Sil: What are your least liked features of Sil?

[Scatha]

Personally I'd be fine with (and would prefer) the simpler and more intuitive system of preserve=on and no fudging of the artifact rolls depending on number already generated.

I think you are fooling yourself that Preserve mode is intuitive, perhaps because you're used to it from Angband. A true Preserve mode, which allowed artefacts to be generated again whether they'd been identified or not (so long as there was never more than one on the level), would at least be simple, if somewhat odd. Having differing behaviour based on whether something has been identified is really quite strange.

This might make the standard deviation of number of artifacts found greater, but the average the same. That sort of randomness is fun in its own right, and is a core gameplay principle of roguelikes, to me.

I'm not sure you're engaging with the answers we're giving. I explained that the problem is not the luck-based variance, but giving regular characters enough artefacts to have fun with while not making things ridiculous for the power characters.

[Starhawk]

I have mixed feelings on this. I don't like the realism issue of quick armour switches, and I don't like the gameplay encouragement to micromanage. On the other hand, with the single exception of forging, every action in Sil takes one turn. There is a simplicity to this which makes it easy to understand, and an elegance which makes it appealing as a mechanic.

There are some actions which are compound, and so in effect take more than one turn, such as tunnelling through quartz or granite, but I don't see an easy way to break up the armour exchange without demanding that you first spend a turn taking off your old armour -- which would get annoying extremely quickly.

If you break armor exchanging into two turns (remove A / wear B), the next question you get will be someone asking why you don't have to put away your weapon and shield to use a bow.... or why you don't have to have a free hand to drink a potion.... or how many turns it should take to chew an herb.....

Don't give in! If I want ultra realism, I'll go run around the woods chasing deer.

[taptap]

+1 to starhawk. Convenience tops realism concerning armour swapping. It also encourages strategic thinking and varied playstyles = more meaningful decisions in the game.

Nobody "plays" pacifists, they "work at" pacifists.

But yeah point taken. Artifacts could still drop from chests though.

Personally, I rank my (semi-)pacifists among my most tense and exciting characters. How can it get more exciting than a char that can be killed by literally everything?

Artefacts: When I played with a staff of treasure and repeated the 900/950ft once and probably got lucky - I already complained about too many artefacts in the forum. Preserve on wouldn't help the players close to the turn limit but those very powerful who have plenty of time to scout 900/950ft. repeatedly and don't need more help. But you can say, the automatic drops are in a way preserve on, so if you want to build around a certain artefact take one of those, as you are more likely to get them (They are: Spear of Boldog, Glend, Delmereth and Galvorn Armour of Maeglin.) or wait until christmas.

[BlueFish]

I think you are fooling yourself that Preserve mode is intuitive, perhaps because you're used to it from Angband. A true Preserve mode, which allowed artefacts to be generated again whether they'd been identified or not (so long as there was never more than one on the level), would at least be simple, if somewhat odd.

Well obviously the stuff about "true preserve mode" is a strawman since I'm talking about the preserve mode of vanilla.

Having differing behaviour based on whether something has been identified is really quite strange.

I strongly disagree, in that I think this idea favors theory over practice.

The most intuitive notion of searching for items, including artifacts, is that you should explore the dungeon and kill monsters. Without regard to trying to fully explore levels. It is completely unintuitive to think that taking the stairs could have any conceivable effect on your chance of finding artifacts, other than changing the depth at which you're searching.

That clear intuition is broken by preserve=off, for the reasons I've already stated, having to do with the (counter-intuitive and non-obvious) "celestial compass" which guides the player to levels on which artifacts can be found, but then penalizes the player by not allowing those artifacts to ever be found again, if they are not found there. This is the effect of preserve=off, and this is why that's not the default in Vanilla.

To find something wen you have zero clue about its whereabouts, you simply cover ground and search. You don't pre-occupy yourself with gaming the system by fully exploring one partion of an infinite space before exploring another partition of that infinite space. One part of teh dungeon should be identical to another for the purposes of finding stuff. That assumption is what is broken by the theoretical "logic" of preserve=off.

I do get the theory - that the same thing can't be in two different places. But in practice, the player isn't confronted by that idea often, and when they are, it's hardly frustrating.

Point taken about the odd tactics of not IDing artifacts you're not interested in - but those tactics would not be encouraged by preserve=on. They'd be encouraged by the fudging of artifact rolls based on number that had already been found. That fudging could also easily be done away with.

It's been made clear that artifacts aren't that powerful in Sil to begin with. It's ok not to over-design the artifact system to make sure a certain number are found in a game. It's ok to be more random. This would not be rocket science to balance. The game would stand on its own with no artifacts at all. There's huge wiggle room between that, and making artifacts so common that they make the game too easy.

I'm not sure you're engaging with the answers we're giving. I explained that the problem is not the luck-based variance, but giving regular characters enough artefacts to have fun with while not making things ridiculous for the power characters.

That's exactly what I already said (that the point of the system was to make the first couple artifacts more common rather than making the last couple less common). I didn't expand, for the sake of brevity, but if I had, I would have noted that the fudging of artifact rolls based on the number that had been generated is presumably just a hack which allows the initial artifact generation chance to be abnormally large.

All of this has been over-designed, IMO. (In a game that you and half did a remarkable and impressive job of avoiding that pitfall.) The normal vanilla system with preserve=on would work perfectly in Sil, and would remove the counter-intuitive and frustrating consequence of leaving levels unexplored - that you can permanently lose artifacts by doing so. While that makes some logical sense from a shallow perspective, actually, in an effectively infinite dungeon in which you're searching for stuff, changing levels should have zero effect. Logically. And that's not a point of pedantry, it's actually how people think.

[taptap]

frustrating consequence of leaving levels unexplored - that you can permanently lose artifacts by doing so

Did you ever notice this in the game - being starved of artefacts in the late game because of stair use earlier? Then it may be a fair argument that chars somewhat below the power curve are further punished by having access to less artefacts. For me pointing out the effect was certainly theoretical as I tend to find artefacts in the throne room and sometimes the best of all during ascent while not even looking - but admittedly I default to explore full floor, go down when I am not in need of stair use.

Btw. I find it far more frustrating when I find strong artefacts when I already committed to a different build (like the wonderful artefact (light) swords with my recent polearm master), when I walk past artefacts without noticing and only see them in the notes afterwards or when I find seriously underpowered artefacts I have no use for (Boots of Irime) than getting a point on some artefact-counter where I don't notice it at all. I wonder how self-smithed or automatically dropped artefacts enter this equation.

Either way, you don't need a full set of artefacts to win the game. Most chars that make it to 1000 ft. have a fair chance of taking home a Silmaril.

[half]

I wonder how self-smithed or automatically dropped artefacts enter this equation.

They have no impact on it, to avoid any weirdness about trying to not kill Boldog or something.

Indeed, the player needn't know about the rule at all.

[Scatha]

Well obviously the stuff about "true preserve mode" is a strawman since I'm talking about the preserve mode of vanilla.

The point was that the behaviour depending on whether items had been in-game-identified, which I found completely unpalatable, could be avoided. I was trying to find a better preserve mode, not build up a strawman!

The most intuitive notion of searching for items, including artifacts, is that you should explore the dungeon and kill monsters. Without regard to trying to fully explore levels. It is completely unintuitive to think that taking the stairs could have any conceivable effect on your chance of finding artifacts, other than changing the depth at which you're searching.

Thanks, this section was really helpful for me to understand the intuition driving the idea of preserve mode: the idea that levels are just an abstraction which realise the idea of a huge dungeon stretching on and on.

This suggests another alternate version of preserve mode which I'd be much happier with: preserve artefacts unless they've been seen (rather than identified). This has much less potential to distort behaviour, since you'll never know what it is you're not seeing. The only distortion I can think of is that it might make you reluctant to use Staves of Treasures on a level if you're worried you might be forced off. You might reduce the radius of Treasures (something we'd wondered about anyway) if this was an issue.

It's not clear to me that is better: Sil has reasonably small levels, and I think they can be taken more at face value than as abstractions, but I appreciate the idea of wanting something in that direction.

That's exactly what I already said (that the point of the system was to make the first couple artifacts more common rather than making the last couple less common). I didn't expand, for the sake of brevity, but if I had, I would have noted that the fudging of artifact rolls based on the number that had been generated is presumably just a hack which allows the initial artifact generation chance to be abnormally large.

Sorry, when you referred to making the first couple of artefacts more likely I thought you meant ensuring that artefacts could reasonably be found early in the game (which is also an issue), rather than making sure everyone has some across playthroughs.

However, I don't think that there's a real distinction between making the early ones more common and the later ones less common (with an appropriate shift in baseline rarity). It's all trying to deal with the problem that you get the best stories when there are somewhere between, say, 4 and 12 artefacts you come across over the game, but that some characters explore a lot more than three times as much of the dungeon as others, so we don't want to make the number of artefacts found linear with the amount of dungeon explored.

This does follow a story-logic. It's not that the dungeons are actually infinite, but that we don't know how large they are. In stories (characters) where they're larger, you have to cover more ground to find the artefacts.

The normal vanilla system with preserve=on would work perfectly in Sil, and would remove the counter-intuitive and frustrating consequence of leaving levels unexplored - that you can permanently lose artifacts by doing so.

I think this mindset is ported over from Angband, where it's more correct. It's a little weird talking about losing things permanently in Sil, since nothing is permanent -- it only lasts a few tens of thousands of turns. Sure, you might never find Glamdring because it was on the floor on a level at 600' you abandoned, but you're far more likely never to find Glamdring because it just never gets generated. So Glamdring was never really yours to lose.

[Mikko Lehtinen]

This suggests another alternate version of preserve mode which I'd be much happier with: preserve artefacts unless they've been seen (rather than identified).

Yes, this would be perfect.

In my mind, unseen areas of a generated dungeon level are just as "real" as levels that have not been generated yet. Having different game mechanics for these two bothers me a bit in a game with infinite dungeons and mechanics that push you to dive.

It seems really cool that after you've seen an artifact, if you can't step on up and handle the challenge, you will permanently lose the item. It even feels completely fair that you are then further punished by a reduced chance to find other artifacts.

EDIT: Preserve off is not the important mechanic here. Fudging the artifact generation chance down by 10% or more sounds like a similar but much more powerful effect. Ideally you shouldn't do it until the player has seen the artifact. This sounds like it might be a bit tricky to implement, but maybe not.

[Mikko Lehtinen]

It's not clear to me that is better: Sil has reasonably small levels, and I think they can be taken more at face value than as abstractions

Despite what I just said in my earlier post, I like this approach too. I'd just like Sil to telegraph this idea to the player somehow: that individual dungeon levels are more than abstractions. This might not be as intuitive as you think in a game with infinite dungeon levels. Are there other significant mechanics tied to dungeon levels, or is losing any artifacts on the level the odd one out?

Maybe for us veteran Angbanders it is harder than usual to treat dungeon levels as something more than an abstraction. People used to other roguelikes treat dungeon levels with more respect.

[half]

Are there other significant mechanics tied to dungeon levels, or is losing any artifacts on the level the odd one out?

Two other things that are only generated once are the unique forge and the greater vaults (4 out of 5 of which are where 4 very powerful uniques live and would be very odd to see twice on different levels).

Maybe for us veteran Angbanders it is harder than usual to treat dungeon levels as something more than an abstraction. People used to other roguelikes treat dungeon levels with more respect.

I think that is right.

[Mikko Lehtinen]

Two other things that are only generated once are the unique forge and the greater vaults (4 out of 5 of which are where 4 very powerful uniques live and would be very odd to see twice on different levels).

Are there "level feelings" that warn about these, and artifacts? That would make the system completely fair to me, and would be all the telegraphing I'd want. Perhaps the Angband's traditional "You feel there's something special on this level" tweaked to suit Sil's sensibilities.

100% sure level feelings would allow you to game the system, though. Maybe a chance of getting a level feeling that is somehow based on your attributes or special abilities. (And perhaps you've already thought about this and implemented it somehow.)

[Scatha]

Any kind of level feeling encourages stair abuse, unfortunately. We get this a bit already with Staves of Revelations when people are hunting for forges, but it's limited by the number of charges they have.

You could also think about having a preserve mode for the unique vaults, but it's not so obvious there's a clean rule. I thought you could allow them to be re-generated unless you'd seen any square of them, but then realised that if you're loud enough you might lure monsters out of the vault and realise it was there without ever seeing it.

[Mikko Lehtinen]

We get this a bit already with Staves of Revelations when people are hunting for forges, but it's limited by the number of charges they have.

That doesn't sound abuse, but a sensible thing that a person with a Staff of Revelations would do in a fantasy story.

Alright, after changing my perspective on Sil's dungeon levels, I'm pretty happy with the current system. I like how vaults and artifacts use the same system.

[taptap]

One-and-half handed weapons

I am not a huge fan of the change to the STR bonus system. I understand that more simplicity was needed, but it ended up being a rather big reduction in the power of throwing weapons (which aren't exactly overpowered anyway) and it made two-handed weapons (at least swords) more or less completely redundant (unless you use knockback). One of the logical problems before was that despite the +2 bonus to damage sides when two-handed often enough you ended with more damage one-handed than two-handed while now you always do more damage two-handed.

So an idea for the one-and-half hand weapons:

What about capping the two-handed damage of one-and-half hand weapons to the max damage it could do one-handed? I.e. a 4 lb 3d3 Bastard Sword would be limited to 3d7 w/o momentum with STR 4 there is no benefit for wielding it two handed anymore and there would be a benefit for taking a Greatsword instead. Basically, wielding it two-handed would give a STR-bonus of 2 for attacks with the weapon. (I am not sure whether there currently is a +4 bonus for knockback for one-and-half hand weapons wielded in both hands or not, reducing this to the 2 points STR bonus given for wielding them two-handed would be fine with me.)

Related topic knockback: I would limit effective STR for knockback to the possible STR bonus for the weapon, maybe with allowance to add two-handed bonus afterwards? There are some balance issues, but knocking back with a dagger feels just wrong. Before knockback almost completely depended on weapon weight, now it doesn't figure at all.

[BlueFish]

The point was that the behaviour depending on whether items had been in-game-identified, which I found completely unpalatable, could be avoided. I was trying to find a better preserve mode, not build up a strawman!



Thanks, this section was really helpful for me to understand the intuition driving the idea of preserve mode: the idea that levels are just an abstraction which realise the idea of a huge dungeon stretching on and on.

This suggests another alternate version of preserve mode which I'd be much happier with: preserve artefacts unless they've been seen (rather than identified). This has much less potential to distort behaviour, since you'll never know what it is you're not seeing. The only distortion I can think of is that it might make you reluctant to use Staves of Treasures on a level if you're worried you might be forced off. You might reduce the radius of Treasures (something we'd wondered about anyway) if this was an issue.

Yes, this would be good, and would solve 90% of the practical issues with how the current system penalizes players for not fully exploring levels.

It's not clear to me that is better: Sil has reasonably small levels, and I think they can be taken more at face value than as abstractions, but I appreciate the idea of wanting something in that direction.

Thanks Scatha. Sil's levels are I suppose relatively small compared to Vanilla, but the time limit makes the opportunity cost of exploring them larger than Vanilla. I do not believe that Preserve=off works better in Sil than in Vanilla; in fact I believe the opposite. In Vanilla, players who are truly worried about it can, at no cost to them, explore the levels. Even this though was found to be sub-optimal design, because it encourages a boring style of play, for no practical benefit.

Sorry, when you referred to making the first couple of artefacts more likely I thought you meant ensuring that artefacts could reasonably be found early in the game (which is also an issue), rather than making sure everyone has some across playthroughs.

However, I don't think that there's a real distinction between making the early ones more common and the later ones less common (with an appropriate shift in baseline rarity). It's all trying to deal with the problem that you get the best stories when there are somewhere between, say, 4 and 12 artefacts you come across over the game, but that some characters explore a lot more than three times as much of the dungeon as others, so we don't want to make the number of artefacts found linear with the amount of dungeon explored.

This seems an overly complicated design which provides little of practical benefit, as far as I can see. To intentionally design out the intuitive and realistic correlation between "number of items found" and "amount of dungeon explored", seems like it should be addressing a real problem with game-play. I'm not sure this addresses any problem other than a "gut feeling" of the designers that it's more correct or elegant.

Even the way you and half explain the design considerations point to the problem - you're not trying to penalize players for not exploring parts of levels. You're only trying to control the number of artifacts *found*. Whatever artifacts are lost without being seen is a side effect and not intentionally designed. It may be "baked in" to the balancing on some level, but it still introduces a motivation for the player to fully explore levels which is more gaming the system than playing the game and creating a compelling story.

I think this mindset is ported over from Angband, where it's more correct. It's a little weird talking about losing things permanently in Sil, since nothing is permanent -- it only lasts a few tens of thousands of turns. Sure, you might never find Glamdring because it was on the floor on a level at 600' you abandoned, but you're far more likely never to find Glamdring because it just never gets generated. So Glamdring was never really yours to lose.

But a player always wants to feel like there's a chance they'll find it. Preserve=off undermines that, for no practical benefit to the player.

You and half are right that if the mechanic was a complete secret, players wouldn't be upset by it, since they could never deduce for sure that that's what's going on. But it's not a complete secret, and the feelings of players are real, while they're playing.

I understand the elegance of the system from a designers perspective, but a Preserve system (such as the one you propose, or simply Vanilla's system) seems to give you as the designer more control over the players' experience of playing the game. It doesn't seem intentional that leaving levels unexplored has any effect on the number of artifacts a player finds. But I believe it does have a significant effect, all other things being equal.