Ship Concepts

[mkDanger]

Hell if I know where to put this, so it's going here until I'm instructed otherwise.

This is loosely based on the Cobra Mk III. In general, I do not intend to recreate the ships from Frontier. Not only is that Braben's property, I just don't think the old designs translate well in a modern game. In this particular model, I'm working with clean, elegant lines and a very singular design. Other factions, obviously, could be completely different.

Bear in mind that this has not been textured and lacks a great deal of detail. When I have the opportunity to discuss game-specific needs with the community, I'll add those things in. i assume it'll need a variety of thrusters, animated landing gears, etc. I'm offering broad strokes, at the moment.

[mkDanger]

All that said, I'll need some other information before I can get too far. What is the game engine capable of? How low do I need to keep the polygon count? What sort of textures am I using? That will determine how much I model and how much I leave to normal mapping.

[Luomu]

What is the game engine capable of? How low do I need to keep the polygon count

I have no idea. Let's say try to keep it below 10000. The game is played on a wide variety of hardware. This (from an ex-contributor) is 4322 triangles:

And this is 3217:

It is important that you do detail levels so the full detail meshes are not used for far away distances. The game supports up to 4 detail levels, ships should have 2-3, depends how complex they are.

What sort of textures am I using

You can use diffuse textures, and emissive textures for self-illuminated parts.

how much I leave to normal mapping.

There isn't any.

[Luomu]

Another thing: there shouldn't be much animation, if any. The collision system in the game does not support this.

[mkDanger]

You're right about those being cluttered. I see what he's trying to do - reimagining the ships from Elite - but it's not really effective.

I'll work on models for now, and try to nail down the 'feel' of various factions. I'll see what the consensus is at that point and build from there. Opensource programming is great; opensource cohesive design isn't.

It'll work out. It'll look good. I'll keep you posted.

[mkDanger]

Native markup. Crazy. Now I can post images rather than links:

Question: Can I call an animation based on, say, a ship entering atmosphere? Would be sweet to have retractable wings/foils and such.

[mkDanger]

With some color, for my own amusement.

[s20dan]

Hi there, I don't have much to add other than: I really like these designs, especially the last one. That last one is really sexy :)

[mkDanger]

Thanks. I'm learning/relearning as I go. It's been a few years since i touched the software, but it's coming back.

[mkDanger]

Here's the first ship turned into a cargo vessel instead of a single man fighter:

[johnbartholomew]

I really like these. Some things that I think it would be good to keep in mind as you try different styles and designs:

• I would not expect all ships to be capable of atmospheric flight. I would expect the ones that are to be designed with some acknowledgement of aerodynamics, though I'm happy for artistic license here. Space-only ships may be a lot less aerodynamic, and also perhaps be somewhat unclear with regard to their preferred orientation, except for their primary thrust axis.
• Ship size shouldn't be completely arbitrary. At least details like airlock doors should be a sensible size. Let's assume (despite the evidence to the contrary) that humans will remain roughly the same size they are now.
• Ships will have equipment, and some of that equipment (e.g., lasers, missiles) may be visible from the outside, at least on some ships.
• It seems likely to me that the physical requirements of space-travel and the task(s) that a particular ship is designed for will have a greater effect on its overall structure and design than some faction style guide. (*)
• Manoeuvring requires thrusters, so it's good to think about where they should be put. Thrust calculations "cheat" (they don't use the thruster positions specified in the model), so the modelled positions don't have to be ideal, but they should broadly make sense.

(*) I'm particularly against the effect seen in some games where you get the evil faction with spikes on all their stuff and the techno faction whose designs consist entirely of squares and blinking lights and the bird faction with designs that all incorporate a wing motif and so on. I have no problem with groups of ships all sharing the same style: that's easily explained as a bunch of ships all from a particular shipwright. Actual factions are more likely to be identified through paint pattern, decals and how quickly they start shooting at you.

[mkDanger]

I agree on all counts. Those are things I intend to keep in mind while I explore different styles and try to nail down faction and independent ships. Form and then function, as it were. Of course, as you can see above, (and I was starting with fighters, from which I could branch into other ship types from specific factions/manufacturers (though (can I nest parentheses in text? (looks like I can)) if you look at the products of, say, Lockheed Martin, they vary wildly as far as design)) ships that ought to land on planets will look like they reasonably could - ships that ought to stay in space will, generally speaking, look like they belong there. Thanks for the input!

[mkDanger]

I, uh... I promise not to do the parenthesis thing again.

[mkDanger]

[mkDanger]

One of the above remodeled. (Cleaner build - more manageable UV map when I get to it)

[mkDanger]

[mkDanger]

Rough Landing Pad without textures. By request.

[mkDanger]

The shapes and whatnot will make more sense next to buildings. You'll see.

[mkDanger]

[mkDanger]

[mkDanger]

[robn]

The buildings are looking fantastic!

[mkDanger]

Thanks! I tried to give a sense of scale here. I'm not sure about the landing platform. It may work to just scale it up if there are landable ships too big to fit, though I've got to think a giant cargo ship would land in something more like to a dock with cranes and whatnot.

[vaLinBSD]

@mkDanger:Hi, may I ask what program do you use for creating the ship concepts? And BTW, awesome work!

[mkDanger]

Thanks! I'm using Lightwave for the modelling and Photoshop for the textures. http://www.cgtextures.com/ has been a phenomenal resource.

[fluffyfreak]

These are amazing.

[robn]

@mkDanger Hi! We now have a proper modeling system (no scripts required) and a real need for models that use it so we can replace the old system. Are you interested in doing some models for inclusion within the next few weeks?

There's a list of what we need (at minimum) here: http://pioneerwiki.com/wiki/Asset_List

Of course we'd be more than happy to help with anything you need to make this happen. Feel free to reply to this or contact me privately if you're interested.

[Luomu]

pioneerspacesim.net/forum is the place to discuss art, and final submissions should be made to github.

[fluffyfreak]

It's a real shame these designs never made it into the game, between these and those in #1532 there's some really nice models here :(

[robn]

Yeah, it really is :(

[mkDanger]

The community here does seem to have some trouble holding on to artists and their assets. I can't speak to the experiences of the others who have come and gone, but perhaps relaying my own will shed some light on your predicament.

Firstly, the creation of a good model from inception to completion is an obscenely time-consuming process. In a professional setting, the workload is divvied up across a team of artists. Concepts are drawn by one team and approved by the project lead. The generation of a universe's worth of (cohesive) content demands structure. Those concepts are handed off to a team of modelers. Their models (sometimes with vertex maps - sometimes a separate artist unwraps them) are handed off to be textured. The textured model is handed off to be animated. The execution of this pipeline requires communication, clear guidelines, leadership, and accountability. Now, assuming that such a team of artists will never be brought together for this particular project, what you end up with is a small set of individuals handling that pipeline themselves. It is, again, an obscenely time-consuming process. Any artist with his or her salt will, at very least, want to know that their work is a part of something worthy of their talents - that's where cohesion is absolutely paramount. The 'hey, do what you want, no one's in charge' attitude leaves you with a ridiculous hodgepodge of mismatched objects that no serious artist wants to be a part of.

Now, I understand that the spirit of open source freedom and unconfined collaboration is the beating heart of this community. It's wonderful in its own right, making for a project with the potential to truly showcase the best of what each individual has to offer. Problem is, to make the visuals work you need your artists working under one vision - even if that vision happens to be the consensus of the community.

When I stumbled across Pioneer (looking for an Elite clone, as it happened) joining the project was a matter of sifting through countless pages of disorganized information in an effort to figure out what the hell was going on and what the requirements for participation were. Frankly, it's an awful and frustrating process.

All that said, this project has clearly been in visual limbo since its inception and, so far as I can tell, will be until certain issues are addressed.

My suggestion:

Put someone in charge. Start fresh with concept art and collect a respectable amount of it until the community agrees on, at absolute least, some common visual themes that will carry the aesthetic end. Show some respect for the talent and experience of your artists by providing that much.

Once you've got some concrete visual ideas, you'll be able to shop them around and recruit a team of artists and finish something you all know has been damn near dormant for far too long. None of us has any interest in pouring countless hours into something with no direction.

Collect the relevant information (storyline, aesthetic requirements, technical requirements, etc.) and compile it in one easy to reference location. There ought to be a comprehensive list that, in concert with the concept art, provides accurate and thorough descriptions of every asset needed.

Answer questions. When an artist needs information that isn't available in the information i just described, provide it. Then, add it to the reference materials - that'll save you answering the same questions every time a curious designer pokes his head in.

This post is perhaps not as succinct as I'd like, but hey - I'm writing over breakfast and conversation. If you have any questions, feel free to send them my way.

[laarmen]

Quoting mkDanger (2013-07-04 14:14:24)

Put someone in charge. Start fresh with concept art and collect a respectable amount of it until the community agrees on, at absolute least, some common visual themes that will carry the aesthetic end. Show some respect for the talent and experience of your artists by providing that much.

Well, at least for the "put someone in charge", pretty much every time someone comes in with demands regarding art direction the answer is something along the lines "you want direction, feel free to provide it, we won't stop you".

Experience shows that the community will never agree on such things, since it is all a matter of taste, that's why we only need someone to step up and say "OK, that's how it's gonna be". This person cannot be us because, well, we already have a full plate, and adding more work and responsabilities on top of that is the best way to have a quicker burnout of the current contributors.

My 2¢

[Luomu]

Dear Mr. Danger,

I agree with everything you say. In theory.

The thing is, we have never reached this stage yet:

Put someone in charge. Start fresh with concept art and collect a respectable amount of it until the community agrees on

The only way for someone to fill this position is to get stuff done. We don't just appoint people and expect everyone to follow. Sounds cliched, but the ideal candidate must get their hands dirty, figure things out themselves, love the game etc.

We just received two high-quality models, both created without any artistic hand holding, so I know there are capable people out there.

[fluffyfreak]

I think there are some good points, I don't know how many we can actually adopt. We can't have the professional pipeline simply because the professionals aren't volunteering to model/texture/animate other peoples ships and things.

An overall style guide, list of required ships and other things would be welcome if we can get it put together.

@mkDanger For what it's worth we (especially me) would have helped you to get all of the above ships into Pioneer, but as far as I can tell you've never made them public for us to try.