"Butterfly Effect" Sources In The Last Federation

[x4000]

Original: http://christophermpark.blogspot.com/2013/12/butterfly-effect-sources-in-last.html

The Last Federation is a really unique game in that it is a strategy/tactics game set inside a simulation game.  Check out our first sneak preview about it, or our preview discussing simulations in the game.  Also our recent podcast with Space Game Junkie with lots of other details, the description of your role as an independent agent in the solar system, and details on the alien races.  Oh, and the forums for the game are filled with lots of other detailed tidbits, too.

Three posts in one day!  Okay, I wasn't planning on doing this, but madcow asked a really good question, and this also really is the answer to a question that Cyborg asked a week or two ago, anyway.

madcow:

The different political factions and the like sound intriguing, I'm kind of curious if there are coded differences between them. For instance one faction/governor/whatever in insert race always tends to behave one way versus another. Or if the factions are essentially the same but with different random roles/circumstances .

Cyborg:

(Paraphrasing, as I can't find the quote) In what ways is the game randomized each time.  In other words, how does the "butterfly effect" come into play from the simulation?

My response to madcow:

The factions are extremely different in code.  Specifically:

1. They each have various modifiers that make them react to circumstances differently.  Aka, Burlusts always do better on hot planets, and poorly on cold.

2. They each have a variety of specialized race actions that no other race can take (e.g. the Evucks unleashing a virus as a weapon, versus a virus just unfortunately happening).

3. The system of government is entirely different between each race, and has its own completely unique data structures.

4. The political deals between most governments are very much the same about 80-90% of the time, but the circumstances of when and how you can use them is different.  And there are key additions or omissions from certain races.  So you can use the Skylaxians to backdoor in another race into the federation, but for the most part you cannot do that otherwise.

5. There are certain inherent attitude-adjusters that various races have.  For instance, if too many of the "evil" races get into the federation with no other races there to balance them out, the "good" races start having negative sentiment toward the federation and toward you.  And vice-versa.

6. There are various alliance types that can spring up outside of the federation that are race-specific, like the Solar Axis Pact.

7. There are some other special things that can happen, like the federation actually betraying you and becoming a different hostile alliance if you leave it with just the Thoraxians, Burlusts, AND Acutians alone in it as a trifecta for too long.

8. There are various behavioral modifiers on each race, such that only the Andors and Skylaxians have the Honorable flag that makes them go to help other folks, whereas the Boarines and Burlusts are completely anti-trade.  And so on.

9. Bribes have different effectivenesses on various races depending on the type of bribe, and blackmail only works on burlusts.

10. Only some races can have leaders assassinated, and the consequences vary.  Also various other things politically vary in terms of how you navigate each race.

11. There are certain normal racial actions that some races will NEVER take, such as Andors will never capture a planet or turn to piracy, privateering, or raiding.

12. The attitudes of each race toward each other race is randomized at the start, but the range of randomizations and how many they are positive/negative/neutral towards varies by race type.


And... yeah.  On and on.  I know I'm forgetting a bunch of things.  There are also things that are randomized per planet, which have an effect on the circumstances of the race (but the ranges of randomization on the planets make sense for which of the 11 planet types it is).  And there are also various racial things that are randomized per game, such as the starting attitudes already mentioned, and so forth.

Which race is first spacefaring you get to choose at game start, but how close each other race is to becoming spacefaring depends on how close in distance their planet is to the planet of the starting race you chose, and that's random.  Oh, and what sort of action the races will take when they hit equilibrium population is random per game, but weighted based on race type.  As is there percentage of ships that they try to keep on guard, picket, and raiding duties.

Gosh the list goes on and on.  Basically the model is incredibly detailed, we literally have around 200 pages of internal documentation on all this.  But there are also so many DIFFERENT models:

• aliens model
• planets model
• trade model
• randomized ship design model
• fleet tech model
• planetary tech model
• construction model
• economy/order/medical/environment model at planets (our "RCI" bars)
• birth/death rate model (which is per race, but also affected by planet type)
• the spread of racial minorities
• the bouncing around of how attitude adjustments over time go
• pirates
• semi-randomized "events" that happen to planets and space installations (weighted based on the situation at each place)
• semi-randomized racial actions that races take, again based on circumstances but with some randomization thrown in.

And lastly, of course, things change based on your influence as a player.  How you choose to influence the complex situation you are thrown into, and what ramifications that feeds back in to all of the many subsystems mentioned above (and some others I've forgotten about, I'm sure.

How You Go About Dealing With All This
Basically, the above sounds like it could get prohibitively complex, right?  I mean, this is definitely our largest and most complex game ever at this point, design-wise.

Where you as the player come in is that you look at the situation, and you decide what you want to change, and then you see what happens when you do.  Here's how an early game might play out:

1. Okay, so the Evucks stink this game, and the Burlusts are awesome.  The Burlusts are close to being spacefaring, but are not yet.  You can smuggle them that tech and they will like you a lot, but the Evucks (which, let's say, was the first spacefaring race) will like you a lot less.  So you smuggle the Burlusts their tech early, and suddenly they like you a lot more.  The Evucks like you even less -- in fact, they hate you more than anyone else in the solar system does, but that's okay because it is still early going.

2. All right, now let's ignore the other races because they are not yet spacefaring.  Let's instead run some contracts for the Burlusts, killing some pirates and helping them improve infrastructure, etc.  We'll bribe and blackmail away until we get a warlord that really likes us, and we'll use that to get favorable political deals out of them.  Such as preventing them from going on a war of conquest with the Evucks, which would be really dangerous.  We don't want our Burlusts being sent Plumping Tubers or something.

3. Now the Burlusts are doing really swell, and our own little mercenary fleet is getting pretty sizeable and strong under their umbrella.  We are peas in a pod, and thanks to my influence they aren't murdering anyone yet.

4. Now the Peltians become spacefaring, purely by the chance that they were on a really good planet for them, and were third-closest to the Evucks from the start.  Okay, they really stink right now, but I want to get this federation up and going sooner than later, I've decided for this game.  I could play long ball and let things go, but I don't want the Burlusts to get any more hostile and I don't want to have to keep spending BP to pacify them, either.

5. All righty, so time to go visit the Peltians and run a bunch of contracts for them.  Voting proxies in hand, I'm able to really help their economy and get them constructing a ton of buildings and techs, and now they are doing pretty decent.  I can't set up trade with them and the Burlusts, because the Burlusts aren't willing to trade.  But I can set up trade between the Peltians and the Evucks, and that makes them like each other better, even though the Evucks hate me.  Having the a race that is friendly to me also be friendly with the Evucks might be useful, so let's do it.

6. Okay, so now it's time for me to start bargaining personally for hull tech and fleet tech off the Burlusts, and I then start gifting that to the Evucks.  Suddenly they are just as powerful as the Burlusts militarily (in space, anyway), although their fleet is smaller.  They are loving me for all these big goodies that I'm giving them.  Meanwhile the Skylaxians become spacefaring.  I'll deal with them later, right now they mistrust me but won't bother me.  Let's see what they can do on their own, and just make sure war doesn't break out.

7. All right, I finally have enough goodwill with both the Burlusts and the Peltians that I can convince them to form the federation.  They sign the papers, and then boom -- alliance!  The federation has formed.  I'm off to a roaring start; this is the zerg rush of federation formation, heh.

8. Now, however, I start having to deal with Anti-Federation Alliance sentiment.  That's building rapidly on the Evucks homeworld, and in a very minor way with the Skylaxians.  The Thoraxians just became spacefaring, and are looking hungrily at the rest of us.  But they have some teching up to do before they are a true threat.


Now what?  The game goes on from here, we're just getting started even though we did get the federation up and going quickly.  That may or may not turn out to be a good thing in the end.  It did get the Burlusts on our side (and, well, the Peltians for what that is worth), but it's putting us in a worse and worse state with the Evucks.

From here I might:

- Try to get the federation, or even non-federation races, to quickly kill the evucks and take their planet before they blow it up or infect us with some horrible disease or whatever.  That's going to piss off the Skylaxians and the Andors, though.

- Make nice with the Skylaxians, and broker good deals behind the scenes with them and the Evucks.  Then try to get the Skylaxians into the Federation, and then use them to backdoor in the Evucks.  Doesn't matter how much the Evucks hate me.  But if I take the time to do this, the Thoraxians are likely going to sweep a few other planets and suddenly be really a big threat.  Whether or not they'll have time to get up a Protectorate or Fear Empire remains to be seen, but we might wind up with a divided solar system that then has to be repaired.

And so forth.  There are a lot of other options, too, from a grand-strategic perspective.  And how to achieve the grand strategic goals involves a lot of more subtle smaller choices, such as which political deals to take and which contracts to take.  And then inside that, there are tactical decisions in each battle, and the personal fleet composition choices, which play both into what kinds of contracts I can expect to survive, and so forth.

Meanwhile, there are all sorts of other factors banging around in the simulation, as noted above, and the situation very possibly is going to change drastically before I finish getting through with either of the two options above.  At that point, I have to kind of sit back and think of what to do to deal with whatever fresh situations have come up.

That's a portrait of how things would play out in one hypothetical starting scenario!

[nas1m]

W - O - W 
Never played something remotely like this. Can't wait .

[x4000]

It's pretty novel.

[Teal_Blue]

you know, as a stand alone single player game i would guess this rivals many of the strategy titles that Paradox has in its stable. Though if you do get the 'joined single player games' up and running it could be even more complex than SP alone.

I don't know, but thought i would ask. If i have a single game and it has its own set up, and i joined a game with say Madcow, who has a game with his own different setup, then the 'things' that get traded across from his game to mine, or mine to his are things that our own games could not have, or possibly could not have generated by themselves, yes?

In that way the joined games become even more complex and perhaps require a different approach than either my game or his game alone would use.

You know, if you had this 'joined single player games' mode for several, lets say, 2 or 4 or 6 players, then that could lead to all kinds of new levels of complexity and of course, perhaps co-operation (or antagonism) in the other players.
   This sounds, if i am not making some basic mistake in understanding, like it could be quite something to look forward to if you decide to go that route. I for one would suggest add the 'joined' mechanic. Though that is just my opinion. 

-Teal


 

[tigersfan]

you know, as a stand alone single player game i would guess this rivals many of the strategy titles that Paradox has in its stable. Though if you do get the 'joined single player games' up and running it could be even more complex than SP alone.

I don't know, but thought i would ask. If i have a single game and it has its own set up, and i joined a game with say Madcow, who has a game with his own different setup, then the 'things' that get traded across from his game to mine, or mine to his are things that our own games could not have, or possibly could not have generated by themselves, yes?

In that way the joined games become even more complex and perhaps require a different approach than either my game or his game alone would use.

You know, if you had this 'joined single player games' mode for several, lets say, 2 or 4 or 6 players, then that could lead to all kinds of new levels of complexity and of course, perhaps co-operation (or antagonism) in the other players.
   This sounds, if i am not making some basic mistake in understanding, like it could be quite something to look forward to if you decide to go that route. I for one would suggest add the 'joined' mechanic. Though that is just my opinion. 

-Teal
 

I think it could be quite good, if we can figure out a good way to pull it off.

[Cyborg]

Or, have each player running contracts, and see who is in the Federation at the end. One of them could be the leader, one could just be in the Federation, either or none at all. With the possibility of backstabbing. Could be fun.


Hopefully the interface will enable all of the interesting notions above so that each situation is clearly understood by the player.

[x4000]

Right now the focus is definitely just on solo play, but if this takes off then looking at multiplayer is something I'm really interested in.

In terms of the interface, that's something that we're starting to work on increasingly heavily.  We've been beating the heck out of the combat GUI and basically have that where we want it now (I'm super excited about it).  And the simulation design/implementation itself is very very close to done (though we will add more to it later, it is already enormous).  So with those two things out of the way, we're turning increasingly towards visualizing the simulation in a variety of cool and productive ways (as opposed to just having it be spreadsheets of bland data).

There's still plenty to do with the interface outside of combat, but I'm really excited by the concepts that we're working with on several fronts there.  Josh is working with detailed text-based "intelligence" screens, which are kind of computer-y reports on detailed subjects.  Basically like when you read the description of a dwarf in Dwarf Fortress.  That's one way in which we're bringing the data forward, but it's just one dimension and for people who want to have it in one form.

The other way we're bringing the data forward is through a variety of visual interfaces that give you a very artistic and also very visceral/intuitive feeling for what's going on.  That way you can essentially get overviews of things at a glance, and then move into detailed text or charts if you want to.  It's a big task, and we won't be done with it when alpha starts, but the goal is to have all of the most important things in there so that players can make decisions clearly and understand what is going on clearly.

Anyhow, those aspects are still a work in progress, but the simulation itself works really well.  I still am addicted to adding more cool things to the simulation, of course, but frankly there is more than enough there already to be really compelling for a long time, heh.  But there's just no shortage of ideas for new and interesting things to add with it.

[Cyborg]

I'm excited for your whole team. Every update has sounded great. Hopefully, the press will get going once you have a screenshot of the gameplay.

[x4000]

Thank you very much!  The first screenshot of the combat should be ready by the middle of next week, I think.  Which is exciting.

The press has basically checked out for the most part at this time of the year, which is fine; right now we're kind of doing the early fuel-stoking, and once we hit January we'll start doing a bunch of stuff with more screenshots as well as videos.

Incidentally, we have released one partial screenshot from the game, all the way back in our second post about the game (the preview of the simulations in the game).  This is it:



That image is a cropped screenshot with the logo overlaid, but that's it.  Beyond that, it's just an actual semi-zoomed-out view of the solar system map.

There have been bits and pieces of other art from the game in every image we've shown, too -- all of that is in-game art, not concept art, even the still scenes.  But there aren't any pieces that were literal screenshots beyond the above one.

As we're getting more into January, we'll be able to really start fixing that rapidly, I think!

[DrFranknfurter]

Hey, I've posted before and was impressed by the detail and passion of the responses and the vision this game has. Question about the butterfly effect:

Have you actually tried to see how one event changes the simulated world, by saving a game at a set point say just before making an alliance and then running ahead say 20 years in-game to see how drastically different the two paths are? (and generally any more good stories from that sort of testing?)

(I'm just curious if major choatic divergence points occur all over the place or if a few things are overwhelmingly important. Does curing a disease really make a huge difference (preventing a race from dying for example) or are the alliances and technology trades the biggest factors in determining the balance of power?)

[x4000]

We haven't explicitly tested that, but based on the design I can tell you those make big differences. I mean, if the acutians are weak and they get a computer virus, then they are almost definitely going to die if someone wants to attack them. If you cure the virus, they may or may not be able to save themselves depending on events. If they get a virus while strong, that can just decimate them. But curing them might leave them strong still.

So it kind of depends on which levers you choose. At any rate, this is not chaotically random in the sense of, say, the woes in skyward collapse. Those were much closer to pure random, whereas here the weighting is exteey high. You are unlikely to have a volcano erupt on a planet if its environment score is good, etc.

[Mick]

Please excuse me if this has been mentioned elsewhere, but does the player have an ultimate goal in this game, or is it more like a sandbox?

[x4000]

You're trying to form a solar federation spanning all of the planets -- with whatever races remain on them (and what planets remain, in some cases).

I think that, for the first few times a player is in the game, it's best to mostly treat it as a sandbox, though.  I mean, to fly around and visit people and experiment, and work vaguely towards the goal of a federation, but not get too hung up on succeeding.  The journey is fun whether or not you win, and this is one of those games that lends itself really well to crazy Boatmurdered-style LPs.

So in that sense you can treat it like a sandbox and still have a grand old time.  But eventually if you just keep sandboxing it up, it will kill you.  Around about 6 hours into the game, new racial actions and buildings start unlocking for the various races, and they start upping the ante.  You can still play indefinitely if you're really good, probably, but that's kind of like the players in AI War who take 120 hours of gametime for one game and try to capture all 100 out of 100 planets.  It can be done, but it's not typical.

[Mick]

What kinda game length are you expecting for an average playthrough?

[x4000]

Typically in the 5-6 hour range, for the most part, though I suppose we'll see.  Something that is more short and condensed, without any real dead space.

And for those who like something longer, the ability to reload any existing savegame as a new starting solar system will be there, too.  So if you wind up creating an awesome and hilarious history of Boatmurdered, and then you are sad when it ends, you can reload that as a new solar system set 1000 years into the future, and it's the Return To Boatmurdered.  Some stuff gets wiped out and reset so that you are back and starting over anew as a new mercenary, but for the most part it is the same as it already was in the prior game.