Hello! I'm Tom. I designed a game called Gunpoint, about rewiring things and punching people, and now I'm working on a new one called Heat Signature, about sneaking aboard randomly generated spaceships. Here's some more info on all the games I've worked on, here's the podcast I do, here are the videos I make on YouTube, here are some of the articles I wrote for PC Gamer, and here are two short stories I wrote for the Machine of Death collections.
The theme for this weekend’s game-making competition is evolution. As usual, I’m gonna stick to working on Gunpoint but write up the idea I’d do if I had time to get distracted.
I think if you’re going to make an evolution game, you’ve got to actually model evolution. God knows gaming misuses that word enough – we need to repay science for every time a game has claimed some magical goo caused our character to ‘rapidly evolve’ into a superhuman.
I can’t think of a good way to put the player in direct control of that process, because that process is pretty much defined by the lack of direct control. Spore is a good example of what’s not good enough in this context: the player chooses how his creature ‘evolves’, making it essentially a game about intelligent design.
So how about enemies that evolve? That keeps the theme nice and close to the central experience: you’re always going to be thinking about and dealing with enemies, so anything that governs their behaviour and traits is going to be super important to you.
It also suits a game’s natural difficulty curve: you want it to start easy and get harder, which fits well with enemies that start dumb and get better at fighting you. But how would we measure ‘good at fighting the player’? And why would that lead to that enemy reproducing? When and why would you end up fighting its offspring?
You can fudge those questions, but you don’t want to stray too far from how evolution actually works, or again, your game ends up just evolution-flavoured rather than genuinely about the mechanics of it. I tried to think of an idea that doesn’t need a lot of fudging for evolution to fit with it, one where the evolution mechanic doesn’t have to be dressed up or reworked from what it naturally is. The best I could come up with was this:
You play a beast lying in a vast, dark cave. Roaming around the cave are lots of much smaller glowing creatures, each slightly different. As soon as you move, they all realise you’re awake and run like hell.
Any time you get near one, a sticky web emits from your body and ensnares them. You can eat them quickly and easily. As you do, a meter fills up: you’ve got to eat about 80% of the population to fill it. When you do, your movements slow to a stop, and you fall asleep. For forty years.
In time-lapse, you see the cave repopulating. Each creature has its own values for Speed, Strength, and Resilience: random pairs of the survivors reproduce, having ten children each before they die. The children inherit a mix of their parents stats, but each child has a chance to be better at one particular stat. If they are, one or all of their others will be reduced, so their total is the same.
By the time you wake, all of the original survivors have died off, and only their children remain. Again, you’ve got to eat 80% of the population in order to return to your slumber. It’s probably the fastest who survived your last attack, so this generation will probably be a bit faster.
After a few slumber/feast cycles, you develop the power to shoot your web-strings at fleeing prey to ensnare them. The strongest creatures can pull themselves free before you reach them. So in this generation, strength is more important than speed.
After a few more, your webbing becomes acidic, slowly killing prey it catches. Now Resilience is a factor: even amongst the strong, only the toughest live long enough to have a chance of freeing themselves.
From then on, every few cycles you can choose which of your abilities to improve: speed, web production, web acidity. This is not evolution, it’s just your development as a maturing creature. But you can choose these to counteract the dominant traits of the prey population – produce webs faster if they’re all too fast and getting away from you too quickly. And the population will, of course, evolve in response, as only the best at avoiding you will live to produce the next generation.
You could even have a ‘Social’ trait for the prey, which determines their willingness to help trapped friends out of webs. And maybe a fear pheromone for yourself, to counter-act it. You could intentionally kill everything with webs for a few generations, to see if the Speed trait starts to decline as it no longer offers an advantage, so offspring who lose it in favour of other traits start to do better and better. Then feast on the slow generation.
Also there’d be unlockable hats.
More Game Jams
Risser: Oh my God, that's a GREAT idea! I love it!
Jack!: Oh wow. Why is every one of your game ideas the greatest thing ever? You should start your own company and get them all done. It would certainly let you get Gunpoint out faster!
some guy: This is the best evolution game idea I've ever heard of. I love AI and Genetics and have been duly disappointed by games such as Black and White and Spore in the past. Perhaps a kickstarter is in order?
pascal: I'd simulate the creatures at a lower level, like some artificial life research does, as an evolving construct of limbs and sensors. It will take a lot more generations of creatures to get a challenging game, but it would be more obvious to the player that evolution is going on (when the children inherit the extra limbs from their parents, or more specific behaviour like a tendency to steer to the left due to sensor placement)
flaillomanz: I love this idea. I've noticed that there hasn't been a single good, or even remotely original, flash game release on the internet in a YEAR. Sure, there has been some half-decent ones, but the vast majority are carbon-copies of "Generic Platformer: The Game" or "Colourful Click-a-thon: The Reimagining". The worst offenders are actually the old hands - companies that once released a couple of brilliant games and never made anything original again.
I mourn the days of good flash games; from the jaw-dropping Morningstar to the absolutely fantastic "Robot Wants..." series.
Alas, those days seem long gone... and this saddens me.
It isn't even for lack of great ideas, as yours shows - there has to be others out there capable of imagining things for themselves.
I've said too much.
Pope Jamal: Very interesting idea!
Tony: Great idea. Also, this is exactly what Mass Effect looks like if you're a Reaper.
William Russell: Make ittttttttttt!
Christian Walde: This has actually been done already. Look up Warning Forever. :)
Jazzels: If I play it do I get a Web hat in TF2?
Great idea though.
Jason L: Warning Forever is a good touchpoint.
Overall I'm not feeling it though. The constraint about equal points in a total of three niches feels like it would rapidly devolve, hah, into an eternal cycle at best. In such a trivial situation, evolution possesses no explanatory power - the power of natural selection is in freezing interesting environments into genes, with the most interesting elements usually being other genes in arms races, symbiosis, relatives and what have you. And the beauty of it's all about blindly breaking out of that kind of artificial constraint to have someone else's cake and eat it too.
Probably just a change to a more chaotic selection criterion could make it great. Maybe if the player avatar left trails in/pushed against a gelatinous particle fluid, and the targets adapted to camouflage and/or move in that environment, that could lead to interesting interactions like the player constraining their own motions to distinctive shapes or something. I do love both the design power and the poetry of pitting development against adaptation - ontogeny vs. phylogeny if you will. I'm getting sort of a Godfellas vibe, the interaction between one long-lived slow creature and fast ones is great. Maybe bullet-time of some sort is in order?
Me, I'd probably just go the literal route and do the bacterial resistance dilemma. Arena twin-stick or similar genre, or unrelatedly turn-based tiles. Spawn steadily? Waves? Anyway, any enemy that survives being near your attack for a certain time has 'escaped' and is now unaffected by your weapon until you run and get a new powerup. Let them spread immunity by touching if you want to get literal about horizontal gene transfer. The dilemma between watching their forces grow while waiting until you can commit to a sweep that will leave zero survivors from an area is the interesting bit. Maybe their movements could be dictated by GA'd Markov chains, so ideally they become more frustrating to your personal movement style as the game progresses. Fun!
Or something about the Baldwin Effect is always fun.
Chris Pearson: You had me at unlockable hats! (P.S I would definitley play this game)
Popeye Doyle: What would happen if the player continually attacked the fastest, most able glowing creatures? Would that ensure the next generation were weak? Would there be mutations that allowed for even weaker creatures to appear? Could you create a race of sheep?
I think the game would need this ability (to be about evolution), but it would also potentially undermine the levelling up side of things, making it an interesting if (again potentially) short lived sandbox.
John Waters: http://www.youtube.c... ...r_embedded
RakeShark: It's a good start of an idea, and a good core for a flash experiment. However, the evolution experiment I feel isn't just about predators and their preferred prey (and how they hunt with acid-webs), but cognitive evolution. There's another of my kind in this habitat with the same abilities, how do I keep my hunting grounds, do I share them and work together or am I territorial? Or do I leave for greener pastures and let him deal with what I couldn't catch? Perhaps there's an invasive species better than me, how do I find a better link to the food chain?
The one think Spore REALLY didn't get right was competition within a single predator species, not just competition to survive within the prey species.
NounVerber: I had a similar idea to this.
Instead of eating enemies you just have to FIND them (by clicking on them). Imagine bugs on leaves. Every bug has a randomly generated shape and texture governed by a number of parameters. After you find most of them, the remaining ones have offspring. Over time, they will start to resemble the leaves. They will start by being green, then imitating the general form and maybe even the motion leaves do in the wind, just like in real life! Of course, they can't be allowed to become too small they will have to have a minimum area.