Sil-Q Review

[half]

Hi Quirk,

I haven't had time to try Sil-Q yet, but have enjoyed reading this thread and finding our more about it. Like Scatha, there are things I like and things I dislike, but my overall impression is being impressed that you *get* so much of the underlying aesthetic and reasons behind many of the choices we made, so as to be able to improve it at all by our lights. That is no easy feat. It is difficult to pick up someone else's novel and make a whole lot of changes and find that they agree with even a tenth of them, and I think the situation is similar with unusual, opinionated games like Sil. The people behind mpa-sil shared some, smaller, part of the aesthetic with Scatha and I, while you share a different and larger part.

Here are a few quick comments on things that have come up in the thread.

Deathblades: are there for the reasons Scatha mentioned, and also because they are cool. This probably wouldn't be quite enough to include them (especially not with a not-so-Tolkienian name), but the extra reason is that they are a subtle homage to the Dungeon of Doom: the first roguelike I played and one that had a few really nice ideas and hasn't been remembered by the community (perhaps because it was on Mac). I'm willing to bend things a bit more than usual to get in a homage, but future developers should feel no need to bend things to allow a homage they don't personally connect with.

Blunt weapons: I never intended these to be balanced with swords, as I think they simply are less good for combat (prior to plate armour), and they aren't used much in the source material. I included them because they are mentioned in a few places and I'm happy for them to exist and just be worse. The same goes for broken swords, curved swords and filthy rags. To a lesser extent, we were happy to have swords be a little better than other things for most characters so that most characters ended up with some kind of sword (like in the books) rather than spread equally between all kinds of different weapons.

Scaring monsters: Apart from the monster that kills you, you always manage to deal with the monster in some manner (be that sneaking past, fleeing, scaring, or killing). You get half the experience for each monster on first encountering it because that is easier to implement than getting it upon dealing with the monster. I'm happy with killing it being the only thing that grants extra experience. As why is scaring it off the level any better than scaring it into another room? or to sneaking past, or putting to sleep, etc.? If I were changing anything, it might be to drop the extra experience for killing it -- you already are rewarded with the items. (this would obviously require increasing experience a little to compensate)

Early game: As Scatha said, we had an approach of making it feel more magical and more complex as you descend. But the downside is more mundane and simple at the top. The mundane is bad for new players and the simple can be bad for old players. I'm willing to believe that we overdid it and should have left a bit more excitement for the first levels.

Other things: It's great that you found and removed many small bugs, and make various changes that were overdue (e.g. removing the need to be hungry before Morgoth, doing something about Song of Slaying in the throne room, and doing something about Momentum).

As I said at the top, it is great to see that you understand so much about what the game is trying to do, and have continued in that spirit.

[Quirk]

Hi half!

Thank you for all the kind words. Thank you also for making such a great game. I believe a lot of the aesthetic resonance you note in me exists precisely because Sil is such an unusual and opinionated game: the game has principles and it is true to them. Beginning from old Angband, it strides strongly toward a descent that Tolkien might have described, while enabling a rich strategic sensibility with a flavour very unlike its predecessors. Where I have succeeded, it is by understanding the game's principles.

There are flaws in some of my additions, and particularly in some of the earlier ones where I was a little too keen to latch on to whatever feedback I could get. I strive to keep on improving. I've talked at length on most of the topics here, so I won't go back and cover the same ground, though I will note again that the experience for scaring monsters was in hindsight a mistake, and if I do decide in future to attempt to make the Elbereth pacifist a little more viable it will be through less clumsy means.

Lastly, I hope that when you and Scatha find time to return to Sil development that you feel free to lift and improve any ideas in my little fork that you like. I have enjoyed your creation immensely.

[Quirk]

I went to make an account to delurk and comment on this thread and discovered I'd already made one... 5 years ago? God.

Exceptionally good to have you posting. I've very much enjoyed your comments on your ladder characters.

I just tried a few abortive runs with Song of Delvings after Scatha mentioned liking it. I was pretty unimpressed by it's description on paper but I absolutely love it in practice. Having that additional information to work with is nice for seeing where you might get flanked, planning possible escape routes, and spotting special rooms in advance (I look forward to avoiding wolf pens entirely once I survive deeper). I got consistently killed by rushing to it over more reasonable early-game investments because it's so fun to have.

Hopefully a slightly more reasonably costed version can find its way to be a fixture in places other than the ascent!

I would love to see Slaying and Sharpness return to Sil-Q on thematic grounds. I wonder if there's a place for Slaying as a "capstone" song, possibly combined with Challenge as Scatha mentioned so that it proactively drew in enemies to keep up momentum in more sparsely populated areas of the dungeon.

I would like to find a way to bring some variant of Sharpness back - probably at a much cheaper cost. Songs are hard to balance though as you usually gain the benefits of only one of them at a time, while you slowly lose Voice, and you attract more foes. This requires quite a lot of raw power to overcome, but enough raw power and a song tips over into an obvious pick, and a song that is an obvious melee pick in the endgame risks outshining simple useful options such as Trees and Freedom. Mastery also exists at present in a somewhat endangered position: the Song needed to make it good exists primarily on Lorien builds which don't really need it, and on the too-rare Elbereth archer. A more combat effective song in a similar position really would serve to virtually obsolete it.

Song of Overwhelming - Song of Fierce Blows at the start - began as an attempt to provide guaranteed effectiveness without the huge investment needed for Mastery. It failed; it did not offer enough over Song of Staying, and even if it had there was no ladder to get there. The ladders are Elbereth, Silence and Lorien and none of them work well for a non-stealthy melee character.

I may reconsider whether it is possible to make a melee ladder Song in making a fresh attempt at the Song tree. It is a little difficult in that moving on to the next Song requires the last to be outshone. Lorien is a clear upgrade on Silence, Mastery is an expensive and initially slightly dubious upgrade on Elbereth, but Slaying/Staying/Sharpness don't scale all that well over most of the game, and Challenge/Staying/Overwhelming have not done much better. My main idea at present is to keep the territory the ladder needs to reach short, so investing in a new song is not so daunting. Perhaps we can see more prerequisites return in time, but songs will need to be a little more popular first.

I'd also be in favor of ditching blunt weapons as a category. Having sceptres as a dedicated stat-stick weapon type (i.e. Sceptres of Power) makes some sense thematically, but my impression is that Sil design doesn't really favor stat-sticks. I agree that sceptres and crowns should be rarer, and possibly special-only.

On some level I like the notion of descending with robe and staff, but the approach really works only for the dedicated pacifist, and there aren't that many of us.

I would be tempted if I got to playing properly with level generation to reintroduce items which were more explicitly tailored to flavour than gameplay. I would rather though have Angband filled with things crudely made by orcs than broken elf-things. The latter is tolerable as long as it suits gameplay but it irks my inner simulationist.

I have to say I do like consumables as they are from a gameplay perspective, even if they don't make complete thematic sense. I'm still holding out for Herbs of Athelas and maybe Niphredil. I like the Sil-Q trend of adding a bit more variety in items and adding more of interest to the upper levels.

Athelas! Yes. I would prefer to do more with herbs and less with potions, though I do love the way Orcish Liquor works. This would be a potentially hefty rebalancing though, and in truth there's a limited supply of Tolkien-faithful consumable magical substances.

I think the question of ease of detecting and disarming traps is mostly a sideshow to the fact that Sil's trap design, with the exception of roosts and webs, is not great or at least certainly not up to the standards of the rest of the game. I might make a thread on this, actually, if there's not a minimum post limit to do so.

This is largely why I have weakened them so. I feel traps that the player is expecting are fair, provided the consequences are not too ruinous and in alignment with player expectations. If the player can choose to take the risk and has means of mitigating that risk, the player largely has themselves to blame when the trap shuts. Currently, traps spawn in places with no underlying logic, and a player who has fallen through a floor or damaged their armour simply through underinvestment in Perception or a poor roll is entitled to feel a little upset.

Finally, this doesn't quite fit with the current discussion but a thought about Sil I've had for a while is that Morgoth shouldn't be immune to fear, at least until the player has picked up a silmaril, on thematic grounds. His cowardice, and his fear of Elbereth in particular, is frequently noted in canon and I think that cowardice is a major part of Tolkien's conception of evil in general and the way it is often both seemingly unstoppable and pathetic.

This is a not unreasonable stance, though I am not sure he would necessarily flee the player when the player comes into his very own lair. He did not flee Fingolfin at the gates, for all he took his sweet time coming up, and I don't think second-hand Valar would be enough to scare the originator of the attack on the Trees, for all he may have feared Tulkas. Still, I'd listen to other opinions on it.

[Wiwaxia]

Exceptionally good to have you posting. I've very much enjoyed your comments on your ladder characters.

Thank you! I'm flattered.

Exploring the song tree a bit more, I've found Challenge somewhat helpful but bland; opportunist deals with the same kind of enemies it does much more stylishly. It might stand out a bit more if the trick to lure group AI into fighting one-on-one in doorways by closing the door was taken out, which was always kind of a weird exploit.

Thresholds is very fun. It feels super versatile: just playing around with it for a short while I used it to lock away things I didn't want to deal with, make a bolthole to rest, continuously cover my path so I wouldn't get outflanked, and lure archers into a room and trap and kill them one-by-one. I imagine it would also be very nice for defending forges and I can see an edge use in just using it to mark the map at some point you don't want to lose. I can see it having good synergy with Delvings and Listen to proactively control the flow of enemies through the dungeon. I like that the wards remain after you stop singing, because it puts Thresholds in less direct competition with other songs - you can get the benefits of both skills without needing Woven Themes (this is true of Delvings, too). I haven't gotten very deep with it yet, so I don't know how it plays at deeper levels.
My one complaint is the colors: green isn't intuitively "more" than blue, it would make more sense for there to be a gradient of dark to bright with increasing strength. Bright blue is also used for mithril, glowing slays, and lesser jewels and so has strong positive associations and feels like it should be strong rather than the weakest level. On the other hand, bright green has obvious visual associations with glyphs of warding and helps convey what the song is doing.
The 16-color palette makes coming up with a good color ramp hard, here. Dark green -> bright green -> yellow might work, although the directionality between bright green and yellow is still a little shaky. You could also do dark blue -> bright blue -> white, but I'd be concerned about white looking like the floor and thus less powerful rather than more.

I am seeing your frustration with costs on the song tree. I can't buy the more expensive songs off the bat, so I don't know how they play, so I don't have a good idea of how to start a build that will use them eventually, and I may not have anything to do with that song investment until I get there. This is doubly true for me as a so-so player, because my characters often die long before I can actually buy the ability I'm working towards. I don't know if cutting costs is the right fix, but I agree that it needs fixed (and as you've mentioned, lower costs means easier playtesting for new songs).

On some level I like the notion of descending with robe and staff, but the approach really works only for the dedicated pacifist, and there aren't that many of us.

I do think it's worth having some dedicated toys for rare builds like pacifist even if they will never be popular choices, especially because, as you say, descending with only a robe and staff is so evocative.

This is largely why I have weakened them so. I feel traps that the player is expecting are fair, provided the consequences are not too ruinous and in alignment with player expectations. If the player can choose to take the risk and has means of mitigating that risk, the player largely has themselves to blame when the trap shuts. Currently, traps spawn in places with no underlying logic, and a player who has fallen through a floor or damaged their armour simply through underinvestment in Perception or a poor roll is entitled to feel a little upset.

Very much agreed. I think it would be better, and it sounds like you agree here, to rework traps and/or trap generation rather than just weakening them, but I imagine that being pretty involved.

This is a not unreasonable stance, though I am not sure he would necessarily flee the player when the player comes into his very own lair. He did not flee Fingolfin at the gates, for all he took his sweet time coming up, and I don't think second-hand Valar would be enough to scare the originator of the attack on the Trees, for all he may have feared Tulkas. Still, I'd listen to other opinions on it.

Partly I just like the idea of an Elbereth-pacifist win on both thematic and gameplay grounds. Sil also already provides the opportunity for players to accomplish things well beyond canon (returning with all three silmarils, killing Sauron, hypothetically killing Morgoth) with a good deal of extra effort, and making Morgoth turn and run fits into that category, I think.

[Quirk]

Speaking of songs, after a discussion with wobbly: how would people feel about elves having song affinity instead of sword proficiency? Doriath having mastery, of course.

[HugoVirtuoso]

+1 for Noldorians to start with +1 Song

[Quirk]

+1 for Noldorians to start with +1 Song

So this would be a free song skill instead of a Song boost - like archery.

[HugoVirtuoso]

So this would be a free song skill instead of a Song boost - like archery.

What you said - That's what I meant.

[wobbly]

So I guess I'll explain my logic behind the idea. A "mage"-style build with multiple small utility songs has always been considered too expensive to be worth it, if the aim is to make multiple songs more affordable proficiency lines up with the current Sil mechanics. Any song later then trees is a large investment: Delving = 2.6k(6 + song), Staying = 3.1k(7 + song). Very few people are going to try this without using an earlier song as a stepping stone. How much is 2.6k? that's still +1 melee & evasion by the time you have 13 pts in melee & evasion, e.g. further then a lot of people reach in the 1st place. So I had the vague idea that challenge & elbereth could be the base of the song tree in the same way that power/finesse are the base of the melee tree (& trees, silence/lorien remain as alternative paths), but that needs the price of multiple songs to drop, as people often aren't taking songs now, so why would they take them if they had to invest even more?

As for sword proficiency going, that was mostly because 3 proficiencies seemed overboard & it occurs to me I don't like sword proficiency on an elf. Half has always expressed a preference for swords being king. I respect that, I just don't think the proficiency is necessary to achieve that. I think swords are king anyway (up until late game where it's a bit of preference). I often preference a sword on a dwarf despite dwarves having axe proficiency, the extra accuracy, the extra evasion, they are just more reliable. I also suspect it's a touch confusing, players forget that +1 is even there (I certainly do most of the time).

However I don't want to be pushing an idea just because it sounds good to me. After all it was my idea so I have a bias. Ultimately it's up to Quirk, but it would be nice to hear from anyone playing, rather then a change being added that players do not like or doesn't achieve it's actual aim.

Edit: I guess another way at looking at the question is - what would it take for a player to try a singer? To try taking multiple songs? to invest in more then the minimum to grab a favoured song?

[Infinitum]

I could do wihout the extra weapon proficencies tbh, swords are stronger than the other weapon cathegories anyway, and I seem to recall even the dwarfes of Bilbo going with swords over axes. Dwarfes being axe-only is more of a DnD trope (Gimli nonwithstanding). Maybe make swords a Feanorian specialty (instead of the archery affinity) to go with their clans preference for them in the raid of the harbours arc. Speaking of, maybe downgrade the archery affinity to a bow proficiency for the elves (with Falas receiving an archery affinity as well). Elves are plenty strong as is.

I still feel the songs should provide flexibility to a character, and as such shouldn't be purchased separately (since that encourages narrow specialization, especially with the character pregression system Sil uses).

One way to go about it would be to give each race its own set of songs, and make that a requirement for continuing down the song tree (eg an elf would have to learn elven songs at the bottom of the trees before learning Dwarfen or Human songs and vice versa). Each set of songs would be its own ability; eg:

Elven: Elbereth, The Trees, Overwhelming
Dwarfen: Delvings, Thresholds, Staying
Human: Freedom, Silence
Songs of Mastery: Mastery, Lorien

'Course, the elves are effectively two separate languages so if one really wants to geek it up one could move Thresholds to the Elven songs and then split them in Sindar and Quenya variants (Elbereth and Thresholds probably being Sindar what with Elbereth being a Sindar word and their proficiency with enchantments (eg the Girdle surrounding Doriath).

Speaking of the starting races, any chance of removing the Dexterity/Strenght maluses from dwarfes and Sindar respectively? Dwarfes being worse fighters than humans doesn't make much sense, neither does elves being weaker than men. They could both lose a point of constitution to compensate. Maybe give the dwarfes a stealth malus as well.

[Quirk]

So I've discussed this a bit with wobbly already, and won't say much more than: I think the song proficiency on elves is absolutely canon, dwarves and men are rarely seen working magic through song if at all. There is the possibility that it leads to every elf being a singer of some description, but that may be okay, and I do note that not every elf invests in archery at present despite its obvious power.

Speaking of, maybe downgrade the archery affinity to a bow proficiency for the elves (with Falas receiving an archery affinity as well). Elves are plenty strong as is.

This is a perfect and elegant idea. Swap out the archery affinity for song affinity, and sword proficiency for bow proficiency. I very much like this.

With regard to songs being split by race: currently it's been quite hard to balance them without imposing further restrictions, so I'm not sure how I feel about this. It would be somewhat difficult to make it work as you've suggested - Trees couldn't go at 1 elegantly, as many characters wouldn't have the requisite 5 points to get even one light level of benefit, Staying would be monstrously powerful, etc - but splitting songs between races would go some way to resolve the current crush of songs that need the investment to be cheap to be viable (Elbereth, Challenge, Silence, Freedom, Delvings, Thresholds). That said, we then weirdly incentivise the dwarven or human singer, which is dubiously canonical.

With regard to removing Dex/Strength maluses, I think my preference would be to remove a Con point from the Sindar, and a Grace point from the dwarves. Nogrod would then gain back the Grace point to be the same as before, and Belegost would swap the Dex malus for having 0 Grace. This preserves the bruiser nature of Belegost and may even be a small buff. Sindar will lose out some from not having the extra Con, but I think Sindar are currently decently stronger than Naugrim.

I do think it would be interesting to have some skills which were race-dependent, but it's easier said than done. Coming up with one decent skill to fill a slot in a manner which is balanced enough that it gets taken on some builds and not others is hard enough, never mind finding three, and having them all suit neatly into racial stereotypes. Trying to do this for multiple trees or multiple skills in the same tree will get wearing fast. Having a separate tree for them might work but feels like it might prove inelegant.

[Quirk]

What I was sketching out last night is a structure that looks roughly like this:

1. Song of Elbereth / Song of Challenge
2. Song of Silence / Song of Staunching
3. Song of Freedom / Song of Thresholds
4. Song of Delvings / Song of Whetting (prerequisite Challenge)
5. Song of Trees - (prerequisite Elbereth or Freedom)
6. Song of Staying - (prerequisite Staunching or Thresholds)
7. Song of Lorien - (prerequisite Silence)
8. Song of Mastery - (prerequisite Elbereth or Lorien)
9. Woven Themes
10. Grace

Song of Staunching, based on Luthien's song of staunching sung to heal Beren, would shorten bleeding and more importantly greatly speed regeneration.
Song of Whetting I'm currently thinking gives an extra damage side. I don't like the old Sharpness implementation in which each additional Song point gives you 2% of Xd4 more damage against your opponents, it feels clumsy. On the other hand the canonical whetting spell allowed Beleg to sever metal shackles with a sword, which feels like it maxes out at more than just a damage side.

In both cases the scaling with Song is tricky. Songs tend to a linear scaling when for balance purposes this does not always work very well. I have tried to keep this, but in places it can lead to relatively inelegant implementation.

Having Staunching's regeneration scale linearly is problematic: it will do too little early on, and possibly too much late - particularly if Song is scaling a multiplier on the existing regeneration mechanic which also scales with Constitution. Ideally it would do enough early to save a player from death by poison, and still be usable later. I may try something like "Stops all bleeding and regenerates 2 + (Song/6) health per round" (the numbers here will be tweaked after much testing). I think it has to go quite early because it is potentially a useful emergency song that could be grabbed at the point you realise you won't survive the poison you just took.

Whetting scaling to multiple damage sides could easily become very broken, but perhaps something like 1 + (Song/12) damage sides would be plausible. The other mechanic I was debating was it granting the extra damage side to weapons weighing less than the singer's Song score, making it easier to whet a dagger than a greataxe. It might not be broken to have it grant actual sharpness to weapons weighing less than (Song / 5) lbs as daggers and shortswords are very much stealth weapons and the song could be a noisy one; you would need to get to 10 Song to whet a light longsword. This would make shortswords somewhat more viable without backstabs.

Whetting arguably shouldn't work on blunt weapons, particularly if it's granting actual sharpness, but this approach gets us toward the situation where you put aside your warhammer when you have to deal with an armoured opponent, which seems fundamentally wrong - I can't think of a better tool to deal with an enemy made of stone.

I'm not sure how I feel about prerequisites here. Prerequisites block off some builds, and punish skills: weaker higher-tier skills suffer from having a prerequisite, weaker lower-tier skills suffer from not being a prerequisite to a strong higher-tier skill. Delvings is not a prerequisite for anything, which may mean it is still too costly at 4 after the competition has hotted up; it may be better to swap its place with Silence, which will still be viable at 4 because it lives on an all-in Song/stealth build. I don't really want to split Challenge/Elbereth or Freedom/Thresholds because they make nice naturally opposed pairs at the same cost.

[HugoVirtuoso]

Question about Smithing Enchantment ability: When you pick enchantment, does it apply to just to the chosen prerequisite (e.g. Weaponsmith) or both Armorsmith and Weaponsmith? Imo, it should apply to just the already chosen prerequisite.

[Quirk]

Question about Smithing Enchantment ability: When you pick enchantment, does it apply to just to the chosen prerequisite (e.g. Weaponsmith) or both Armorsmith and Weaponsmith? Imo, it should apply to just the already chosen prerequisite.

It applies to Weaponsmith, Armorsmith and Jeweller - but you can only make enchanted versions of these items if you already could make those items. So if you have Weaponsmith and Armorsmith and pick Enchantment you can make both enchanted weapons and enchanted armour, but you still can't make enchanted jewellery.

[HugoVirtuoso]

I realized I asked something else than intended: my corrected question - does the enchantment's identification apply only to those items I can make?

On a separate topic - various other things:

Does the Song of Elbereth work if player is blinded?

I noticed these too:
' command (reversed apostrophe) works as another ESC key.
When accessing the abilities via TAB key, TAB can be used to exit it.
When changing options, 't' can be used to toggle yes/no
Left arrow key can be used to exit out of the Smithing menu