Jump to content

  • Log In with Google      Sign In   
  • Create Account

- - - - -

Slightly ambitious?

Posted by , 01 December 2012 · 835 views

dominion ringworld space game kickstarter
Slightly ambitious?
  • Dominion 2012 12 01 17 16 57 71
  • Dominion 2012 12 01 17 16 36 80
  • Dominion 2012 12 01 17 15 29 68
  • Dominion 2012 12 01 17 16 58 72
  • Dominion 2012 12 01 17 16 05 24

So, one of the slightly more ambitious features I wanted to bring into Dominion is the Ringworld.

I would have attached a video link - but I've posted it as a special "Backers Only" update on the Kickstarter pitch as a thank you for pledging. I also tried embedding images into this Journal - but the gallery/album thing kept breaking, so I'm afraid you'll have to see the (hopefully attached!) album.

This isn't the small scale "halo" you will be familiar with from the game of the same name, but the theorised astronomical scale behemoth that was first(I think?) proposed by the great sci-fi writer Larry Niven in his 1970 book, Ringworld. http://en.wikipedia.org/wiki/Ringworld

Putting one of these beasts into a space-game is a recipe for disaster, surely?

The one in the picture has approximately 184,000 times the surface area of Planet Earth - how are we going to texture that? And Ep2 has landing...where will the terrain come from?

Well, we think we've got that sorted already, in my head at least! But I'm always planning for the worst, so come what may Dominion will have Ringworlds - but the player may have to extend some belief-suspension to benefit from them ;) Some cunningly contrived plot-conceits (again, already prepared!) will ease the process if needed. For now, I've just tiled a planet surface - which is very noticable when zooming along the Ringworld surface under g-drive!

At the moment, you can fly to any point around the Ringworld - it's not a preset scene just for show, it physically exists and is to scale. The "angular" appearance is entirely due to pespective and the FOV - this really is what it would look like in reality from this point of view - something I wanted to check purely out or curiousity. Granted it's suffering from some terrible floating point precision jitter when you get real close - but believe it or not, I've got that in hand and it will be easily dealt with. I want to avoid the use of doubles as Dominion will be x-platform, with mobiles in mind - until I can affirm double performance across these platforms, I'll avoid using them. Besides, so far my approach to constructing the scene (which I chose deliberately to avoid needing doubles) is working just fine.

As my (deliberately) corny catch-phrase goes... Watch this Space!

I've had some experience working with large-scale maps in my own projects. What I've learned is that in-game real estate is a serious consideration. As scale increases, the complexity of the required code increases exponentially. More importantly, the purpose for the extra playing area becomes questionable.

Essentially, you're investing development time to increase player enjoyment. Being in charge of a project, one must constantly consider the return on investment for time spent by the team. As the world size increases, the return on investment quickly decreases. From the player's perspective, how much are they getting from the larger world? At some point, the answer eventually reaches zero and the justification for the larger world (time vs. return) dwindles long before that. I'm pretty skeptical of to-scale planetary surfaces in games (we're still a long way from pulling it off AND making it interesting to players) and 184,000 times the surface area of the Earth is just not happening for under a half-million dollars.

This would be a fantastic project to attempt on your own but it's a risky game to play with other people's money. I'd suggest building a small-scale planet surface with content, collision detection, etc. to get an idea of how difficult it can be to have all the components of a game working together effectively at that scale.
I agree - it's "slightly ambitious" ;) Right now it's more an aspirational artefact - in Ep1 the player cannot interact with a moon, satellite or ringworld directly, entirely due to the issues you've rightly pointed out. But, we can still allow players to locate bases anywhere in the game universe, even on a ringworld surface, despite it being so vast - although only being able to interact with them "remotely" from orbit during Ep1 may prove unpopular and we may have to move this "feature" to Ep2.

When we come to surface detail in Ep2, we can procedurally seed the terrain across the ringworld in a similar way to planets - hopefully without it being too bland as a result. This would allow the player to actually land in/at/by their base anywhere in this stupendous terrain. Procedural techniques for flora/fauna then fill in around them - but overall diversity would be limited so some suspension of belief is needed to enjoy the potential.

In other words - for the Ringworld we'll be leveraging off techniques we'll have to build down the road for planets anyway, and this is a long term goal. For Ep1 - Ringworlds will just be yet another pretty Big Dumb Object to go "Oooh" at from orbit.

We hope ;)

PS. I forgot to mention - I do take your points as well given - Landing on planets is cited as a high risk in the KS project - we don't take the scope lightly. There's no point dropping the player on a textured sphere these days and saying "There you go!" - expectations are higher, tech is more advanced - people expect sub 1m resolution terrain, with overhangs, tunnels, densely lush forests... a daunting prospect to attempt to fulfil!

August 2016 »

2122232425 26 27

Recent Entries

Recent Entries