Sign in to follow this  
Greg Rice

Spacebase DF-9: Voting Discussion Thread

Recommended Posts

Project Lead: JP LeBreton

Build a sprawling flotilla-city in space. Peer into the thoughts of its citizens: see their personalities, prejudices, and relationships play out in unexpected ways. Allow visiting ships to permanently join with your base. Take screenshots and string them together into stories to show your friends.

Share this post


Link to post
Share on other sites

Did you have Startopia on your mind when you came up with the idea? If so, how will this game differ from it?

Share this post


Link to post
Share on other sites
Did you have Startopia on your mind when you came up with the idea? If so, how will this game differ from it?

Hi! Yes, Startopia was great, and it's definitely on the list of inspirations, but I think Spacebase will be more focused on the social simulation, ie the lives of your citizens, than in building of an efficient and powerful base. They're close though.

Also Startopia's soundtrack was fantastic... one of the best of last decade.

Share this post


Link to post
Share on other sites

The contents of the PDF doc on the game's pitch page, reproduced here for convenience. This material is more internal development-focused, but provides some useful particulars.

Project Codename

Spacebase DF-9

Project Lead Name

JP LeBreton

Project Lead Bio

JP is a designer-programmer who worked on the Bioshock games before coming to Double Fine to help Ron design The Cave.

One Sentence Pitch

Build a space station where aliens live and work, and watch interesting stories emerge from their simulated lives.

Team Size for AF

5 (including JP)

Game Summary

As the omniscient overseer tasked with building a single habitation pod into a glorious sprawling flotilla-city, you will be able to peer into every thought and impulse of the citizens who come to inhabit your corridors. Citizens live and work as bartenders, scientists, dockworkers, entertainers, bureaucrats; they have needs, personalities, daily routines, and memories... which means that over time they develop friendships, enmities, prejudices, addictions, phobias, and so on. You'll also have the power to do things like set base policies (which may not necessarily be obeyed), issue orders to maintenance robots, and decide which visiting ships get to merge with your base and become a part of its permanent architecture. Each player's base becomes a unique, unpredictable city teeming with the tragedies and comedies of daily life, with a sense of creative ownership that grows along with it. Players in turn share their bases with friends, and the game expands over time with input from a thriving community.

Design/Development Tenets

Social Simulation-Driven: No pre-authored story or cutscenes. Rather, citizens have personalities, relationships, memories, desires, jobs - all of these drive their behaviors, bringing them into conflict with one another. Drama ensues.

Strong, Low-Fidelity 2D Art Style: We must develop a charming visual style whose every element is thoughtfully designed to help convey what's going on in the simulation. "Information efficiency" is key; SimCity and Animal Crossing teach us that less is usually more. Avoiding lots of fancy animation and highly detailed art has two key benefits: the game can be built far more quickly and cheaply, and players' imaginations can play a more active role in making its events meaningful.

Social Storytelling: Document the daily life of your base with a "photo album" feature where you can annotate screenshots. Share important, hilarious events with the world. People who play builder games have a natural desire to show off their creativity and tell stories - supporting this fully is a huge word-of-mouth marketing boost for our game.

Ongoing Support-Based Development Model: Some TBD definition of a minimum viable product should be released early, with ongoing updates provided by a very small (1-3 people) support team that expand the game based on what players are excited about and what helps further our vision for the game.

Share this post


Link to post
Share on other sites

Oh man im gonna be so sad if this doesnt come in top 4.

This sounds really really awesome!. Gl!.

Share this post


Link to post
Share on other sites
Hi! Yes, Startopia was great, and it's definitely on the list of inspirations, but I think Spacebase will be more focused on the social simulation, ie the lives of your citizens, than in building of an efficient and powerful base. They're close though.

Oh man, if you would introduce Arona Daal into the game there would be so much drama and social interaction :lol:

Also Startopia's soundtrack was fantastic... one of the best of last decade.

I couldn't agree more! I just listened to some of the tracks, so much nostalgia! Damnit, now I need to play it, because there is no space sim as good as Startopia today. Maybe your project can change it? ;-)

Share this post


Link to post
Share on other sites

All I can think off is the weird civilizations the crew of the enterprise would visit, where they had strange laws or all wore a certain type of hat. Is the game inspired by these places and can the player create such a strange world (using these base polices)? Are there going to be strong political themes in the game, will my citizens have a revolution if I end being a tyrannical leader?

Also hats, I'd like to force my citizens to wear a strange hat for absolutely no reason.

Good luck with the game, I hope you get to make a prototype as I'd love to play it.

Share this post


Link to post
Share on other sites

This is an insanely ambitious project, and I really hope you can pull it off!

A few questions though:

1) what perspective will the visuals be displayed to the user? flat top down 2d where you can see the world at different depths, or something more diagram like - side view where you are shifting focus to different "sections" of the station? - can you guess what I think is a better idea ;)

2) how will station expansion be handled? Are we going to be dealing with a pre-fab structure that we fill out, or a single (or few) modules that can be added onto? and if the latter, will each module be a functional building type, or just space to assign meaning to?

3) how much control/influence will we be able to have on the different people aboard our station? Will we be able to say no to different citizens before they join? Eject the unwanted into space? Change their profession to torture them with tasks they hate/fear?

As I said up front, I think this is an insanely ambitious project... but I LOVE it, and want it to succeed!

Share this post


Link to post
Share on other sites

I really really really like the idea. The only thing I don't understand is: what the player does? By the description, it looks very much "sit and watch what happens" (I know it is not like this, I am just saying that I could not figure out what you have in mind). So, could you please explain what are the player's "powers/decisions"?

Share this post


Link to post
Share on other sites

I like this idea. I immediately had to think of Startopia and I love that. So if it is somehow similar, then I will surely enjoy playing it greatly.

Share this post


Link to post
Share on other sites

Remember guys that a lot of prototyping is figuring out just the sort of questions you're asking. Sure, they probably have an idea about how the player will interact with the world, and what they'll spend their time doing, but it's a sort of blurry one that will come more into focus over the prototyping. It's very much broad brush stroke ideas right now, which is a very fun place to be, but there's a reason that only half the games worked on in Amnesia Fortnight tend to end up going forward - sometimes as you continue working on it, it turns out that it just isn't as easy to make the idea work as you'd thought. I know I've experienced that a couple of times during game jams.

Totally looking forward to how this one will turn out, as it looks pretty safe in the top 4 for now :)

Share this post


Link to post
Share on other sites

Simtower in space, i think :) obviously the player will have to build stuff.

It doesnt look good though, those few prototypes that are now in 4-5th place are moving up mighty fast, they are gonna over take this i think, and it will go out of top 4 :( oh noes.

Do no vote for evil competition! lol

Share this post


Link to post
Share on other sites

its great this is currently the most voted, but to me its a very open ended description and it seems like the gameplay will be passive and watching a simulation happen. Obviously I know nothing about this game, but yea I dont really understand the idea yet. Will there be anything to do after setting up the world, besides watching a story unfold somehow.

Share this post


Link to post
Share on other sites

yeah barely, its like been 25-30 votes ahead, its amazing, its gonna be amazing if it stays ahead.

Share this post


Link to post
Share on other sites

Anything with space gets my vote...you know me too well interwebs.

Smiles

Share this post


Link to post
Share on other sites
Anything with space gets my vote...you know me too well interwebs.

Smiles

Indeed, it is difficult to go wrong with space. Some might say...impossible.

Share this post


Link to post
Share on other sites

It's fairily possible to "go wrong with space". People die there instantly.

Speaking of which, will I be fixing oxygen leaks, and stopping my whole base from getting sucked out into oblivion?

Is there like a terraforming angle to this game, where I make a lovely park on the moon? I mean there has to be, right?

Are the bases ON PLANETS, or are they like giant orbiting metal structures?

aaa I guess these are all small questions, the main concept sounds good already.

Share this post


Link to post
Share on other sites

Voted for this, and there is one definitive feature this game must have: The ability to blow any random citizen you want out of an airlock at any time.

It's definitely a classic, well loved feature of all such management games. The reasons why are up the player, presumably something about them complaining to much, or just to see what would happen.

Share this post


Link to post
Share on other sites

This was one of the best pitches, so it immediately earned my vote. So do you think there will be random catastrophes like other sim games that will throw the player's plans in the shredder? There's probably some interesting crazy things that could be thrown at the player with the sci-fi spaceship setting.

Share this post


Link to post
Share on other sites

THIS.

This is a game I really want to play. It sounds terrific and I can already see myself having a lot of fun. We need more awesome simulation games and this one is in space. Not sure how you can top that.

Share this post


Link to post
Share on other sites

I'm more of a linear story guy and sims of any sort are a tough sell, but I've also experienced some memorable user-created stories from just watching and interacting with elements and characters in an environment. This seems like it has the potential to create that sort of compelling environment.

Share this post


Link to post
Share on other sites

I am a sucker for space games, and its nice to see one that isn't purely action oriented. Love this lots!

Share this post


Link to post
Share on other sites

Hey folks! Thanks so much for the support and questions. I've got some answers for some of the bigger ones coming very soon, just give me a few hours (holiday stuff!)

Share this post


Link to post
Share on other sites
Voted for this, and there is one definitive feature this game must have: The ability to blow any random citizen you want out of an airlock at any time.

It's definitely a classic, well loved feature of all such management games. The reasons why are up the player, presumably something about them complaining to much, or just to see what would happen.

Yeah, and you should be able to choose how you answer before people:

- make it look like an accident (there will be people who won't believe),

- set the guy up as a terrorist (people will turn aggressive toward his friends, who maybe play vital role in the station well-being),

- be blunt about it (people will be afraid of you, so they will obey rules more strictly, but then the general anger will rise and there will be more conflicts on board)

I see this as a political simulation, where your power is very indirect - just like politicians': they can set rules trying to direct people, but the choice is always on the lowest level. It's really interesting to try to imagine yourself as a political leader: you can have good intentions, but then what should you do, to direct people into some direction? When if that direction's not so good? When you try to do good and it comes otherwise, people start to hate you. How do you gain their trust again (if it's at all possible)?

...

Yeah, but the game idea sounds really great! ;D

Share this post


Link to post
Share on other sites

I really hope they can find the time and money to flesh out this game, maybe make it a smaller but still full game. :D

People definitely like the idea!, almost cant believe its back in first position again :D

Share this post


Link to post
Share on other sites

Hey folks! Thanks so much for the support and questions. I'll try to answer some of the big ones here. I'm sorry I didn't address any of the forum questions directly, but hopefully this fills in enough detail around them you end up satisfied.

If Spacebase makes the cut and becomes an AF project next week, I don't yet know who will be on my team. Their expertise will inform a lot of the creative decisions, so don't take any of the following as carved in stone. These are just first guesses, directions I wanted to have in place as a project lead, in case nobody had any better ideas.

Scope

Seeing such a positive response has me worried about disappointing people, that the game in peoples' heads when they're voting is as vast as Minecraft or Dwarf Fortress. There are so many things you can't communicate in a 30 second pitch video, and one thing I wish I could've said is, "as much of this as we can make in 2 weeks". I hope we can produce a prototype where something interesting can happen emergently, and you feel like your actions are driving the growth of your base.

In actuality, I think in 2 weeks we could make:

- Base building - adding new rooms in a grid-based layout, and designating what use a room will serve (habitation, industry, research, entertainment, storage, maintenance, etc).

- Basic citizen routines - People go to their jobs, buy food when they get hungry, go to bars to relax, and a few other things that introduce chaos. See "AI programming thoughts" below.

- Basic citizen interactions - People become friends or enemies based on personality traits, do fun stuff together when they're friendly enough or fight if they hate each other enough.

- Citizen memories - Citizens "remember" stuff, ie they keep a log of significant events. You can use the UI to view this, but I don't think it's realistic to have it inform their actions in deep and interesting ways for the AF prototype. This is the tip of an iceberg though.

- New ship docking - Procedurally generate very simple ship layouts, and have them show up and offer to dock. You choose where they attach and then the layout becomes yours to edit much like rooms you've built yourself. The docking and placement UI and code probably share a lot with the normal building functionality.

- Some simple space disaster. Something like fires in the electrical systems, problems with food replicators, maybe an alien parasite that spreads undetected through the base with hilarious results.

- Screenshot annotation feature. HTML export for sequences of annotated screenshots for a very basic "photo album". Maybe there's a smart way to automate getting this online and share-able somehow.

Even this stuff might be way too much... if I get a team, we'll talk about it and figure out what we can pull off in 2 weeks.

Perspective / Style

Spacebase really seems like it needs to be a 2D game. There are clever ways to do low-fidelity 3D - Minecraft, obviously - but almost everything takes more effort to do in 3D.

The question then is 2D perspective: isometric, top-down, side view, something else? And even within, for example, top-down you have different takes: Hotline Miami's oddly flat top-down, Ultima VII's unique 3/4 projection, Zelda: A Link to the Past's "face front" top-down, et cetera.

This is one of the biggest questions and I want to wait and pose it to the team.

Isometric is a bit harder to get up and running code-wise, and graphics tile authoring can be tricky. Potentially the most visually interesting though.

Top-down tile-based is very easy to do, but might be less visually appealing - we want you to look at an impressively intricate base you've built and want to show it off to friends.

Side view I'm less psyched about, just because it seems to compress a simulationally-critical / visual axis and makes characters blend together into crowds, which works fine for management gameplay like in SimTower - games that deal with people in aggregates - but makes it harder to observe and relate to individual citizens.

I'm seriously considering making the AF prototype version only take place on a single layer of depth, even though I'd certainly want to have multi-story bases in a more full-on version. It would make pathfinding, UI, rendering, and content creation somewhat easier... the question is how much depth (literal and otherwise, ha) is lost in this tradeoff. Game development!

As stated above in the "tenets", I want to keep the graphics low-fidelity but expressive and charming... several steps above Dwarf Fortress but not so much that adding a new character type, object, or feature to the game takes an inordinate amount of work.

Gameplay

I want to steer away, somewhat, from the "business sim" mechanics of games like Startopia, Outpost Kaloki X, and the Zoo Tycoon games which I worked on briefly years back. When players have one or a few numbers that measure how successful your base is, they will tend to optimize completely towards that. Compare with the Sims or Dwarf Fortress, where you have a more open-ended definition of success that can include player expression: telling stories, making a huge cool-looking base, doing weird stuff nobody else has done before.

Obviously there still need to be some objective ways to measure progress, and some far-off goals to shoot for, like smelting adamantium in DF.

Rather than base-as-business, I want it to feel like a little functioning society, that you are influencing in "god game"-like ways. You could lock and unlock bulkheads, override the environment regulators, raise taxes, forbid or allow certain behaviors, things like that. And maintenance robots would give you something to influence more directly, but are still part of the unpredictable sim-churn.

I also want to make "success" feel like a very fragile thing - accidents and space disasters happen frequently, the world feels like a messy chaotic place. "Losing is fun", as Dwarf Fortress famously put it. Players shouldn't grow so attached to one base (play session) that they never want to start a new one.

The "new ships perma-dock with your station" mechanic is important to me. Without it I think you're just building out into the void, and you don't get that sense you get in Minecraft or DF, of tunneling into the unknown where interesting things pop out and force you to react. Space is empty, the earth is full of treasure and mystery; Spacebase needs a conduit for the unknown.

Interface-wise, I can imagine the game working on touch devices but the AF version should definitely stick to mouse-driven, though I'll almost certainly hack in keyboard shortcuts for advanced players.

AI programming thoughts

Disclaimer: I've barely done any real AI programming, I've just read enough to be a convincing poser :]

As a general approach, I'd like to start with a simple behavior tree implementation (hardcoded if necessary) with behavior priorities decided by a fuzzy logic system. Citizens will have multiple needs/desires, and some kind of threshold / fuzzy logic system is needed to make things unpredictable but not purely random, and varies according to a personality concept. The "Smart Object" needs-fulfillment-advertising-based system used by the Sims is less than ideal here, as I suspect it would lead to a world where citizens always act rationally and efficiently to fulfill their needs. We want more chaos and strange behavior.

Share this post


Link to post
Share on other sites

I always loved games that only have the players interference as something that dos not need (but can) be more than minor. The creatures games and The Sims come to mind, i like modding the sims 3 in a way that i actualy don´t have to tell them to creat babys or visit frinds. Dwarf Fortess is interesing, but i only played very little of it, i don´t mind old school optics and controls - but DF just took it a step to far (which is some of it´s charm i guess). I´ed love to see a game in that general direction with more user accessibility.

I also would love to work at dubble fine to be able to pitch my superhero game. >_> But i could not even code it if it was picked. :D

Share this post


Link to post
Share on other sites

This idea sounds cool!

I'm really looking forward to seeing your team's approach to the AI logic - I've done a bit of NPC behaviour modelling in the past and have found it really difficult to find the right parameters for the AI personalities, and the right level of simulation - do you use basic physiological parameters such as "hunger", "sleep" and "stress" or will you head into the fuzzier territory of feelings like "love" and "fear"? What tends to trip me up is the constant temptation to model the simulation one level deeper, and end up creating infinite work for myself.

The other thing that can be really difficult is how the AI parameters map to the actual actions performed by the characters. You need to build two levels of simulation - creating and implementing the actions that the characters can perform and the consequences of those actions on the system, on top of which you layer the AI logic.

I really feel like there exists huge potential for games that simulate human interaction, and the more we explore it in our medium the better we'll be able to harness that potential. So thanks for trying to make this happen, and good luck - you've got my vote :)

Share this post


Link to post
Share on other sites
Sign in to follow this