, so I will now:
what about it? Even if pan required superpowers to scum (it doesn't), Abyss is even easier and contains plenty of floor trash and free exp, but is just a little less convenient to scum (particularly for exp since there are far fewer monsters and even among those not much is tough at all).

[KoboldLord]

If you want endless challenge, there's always the New Game option. You can even use it repeatedly, and it never wears out or damages gameplay.

I don't think the devteam should self-inflict the MMO raid problem, wherein 90% of development time goes toward creating yet another set of endgame raids. Infinite and interesting play with a single character is a bottomless hole that will suck up everything you pour in and never be filled. If somebody has an interesting thing to put into the post-endgame, then that's great, but numerical hard for the sake of numerical hard is not interesting.

If we absolutely must have arcade mode, it would be more suited to being an alternate minigame like Sprint or Zot Defense. And if it absolutely must be in the main game, it should at least be confined to the Ziggurat ghetto where it won't bother normal gameplay.

The Abyss is a more meaningful place to scale, since it can be visited with relatively low-level characters who should have some sort of chance to survive, but it should also be a meaningful challenge to get the Rune from. But in this case, since the goal is not to prevent scumming the challenge can cap off at some point, meaning the task should require a finite amount of development and testing.

[galehar]

If you want endless challenge, there's always the New Game option. You can even use it repeatedly, and it never wears out or damages gameplay.

I don't think the devteam should self-inflict the MMO raid problem, wherein 90% of development time goes toward creating yet another set of endgame raids. Infinite and interesting play with a single character is a bottomless hole that will suck up everything you pour in and never be filled. If somebody has an interesting thing to put into the post-endgame, then that's great, but numerical hard for the sake of numerical hard is not interesting.

If we absolutely must have arcade mode, it would be more suited to being an alternate minigame like Sprint or Zot Defense. And if it absolutely must be in the main game, it should at least be confined to the Ziggurat ghetto where it won't bother normal gameplay.

The Abyss is a more meaningful place to scale, since it can be visited with relatively low-level characters who should have some sort of chance to survive, but it should also be a meaningful challenge to get the Rune from. But in this case, since the goal is not to prevent scumming the challenge can cap off at some point, meaning the task should require a finite amount of development and testing.

I think you missed the point. And I certainly didn't get yours.
The point is to prevent scumming. We want to limit both Abyss and Pan, because by being infinite, they allow the player to maximise their power, thus rendering harder challenges (Ziggurat) difficult to balance. Optimal play for a ziggurat runner is to bore himself to death in Pan until he has an invincible character with maximised skills. By making the difficulty of infinite branches scale with the time, we fix this in an elegant and interesting way.

[KoboldLord]

I think you missed the point. And I certainly didn't get yours.
The point is to prevent scumming.

You can't prevent scumming by adding extra challenges that can only be completed by scumming. The harder you make it to scum, the greater the rewards for doing so efficiently.

We want to limit both Abyss and Pan, because by being infinite, they allow the player to maximise their power, thus rendering harder challenges (Ziggurat) difficult to balance. Optimal play for a ziggurat runner is to bore himself to death in Pan until he has an invincible character with maximised skills.

This is a problem with the Ziggurat, not the other places. If it was possible to complete a Ziggurat without the perfect set of spells, players would just do that instead of doddering around in Pan for tens of thousands of extra turns.

By making the difficulty of infinite branches scale with the time, we fix this in an elegant and interesting way.

What elegant and interesting way is this? Slowly raising the numeric statistics of random Pan Lords is neither elegant nor interesting, and it probably wouldn't work anyway because there's no reason to actually fight random Pan Lords. Jacking up the demon spawning rate just makes grinding faster, and forces the use of Makhleb, TSO, or Vehumet so you can auto-mulch everything with Tornado or Fire Storm the moment it steps on screen.

If there's an elegant and interesting idea to improve Pan, it should be added because it is an intrinsically elegant and interesting improvement, not because the community is scandalized by some random player's Ziggurat streak. If Ziggurats are bad for the game, cut them out.

[danr]

I think the idea is not to make scumming more efficient, but to make it impossible. You can only run away from pan lords when there isn't one generated in every room. Generate enough, and it's not called "efficient scumming", it's called "dying".

However, here is an alternative: what if there was only so much experience you could gain from fighting each type of monster / demon? E.g. once you've killed 5 ancient lichs, there's not much to be learned from killing another one.

This would eliminate the reward for scumming pan, as eventually the player has killed every type of monster 10x over and can't gain any new experience.

That would work best coupled with the slowly increasing difficulty, so that the reward / risk ratio really starts to drop off.

One easy way to combine these would be to treat increasing percentages of pan creatures as if they were summoned - no XP!

Of course, I've never set foot in Pan or a Ziggurat...

[KoboldLord]

I think the idea is not to make scumming more efficient, but to make it impossible. You can only run away from pan lords when there isn't one generated in every room. Generate enough, and it's not called "efficient scumming", it's called "dying".

If you get Pan Lords in one of the last few levels of a Ziggurat, you'll get about 30 Pan Lords in a tiny room with minimal room to maneuver. Plenty of players kill them all anyway. The wiki recommends you bring manuals of any skill you haven't mastered yet, because it's hard to use up all the free xp without them.

It seems strange to me that so many posters on this thread think the problem is going to be easy to solve. Ziggurat runners wouldn't be playing a roguelike in the first place if they didn't specifically want to show off by beating something that is reputed to be hard, so throwing in a bunch of changes to make it harder for them will only encourage them. Better to concern ourselves with normal players, and how the experience is for them.

[danr]

There seems to still be a disconnect. If a room full of pan lords is popcorn, then the solution being discussed here has not been implemented.

I don't understand, is there something else players can scum to make pan scumming viable if it is made harder?

Of course, if someone wants to scum pan forever, who cares?

[MrMisterMonkey]

written before danr:

Nobody ever said anything about slowly raising the numeric statistics of random panlords, you silly goose. Obviously a proper implementation would need discussion, but you seem to be more intent on pretending the only way to do this is to give panlords more numbers.

On tornado being overpowered, if it can trivialize anything, that's its problem, and it needs a nerf.

On players seeking challenge, this is a good thing.
As long as this "scumming" is an interesting challenge, it isn't scumming (it's an interesting challenge).
If loving challenge makes people horrible, I'm not really sure what to say. Hanging out in infinite areas to reduce challenge in any way is horrible, sure (and I must admit here that I don't entirely agree with galehar's example about scumming pan for ziggurat preparation, as these players are indeed fully doing it to themselves, as I doubt the player would have much of anything to do with the ziggurat loot (contrast with legitimate ziggurat clearing done purely for the challenge, and with premature ziggurat attempts with the loot in mind). it'd be really nice to ensure ziggurat clearing is legitimate, though), but increasing difficulty over time would make it so that hanging out in itself would be a legitimate challenge.
And it's not about beating something that's reputed to be hard, it's about beating something that is hard. Ziggurats, for instance, get boring when trivial, but a feat so simple as winning can be great fun when done with interesting constraints such as turn count minimization.

Note again that this would not affect "normal players" if implemented properly.
If you're only arguing because you think it's not worthwhile to bother, what is wrong with you? Seriously.


written after danr:

because coupling of tornado and TSO/Makh HP/MP regain is ridiculously overpowered and badly needs nerfing

[KoboldLord]

Nobody ever said anything about slowly raising the numeric statistics of random panlords, you silly goose. Obviously a proper implementation would need discussion, but you seem to be more intent on pretending the only way to do this is to give panlords more numbers.

It has come up several times on the Pan page of the development wiki, which is surely relevant. There's also recurring reference to 'scaling Pan difficulty' that is generally not explained, which I may be predisposed to hear as 'making Pan get numerically hard without changing anything else' because that's something I particularly do not want, so perhaps I'm unfair there. In any case, raising the statistics of random pan lords also has the distinction of being the only Pan buff listed there, here, or anywhere else that I've seen that can be accomplished procedurally. If raising the statistics of random pan lords as time passes in-game is NOT on the table, then the issue is simply dead because nobody has figured out a solution yet, meaning I have no problem.

Making Pan harder and more interesting, and nerfing overpowered builds and abilities, are of course desirable as ends in and of themselves.

[galehar]

Nobody ever said anything about slowly raising the numeric statistics of random panlords, you silly goose. Obviously a proper implementation would need discussion, but you seem to be more intent on pretending the only way to do this is to give panlords more numbers.

It has come up several times on the Pan page of the development wiki, which is surely relevant. There's also recurring reference to 'scaling Pan difficulty' that is generally not explained, which I may be predisposed to hear as 'making Pan get numerically hard without changing anything else' because that's something I particularly do not want, so perhaps I'm unfair there. In any case, raising the statistics of random pan lords also has the distinction of being the only Pan buff listed there, here, or anywhere else that I've seen that can be accomplished procedurally. If raising the statistics of random pan lords as time passes in-game is NOT on the table, then the issue is simply dead because nobody has figured out a solution yet, meaning I have no problem.

Making Pan harder and more interesting, and nerfing overpowered builds and abilities, are of course desirable as ends in and of themselves.

The fact that Pan is infinite and scummable is not the only problem about Pan. There's also the fact that it lacks diversity. This problem is acknowledge too, but this is not the one we're discussing here.
Making Pan progressively harder so that it can't be scummed can be done in a lot of different ways. Most of them involve some numbers raising, sure, the question is which numbers and how. We can scale Pan Lords power and frequency. We can increase monster difficulty and density (less popcorn). We can even reduce the loot, so that food becomes an issue. It has also been suggested that god powers doesn't work as well in Pan. This can be scaled too, so that over time, it becomes harder to use the powers and gain piety.
Do you have ideas of your own? Or do you prefer to just criticise other's?

[galehar]

Criticism is useful.

Of course it is, and it's welcome. But some people like to write long and wordy posts explaining in details why such and such ideas are bad. It's even worse when it's mostly opinion not really backed up by arguments. When you're not proposing anything, then you're not really helping to find a solution. You're just arguing for status quo or for the sake of arguing. And I've read too many posts like that recently.

Also this:

Now, I would like to see some Pan or Hell alternatives, and I think there's some pipe dreams already up on the wiki. It would be nice to get a "Primal Heart of Mount Zot" instead of Pandemonium in some games, from which springs the mutagenic power that warps animals into the monsters in the main dungeon. Infested with twelve-headed acid-spitting blink hydrae. Or maybe the classical elemental planes, where the air itself tries to kill you by randomly turning into fire. Or the Celestial Bureaucracy, in which you are bound by red tape and must follow arbitrary and rapidly changing conducts if you want to avoid being punished as a law-breaker. Torment and hellfire are fun, but they don't need to inevitably show up every time.

Come on, this isn't a proposal. Throwing some random ideas can be fun, but they are useless like that.

[danr]

I think this topic has run its course. I'm gonna lock it and see what happens.