Adapting mechanics from Sil to other (longer?) games

[Scatha]

I have seen the XP fall-off suggested before (and was planning to use it in Beleriand); the bit I think is genius is the experience for encountering monsters. That I am pinching.

That came into the game not long before release (you can see in the pre-release change notes), when we were looking for more non-combat sources of experience to make a wider range of builds viable. We're happy with how it turned out!

One mechanic I'd direct your attention towards if you're looking for things you'd consider putting into Beleriand is the rules for light and magical darkness. It's very simple, plays well, and makes light and dark a serious part of the game, which feels perfect for the Tolkienian setting.

Edit: Adding a description of the mechanic, since it's not in the manual.
Dungeon squares can by default be lit (light level 1) or unlit (light level 0). If you have light radius N then you add (N+1-k) to squares at distance k (where k < N+1). If you have dark radius N (= "light radius -N") you do exactly the same but subtract (N+1-k) instead of adding it. As well as the player, some monsters have a light or dark radius (as do some items on the ground). If the total light level on a square is positive, it's lit and you can see what's there. If a light-averse monster is on a square with light level L > 2, they get a penalty of (L-2) to morale and all of their rolls.

Edit 2: That's actually a very clunky description for such a simple mechanic. I hope the idea is clearly conveyed.

Edit 3: Perhaps a picture.

Shadow Spider (not moving), light radius -2; player, light radius 2; dark room:
######## ######## ######## ########
#******# #******# #******# #******#
#**M***# #******# #******# #****..#
#******# #******# #****..# #***@..#
#****.*# #**....# #**.@..# #**....#
#*.....# #*..@..# #*.....# #**...*#
#*..@..# #*.....# #**...*# #******#
######## ######## ######## ########

[Scatha]

The one aspect of Sil I don't like (and remember this is on paper/in theory, since I haven't actually played it yet) is the twin sets of rolls for combat. I've played a lot of combat games, and I prefer there to be one roll for attack and a constant value for defence (AC or evasion or whatever). Two opposed rolls makes the mental arithmetic of estimating your chances much too hard. Of course this may be the idea!

There are definitely some simplicity advantages to a single roll. However an advantage to using opposed rolls (or indeed a single "roll" drawn from a distribution symmetric around zero) is that it makes the different skills directly comparable. If your melee is higher than their evasion, you will hit over half the time. And sometimes we roll, for example, Will against Will. The opposed rolls also give you more of a bell-shape, which reduces the incidence of weird effects where extra points in a skill are useless to you.

That said I think we'd like it if there were an aid in-game to help you with the combat arithmetic.

[fizzix]

That said I think we'd like it if there were an aid in-game to help you with the combat arithmetic.

In Vanilla once you learn the AC of a monster you will see something like, "You have a x% chance to hit this monster with your current weapon" on the creature recall screen. I could imagine putting that information there. Maybe something like:

Your chance to hit monster: x%
Your average damage:  d

Monsters chance to hit you: x'%
Monsters average damage: d'

The calculation would include your current weapon and armor of course, so it should change as you switch around weapons. If you really want you can include some measure of variance for the damage as well. Extra lines for "range of possible damage" and/or some confidence interval for likely damage values.

[Magnate]

There are definitely some simplicity advantages to a single roll. However an advantage to using opposed rolls (or indeed a single "roll" drawn from a distribution symmetric around zero) is that it makes the different skills directly comparable. If your melee is higher than their evasion, you will hit over half the time. And sometimes we roll, for example, Will against Will. The opposed rolls also give you more of a bell-shape, which reduces the incidence of weird effects where extra points in a skill are useless to you.

I accept that opposed rolls provide a symmetric distribution, but I think an asymptotic single roll with diminishing returns is a nice hybrid of smooth distribution and simple mental arithmetic. So every skill point continues to get you *something*, but each is worth slightly less than the last.

It's worth noting that V spectacularly doesn't do this: there are huge breakpoints at the upper ends of the STR and CON bonus tables. Then of course there's the significantly un-asymptotic 0% failure ...

[getter77]

At the very, very least then----exp for successfully Identifying things and making it to a newly reached decent depth or other metric is a no-brainer and would essentially move mountains towards making stealth and other more unorthodox playstyles more viable across the entire landscape of such, especially alongside the "first encounter" exp---since otherwise all you have is combat kills and that funnels some tunnel vision into the entire game design something fierce.

[Nick]

One mechanic I'd direct your attention towards if you're looking for things you'd consider putting into Beleriand is the rules for light and magical darkness. It's very simple, plays well, and makes light and dark a serious part of the game, which feels perfect for the Tolkienian setting.

I like that, and thanks for the description. FA has Oangband's Unlight ability, which enables the player to see in the dark as if by lantern light - I can see that also fitting in nicely with your system. So I will probably include that or something very similar eventually (I'm not sure whether I'm comforted or depressed by that fact that you guys were working in Sil for 8 years before releasing it...)

[Scatha]

(I'm not sure whether I'm comforted or depressed by that fact that you guys were working in Sil for 8 years before releasing it...)

I'm not sure where 8 years comes from. I think half first started thinking about it before that (he mentions 2001), and I began talking to him about it some time in 2007, and contributing ideas a little after (the alpha version numbers mark the point where he shared the game). Maybe 8 is the average of the two?

The Unlight ability sounds pretty fun. Would it render you unable to see lit squares?

[Nick]

IThe Unlight ability sounds pretty fun. Would it render you unable to see lit squares?

Visually it acts as if the player has a light radius of 2 (although you don't get the special torch light colour), because it has to be actually a good thing for the player. And it gives bonuses in the dark rather than hampering in the light - it's like the player can put darkness to good use.

[Scatha]

Visually it acts as if the player has a light radius of 2 (although you don't get the special torch light colour), because it has to be actually a good thing for the player. And it gives bonuses in the dark rather than hampering in the light - it's like the player can put darkness to good use.

Two uninformed and perhaps mutually opposed comments, I'm afraid:

Firstly, is it obviously necessary that it be a good thing for the player? If it comes together with various other bonuses it could work well even if it's overall bad in itself. (I often like effects like this.)

Second, do I correctly understand that they can only see out to distance two in the dark? That sounds almost a disappointing handicap on an awesome-seeming ability. Would anything break if it let you see the whole dungeon as though floodlit? (possibly with an exception for lit places if you want to play with that route)

[Nick]

Firstly, is it obviously necessary that it be a good thing for the player? If it comes together with various other bonuses it could work well even if it's overall bad in itself. (I often like effects like this.)

Well, it's a specialty ability, and the player only gets to choose 3 or 4 of those - so it has to be competitive. There is actually (IIRC) an ego-item version which is a bit more two-edged.

Second, do I correctly understand that they can only see out to distance two in the dark? That sounds almost a disappointing handicap on an awesome-seeming ability. Would anything break if it let you see the whole dungeon as though floodlit? (possibly with an exception for lit places if you want to play with that route)

Of course this is what should happen - thank you.

[debo]

I told you I'd revisit this someday

I've had a crazy idea I can't get out of my head lately, which is to borrow Sil's mechanics for use in a story-driven tactical JRPG.

The idea would be that you walk around like in a normal JRPG, except the encounter rate is a bit lower. When you're attacked, it switches to 'battle mode'. I want this to be a throwback to old brick GB style gaming, so small screen (would also be good for mobile). Maybe 6x8 tiles, with random terrain generation. This battle screen would be similar to Sil/*band combat in general.

Some hitches:

- Up to 4 PCs would be found in the quest. This means turn based combat. Simplest thing to do would be to have the players move first, then the monsters. Actors with greater speed could get 2 actions a turn instead of one, lesser speed would 'miss' a turn.

- I'd like to have the same use of XP as in Sil, sort of -- stats are fixed, HP+Voice are fixed, but you can permanently increase stats via buffs of some kind (potions, pills, whatever)

- I was thinking of having 8 skills:

o Melee (Dex)
o Evasion (Dex)
o Gunnery (Dex)
o Stealth (Dex)
o Perception (Gra)
o Will (Gra)
o Song (Gra)
o Wizardry/Magic (Gra)

- And 4 character classes, each with 4 'prime' skills

Gunner:
o Evasion, Gunnery, Perception, (Stealth or Will, not sure -- depends on backstory)

Songsword (name required lol):
o Melee, Evasion, Will, Song

Assassin/Survivalist:
o Melee, Evasion, Stealth, Perception

Mage:
o Evasion, Will, Wizardry, Perception


You can spend XP to buy points OR gain abilities on Prime skills. You can only buy points in secondary skills, you cannot gain those abilities.

After each battle, you get fixed XP. (No XP for seeing a monster the first time, discovering a new item, etc, this is more trad JRPG stuff.) All dead characters are restored to life and maybe to full HP or something.

There's a bunch of open questions here (big one: How do character abilities interact?), but this is sort of what my pen-and-paper RPG evolved into. I always hated how JRPGs, even the "strategy" RPGs, really lacked in tactics. Sil's tactical gameplay + mechanics could be super fun for muggles (e.g. non-roguelike people) in this context, especially if you take away the permadeath aspect but make battles legit hard.

I see several challenges in actually balancing XP expenditure and combat rolls, some of which Scatha has addressed in the above posts. My question to you all is, how dumb does this idea sound at first glance? There are a lot more details I might fill in as time goes on (perhaps in a different thread), but would love to hear first impressions of my rambling lol

[half]

My question to you all is, how dumb does this idea sound at first glance?

I'm not really the JRPG type, but this sounds pretty cool to me. The combat sounds a bit like Ultima III (and perhaps some other early party-based Ultima's) but with better tactics. Keeping the 'arenas' very small (9x9?) would help.



Anyway, sounds good to me, especially as the reason I don't like JRPGs much is the lack of tactical combat (and the grinding!). Feel free to use any part of Sil you like in such a project. We'd be very happy to see its design innovations spread.

[Scatha]

I've had a crazy idea I can't get out of my head lately, which is to borrow Sil's mechanics for use in a story-driven tactical JRPG.

Sounds interesting! A couple of comments:

- Up to 4 PCs would be found in the quest. This means turn based combat. Simplest thing to do would be to have the players move first, then the monsters. Actors with greater speed could get 2 actions a turn instead of one, lesser speed would 'miss' a turn.

A common trope in games where you control more than one unit is to make the turns blockier as a function of time than they are in roguelikes. This is probably sensible, because it means you can focus attention on each character in turn. You want to avoid the meaningless turns like wandering down a corridor in a roguelike.

You could probably cope with speed, to the extent that it's needed, by varying movement allowances per turn.

- I was thinking of having 8 skills:
...
- And 4 character classes, each with 4 'prime' skills

Aesthetically I'd want to have each of the skills be a 'prime' skill for 2 of the classes, given this set-up! I don't know if you could arrange that. There doesn't seem to be any need, for example, to have everyone so good at Evasion.

Sil's tactical gameplay + mechanics could be super fun for muggles (e.g. non-roguelike people) in this context, especially if you take away the permadeath aspect but make battles legit hard

I'm not sure it would be right for this game, but just a comment that I think permadeath could have good uses outside of traditional roguelikes. Faster Than Light is a recent example. It depends on things like how much you want to generate content procedurally.

Anyhow, good luck if you take this further!

[Derakon]

I've been spending a little time thinking about how party-based combat could work within a roguelike framework. I do think that it's important that moving a single tile not take your entire turn. If you look at party-based combat games, practically all of them either allot multiple tiles' worth of movement per turn (most tactical RPGs) or make positioning irrelevant to a greater or lesser extent (c.f. Final Fantasy for complete irrelevance, Chronotrigger for mostly-irrelevant).

Most games also conflate movement and attacking, so you can do both in the same turn. Some require attacking to use up some of your "movement points"; usually attacking also ends your turn.

Of course this makes it harder to make tactical decisions that depend on fine positioning because things can change very quickly -- that balrog that's just out of LOS of your squishy mage could run around and suddenly be in melee range -- but it seems preferable to work on rebalancing that aspect than it does to deal with characters and enemies that can only move one tile, xor attack, on their turn.

[fizzix]

or make positioning irrelevant to a greater or lesser extent (c.f. Final Fantasy for complete irrelevance, Chronotrigger for mostly-irrelevant).

Check out Final Fantasy Tactics. It's not at all the traditional FF game, but one that really played with terrain and positioning in a tactical combat situation. Speed and movement were separated. Speed determined how often your turn came up in the combat list. Movement determined how far you could go each turn. Lastly, making an engine from scratch to mimic the gameplay seems possible.