Art Style Guide for Pioneer - brainstorming

[bszlrd]

Luomu asked me if I'm still interested in creating an Art bible for Pioneer. Of course I am. Thank you again for asking me.

I think first there should be a proper brainstorming and discussion about this, to provide a good foundation. Especially among the developers who are propelling the game forward and already have a vision about this great game.
My main question is: What are my limits? How much can I tamper with gameplay and world building aspects for example? Only visuals, or I can dip my toes in the world building/writing aspect of it? Backstory and such. Or maybe some gameplay, like adjusting overall ship power, or mission type requests and such for example. Anyway the following questions are valid in any case, even if I'm allowed to working on strictly the visuals, because they will help me a lot establishing a coherent style.

What is the main direction you want Pioneer to head? Realism vs abstraction ratio for example. How current artwork overlaps with the vision? (I have no aversions about throwing out every ship, including what I've already designed, if we come to a conclusion that these aren't fit properly, but only if they really need to go. There are some very nice ships in the game already.)
So this way I will know what look to aim for and where can I stretch my legs (even feature request-vise if the guide reaches a that developed stage).

I gathered that one aim is to provide a an engine for others to build on. If that's correct, then how detailed the artwork should be? One or two ships for each role, or a more extended and lively ship stock for example.


My observations/opinions/questions about the game right now:
-It's vast. Also a bit desolate right now. I think it needs quite more diversity.
-I'd tone down overal ship performance to further pronounce the vastness, and emphasize more in-system play. Multiple factions/nations in the same system and such, if you think it worth the effort to put work that into the game. System exploration missions would be a nice thing to do for example.
-I think overall ship realism is on the spot mostly. Maybe I would tone it towards the realistic side of the spectrum but not much.
-How valid the history/world guide page on the wiki? I think it's a good base, but maybe a more pronounced Cold War state would be interesting for these freelancer-like occupation of the player.
-How advanced the technology is? How much is the handwavium ratio? Where should that ratio be? It seems to be good in this regard for me, like I'd say there isn't any artificial gravity in the game right now and that should stay that way. In most times a handwaved performance increase for ships would be good, to allow more then 10-20km/s deltaV (more realistic) budgets to a level of 5-10000km/s. But fuel/propellant should be more expensive in my opinion. Apart from this and the hyperdrive, I think the game should be rooted in reality, but in a way to suggest a civilization with quite a few centuries of space travel. Maybe I would put the time-frame back from 3200 to 2300-2400.
-I would separate atmospheric capable/only space capable ships a bit more.
-I would balance ships in a way to allow for quite a lot of styles of play, but also to feel organic. There should be crappy ships no one likes, there should be crappy but very likable ships (Trabant car is a good example). There shouldn't be a clear line from the worst ship to the best, like the way Freelancer had it. Not that it's the case with pioneer right now, I think it's quite organic.

Also: what visual styles do you like? I'd say, we shouldn't aim for a too photo-realistic look. A hint of coherent stylization can help a lot. (My two ships - making the Pumpkinseed and Mola, I was trying to have a believable but characteristic syle). Cowboy Bebop Planetes or Memories:Magnetic Rose has some very good style regarding to space hardware in my opinion. Also there were some very good western cartoons too. But I don't want to imitate any of those. It's always better to start from the real thing in basically any form of art, then to start from somebody else's work. But inspiration is a good thing.
What kind of character portraits we should use? I'de aim for a bit comic-book look with very diverse silhouettes, but we should avoid any too symbolic style (like anime or grounded western comic styles). But this alone will be a very exciting topic to research. Hergé and Moebius comes to my mind first, but maybe some time we can test out very distinct styles on the players and have some feedback. I think it worth to give shot for a style loose, you see in Joan Sfar's Rabbi's cat animation and comics for example. To see how the players react to those. Again, no imitations.

(And I think there should be a kind of quality guide. Imagine something like a coding guide to ensure that future modifications don't need full remodeling of a ship for example. But that will be on another topic when the need arises)

These are which comes to my mind right now, but I think this will grow to a useful thread anyway, and questions will pop up during discussion.

Sorry for the long rant, but I feel these topics should be discussed properly before a grab my pencil. And it's usually more inspiring if an artist has some boundaries to work with, than if there's unlimited creative work. And apart from a visually pleasing game, a good framework will provide a good environment for artist to do what they enjoy, nurture their skill and have quality feedback on their work too, and good motivation from it. And I hope the developers will get back even more motivation from the evolving art aspect of their game.

Thanks again!
Szilárd

[Luomu]

I have some opinions :)

I gathered that one aim is to provide a an engine for others to build on

I'd like to make a game, but we need to build the engine all the same.

What are my limits? How much can I tamper with gameplay and world building aspects for example?

Can't really tamper with the gameplay. We wrote this short scope page a long time ago, it's good to read it first if you haven't http://pioneerwiki.com/wiki/Design_Scope
With world building, there's more freedom... But my experience is that the more words there are, the less interested other people are. It's better to show than tell.

How current artwork overlaps with the vision?

Nothing feels out of place so far, just a bit generic. The "vision" also includes more customizability, but not all of the engine/UI is there.

How valid the history/world guide page on the wiki?

...probably not very. I don't think storytelling is easy for an open-source game, so I try to avoid the subect. Still, I predict the basic setting will always remain similar. Humanity has spread into stars, rapid expansion made possible by the hyperdrive. Perhaps the expansion has been too rapid, judging from all of these quarreling factions that have sprung up! We'll have a few predefined factions, and many procedurally generated minor ones*.
I don't think the game is trying to predict the future, just provide a setting with places to go and things to do. The tech and society presented are a bit backwards for something to happen in year 3000, so I wouldn't mind adjusting the year.

For technology, I have a couple of things I am fixated on. First, there is no magical antigravity because we just don't need it (unlike TV series). This is why stations need to be rotating to produce gravity, and larger ships may have a spinning section (not possible in the engine, but let's work towards it). Humans can spend a long periods of time in zero-g due to unspecified medical advancements**.
Propulsion is achieved using thrusters. They are implausibly powerful, but let's just not talk about that aspect. Visually it means that all ships should have visible thruster ports, and the propellant must be stored somewhere (does not require visible tanks, but an appropriate ship volume).
For energy production, I guess we have fusion. That means solar panels would look out of place, except maybe on some low-cost housing.
There is no sentient AI or replicator-style nanotechnology, I don't know if that affects art but I just like to rule it out as being too unpredictable.
There are no aliens (live ones, at least).

Since it's relatively effortless to get into space, one would assume that space habitats will be more effective than colonizing a planet. This, coupled with rapid expansion, can neatly explain why there are only few small cities per each planet :) I'd like to think terraforming is not plausible for our tech level/timeframe.

I would separate atmospheric capable/only space capable ships a bit more.

Yes, this is in line with previous discussions. Atmospheric ships should be vaguely aerodynamic and not be insanely large. Space-only ships may be "boxier" and the largest ones may house a shuttle or two, but there are no good reasons for multi-km long ships.
We've talked about docking bays and landing pads having a 100x100x100mm upper limit. Stations should also have external docking ports, and the largest ships whould not dock at all, but transfer cargo by shuttles/drones. Let's just assume we have these docking capabilities in the engine someday.

It's vast. Also a bit desolate right now. I think it needs quite more diversity

It is, and of course we need more life in there. Still, a feeling of vastness and isolation is something I'd like to communicate through the game. Space is large and uncaring. I don't know how possible this is while providing fun activities for players.

Also: what visual styles do you like? I'd say, we shouldn't aim for a too photo-realistic look.

I would like to have something stylized, although photorealism is easier when you will receive contributions from multiple people over the years. Your two ships are a good direction so far. They are curvy but not *too* retro.
Because space is black, I want colourful ships. Bright, flat colours with stripes/patterns/tribal markings. "Chris Foss mixed with Homeworld" is a good summary of what I like. What I want to avoid are uninteresting ship shapes with detail for the sake of detail, only available in silvery/grey/black colour shemes.
You will not often see other ships up close, so it's good if shapes and colours are recognizable from distance. Fine detail will also often go unnoticed.

Finally, some tech bits:
- The pattern/colour system is in the engine so we can achieve variety and customizability. There are some aspects of it that suck, but I'd still stick with it (and surely it can be improved...)
- I still don't want to introduce normal mapping, because the quality requirements required will rise sharply and we want the core ships to be somewhat even quality. We'll see...

*with randomized insignia, someday
**handwaving is ok, technobabble is not. So I'd prefer not to make up too much tech.

[bszlrd]

Yes, I've read the Game Scope. (And wasn't even thinking about anything large about gameplay anyway.) I think we are on the same page mostly. :)

With world building, there's more freedom... But my experience is that the more words there are, the less interested other people are. It's better to show than tell.

I agree, too much writing is not a good idea, but a general good, coherent background, which provides good room for the player to bring in his own imagination is a good thing in my opinion. But I agree, that this shouldn't be too specific. But to have a good background, there are quite a few things to consider and think trough, even if the player never sees a word about that background world.

I like the basic premise of SolFed - CIW situation. I'm imagining it as some kind of Cold War situation with internal struggles as well. (Haber world domination is a bit of a stretch for me, but I don't have any real problem with it)
I think adjusting back the year would be a good thing. (Might even be good idea to give a bet of randomness to it, like a 2-3 year interval the game picks from randomly when starting a new game. It would loosen up things in a very subtle way.)

I don't think the game is trying to predict the future, just provide a setting with places to go and things to do.

This is why I think there should be more focus on intra-system affairs/possibilities, like more in-system missions, in system trading possibilities and such. That would create an even more vast, and more lively feel to the game. But this would need somewhat nerfed ships, which I'm in favor of anyway. I think the current ships are way too powerful. It would add some very subtle richness, if a player needs to plan his travel at least a bit, when picking destination. Not on a fun-killingly hard level anyway.

I'm all for excluding magical technologies, like antigravity, or artificial gravity. There's gravity in your ship during acceleration anyway, so.... Ant this is another reason for weaker thrusts. If a transport has constant 10Gs of acceleration, then the passengers will resemble meat pulp at the end of the trip :D. Medical and technological advancement can alleviate that on a level, and these explanations might never be seen by the player, but I think this needs to be considered. (I've already created a mod where all ships are toned back quite a bit. haven't really tested it properly yet, but it already caused crash when dsminers were constantly chrashing to planets, getting damaged and panic hyperjump, so the tradeship lua freaks out when it can't reach them.)
If I'm correct, the Mig 29 has about 1.3G acceleration. I know that this could kill all the fun from the game, that's why I made that mod, to check if it there'S any point doing it.
I'd even tone back the deltaV capabilites at least a magnitude to further aid intra-system play.

About spinning stations. I was already thingking about, how would I make an huge O'neil Cylinder station type. How would it look and behave regarding to docking bays and such.

In my opinion, interstellar travel should be at least a bit of an undertaking to enrich the game even more.

Excluding huge, several km ships is a good thing in my opinion too.

I have the same opinion about sentient AI and nano-replicator technology. I would even go as far as excluding any humanoid type robo servant, robo-soldier or anything. Even if they aren't show any hint of even imitated sentience. Industrial robots are another thing.

I'd like to think terraforming is not plausible for our tech level/timeframe.

I wouldn't rule out terraforming. If it's not a magical two week process, but a several centuries long undertaking, then it would also give some sublteties to the world. It could create unique trading opportunities and such too.

It is, and of course we need more life in there. Still, a feeling of vastness and isolation is something I'd like to communicate through the game. Space is large and uncaring. I don't know how possible this is while providing fun activities for players.

I agree with you on that. I was saying desolation on another meaning I suppose. I'm thinking about the too uniform world. All space stations look exactly the same, very few ships for a huge universe. Obviously that's why I've already contributed that two ship, to give some more detail to the world.

Because space is black, I want colourful ships.

I absolutely agree with this too. All space games have gray and dark ships, and it's quite depressing. (not to mention, how much more heat management problem a darks ship would have :D ) And I kinda like both the style of Criss Foss and the Homeworld/Hardware - Shipbreakier franchise too.

handwaving is ok, technobabble is not. So I'd prefer not to make up too much tech.

Also, I couldn't agree more.

[FluffyFreak]

I gathered that one aim is to provide a an engine for others to build on. If that's correct, then how detailed the artwork should be? One or two ships for each role, or a more extended and lively ship stock for example.

My personal take is that the best way we can make an engine for others is to make the best game we can with it (http://scientificninja.com/blog/write-games-not-engines). This is because when you make a game you end up with a game and an engine, when you make just an engine you generally end up with an unworkable mess ;)

[Tichy]

I like the idea of having more stations than terraformed planets. It also justifies the few cities we find on the planets. About those, I think we should have different kinds of habitats or installations depending on the characteristics of the planet. They should also be placed in a rational way, according to their purposes. Underground cities or domes on planets with different atmosphear from that of the Earth. Indistral or mining installations on asteroids. Normal buildings on Earth like planets. O'Neill cylinders, torus stations, ringworlds and orbitals (even Dyson spheres?) and similar habitats orbiting around a star at appropriate distance. Outposts, industrial installations, embassies, customs, trading stations around the planets. I think that if a player prefers to use a ship without planet landing capabilities, he should be able to do most of the activities.

About the graphic style, I agree about having a "comic book" (I also like Moubius style. and Gimenez, in the saga of the meta-barons) aspect with plausible characteristics. One think that I'd like to see is the aging of the ships and stations. To be able to find old or poorly mantained ships or stations.

[FluffyFreak]

How valid the history/world guide page on the wiki?

...probably not very. I don't think storytelling is easy for an open-source game, so I try to avoid the subect. Still, I predict the basic setting will always remain similar. Humanity has spread into stars, rapid expansion made possible by the hyperdrive. Perhaps the expansion has been too rapid, judging from all of these quarreling factions that have sprung up! We'll have a few predefined factions, and many procedurally generated minor ones*.
I don't think the game is trying to predict the future, just provide a setting with places to go and things to do. The tech and society presented are a bit backwards for something to happen in year 3000, so I wouldn't mind adjusting the year.

For technology, I have a couple of things I am fixated on. First, there is no magical antigravity because we just don't need it (unlike TV series). This is why stations need to be rotating to produce gravity, and larger ships may have a spinning section (not possible in the engine, but let's work towards it). Humans can spend a long periods of time in zero-g due to unspecified medical advancements**.
Propulsion is achieved using thrusters. They are implausibly powerful, but let's just not talk about that aspect. Visually it means that all ships should have visible thruster ports, and the propellant must be stored somewhere (does not require visible tanks, but an appropriate ship volume).
For energy production, I guess we have fusion. That means solar panels would look out of place, except maybe on some low-cost housing.
There is no sentient AI or replicator-style nanotechnology, I don't know if that affects art but I just like to rule it out as being too unpredictable.
There are no aliens (live ones, at least).

Since it's relatively effortless to get into space, one would assume that space habitats will be more effective than colonizing a planet. This, coupled with rapid expansion, can neatly explain why there are only few small cities per each planet :) I'd like to think terraforming is not plausible for our tech level/timeframe.

I quite like the story work done by Thargoid(?) I think it was. It didn't add much gameplay value but some of the setting and setup stuff was really good.

Also we don't need to have so many factions, that wasn't my original intent, it's more a side effect of the flexibility of the Lua system and Skapusniak going crazy with it (in a good way).

So we could just have a core of really focused factions, and an outer rim of random factions, there's been a request for more scripting access and information that I need to get around too for that soon.

For gravity/anti-gravity I agree, although much like my opinions on exposed cargo I don't think you'd actually see the spinning bits on ships, except where they're meant to look cool. Far too dangerous to have them exposed. That's why cargo haulers wouldn't have exposed cargo too, nothing would ever reach it's destination intact! :)

I don't mind AI or nanotechnology, I just don't think it's all that visible, it's building microprocessors and the super-advanced engines that are hilariously powerful ;) So nanotechnology might be tradeable, as in the big machinery it lives inside, but it's not a thing coursing through peoples blood. AI I think we might need to explain away as some kind of social backlash against it?

[bszlrd]

AI I think we might need to explain away as some kind of social backlash against it?

Or there could be a plateau in the rate of increasing computing power, before some critical mass of computing power that could support it.

I like the idea of having more stations than terraformed planets. It also justifies the few cities we find on the planets.

I think there should be much more space habitats regardless of the number of cities. Ranging from very small outposts, to those toruses and cylinders.
I was also thinking about domed cities. I kinda like that trope.

[FluffyFreak]

We've talked about docking bays and landing pads having a 100x100x100mm upper limit. Stations should also have external docking ports, and the largest ships whould not dock at all, but transfer cargo by shuttles/drones. Let's just assume we have these docking capabilities in the engine someday.

We can do the internal/external docking bays right now :D The docking system work, also we can limit the bays that are available to certain ship sizes, so you can have 100x100x100m bays, then bigger external ones that will only accept larget ships. That *should* all work right now, I think Paragon are using it but it's just the same system I wrote.

It's vast. Also a bit desolate right now. I think it needs quite more diversity

It is, and of course we need more life in there. Still, a feeling of vastness and isolation is something I'd like to communicate through the game. Space is large and uncaring. I don't know how possible this is while providing fun activities for players.

I'd quite like to customise each planet a lot more (This needs major... no, total rewrite), add things to find like ship crash sites from earlier waves of exploration/expansion, alien artifacts (I'm getting my damned Orbital in somehow!), failed (human) colonies.
We're not really exploiting Luomu's pattern, decal or tag system as much as we could be either. Buildings, spaceports & stations could all do with reworking to get more colour and style variations.
It'd be really cool to populate all of those "Small scale mining operations" with small spacestations and hidden pirate bases (basically junked looking spacestations made from old ships) and that kind of thing, something to add flavour.

Also: what visual styles do you like? I'd say, we shouldn't aim for a too photo-realistic look.

I would like to have something stylized, although photorealism is easier when you will receive contributions from multiple people over the years. Your two ships are a good direction so far. They are curvy but not *too* retro.
Because space is black, I want colourful ships. Bright, flat colours with stripes/patterns/tribal markings. "Chris Foss mixed with Homeworld" is a good summary of what I like. What I want to avoid are uninteresting ship shapes with detail for the sake of detail

Emphasised for the bit that made me want to have your babies Luomu :D

I agree with this in general, but also stylised things are often easier to get feeling right, everyone can quibble about what's real but if you're not trying to be "real" then you've got a lot more leeway.

- I still don't want to introduce normal mapping, because the quality requirements required will rise sharply and we want the core ships to be somewhat even quality. We'll see...

Ah so that's why you don't want normal maps, I never realised before.
I'd add that it would be good to have the ability to select from a palette of geometry/sub-models. Maybe not for the ships at first, but for buildings and spacestations it might help to have solar panels and modules change around. The interior or our current spacestations are vomit inducingly boring, mostly because I just can't model for shit :/
That'd be a great place to have machinery, cargo containers, ship parts, or control towers changed - along with the existing decal, patterns and tags systems.

[FluffyFreak]

O'Neill cylinders, torus stations, ringworlds and orbitals (even Dyson spheres?) and similar habitats orbiting around a star at appropriate distance. .

Having tried some of this I'm immediately going to rule some of them out, Ringworlds and Dyson Spheres are a no-no baby ;)
I'm sure we could make them work, technically, with some ... well a lot, of compromises, however ever a small Orbital just 35,000km across is a challenge already.
I wouldn't consider something an Orbital until it was ~6 million km across either, so: O'Neill cylinders, torus stations, even an Orbital should be ok. The first two as human constructs, the Orbitals as dead alien artifacts.

[Tichy]

I wouldn't consider something an Orbital until it was ~6 million km across either, so: O'Neill cylinders, torus stations, even an Orbital should be ok. The first two as human constructs, the Orbitals as dead alien artifacts.

They could be broken, aged or uncompleted. Maybe very rare. Maybe archeological sites... which means more kinds of missions :)