For the Love of God, Choices!

[Aaron]

One thing I'd like to address is how "optimal" a choice or set of choices are.

In the above image the conflict begins at "start", and this has set an inevitable series of events in motion. The three dots below "event" denote that you may influence it in three different ways. Maybe you get there early and address the problem before it gets out of control. Maybe you arrive with not enough civilians to influence the event, and you miss a chance to get the best outcome. Maybe you show up so late that it's already happened and left things in the worst possible state. These three options are just examples, each key event can have any number of different states.

The talk of best and worst outcomes suggests that there are right/optimal and wrong/suboptimal actions and choices. I tend to be suspicious of such weighted choices and branching because it suggests there's a "correct" set of choices. These correct choices and the best overall outcome may involve actions that are more difficult, but then the choices are less about role playing and more about getting the equivalent of 100% completion.

[Ratatoskr]

My hope for the optimal choices comment is simply that he meant it the way I tend to think of it, which is how can you keep the most of your people alive, which is always going to be a trade off. Since this is a post-apocalyptic scenario I figure that's going to be the main goal, and so every choice is going to be a balance of who dies and who lives.

But I do agree that it would be nice if there were a few different ways to get to a "most of your people survive" scenario, rather than having one perfect path and a whole bunch of less than perfect ones. I actually think having a way to save everyone would be rather unrealistic.

[Torgamous]

The talk of best and worst outcomes suggests that there are right/optimal and wrong/suboptimal actions and choices. I tend to be suspicious of such weighted choices and branching because it suggests there's a "correct" set of choices. These correct choices and the best overall outcome may involve actions that are more difficult, but then the choices are less about role playing and more about getting the equivalent of 100% completion.

And then you get to the end and discover that keeping lots of people alive made your caravan big and noticeable, making it easier for Black Armor Dudes to find and slaughter everyone.

While that, specifically, would be a really dickish move on the part of Stoic, I'm pretty sure they've said somewhere that they don't want to make it immediately apparent what the best choices are. I doubt it'll be as simple as the "OOH OOH DO THIS, or you technically could do these things too, but they suck" that pops up from time to time.

[Alex]

To mention it, we are intentionally trying to make it so that there is no "best" path where you save everyone. But more importantly, like real life you don't necessarily know what the best path would be anyway. You may save half a town and wonder if you could have done better. If a town drags their heels and takes days to pack up and leave you may not know how that is going to affect the next event. It definitely goes beyond "saving people", however, into the realm of ethical dilemmas, revenge, no-win scenarios and everything else that comes with the end of the world. We're in love with the idea that you accept the result of your actions and keep going.

[knivesalot]

The important thing to mention is that your influence on event 2 can change the options available to you for event 3.

This system seems unique to me not just because time is a factor, but because an entire civilization is at stake and influences the story, not just one character. A choice you make may determine how many people survive an event, which in turn influences the next event that is contingent on how many people are traveling with you. As the player, you'll have to decide whether it's acceptable to let people starve, or be left behind, or to sacrifice a few for the many. We don't tell you what the good or bad choice is, it's just a matter of what you feel is right.

The last thing to mention is that this does mean the finale is largely in your hands. I can't get into too much detail now but as you can extrapolate from what we've talked about here, your options during the finale will be a result of your combined decisions, not one of three inevitable choices.

...

To mention it, we are intentionally trying to make it so that there is no "best" path where you save everyone. But more importantly, like real life you don't necessarily know what the best path would be anyway. You may save half a town and wonder if you could have done better. If a town drags their heels and takes days to pack up and leave you may not know how that is going to affect the next event. It definitely goes beyond "saving people", however, into the realm of ethical dilemmas, revenge, no-win scenarios and everything else that comes with the end of the world. We're in love with the idea that you accept the result of your actions and keep going.

YES CHOICES!

It's all about the aggregated tree structure. In Alex's previous post it's clear that the model posted is a tree structure (branching choices that go off in different directions) and contains aggregated parts (the culmination of those choices into a single result). I am almost completely satisfied thank you Stoic!

The only part of the model purposed I'd be cautious about is the finale structure. In your model both the red and blue story lines end up with the same result . . .



This is not an aggregated choice but one where your choices end up at the same ending point. For it to be an aggregated choice then the blue line would drift to its own finale point (the one below it) rather than the red line. . .



This is important because if it's not aggregated and the end results are the same then the choices made previously don't matter! However if they are aggregated (arriving at the same point but resulting in differences based on previous choices) then they do matter! Thus, the significance is clear, aggregated choices are important because it allows the previous choices to MATTER and (for example) will cause the blue and red story lines to be important rather than 2 different trails to the same place. But I assume that this is what is meant.

Furthermore, if the game turns out like I assume Alex meant, then I think all my questions/ concerns are answered. If it's done well I think we'll have a game that people will want to play and remember!

[Aaron]

But more importantly, like real life you don't necessarily know what the best path would be anyway.

You might though when you replay the game. Or look up a guide on GameFAQs.

[JokerAR]

That depends upon the game's logic when handling your decisions: deciding to fight, for example, may lead to a tree branch- or it may lead to a behind the scenes dice roll with a variable loss of followers, soldiers, food etc. This way decisions could have unpredictable outcomes alongside standard branch points.

[Torgamous]

This is important because if it's not aggregated and the end results are the same then the choices made previously don't matter! However if they are aggregated (arriving at the same point but resulting in differences based on previous choices) then they do matter! Thus, the significance is clear, aggregated choices are important because it allows the previous choices to MATTER and (for example) will cause the blue and red story lines to be important rather than 2 different trails to the same place. But I assume that this is what is meant.

The choices made previously in red and blue led to the same situation before the finale. That's not the game making your choices not matter, that's your choices logically negating each other. If you save more people in one area but less in another, and move faster in one area but slower in another, then in the end you can have saved the same number of people in the same amount of time, and that won't mean your choices don't matter.

[Alex]

I get what Knivesalot and Torgamous are both saying. What I wanted to illustrate in the example is that you can make different decisions along the way, but you can potentially make the same decision if the variables align that way.

What I did not illustrate is that the red path may have different options than the blue path at the finale, but if the player chooses they could end the same way. Imagine that instead of three options there were 10 (for example), and the red path locked out the top three and the blue path locked out the bottom three. There is still some overlap. Not the easiest concept to illustrate, I'll admit, and not the easiest thing to implement, either, but that is definitely our goal.

[belamoor]

To mention it, we are intentionally trying to make it so that there is no "best" path where you save everyone. But more importantly, like real life you don't necessarily know what the best path would be anyway. You may save half a town and wonder if you could have done better. If a town drags their heels and takes days to pack up and leave you may not know how that is going to affect the next event. It definitely goes beyond "saving people", however, into the realm of ethical dilemmas, revenge, no-win scenarios and everything else that comes with the end of the world. We're in love with the idea that you accept the result of your actions and keep going.

The way Alex talks about choices is very promising to me. Correct me if i am wrong, do players making different choices on one event can still sort of "make up" for it on another ? That would be awesome, in a way that it may allow to choose what you are able to sacrifice, and what you wish not to, but having a possibility to arrive in largely the same result, had you taken another route. That is of-course if players will have an idea of what they are sacrificing, atleast in a short term.
The other way would be stumbling blindly in a harsh world, trying to survive and make a right call. Sounds like a definition of mature storytelling.

[lamaz]

Yes, yes and yes, this sounds excellent. I love it when you have to make a decision - the tougher the better - and then stick with it. There's always a situation that I would like to achieve, but I never save and load to make it happen. It's great to stick with the decisions you have made and see how it all comes back at you later.

I also like the fact that you don't necessarily know what you want to decide. I mean you know it's not good if people die, but you have no idea if saving them will cause a lot more suffering on some other occasion. Maybe you left someone behind earlier because he was lazy or stole from people, and it turns out he could have helped you a lot later, or better yet, you have no way of knowing.

[Flickerdart]

I played a game (Yawhg) at the Toronto Comic Arts festival that was very similar to this structure in concept. You have 6 turns to go to places, where you will gain certain stat-ups and a random event where your decision influences other stuff that you might get. After the end, there are two more turns where you can use your stats to help the survivors survive by doing what you've trained to do. It was quite fun, and really replayable, because there were many choices to be made. If a Banner Saga episode is 10 hours long with a structure like this, then it will have absolutely insane replayability, just to see all the choices play out - even the couple minute long Yawhg, I'd wager, could be replayed at least five times and you'd still find something new every time.

[Alex]

Oh man, I am truly jealous. I'm a huuuuge fan of Emily Carrol (her art hangs out on my desktop roughly half the time) and the concept for Yawhg looks freaking cool. Impatiently awaiting the chance to try it out!

[LoneGunman]

In the end, I suspect if you make each choice have the impact you mentioned but in effect offers different sorts of benefits such that you don't necessarily win or lose with each decision but you simply affect the finale, it'll work just fine. For example, taking the town example, you could argue that waiting means you lose time but gain more people that can be used to gather/hunt later or even be used as a potential solution for a future choice (i.e. the sacrifice the few for the many) costs you time but gives you those people. Meanwhile, just leaving town and leaving them behind, costs you the extra people (and potential help with future choices) but gives you more time for a future choice where time could be the currency (i.e. you'd have time later to wait for people from another town that happen to have more warriors or you have the time to follow a sub-quest that you wouldn't have had before). In effect, instead of having set events based on game time, they might be based on another progress metric while allowing time to be a currency along with people, weapons, food, etc.

I remember loving Castles (old game) because of the random chained events that tied to each other. You'd get this one event talking about some noble visiting you and offering you some choices (send him away, give him gold, let him stay in the castle). After a little time passed, you'd get another event that worked off your decision from the previous choice (e.g. you let him stay, now he demands you help him with something). Depending on the events and the choices you made, it could end early or continue on for a while. But in the end, you got something for it, always. Either you got gold, reputation, resources, etc. So, in effect, you could make a "bad" decision but it wasn't an all or nothing decision. You'd still get something (or maybe you wouldn't but it would be retained in the game state and might come back to haunt you later) regardless of the decision.

The great thing about it is it doesn't prematurely end the game (unless you just screwed up the other aspects of the game) but really added to the atmosphere of the game (you felt like a ruler having to make decisions other than moving sliders and managing resources). Plus, I feel like it makes the development a bit easier because you don't necessarily have to plan out every event, just major ones like the ones you illustrated. Sure, there might be some balance issues and it would be nice if you can factor in the type of previous event so that you don't constantly get similar types of events delivering the same rewards every time.

My two cents.

lg

[knivesalot]

I get what Knivesalot and Torgamous are both saying. What I wanted to illustrate in the example is that you can make different decisions along the way, but you can potentially make the same decision if the variables align that way.

What I did not illustrate is that the red path may have different options than the blue path at the finale, but if the player chooses they could end the same way. Imagine that instead of three options there were 10 (for example), and the red path locked out the top three and the blue path locked out the bottom three. There is still some overlap. Not the easiest concept to illustrate, I'll admit, and not the easiest thing to implement, either, but that is definitely our goal.

Bam, that's exactly what I thought you meant Alex . I was just making sure because the assumption I was making was that each story choice wasn't mutually exclusive (if you went down the red path for awhile you could still jump over to the blue path). So when I saw the red and blue paths =ing the same result, I just had to be sure .

But just being thorough, I would like to make clear the importance of actually having mutually exclusive choices (choice aggregation).

For example, if choices are non-mutually exclusive and allow for the character to jump between paths then this makes choices NOT MATTER in favor of more story-line flexibility.



But if choices are mutually exclusive (choice aggregation) then choices DO MATTER but the story-line is less flexible.



This is important because most games fall into a story-line pattern where their choices are COMPLETELY independent and then we end up with a braid story like rather than a tree. Thus, my point is that, while allowing characters flexibility is all well and good, without mutually exclusive choices (choice aggregation) then you just have a braid story line because the choices are independent (even if it is made relatively complex).]

I get what Knivesalot and Torgamous are both saying. What I wanted to illustrate in the example is that you can make different decisions along the way, but you can potentially make the same decision if the variables align that way.

What I did not illustrate is that the red path may have different options than the blue path at the finale, but if the player chooses they could end the same way. Imagine that instead of three options there were 10 (for example), and the red path locked out the top three and the blue path locked out the bottom three. There is still some overlap. Not the easiest concept to illustrate, I'll admit, and not the easiest thing to implement, either, but that is definitely our goal.

But in response to this, I think Stoic already has a handle on what's going on with a semi-mutually exclusive story-line. Genius compromise (choices matter and they allow more flexibility)!

So anyways, I am now super excited!

[Mudfly]

I really like what was written in the yellow posts! Thank you guys at stoic for being awesome!

Just a quick question though... What if i make a lot of bad decisions and i end up in a spot where i can’t beat the game? Will i be forced to restart the game entirely (which i guess is what i would like) or will there be some sort of reverse function to move the story back enough to allow me to correct my mistakes?

[lamaz]

I really like what was written in the yellow posts! Thank you guys at stoic for being awesome!

Just a quick question though... What if i make a lot of bad decisions and i end up in a spot where i can’t beat the game? Will i be forced to restart the game entirely (which i guess is what i would like) or will there be some sort of reverse function to move the story back enough to allow me to correct my mistakes?

That's a good question. It would be cool if you had to start from scratch, but it could get really tedious if you failed more than once or twice. That would almost surely mean that the game wouldn't do very well financially. Starting from a save file could help or it could be that you had saved too late and then you wouldn't be able to change the fact that you are going to lose at some point.

I'd like to see some solution where you could still lose the game, but it wouldn't feel that much of a failure. For example you got to see some cool ending scene, which would change based on how you lost and what it meant for the rest of the world. That way you could lose in the game, but you'd get to see something that made it worthwhile for you. I don't know if that would be enough to keep players interested though, it depends so much on the execution. Interesting to see how they will handle this.

[Mudfly]

I'd like to see some solution where you could still lose the game, but it wouldn't feel that much of a failure. For example you got to see some cool ending scene, which would change based on how you lost and what it meant for the rest of the world. That way you could lose in the game, but you'd get to see something that made it worthwhile for you. I don't know if that would be enough to keep players interested though, it depends so much on the execution. Interesting to see how they will handle this.

Great idea! that would be really cool! It's almost how Sid Meier’s games usually end... If you've played really well you get a nicer ending than if you played poorly...
(and now i really fell like playing pirates!)

[JokerAR]

If that's the case I'll almost certainly never see the Banner Saga's ending. It'll be Oregon Trail all over again, my wagon's axle will break and my army will all die of dysentery. (I blame everyone else for my poor future strategic decisions)

[Torgamous]

If that's the case I'll almost certainly never see the Banner Saga's ending. It'll be Oregon Trail all over again, my wagon's axle will break and my army will all die of dysentery. (I blame everyone else for my poor future strategic decisions)

It's your fault for packing lots of bullets and nothing else. Why did you even bring any bullets at all? Vikings don't have guns. That was a horrible choice.