Will Wright's Next Game

I’m here in San Francisco at the Game Developers Conference.

This morning Will Wright gave his session. Will’s sessions at the conference are always standing room only – he’s a very entertaining and creative guy and almost always has something to say that pushes the boundaries of the industry. In fact, I usually miss his talks because people start lining up for them before the previous round of sessions has ended, and by the time I show up there’s no more room. But today the previous session I attended happened to be in the same room as Will’s talk, so I just kept my seat.

The title of his talk was “The Future of Content”. As soon as he started he admitted that he wasn’t really going to talk about that. He’d just given the conference organizers a vague title because they were pressuring him for a name.

Instead, he did a leisurely, hour-long demonstration of his next title.

Oh. My. God.

Now bear in mind that this was a room filled with several hundred professional game developers. Personally I’ve been in the industry for ten years. As a group we tend to be fairly resistant to hype about new titles. Things that get called “revolutionary” almost never are.

But … Oh. My. God.

It’s called “Spore”. It starts off with you playing a microscopic organism in a tide pool. You swim around and eat stuff. Stuff tries to eat you. If you accumulate enough energy you can lay an egg. This gives you access to a creature editor that lets you improve the next generation of microbe.

The cool thing about this is that the whole simulation is physics-based. You can actually tinker with the organism’s “skeleton” to create different swimming mechanics that are procedurally generated by the simulation. It’s a very cool little evolutionary toy.

But it gets better.

Once you get big enough the view switches from 2-D to 3-D. Now you’re playing a fish-like creature in an undersea habitat. You’re still swimming around and competing and evolving. And eventually you can develop legs and move out onto the land. And everything still is physics-based. The simulation reverse-engineers a walk cycle for your creature based on its leg configuration for example. Very cool.

Even cooler is where the other creatures in your world come from. Every time a player creates a new creature, it gets uploaded to the Spore server. So when the game needs to populate a new ecosystem, it draws from this huge pool of thousands and thousands of user-engineered creatures.

(You can play offline too, using Maxis-supplied content on the CD.)

Okay, wow, at this point I was already really impressed. A complete evolution sim with seamless downloads of user-generated content. Very impressive.

But it gets better.

If you can evolve your species’ brain so its large enough, they develop sentience. Now their physical form is locked, but you can begin playing around with their culture. You can give them tools (also user-created) and wage wars with rival tribes using an RTS-like interface.

But it gets better.

Once you’ve progressed far enough in the tribal phase you can start building cities. Now instead of an RTS, you’re playing Sim City, building roads and buildings, trying to create an efficient city for your creatures to live in. And still all the content is completely user-editable. You can create your own buildings, or download buildings made by other people from the internet.

But it gets better.

Once your city is big enough the game turns into Civilization. Now you’re competing with rival cities across the globe. You build little military units and fight it out for global domination. Again, you can use standard units, create your own, or download what someone else has made. And everything is still physics-based.

But it gets better.

Once you conquer the globe you can build a spaceship. You can pull back away from the planet and see your entire solar system. You can go to other planets and establish colonies, terreform them, and populate them with creatures from your homeworld.

But it gets better.

Once you have enough colonies in your solar system you can develop an interstellar drive. Now you can pull the view back again and see your solar system and the hundreds of other solar systems around it. You can travel to these other systems. These systems (if you’re connected to the internet) will be solar systems developed by other players. (You’re not competing against them in real time. Your computer is just using their content for it’s simulation.)

Now the real game begins.

Now you can do a whole bunch of different things. Fight interstellar wars. Make first contact with primitive aliens. Tinker with evolution on different worlds. Blow up planets. Spread civilization throughout the universe. Ultimately you can pull the camera back far enough to see your entire galaxy at once.

Transitions between the different levels are all seamless. It’s like watching the old “Powers of Ten” movie. You can literally zoom smoothly from the galaxy view right back down to a tidepool on some alien world.

I haven’t been this amazed by a game since I first played the original Civilization back in the early 90’s. And believe me, I’m a very hard guy to impress.

Will got a standing ovation.

He’s going to sell MILLIONS.

Sounds tres cool. Any hint of a release date that will be missed by 6 months?

I already have my SC4-Rush Hour and The Sims2 addictions battling for my limited attention span. How the hell am I going to fit THIS in, too?

And thank you for giving me the timetable for my next new computer.

Will Wright is a maniacal genius! :smiley:

I think I just shit my pants. Just from the text description, this is jaw-dropping.

I can’t wait.

Sounds far too ambitious to pull off successfully. Count on at least 3 of these ‘levels of zoom/gameplay’ being cut from the final product, AT LEAST. Probably more.

I’ve been fooled before too many times by game developers who want to put more into games than the industry is willing to invest. Fable, Half-Life 2, Black & White, countless vaporware titles that never came close to going gold…

It sounds like a dream game, and if it turns out to be more than a dream, I’ll eat my shorts. Or at least chew on them for a while.

Yes, but two of the games you’ve cited are developed by Peter Molyneaux, a man who has become a laughingstock because of continually promising the moon and delivering a muddy hill.

The man in question here is Will Wright, who really HAS continually pushed the envelope when it comes to simulations. The Sims started off as an architectural tool!

I’m confident in my anticipation! :cool:

Exactly what kind of a computer would you need to run such a monstrous game?

Wright is one of the great ones, but there is such a thing as too little focus. Every phase of the game described would be difficult to do right by itself. “SimCity 4” was a flawed game in a lot of respects, and went a bit too far towards mechanics and away from fun. Trying to geet it right five times over would not be easy.

Where Wright has succeeded in the past is NOT in making huge, ambitious games. “The Sims” is not a complicated game. Where’s he’s succeeded is in making software that’s DIFFERENT - stuff that isn’t the usual game fare. Lots of people can make another RTS game, another war game, another first person shooter; Wright was the man who came up with the idea of making city planning into a game, or just running a household.

I’d love to try it but I’m afraid you’re going to need a monster computer to run it.

Unlikely. This wasn’t just a talk. It was a live demo. I saw actual gameplay of each of the levels and smooth transitions between each.

Now they still might screw up the play balance. But as for the implementation they seem to already be post-Alpha.

Nothing I saw struck me as being out of line for the current state of the art. I’ve seen the sort of physical simulation they’re doing already in games like Magic Pengel and Graffiti Kingdom. The cross-fades of different scales have already been done in Katamari Damacy. Several of the simulation levels cover design ground that Wright knows well from products like Sim City, Sim Earth, and Sim Ant.

Another key element in what he’s doing is making heavy use of procedural content generation. So, for example, you don’t texture your creatures by painting a bitmap. You do it by tweaking sliders in an animal camouflage pattern tool. This is key because it allows ungodly compression of assets – for example, Wright claims that all the data for a creature can be stored in less than 1k. This makes it possible to handle large amounts of game content without requiring a wide network or system bus data pipe.

Wright didn’t announced the min spec in his talk. But except for a few glitches here and there it was definitely running at an acceptable frame rate. And, in my professional opinion, nothing I saw was out of the question for a current high-end PC with a good video card.

http://www.1up.com/do/previewPage?cId=3138792&did=1

Will Wright talked about this game briefly in a online chat for The Sims 2 at the Sims official page a year and a half ago, refering to it as “Sim Everything.” Most people thought he was joking.

Doesn’t matter. If that game is half as cool as it sounds, I’d buy whatever it takes.

Does anyone know when NASA’s next garage sale is?

I’ve been looking this up on the web since reading the thread, and the SOONEST projected release date I’ve seen is 2008. That’s not post-alpha stage.

I’ll grant Wright is a genius and he can’t possibly make a game worse than “Master of Orion 3” or “Black & White.”

I’ve written up the notes that I took at Will Wright’s demo of Spore:

http://www.DonHopkins.com/home/WillWrightSporeDemo.html

I’ve been fortunate to know Will for 13 years, and I’ve worked with him on The Sims and robots. Spore is his latest Magnum Opus!

-Don

Honestly, I think I’d hate playing the game. I plan to buy it anyway. Because it sounds so fucking cool.

I like the idea of getting content created by other players (without directly comping with them)

Brian

Uh, cite?

According to Will they’ll be formally unveiling the game at E3 this year.

No one shows a game at E3 three years before their projected ship date. That’s commercial suicide. Now that I’ve seen the demo if you gave me a team of 100 people I could reverse-engineer the game from scratch in 2 years.

(That’s not to say that games don’t slip and wind up shipping three years or more past their public debut. But no one plans a three-year marketing lead.)

From the state the demo was in I’m guessing a street date of no later than Christmas '06.

For the curious, pics of the demo:

Spore Pics 1
Spore Pics 2

Will Wright is as clever of man as you’ll find when it comes to building interesting software toys but I don’t see how he’s going to pull this one off. The problem is that he’s not building one or two or even three simulations (I can’t call them “games”’ Wright himself has been hesitant in the past to refer to his projects as “games” since they lack goals beyond what the user puts into them) he has roughly a dozen to put together there and making them all compelling will take nothing short of a miracle. I fear that the results of this is going to be some cool parts dragged down by some weak parts and not a very entertaining program.

Of course, if anyone can do this it’s going to be Wright and I hope he pulls it off since it would be so cool. There’s no technical reasons this can’t be done (the vivid descriptions in the first post make it clear that he’s not really doing anything more complicated than the actual act of smooshing a dozen simulations together) it’s just those design hurdles.

Me wantee!

I mean, it’ll be tricky to go from Pac-Man to Diablo, to SimCity, to Civilization and beyond without getting bored at one stage. But, still, me wantee!