but how the bloody hell do you procedurally generate complex non-liner character interaction/story!?

Hi Demented! Glad I've got your attention

The short answer to your question? With difficulty.

The long answer is... I've been thinking about this for the last few years, and doing a fair bit of research. There's been a few different approaches, and my own ideas seem to tie in very closely with some one paper in particular, Chris Fairclough's OPIATE System, which is kinda promising.

I want to make something that's interesting and fun to play, which means making it into a game, rather than a stand-alone experiment, so most of my time is going into that first. I'm working solo and part-time on this, so I'm having to make compromises with the scope, but soon I'll have enough of a world to start building the narrative system.

If you want to play something similar to what I'm attempting, check out Bay 12's amazing Dwarf Fortress. It's a full-on simulation of a stylised society, though not as user-friendly or story-focused as what I want to end up with.

Wow, saw this on the procedural Generation thread and I can't wait to see some gameplay!

Thanks. It's coming, I've been slowed down by having to get a proper job, but I'll try and post something up very soon.

Great stuff, loving this thread so much.

Thought I'd share the procedural building exteriors I'm working on for Cold Metal (http://forums.tigsource.com/index.php?topic=17841.0).



Each building can have the size and height set independently, plus there's variables for the degree of symmetry and irregularity. They're generated using some simple cellular automata to carve away at the edges, then 3D meshes are spawned in order of ground, mid-levels and roof components. It's pretty quick at the moment, even without any optimisation.

The idea is that a large area of the city is generated, with simple building prefabs, and then the nearest buildings to the player are swapped out with these procedural ones.

Today I'm trying to get the interiors working, so you can go inside the buildings as well.

When this is finished, I'm thinking about putting it up on the Unity store, with a bunch of different styles (sci-fi, modern day, 1890s London, Medieval). Anyone think that's a good/bad idea?

Thanks so much. I've been putting off biting this particular bullet for months now.

landscapes in real life are formed as the sum of different forces acting upon it (say, weather, tectonic plate movement, plants and animals), all influencing it and each other in a different way.

It's a question of balance. Simple procedural generation can give some interesting and varied 'good enough' results, but they'll never be very realistic. Or, you can go to the other extreme and use multiple passes to calculate erosion, etc. It all depends on how much time, knowledge and processing power you have. I'm guessing the reason Notch fakes stuff in minecraft is because doing it the 'right' way would take forever.

I can't remember who said it, but all game programming is pretty much cheating.

Just to let you guys know, Cold Metal hasn't gone away, it's just been stalled somewhat by College and work.

But now I'm done with everything else, and working on Cold Metal full time, so you can expect a playable version soon(ish).

This is my (pretty rough) schedule:

1.1 Procedural City Generation (February 2011)
Research into procedural generation techniques and scripted mesh construction in Unity3D, Procedural landscape Generation, Procedural street generation, Procedural building generation

1.2 HUD, controls, look & feel (March 2011)
Initial research of custom shaders, HUD interface (static), Movement and interaction controls, Design and implementation of custom shaders, HUD interface (active)

1.3 NPCs, combat (April 2011)
Simple NPC navigation, Basic projectile combat, NPC combat behaviour, Advanced projectile combat, Melee combat

1.4 Vehicles & Traffc (May 2011)
Simple player vehicle movement, NPC vehicle traffc behaviour, Advanced player vehicle movement, NPC vehicle dynamic response behaviour

1.5 Mission generation (June 2011)
Mission test with simple interface, Improve mission interface, Mission generation system, Add missions, refine interface, Add more missions

For the moment, I've dropped the procedural landscape idea, and I'm using grid-based cellular automata to make the city layout, mainly so I can get it finished in time.

Right, back to work. Demo soon, I promise.

Expect the demo pretty soon

My name is Dan, and I approve the hell out of this message.

I seriously can't wait guys, it's looking awesome!

This looks amazing, guys. Looking forward to seeing the demo.

When you visit a galaxy it generates the basics of the clusters, and if you zoom in on a cluster, it generates the basic information on that systems in that cluster and so on. When leaving, enough information is retained to reproduce what you visited and nothing more. And I was intending to implement a system so that areas that you visit and don't come back to for a long period of time would eventually start removing the saved information from the smallest level first.

That sounds fantastic, and coincidentally it's exactly what I'm planning to do on a city-level scale with my own game http://forums.tigsource.com/index.php?topic=17841.0

Hope it goes well, I'm looking forward to seeing how you progress.

That's looking really amazing. I'm curious to see how you're going to develop it. Personally, I'd love to see it as the basis for an Elite-style game, with galaxies, races, planets and cities generated procedurally.

Nice work.

Sure sure, but since this is a Feedback thread, I figured fixing the T button would be a good idea as well.

Yeah, of course. I'm torn on the New Game option as well. I love the abstract, dreamlike feel of the game, but I also think that some players are going to get lost because some of the more customary tropes of games are not present. It took me a while to get into the feel of not having a permanent main character, or a standard front end, but I really like the dreamlike aspects now.

And seeing as this is a feedback thread, there are things about the game I would like to see changed. The music is bugging me a little. Not that it isn't well done, but there are time when the gameplay resembles a bullet-hell shooter, and I find myself missing the kind of heavy, driving, energetic music that usually accompanies them, so it would be great if the music got heavier as the action did. That may just be me though.

The other thing I would love to see in the game is Bosses. I know one of the bullet points on the site is that the game doesn't have them, but I have a strong feeling that they'd work really well with the game. It could be an ability that was present on, say every dozen levels. Just an open area, and a larger creature with no abilities except high health (with no possibility to enter, or no way until its health was reduced), and a host of other 'satellite monsters' (which could be entered), with identical abilities and movement mirrored rotationally around the boss.

I'm having the hardest time understanding what's going on in this game. For one, pressing "T" isn't giving me the instructions again, and I don't think I quite understood them the first time.

Someone's posted up a video of the first few minutes (though they've misleading said that the game is only $1.79), if that's any help.

I'm guessing deleting the files and reinstalling would also reset it.

Interesting what you said about Inception. I'm wondering now if you've seen Paprika, which was, of course, one of Nolan's strongest influences http://www.sonyclassics.com/paprika/

Also, every game designer should read Godel, Escher, Bach

Thoughts, criticism, gut reactions, etc?

That's just perfect. All the best with the launch, Jason. I know I'll be buying it.

Dan

I may do. I'm working on a procedural landscape routine for a city generator at the moment, and I've already thought of half a dozen ways I could use the blob thing.

That's in Unity, so I'll bung you my code if you like.

Cool, glad you liked the idea.

If the blob has complex edges, then difficulty/complexity could be as easy as picking points which are close or distant, and smoothing them or forcing straight lines between them, which you'd probably want to do for the start/finish straight anyway.

It shouldn't be too hard to allow for tracks that cross, either. Just have a random chance of a secondary blob, and give them some kind of surface attraction so they have a single point which touches, then either leave the height of that point flat when you generate the spline, or force the height of the two road sections apart and drop a bridge mesh in there.

I really would love to see this now. If you don't use the idea, I might have a go.

Basically it takes the frustration I feel whenever I play a third-person 3D game, magnifies it, then says "That's the point, actually. Somehow this works for me...

For me too. It reminds me a little of Sumotori, the physics-based fighting game, where you were battling to stay upright as much as you were fighting an opponent. The resident Evil games famously give you a 'broken' control system, but do so in order to exaggerate the horror rather than the comedy.

Congratulations on Octodad. It's definitely one of the most enjoyable games I've played. I'll have to grab the latest version (with my 8 wobbly arms) and give you some more constructive feedback.

By 'framing device', I just meant a small story to frame the action. The Da Vinci Code (book, not game, sorry I didn't make that clear) uses the device of solving a murder to set the main character off running around and solving problems. The execution was dreadful, but the idea is sound: to give a clear reason for the player's actions, beyond the simple satisfaction of solving a problem.

You could cast the player as Jean-François Champollion, or set it in 2011 and cast the player as a detective who is left a series of clues, but no access to any advanced translation tools. The actual device doesn't really matter, but will give shape and motivation to the puzzling, which at the moment is, well, puzzling.

With the game itself, I think the problem I was having was that the possible actions were not clearly defined, so I was resorting to clicking around randomly until something happened. If you give the player some clear feedback (cursor changing to different icons, on-screen prompts, etc.) on what it is that they're doing/about to do, and spell out exactly what the possible actions are, then I feel the game will be much more enjoyable. I also felt that, with the fading text, I was not in control of what was happening. In a game like yours, everything that happens must be as a result of player action, and be clearly shown to be the result of player action.

Good luck with developing the game. I think it has a lot of potential, but just needs to be made more accessible.

Looking great! Sounds a bit like a game I started making a couple of years ago, with a melting snowman character. Yours looks much better than mine ever did though.

Well done.

Just played it, and I really can't think of a single thing to criticise. You guys have made my new favourite Unity game. Well done!