2 Runes for Zot

[all before]

--Require only two runes to enter Zot. One of the runes must be from a non-S-branch. (If you like, give flavorful names to distinguish S- and non-S-runes, like "beastly runes" and "ancient runes," and divide them on the "}" screen. A message upon branch entrance can tell new players what kind of rune the branch holds.)

This change makes it unnecessary to play through both S-branches, which is tedious.
It allows a player to complete both S-branches if she so desires.
It allows a player to abandon one S-branch if it's too difficult and instead complete the other.

This proposal is in response to feedback on my earlier suggestion to only generate one S-branch per game: viewtopic.php?f=8&t=14850&hilit=generate+one+s+branch

[Rast]

It's been suggested that lair drop down to one S-branch per game..

And then we can finally have an elven rune!

[all before]

It's been suggested that lair drop down to one S-branch per game..

I not only like this idea, I've proposed a version of it myself! It's linked in the OP. It got a lot of feedback, and this proposal is meant to address the concerns people had.

[mps]

I like one lair branch and three runes. This way you get three clearly distinct rune branches in a three rune game and force the player into more endgame-y content.

[johlstei]

I don't see why Zot has to be rune limited at all - let there be a whole spectrum of wins where you just try to get the orb and get out. It's not really a routing concern like the vaults rune lock - no one's gonna dive to zot to mop up "easy" XP before doing snake:5 (I hope).

3 runes can stay the speedrun standard maybe, I dunno.

[bel]

Problem with one lair branch, as noted before, is that sometimes you don't have the necessary (or at least helpful) ingredients for one, but can do the other fine. Like lacking rPois and sInv for spider, flying for shoals/swamp etc, clarity/high MR for shoals etc. Since there are not too many easier branches after D, Lair and Orc, you are screwed if you don't have the items.

Now, lair branches are down to 4 levels (could be made even 3, I don't mind). But you can also just dive to the bottom if you don't want to slog through it. So I don't see much issues. The final floor is fairly tough or at least challenging even for characters who have one rune.

[Sar]

Well, you can do these branches without those items. Shoals is very noticeably harder than the rest of the 4 "first rune" branches, though. It would probably need a nerf.

[mps]

The speedrun standard is 15 runes, since that's the only way to high score. 3 rune speedrunning is pretty dead, as far as I know.

Without a rune requirement for Zot, no one would bother with runes and routes would look radically different. Shoals would be extended endgame. People would do lair, orc, maybe elf, then depths and win. If there's anywhere a rune lock is silly and unjustified, it's vaults. If the xp is "too easy," make it less easy?

3 runes for Zot with one lair branch would be a substantial improvement. Having to do two of abyss, vaults, and slime in a typical game is both more challenging and more varied than the current two lair branches, which are either substantially similar (swamp + whatever poison branch) or really annoying (shoals), and most likely vaults or much less likely abyss or slime.

[all before]

bel: this proposal still generates 2 S-branches per game. It just does not require you to complete both.

[Rast]

Flying isn't needed for the water branches any more. (it never really was).

Snake and Spider should be refactored to be less poison-centric, and more interesting in general

[archaeo]

The speedrun standard is 15 runes, since that's the only way to high score. 3 rune speedrunning is pretty dead, as far as I know.

Not dead at all. In fact, the game's best speedrunner is trying to take the 3 rune high score, though I'm not sure even Sapher can beat DDHe at the height of its power.

Of course, we all thought the same about beating NaWz for the longest time.

[bel]

bel: this proposal still generates 2 S-branches per game. It just does not require you to complete both.

I know. I was just responding to the earlier proposal because someone mentioned it in the thread.

[bel]

The speedrun standard is 15 runes, since that's the only way to high score. 3 rune speedrunning is pretty dead, as far as I know.

I only do 3 rune speedrunning. Anyway speedrunning is not too important for crawl design generally.

[mps]

The speedrun standard is 15 runes, since that's the only way to high score. 3 rune speedrunning is pretty dead, as far as I know.

Not dead at all. In fact, the game's best speedrunner is trying to take the 3 rune high score, though I'm not sure even Sapher can beat DDHe at the height of its power.

Of course, we all thought the same about beating NaWz for the longest time.

Oh, I thought Sapher was doing 15 rune runs still. I guess you can't really tell from the typical game o.o;

@bel: speedrunning may not be important to crawl design now, but it should be.

[Sar]

speedrunning may not be important to crawl design now, but it should be

What, why?

[mps]

Because normal play doesn't test the limits of the game in the way that speedrunning does. Unnecessary constraints (of which there are many) on speedrunning narrow its scope and leave aspects of the game unexplored.

[Sar]

Ehhh, I don't think the game should be designed with speedrunning in mind (but I am biased obviously), but I'm not a fun of runelock as well (though I am not sure it influenced speedrunners at all).

[all before]

This thread is being derailed by comments on speedrunning and proposals for one S-branch. Speedrunning does not matter to crawl design and one S-branch has its own threads elsewhere. Any criticism on the proposal?

[Rast]

It's clunky -- now runes are divided into two types -- and unappealing.

Also, diving the second S-branch is fast and easy. The final floor is less easy, but that's a good thing.

[archaeo]

I'm not sure I see a lot of benefit to switching around the rune requirements. I don't mind the "move demonic to elf" idea so much, but I otherwise feel like it'd be preferable to have more Lair branch variety (which is really hard) rather than forcing players to do more of extended. 2 runes just seems kind of paltry, all things considered; once you get to two, you may as well make it one or zero, I feel like.

but I'm not a fun of runelock as well (though I am not sure it influenced speedrunners at all).

Obviously rune lock didn't keep people from setting new, extremely high scores (including Sapher's 75-million point win), but I imagine diving to V:5 would return to quick prominence if rune lock went away.

[chequers]

You could allow players to only enter one S branch, but still have two in a game.

[bel]

To all before, as I said, in my last sentence in the earlier post, lair branches are now 4 levels (could be made 3, I don't mind). And one can always dive to the last level for the second rune branch. I don't think it is too tedious myself. The last floor is generally tough enough for characters who have done the other branch.

[all before]

I agree that 4 floors for S (or even 3) is an improvement. Still, though, it forces the player to do two branches of equal difficulty, which imo gets repetitive and unchallenging. I thought this change allows for the best of both worlds: players who want to do both S can, but if they don't they can just move on to V/M/A.

[mps]

This thread is being derailed by comments on speedrunning and proposals for one S-branch. Speedrunning does not matter to crawl design and one S-branch has its own threads elsewhere. Any criticism on the proposal?

Yeah, I think the proposal only makes sense with one non-slime lair branch, which is why everyone is talking about one non-slime lair branch. It would also have a major impact on the tempo of the game, which is why people are talking about speedrunning.

If you want to talk about the proposal in isolation, I think it's a bad idea. It means you need to do a much smaller fraction of the game to win and a lot of content would no longer be a factor. For example, I'm not sure I would ever play a shoals branch outside of a 15 rune game. On the other hand, games with swamp would be extremely easy. It would shift the focus of the game to Zot and winning a game would revolve around assembling an orb of fire kit.

I think a typical route would look like this: lair -> orc -> poison branch -> swamp if it spawns -> elf, but only if you really need loot -> depths -> zot. If shoals spawns, it would be more like lair -> orc -> poison branch -> maybe elf, as above -> slime (using one of many cheap tricks for killing trj) or abyss (if you're stealthy) -> depths -> zot. Vaults would be much more of a niche thing, because there are few characters for whom vaults would be an easier second branch than both slime and abyss. That would be bad, because vaults is one of the better branches.

[njvack]

For example, I'm not sure I would ever play a shoals branch outside of a 15 rune game. On the other hand, games with swamp would be extremely easy.

Unless, of course, one were to rebalance shoals and swamp to be closer in difficulty.

[all before]

It's already the case that games that generate Shoals are more difficult than those that generate Swamp.

A comparison: In the current standard three rune game, the player chooses between V, M, and A for the third rune. Imo this is an example of how branch decisions should work--each offers its own challenges, they're well-balanced enough that all three are viable choices, and most importantly, the player is not forced to complete the other two after having finished one.

Why not make the choice among S branches work the same way?

It's true that the S's aren't as well-balanced, but it's still an improvement over the current situation, and just shows that the S's need tweaking.

[PleasingFungus]

Snake and Spider should be refactored to be less poison-centric, and more interesting in general

This happened to Snake in, what, 0.14?

[mps]

Well, if we're talking about shoals vs. swamp now, I think it's hard to fix. The point with shoals is that you have a large number of monsters with ranged attacks and polearms. I don't think you can easily replace that kind of damage output with monsters in swamp. Swamp has some good monsters, notably the mangrove guys, and the spriggans fill a niche similar to merfolk with polearms, but there aren't enough of them to have the same effect. The hydra gimmick is easy to defeat by that stage of the game. Increasing the number of spriggans would be a start in making swamp more comparable to shoals.

Even if shoals and swamp were of comparable difficulty, I still don't think it's a good idea to let the player go to zot on the strength of just two lair branches. A swamp game still has the problem of not having enough variety. The player should have to do vaults or something more exotic, imo.

I like the idea of giving the player a choice of lair branches, but it should be an exclusive choice, i.e. entering one closes off the other.

[Rast]

Snake and Spider should be refactored to be less poison-centric, and more interesting in general

This happened to Snake in, what, 0.14?

It was a step in the right direction. It's not enough.

Honestly, those branches should probably get Hive'd.

[all before]

Even if shoals and swamp were of comparable difficulty, I still don't think it's a good idea to let the player go to zot on the strength of just two lair branches. A swamp game still has the problem of not having enough variety. The player should have to do vaults or something more exotic, imo.

I agree. I was pushing in OP for players having to do at least one non-S branch to enter Zot.

With making the branch choice exclusive, other people worry about the scenario where you get 5 unstoppable uniques on a floor and can no longer choose to back out and do the other branch instead. Personally I don't think this is so bad (it can happen currently in D and no one seems to mind), but this proposal allows people to still have that option, while also not forcing the player to complete both branches.

[dpeg]

Snake and Spider should be refactored to be less poison-centric, and more interesting in general

This happened to Snake in, what, 0.14?

It was a step in the right direction. It's not enough.

Honestly, those branches should probably get Hive'd.

I am happy to see that our players encourage radical cuts/nerfs, but this one will probably not happen.

In contrast with Hive, all Lair subbranches have a lot more content, and have seen much more care/work applied to them. This by itself is not a really good reason to keep some feature in the game, but perhaps it's psychologically understandable. Just to be clear, I am talking about very many monsters, many of them specific to the branches, and a truckload of vaults, and special layouts.


What this thread should be about (in my opinion), is a broader topic: what does non-linearity mean (in roguelikes, say), and is it good for Crawl?

Games like Rogue or Brogue are linear, and they fare very well with that: gameplay is tight, and that's good. Nethack has branches, and already suffers from it. (I don't know whether Crawl inherited branches from Nethack or Angband or both.) Obviously, branches are cool in various regards, but I am only concerned with gameplay: branches allow you to choose between different threats and rewards. That's a strategical choice, and hopefully interesting (one check to see whether it's interesting: will player decide differently, depending on local circumstances?). The drawback is that if the two branches pose comparable threats at the outset, then doing one will make the other one much easier afterwards.

The rune lock was a very mild attempt to address the worst part of that -- players postponing the most interesting threats (branch ends) as long as possible. With the lock, they now sometimes have to do a Lair subbranch end a bit earlier (I said it, the lock is really mild ). What is discussed in this thread is much more radical. If you follow that train of thought to its logical end, then you'll get: do away with the runes, offer some side branches, and let players decide when they think they're ready for Zot. (This is the Rogue plan, with the addition of optional branches and a safety vs score tension around them.)

I think that's too far a departure from what Crawl is. Runes are a decent concept, because they serve a a yardstick to progress (there are big differences between players who have "never got a rune", "collected a rune but didn't win", and "winners", for example). Also, the rune vaults often lead to interesting situations, and often death. Just on the Lair subbranches, I can see the following less radical ideas:

• Each game features one S-branch out of the four. (That's the current branch roulette, restricted to just a single branch.)
• Offer two branches as now, but players can only enter one. (If you enter one, the other is closed forever.) This sounds intriguing, but I am not sure this will often lead to interesting decisions: very often, the choice will be clear. Also, that approach forces us to always keep an eye on pretty strict loot & threat comparibility among the four branches (much more than now). Note that the runelock suffers from the same problem (often quite clear which rune to fetch) but that its primary goal was something different: to get players to fight one branch end a bit earlier than they'd do otherwise.
• Offer two branches as now, but entering one will change the other: it has the same amount of loot and xp (on average), but in a much more concentrated and dangerous form, e.g. monsters fully concentrated in more open branch ends. Again, the idea is not so much to aim for an interesting strategic choice (will still be quite clear which branch to attempt first), but to make the other branch more challenging afterwards.

Just to be sure: I want to keep all four S-branches; I would be very happy if the choice of runes is not clearcut (some posters indicate this is already the case); I am interested in discussing rule changes to make the various branches work better (i.e. don't become too shallow if tackled late). On the last point: there is some value in being able to overpower the opposition -- afer all, you've built that character for so long, and it's cool to reap the benefits and have it trivially crush the weaklings. There just shouldn't be too much of that.

[bel]

Snake and Spider should be refactored to be less poison-centric, and more interesting in general

What's wrong with Snake? How is it less interesting than, Swamp or Shoals? I like Snake the most out of all the S-branches.

[all before]

dpeg: thanks for the in depth reply! I want to be clear that I in no way support getting rid of any of the existing S's or of having to make branch order decisions. I disagree that my proposal taken to its logical conclusion is an argument for that.

As I say above, I really *like* the place the current third rune choice is at. Choosing among V/M/A is a meaningful decision, and doing so adds variety to my games. I want more of that.

My OP is really saying: let's get the Lair rune choice to be more like that third rune choice. Don't make the player grind through the branch she didn't choose (unless she wants to! as you say, power gaming can be a fun reward). And yes, for this to work optimally, the existing S branches will require tweaking to be closer in difficulty. But that's something that should happen anyway, and I don't think this proposal makes the status quo of unbalanced branches any worse.

[mps]

You could make choice of branch less clear cut in the two branches, choose one scenario by scaling the difficulty of the branches based on what level of lair they appear on, somewhat the way ice caves seem to work. Adjust the frequency of top tier spawns, etc.