Will Wright's New Game: Spore

[Grumbold]

Originally posted by vovan
I don't think so, because there was nothing to indicate any autonomous evolution in the demo, however, there was nothing to indicate that there won't be any such thing... Everything is possible, but I imagine all the change will be done by the player.

I think you are right, since it is more fun to evolve your creature than to be messed up by a random evolutionary stage. The computer may have to evolve other creatures that share your world (later galaxy) if it is allowed to do more than just seed the place with other player created constructs.

[vovan]

Originally posted by Lord Nuclear
Care to make a wager?

On what?

[DrSpike]

Originally posted by Kuciwalker


Which one?

In this thread.

[Tassadar500]

Originally posted by vovan


On what?

On the success of the game. Although, I'm not sure how to base it on exactly. Ratings, or how many games sold...

[child of Thor]

Is it going to get a console release?

[vovan]

Originally posted by Lord Nuclear
On the success of the game. Although, I'm not sure how to base it on exactly. Ratings, or how many games sold...

Oh, I'm not sure about the success of the game at all, actually. The wager I was proposing to DC (or someone) was that it wouldn't get cancelled. As far as the reception... I don't know, I hope it will be well-received, but who knows... Innovative games have a tendency to not be that well-received.

One could certainly base a bet off, say, the number of games sold during the first week, or something like whether the game hits the top seller for a week or so... But that depends too much on the competition, IMO, so basing a wager on the number of units sold would have to span a long time. As far as reviews are concerned, they could be pretty well off, too: the first review of the Sims bashed it as mediocre and lacking, yet it sold like hotcakes, showing the reviewer had no clue what he was talking about.

[Urban Ranger]

Originally posted by vovan
And really it's not true in the general sense. For example, with self-programmable hardware, a computer can change its own hardware to learn things it was not technically programmed to do.

Yeah, but you need somebody to design that hardware with sufficient built-in smarts.

Originally posted by vovan
Hmmm... I think I already kind of addressed this point above, but I will reiterate. All of the things in the game - creatures, buildings, etc, are not created by the designers. They are purely made by the players of the game through the game's own interface.

Hm, does the player need to create the things from scratch or just do something a la Unnatural Selection (a head from here, an arm from there, etc.)?

Originally posted by vovan
Now, some of these players could be designers, but they don't technically have to be. The engine is smart enough to figure out how a thing should behave no matter how it looks like.

That's a bit implausible, since not even experts such as paleotologists could do that all the time.

[Urban Ranger]

Originally posted by Ari Rahikkala
What I'm claiming is a bit more direct: I simply don't see a reason why it should be impossible for a computer program to calculate plausible-looking IK for a plausible human-created model. The reason why I mentioned text layout is not that I think its solvability makse this problem solvable, too, it's that you should tread carefully because someone might think that you implied that all nontrivial data analysis problems are unsolvable.

Certainly I didn't imply that nontrivial data analysis problems are not solvable, at least not in theory.

You need to consider the context, though. This is a game, written for your average Wintel box, with a deadline and a demand on the return of investment.

Nah, it sounds a bit implausible within the context.

[Kuciwalker]

That's a bit implausible, since not even experts such as paleotologists could do that all the time.

Yes, but the program will have the advantage of knowing exactly where everthing goes, and it's function. And it only has to figure out a possible way for it to move.

[RGBVideo]

After seeing the preliminary screenshots, I think I'll stop being so cynical about this project and agree along the lines of Kucinich:

I take it back; this looks like it could work and not be stupid. It just doesn't attempt a quite *serious* imitation of each level.

[vovan]

Originally posted by Urban Ranger
Hm, does the player need to create the things from scratch or just do something a la Unnatural Selection (a head from here, an arm from there, etc.)?

I believe both options are available to the player.

Originally posted by Urban Ranger
That's a bit implausible, since not even experts such as paleotologists could do that all the time.

As said above, nobody said that the simulation would make the animals move as they would in nature. The only goal is to make it sort of realistic, so really nothing much implausible there.

[Ari Rahikkala]

Originally posted by Urban Ranger
Certainly I didn't imply that nontrivial data analysis problems are not solvable, at least not in theory.

You need to consider the context, though. This is a game, written for your average Wintel box, with a deadline and a demand on the return of investment.

Nah, it sounds a bit implausible within the context.

Ah, now *this* makes sense . And since I know a bit more about the game mechanics this time, having listened to GamingSteve's podcast #3, I might actually have something senseful to say this time...

I don't really know how the creature editor works, but it does seem that you can't just model a hand and expect that the game will somehow figure out how to use it functionally. Instead, you'll pick up a hand object and drop it at the right place. The game will then - I *guess* - give you a certain amount of freedom to do modeling, but the functional parts will stay there unless the player explicitly takes them out.

Also, it seems that a skeletal structure is defined for these creatures. I don't know how it works, Steve just mentioned it quickly at some part, but you know what skeletons are good for: movement . I'd say it's quite likely that the skeleton is defined by the functional parts, and the possible movements are defined by the skeleton (perhaps with constraints based on where the player-created model would self-collide). It really doesn't sound too difficult for me on today's hardware. After all, they probably are doing something smarter than trying out every possible combination of movements and picking the one that gets the creature closest to the end state that was asked for

[kolpo]

When I hear about this game do I fear that the scope is soo big that the actual implementation of each stage shall be shallow.

"The top 3 vehicles are Will's creations for his civilization.

The bottom 2 were created by the computer, an industrial type civilization.

The site also says that each civilization will develop its own culture."

I thought at first that that was a photograph of some toys instead of game renders They look very cool and diverse, even while the game might be shallow would making things like that be cool. Thought I wonder how many decades am I going to need to make vehicles like that?
Wouldn't making 3D models like that be too complicated for kids?

[snoopy369]

Originally posted by kolpo
Wouldn't making 3D models like that be too complicated for kids?

As opposed to these 3d models i made for the better part of my childhood?



You just need to give them the right tools, is all.

[Adagio]

Found a few new pictures here:

http://www.simvision.net/gallery.php?cat=Spore%20Preview

Not many, but still...