IMPORTANT LINKS FIRST
Here’s hype: http://www.gordonmcneil.com/VGCW_Orch.mp3
Here’s the Twitch: http://www.twitch.tv/bazza87/new
Here’s a Wiki: http://vgcw.wikia.com/wiki/Video_Game_Championship_Wrestling_Wiki
Here’s a guy uploading prior matches with chat: http://www.youtube.com/user/VGCWclem/videos?flow=grid&view=0
Okay so here’s the deal: Every now and the a dude named Bazza (AKA Baz McMahon) streams something called Video Game Championship Wrestling on Twitch. The idea is that he sets up matches in THQ’s WWE 2013 where all characters are AI-controlled, and then he lets it go from there while he plays appropriate vidya music and chat goes insane.
If this sounds incredibly stupid, good, because it is. It’s also incredibly entertaining and amusing, but it’s really hard to convey it without watching it, but I really suggest you do so because it’s the highlight of my day when it comes on. Bazza’s got an irregular schedule but he confirmed that there’s a show tonight (Season 2 finale!) that’s going to be seriously epic. So if you like the idea of Donkey Kong and Duke Nukem beating the hell out of each other while Vegeta jobs and Adam Jensen rides into battle on an augmented teleporting motorbike, then THIS IS FOR YOU!
This is Captain Falcon FALCON KICKING Protoman off the top of a cage and through the announcer’s table.This is Ghost Nappa superplexing Gaben so hard the goddamn ring breaks.
Look just take my word for it: You want to get comfy and watch this. Pre-show starts at 10pm UK time, which is 5pm Eastern/2pm Pacific. The show proper starts an hour later.
As Pike revealed yesterday I’ve been working on a mod for Victoria II lately. It came to me while I was playing the rather good Fallout mod for Darkest Hour, because one of the challenges the modder faced there was dealing with the need to have a lot of land empty as ‘wasteland’, for the powers to colonize and claim. If you’re familiar with HoI2 and derived games you’ll know that this isn’t an easy feat, because HoI does not work like that, there are no empty provinces as in EU3 or V2 to colonize.
But hang on, there are empty provinces in those games and mechanics for claiming and settling them. So I thought, why not make a mod for one of them revolving around a similar idea? And now here we are, working on a post-apocalyptic setting for Victoria II, starting shortly after said apocalypse and covering the reclamation of the ruined Earth, the development of new technologies, and ultimately the emergence of new political ideas.
A planned political faction is a cybernetic one.
It’s a lot of work, even the fairly simple stuff like putting in new countries and editing provinces. But what’s really struck me on this project is how tough it is to keep things balanced. Now partly this is because I’m in no position to mod the AI, so I’m working with the thing as it stands, but it’s really difficult to ensure that people don’t just dominate. In the regular game the UK is the dominant power and unless you’re a decent player or the USA they’re staying that way. With so much more land available to colonize however, the ability of countries to simply run away with the game by claiming more and more land is acute here, and my biggest challenge has, as I say, been working with that.
As I work on the mod the problem decreases but it’s still something I’ve been struck by; balancing is something of a totemic idea that holds less real value than might be assumed at first blush. To take vanilla V2 again, playing as Cambodia should not be as easy as playing as France and there’s nothing wrong with that. Still, there needs to be a semblance of the ability to compete even if you can’t expect to conquer the world, and it’s both interesting and daunting to experience first-hand what developers must struggle with every day.
In videogames, difficulty is a difficult thing to get right. It’s one of the reasons multiplayer is so popular after all; to date we’ve not come up with an AI that comes close to a human opponent, outside of chess at least. Now, it’s not hard to just make an enemy hit harder, have more health, or shoot with greater accuracy. Those things aren’t difficulty in a meaningful sense, but they do make the game harder.
Still, there’s not a lot original to say about this tendency to take the easy route and bump up the enemy’s pure abilities rather than their intelligence. What I want to talk about is a different aspect of AIs, which is something I’ve not seen often addressed, but which will ultimately be core in creating convincing enemies who are challenging, but can be defeated.
That aspect is making mistakes. Making believable mistakes, based on oversight, or failure to account for something by accident, and so on and so forth, rather than the result of glitches or the programmer’s failure to account for something. This may not seem like a huge concern while we’ve still got to figure out a way to be outwitted by the AI, but as we do get better at that this sort of thing is going to be crucial to correct for it in order to keep the game both fun and engaging.
A lot of victories in real conflicts are borne from taking advantage of mistakes the enemy makes. Sometimes this is a tactical error, sometimes strategic, and sometimes it’s more deeply rooted and occurs in the years before the war breaks out, when someone’s guess about the important factors of the next war prove to be incorrect. Oftentimes these things will be corrected over the course of the conflict, but sometimes not. In any event the point is that for the player to remain engaged and interested there can’t be an optimum strategy in all situations, which a ‘good’ AI would seemingly be prone towards, and which would thus force the same degree of efficiency from the player.
Of course in the real world there are all kinds of factors that are very hard to emulate. The Confederacy’s best option was probably a Fabian strategy – ceding land for time, and winning by attrition. But the political nature of the CSA meant that border states could not be sacrificed in such a fashion, and they had to be fought for (Well, except when McClellan was in charge of the Army of the Potomac, then not much of anything needed to be done by the Confederates). You can, to some extent, work with this in a game through mechanics like supply lines, dissent, and partisans, but it really has trouble with the nuances of the situation.
Had Lee had this little filly on his side, things would have been different.
Now, getting games to that stage would be a tall order of course. Nevertheless I think we could stand to start thinking about how AIs might make believable, varied mistakes. Things that an astute player can see and exploit, but which the AI might realize and fix very quickly as well. This isn’t a completely untried concept of course, Galactic Civilizations 2 is the obvious example of an AI being designed to do this sort of thing, and it’s a commendable attempt, especially because the AI is actually pretty darned smart without cheating. Halo likewise had some clever foes, for its day, and their dynamic nature meant mistakes on their part could emerge pretty naturally and an observant, smart player could exploit those very well.
What do you all think about this idea? Am I getting too far ahead of our current, braindead AIs, or is this something we should look towards?
I’ve written previously about how strategy games give you a pretty weird angle compared to reality due to how they function, specifically that because they put you in charge of a state and they have a win condition, you become pretty psychopathic with regards to your state. It is only a means to your end.
I’m going to come at this from another angle today. I was thinking about it when I was playing GalCiv, because as I am playing as the Humans I’m sort-of-but-not-quite RPing them as they’re written in the backstory; canny traders, excellent diplomats, with an iron fist in the velvet glove. Now GalCiv has election events that are incredibly trivial. You choose a political party and have regular elections. If your party wins you keep their bonuses (Say, +20% to your influence). If they lose, the bonuses go away until you reclaim control. But if they lose you are still in control. Now from a gameplay perspective this makes perfect sense. Nobody wants to sit back and watch your civ get run into the ground by the AI over the next 30 turns or whatever. That doesn’t make it any more sensible or less jarring; ultimately in strategy games you are your state/country, and anything along the lines of elections, changing dynasties, or anything else is entirely secondary at best.
What’s weird isn’t that they do this, it’s that they try and pretend they don’t. I don’t mind being told “You are the overarching driving force behind the French Empire rather than any particular leader or government therein”. But then a game will turn around and I will be presented as being the particular leader or government, such as EU3 where every notification is addressed to “My Emir” or “My King” or what have you. But how can you address this?
Of course the problem is lessened if you're an immortal Goddess-Queen
The Tropico series has possibly the best approach. You are a tinpot dictator and one of the ways in which your score is evaluated is by how much money you have embezzled from your own country over the years. This is a brilliant little mechanic, because you are actively reducing your abilities in one field in order to bump up your endgame results elsewhere. You’re still just going for the nebulous “score” but it’s something. One idea I had was to essentially provide you with ostentatious monuments to build, of truly obscene scale (Think Bender when he becomes Pharaoh), and the larger you build it the better you are. Civ used to do something vaguely similar where a good performance would make your palace or throne room better, a nice sidebar to the main game, and there’s a mod for Civ IV where you really can lose control of your empire to the AI for a number of turns, an interesting if frustrating feature.
Do you have any examples of this issue being done well? How might a game merge leadership of an in-game actor like a country with being an individual leader? Thoughts and ideas!
No. No it does not. Especially when you’re me, and you love Civ and you really love robots. That’s probably why I’ve had a few people recently link a couple of articles to me.
Basically, some geeks (and I use that word with utmost respect, as I always do!) have taught an AI how to read. Not some programming language, but English. Then, to test their AI’s reading ability, they had it play Civilization II both before and after reading the game manual.
Before reading the game manual, the AI won a little less than half of the time.
After reading the game manual, the AI’s win percentage jumped up over thirty percentage points, to a nearly 80% win rate.
This is super exciting to me on a few levels. Firstly, I’m superhuge into the idea of robots and AI. I envision a future where intelligent creatures of all sorts and designs live together in harmony (I ain’t ‘fraid of no Skynet). I think it’s really exciting that an AI has been taught not just to read, but to learn from a human language.
On a slightly less scholarly level, I’d love to see something like this adapted to make better AI in games. I’ve talked before about how I’d like to see improvements to the AI in Civ; bring it on, Civ-playing robot!
So yeah. As I was saying, it really doesn’t get any better. Unless you get a robot playing SMAC. Then it could be better.
Although this would hurt the poor robot's feelings, I've no doubt. :(
