next »

[starsrift]

Procedurally storytelling is something very hard to do. I've been working on the problem, but I haven't come up with a really good solution. I thought I came very close last year, and I might even finish the project sometime this year and publish it/my findings, but the big problems are the storytelling essentials - foreshadowing, and avoiding the 'samey-ness' of procedural generation or alternatively, the result of a 'static' world that doesn't really respond to the actor(player).

The permutation idea of starting a thread and then randomizing for possible permutations that lead into a short story or chapter is a really good idea, and a not too awful attempt at a solution to the problem. Runs right into those pitfalls of foreshadowing and sameyness, of course, but it's quite elegant and organic, and helps to engage the actor by dynamic creation.

/ I keep thinking Eldrith Horror is Elder Sign because so many of the game terms are the same

[s0]

welp, elder sign is like the little cousin to eldritch horror and arkham horror

[gimymblert]

@starsrift
I think I solve that part using the concept of "stakes", most story generator focus on "plot" or "structure" but they fail to account for the inherent production system that allow for affordance. Having a string of stakes allow you foreshadow and direct scene because you have clear boundaries and a measure of what is important. Good stakes limit the propositions and affordances in a way that make sense on a scene basis while keeping dynamism within that scene, it's also a good framing device to help events have meanings.

[starsrift]

@starsrift
I think I solve that part using the concept of "stakes", most story generator focus on "plot" or "structure" but they fail to account for the inherent production system that allow for affordance. Having a string of stakes allow you foreshadow and direct scene because you have clear boundaries and a measure of what is important. Good stakes limit the propositions and affordances in a way that make sense on a scene basis while keeping dynamism within that scene, it's also a good framing device to help events have meanings.

Foreshadowing is an innate problem. As a writer, foreshadowing is very important, as it gives a lead-in into the direction of the story, and if you're halfway competent as a writer, it provides a way for the reader to feel accomplished by achieving the plot climax.
As a game designer, "foreshadowing" is easy - but the result is that the player can tell by the "foreshadowing" what the end solution is, especially in a 'repetitive' video game. You can modify this by widening the 'range' of foreshadowing to a number of results, but the end equation is remarkably similar. This again, brushes against issues of 'samey-ness'.

Theory is easy. The proof is in the pudding... And the pudding is remarkably hard. I don't think the idea of a procedural, quasi-unique story generation game is impossible, just hard. I'll keep on working on a solution, myself. Maybe someone else will get there before I do - congrats to them, and I'll buy the shit out of that game.

[gimymblert]

I would just invoke contrastive juxtaposition as a way to foreshadows in mass effects 2, ie it has foreshadows for all potential consequences of a given scene before a big choice. Works well Imho and can be turn into a system (system of emotion and narration).

I would sameyness isn't much a concern when it's only the structure, series, genre, have relied on predictable structure where the result are stochastic dfor a very long time.

The key with stakes is that:
1. it create immediate tension and sense of consequence but also sense of narrative momentum
2. by laying down the need and the distance to solutions/failures the build up is organic, the suspense is automatic.
3. It give relative meaning to all events as in how the momentum is affected
4. it frame the situation in a failure/solution boundaries that also act as transition toward the next narrative point. It also frame th eplayer action such as there is no "escape" that will break the situation if well done.
5. it solve the problem of the undefined character player by being literally an exposition device toward the narrative system by quizzing the player to a specific subject evoked within the stake situation. The game will remember that.

Most problem arise because the emotional/moral/psychological dimensions aren't design as stakes affordances in game design (for example character can't do thing because he hasn't courage enough, building courage is through events, ie what a character can or not do is both internal exposition and gameplay affordances)

Now there is two kind of stakes, stakes about choices or about challenge. Challenges as only one good solution at the end of the progression. Choices is like challenge but instead of having clear win/loss consequence they are mixed and going toward one resolution is a choice though there is resistance toward all conclusions, and progression can reveal elements that tip the direction.

Also works well in paper test with people so ...

[starsrift]

I would just invoke contrastive juxtaposition as a way to foreshadows in mass effects 2, ie it has foreshadows for all potential consequences of a given scene before a big choice. Works well Imho and can be turn into a system (system of emotion and narration).

Exactly part of the problem. If all roads lead to Rome, to quote the proverb, is the choice meaningful? Of course not.

[gimymblert]

how would you define meaningful? what do you expect out of procedural story generation?

[starsrift]

Somewhat more personally meaningful than the color of finale fireworks.

Edit: Or to expand upon the previous proverb, maybe some roads leads to Sparta, or Athens, or Alexandria, or Carthage.

[gimymblert]

Well I think you misunderstood my example lol. Mass effect did great thing but also bad thing, look at the whole genophage subplot and the finale, it's way more ambitious than color coded ending. It's about whether or not there is a legitimacy to the genophage virus, and even though there is clear consequence (death of mordin, hope or not for the krogan, peril of the galaxy) the moral dilemma is still unresolve in the "absolute" because the real consequence are outside the scope of the games, which allow the player to both make and own its interpretation on the broad stake yet feel the consequence of the smaller one.

The mission in tuchanka trying to save the last fertile female krogan is a great example of that, the stake is set at the beginning, during the mission you are shown both facts about the finesse of the past krogan's civilization and their downfall, essentially asking which one define the krogan, the culturally fertile or the brutal warmongering civilization? And the female krogan is teased as being able to have strong leadership toward the resurrection of culturally rich krogan, but it's a bet on the future, right now they MIGHT be a menace to the galaxy. What do you decide, you condemn them or not?

Now mass effect is scripted, it's narrow by definition, a pgc story game should be able to circumvent this by effectively analyzing and creating relevant stake on the spot.

For a simple non emotional example: imagine you are a serial killer, you are in the street and across you is a witness shopping unknowingly, he will be interrogated in 3 days, unless ... The stake is clear and the pressure too, if you decide to go to Carthage there will be clear consequence in 3 days and the stake will change organically, if you decide to kill him on the spot you create more witness, if you kill him stealthily he will not testimony and the next stake will be something else.

The distance between him and you is the progression metric, there is a clear direction away from him or close to her as she is the key to the tension, you don't escape just because you are away.

Let's say you decide to chase him, the closer you get the more tension there is, but as you close the system introduce cops that are looking for you, they hinder your progression but also the closer to them you are, the closer you might be from negative consequence. When the witness chit chat, you close to him easily, when you hurt a little boy by being too hasty attention are on you and cop can get closer, all actions, even the mundane one are clearly contributing to the whole scene as hinder or accelerator toward conclusion.

The system work because there is a lag between exposition and choices, maybe you get data that tells you should not kill the witness or maybe you can choose to romance him (you are a woman serial killer). The delay allow you to absorb information and make up yor mind and experiment and ponder the consequence.

Most PGC story fails because there is a lot of histake hi frequency actions with far reaching consequence, it turn the story into chaos.

Stake frame story with hi frequency low reaching action within a scene, and low frequency hi reach action within a stake. Stake are place agnostics, a scene can be diffused, it's mostly a presentation and focus mechanics that arrange affordance within a framework so that there is not as many stuff happening that haven't been decide.

[InfiniteStateMachine]

So much Duelyst. I think I already burned myself out on the game after 2 days.

[ecplav]

Shin Megami Tensei IV, on the 3DS.

I really like it, I feel like the rhythm is better than most of the JRPGs I played. Plus, the fusion system is really really cool.

The only thing that bothers me a little bit is the lack of cohesion concerning the way the demons' portraits are drawn. They had several illustrators for that, the problem is that their styles are really different from one another and it's really noticeable at some points.

[s0]

smt4 is really good, yes.

[Türbo Bröther]

Arcane Dimensions got another patch that along with various fixes adds two more maps one of which looks slightly familiar...

[Schoq]

I tried playing it two weeks ago(?) and it looked cool but then it crashed when I shot an enemy

[Türbo Bröther]

Gotta git (a) gud (PC).

Wait, come back here. Maybe they fixed whatever caused that with the latest patch or maybe increase the heapsize? I remember having to set it to 256000 for the last really big mod release just to be safe so it should apply to this too.

[s0]

Magic The Gathering: Puzzle Quest

Some people recommended this to me as "surprisingly good" and because it's F2P I thought why not. It's definitely not BAD or anything, but it has some problems. It plays like a sort of MTG-lite where you generate mana through a match 3 game rather than by playing lands. Either that or a Puzzle Quest with deckbuilding where instead of damaging your opponent by matching tiles, you match tiles to summon creatures to automatically attack your opponent.

The game's main issue is its awful interface and the main issue with that is the bizarre hand management. See, your hand in this game is actually a queue. Whatever mana you gain is automatically assigned to the top card in the queue and when the that card has enough mana, it's automatically played (no colored mana costs in this). Excess mana is assigned to the next card. All of this seems ok until you realize you can freely move cards around the queue and even "deactivate" them to keep them from being played when they reach full mana. So mechanically it's just standard card game hand management, except presented in a really convoluted and awkward way for no reason. Why they didn't just give you a mana pool and let you pick cards from your hand to play manually is anyone's guess. Add to that the fact the interface is unresponsive as heck, at least on my iPhone. I realize this seems like a nitpick, but it really makes the game about 10 times more tedious to play than it needs to be.

Outside of that it's decent fun tho and even has some neat ideas, such as Support cards (basically enchantments in MTG terms) being attached to random gems on the board once you play them. They can be destroyed by matching the assigned gem. I don't know how bad the F2P grind is.

EDIT: I'd say the game in general feels a little convoluted, like what with having to play a match 3 to summon creatures and stuff. Granted, it DOES have somewhat more strategic depth than other Puzzle Quest games, but idk if it was really necessary to add that many mechanics considering the amount of depth added. A lot of them were seemingly tacked on to make it feel more "MTG-like".

Since there's no real colored mana (you just get a bonus for matching some colors depending on which character you play as), the tactical layer (the match 3) and the strategic layer (the card game stuff) only feel very thinly connected. There don't seem to be many decisions you can make in the Match 3 that actually have a substantial influence on your strategy, beyond just pulling off big combos to get more mana. Imo, Puzzle & Dragons pulls this sort of idea off a million times better.

[Cobralad]

played some Legends Of Eisenwald.
Initially i was not really exited about it because i thought its some historic 4x for nerds but then i heard that its a spiritual successor to Disciples 2 which is a game for the coolest of people.
Its good, but its also very eastern european. Interface has some serious issues and english translation can be gibberish, although russian writing is not really coherent either.
Game boasts itself as "low key" fantasy, but its basically soviet school perception of medieval times, which is basically Ivanhoe but with witch burning. Its not helped by stylistic errors and authors not understanding differences between orthodox and catholic Christianity. That kinda reminded me of Disciples 3: original was developed by french people and it was cool animeish rennaissance with some pretty authenthic looking stuff, while 3 was developed by russian studio that put 11th century crusaders hunting katana elves with wh40k style inquisition.
That said it does some amazing stuff with quests and storytelling: missons rarely give you markers and one of the key features is reading stories that hint at quests and places. One quest involves you using tower shadow as a clock which is not something you can see in a modern game.

[Tanner]

Magic The Gathering: Puzzle Quest

Some people recommended this to me as "surprisingly good" and because it's F2P I thought why not. It's definitely not BAD or anything, but it has some problems. It plays like a sort of MTG-lite where you generate mana through a match 3 game rather than by playing lands. Either that or a Puzzle Quest with deckbuilding where instead of damaging your opponent by matching tiles, you match tiles to summon creatures to automatically attack your opponent.

The game's main issue is its awful interface and the main issue with that is the bizarre hand management. See, your hand in this game is actually a queue. Whatever mana you gain is automatically assigned to the top card in the queue and when the that card has enough mana, it's automatically played (no colored mana costs in this). Excess mana is assigned to the next card. All of this seems ok until you realize you can freely move cards around the queue and even "deactivate" them to keep them from being played when they reach full mana. So mechanically it's just standard card game hand management, except presented in a really convoluted and awkward way for no reason. Why they didn't just give you a mana pool and let you pick cards from your hand to play manually is anyone's guess. Add to that the fact the interface is unresponsive as heck, at least on my iPhone. I realize this seems like a nitpick, but it really makes the game about 10 times more tedious to play than it needs to be.

Outside of that it's decent fun tho and even has some neat ideas, such as Support cards (basically enchantments in MTG terms) being attached to random gems on the board once you play them. They can be destroyed by matching the assigned gem. I don't know how bad the F2P grind is.

EDIT: I'd say the game in general feels a little convoluted, like what with having to play a match 3 to summon creatures and stuff. Granted, it DOES have somewhat more strategic depth than other Puzzle Quest games, but idk if it was really necessary to add that many mechanics considering the amount of depth added. A lot of them were seemingly tacked on to make it feel more "MTG-like".

Since there's no real colored mana (you just get a bonus for matching some colors depending on which character you play as), the tactical layer (the match 3) and the strategic layer (the card game stuff) only feel very thinly connected. There don't seem to be many decisions you can make in the Match 3 that actually have a substantial influence on your strategy, beyond just pulling off big combos to get more mana. Imo, Puzzle & Dragons pulls this sort of idea off a million times better.

I reviewed this for PT and gave it 4 stars. I fell out of it shortly after the review (mostly because traveling screwed up the daily rewards thing and I lost my streak haha) but I agree with pretty much all of this. It made a really good first impression, which I think led me to score it a little higher than it should have been. I might check it out again now that it's been updated. Definitely the best PQ spinoff, though.

[s0]

ohhh, so you wrote that review? that's actually what made me check it out lol. didn't know you wrote for pocket tactics.

[Tanner]

Yup, started at the end of last year. Still can't believe it, tbh. Unfortunately, the whole Owen going MIA and the Slitherine buyout have tinged the atmosphere there a little, but there's some pretty exciting stuff that should be happening soon.