🎉 Celebrating 25 Years of GameDev.net! 🎉

Not many can claim 25 years on the Internet! Join us in celebrating this milestone. Learn more about our history, and thank you for being a part of our community!

SCIENCE (biatch!) for Big Games! (part 1)

Published May 05, 2017
Advertisement

(For the full series, click here)

The image atop the entry is from my last game test. Although the polygon style is becoming problematic (it's an optimization issue), it shows an unexpected result of the world generation. I have programmed in mountain ranges along tectonic plates, shield-volcano mounds, simple craters, sea level depicted by vertices turning blue, and a lot of other things. Right now, I'm throwing every science I can grasp at the dart board and seeing what sticks. Fire and forget. Clean up later. Which I guess means you didn't forget, but... Anyway, I did not foresee this particular result, although in hindsight, I should have: It's a mountain ridge that snakes up out of the water and rises to the point that first its top, then the whole thing, become covered in snow. I never saw mountains do that in real life, but I'm not a geologist, I just play one on TV / my blog. It was not deliberately programmed into the game, but in the process of doing a ton of things, it appeared.

This, to me, is the essence of something being 'natural'. It was not designed to exist, it just became the inevitable result of the universe (in this case, my virtual game universe) existing. I love going for nature walks (stop laughing, some of us geeks do), and I love looking at trees or rock formations and trying to spot how they came about without deliberate sculpting. It amazes me. When I started research for my game, it quickly came to involve every science I could find; starting at quantum mechanics and nuclear physics, going up through chemistry, geology and climatology, through biochemistry and larger bological issues like evolution, up to neurology and springboarding it through ethology (the study of instinctive animal behaviors as a biological phenomenon) and psychology to land in sociology, economics and political sciences, rounding out with various engineering disciplines. Study any one of those fields or their subfields, and you have plenty to deal with. But for me, studying any field is just foreplay. The real meat of it is in placing the things you learn in a greater, continual perspective. How did dead chemicals become life and then form brains, and how did those brains get to the point of inventing corporations to build microchips? Each science is a node. My passion is the connections between the nodes, the red thread that runs through it all.

The thing is, just like my snake ridge, reality is a chaotic mess. Believe in whatever god you will or won't, but from the human perspective, it's chaos out there. No clear master plan, just eons of "this apparently works, so it will continue, and the rest won't". In games, it's called "emergent gameplay", the stuff that is made possible by a game without being planned. It's how Minecraft became a tool for homemade parkour courses, and how Halo games became trick shot simulators. It's what happens when people know the rules, but do not care one bit, and make up their own. When we look at nature, or at civilization for that matter, we think we see purpose. That branch grew there because the tree wanted to grow wider. That company wants to make money from this or that idea. The human brain is designed for detecting patterns, so ancient people saw trolls and gods controlling (is the word controlling related to just trolling in some old way? #SuddenTangent) the weather or the earth, and conspiracy theorists see hidden agendas in oatmeal.

That lust for meaning seems to be guiding how modern games are made, and I think it's because we have the tools now to do so. In the early days of video games, you had far less to work with, so you made up crazy things to turn it into a game. Just try to explain Pacman, Donkey Kong, or any similar old games in a way that sounds like a logical backstory, I dares ya! One amazing emergent element of early games was that as there were fewer enemies on screen, machines could process the game faster. Which is why Space Invaders originally sped up as more enemies were destroyed, raising the difficulty in a way that seemed, like genius, to match your skill. Coincidence. Chaos becoming the next big thing. Natural evolution. But today we have every pixel focus grouped and every character checked for realism (mostly visually, but also narratively). It is entirely controlled, entirely artificial. We run it, with no room for chaos.

I'm working on going the other way. Some people ask me why big companies haven't done the things with procedural generation that I'm working on, if it can do what I keep claiming can be done with it. My answer is simple: I do the one thing they will not, which is to work towards relinquishing control. A company must control its product, micromanage every facet of it. I want to create a chaos that works, and 'set it free', so to speak. So I try to implement sciences and let them do their own thing, rather than work towards a strictly defined result.

But considering the massive volumes of science easily available to implement in a game like that, I often have to stop and wonder how far it can be pushed. Elon Musk has (in)famoously spoken about his beliefs that we may live in a simulated reality already, and on some level, his thoughts make sense. The current advances in game technology (and the lingering popularity of The Matrix) seems to guarantee that a game system will come along that can pass for reality. Add some clever medication and people may forget that it's not real.Or the entities being simulated will believe themselves to be real. When we think sentient machines, we think physical robots. But AI is a part of games, too, so why not. Maybe we are those simulated entities... or maybe just you, and you're the only real person, because this is your game. Don't forget to save.

With the math I already have at my disposal, I am convinced that a game can be made that mimics nature to a frightening degree. And that game would then be able to tackle the best argument against the idea of us living in a simulated world: It's so goddamn boring! A game, on the other hand, can focus on the parts of reality that are interesting, the fun stuff. But to do that, to beat nature at its own game (pun totally intended), we first need to dissect nature. We need to know ALL THE SCIENCE and turn it into algorithms and snazzy graphics. I'll try to give my take on that, in the next part(s)....

 
2 likes 0 comments

Comments

Nobody has left a comment. You can be the first!
You must log in to join the conversation.
Don't have a GameDev.net account? Sign up!
Advertisement
Advertisement