Wednesday, 12 September 2012

A Future of Polygonal Friends: Final Comments

This piece is split into three parts (Part 1 is here, Part 2 is here), and was written for the Critical Distance Blogs of the Round Table. 

Thanks for taking the time to read 'A Future of Polygonal Friends'!
This was actually an idea I'd had for a long long time, inspired by an article I read in EDGE magazine, talking about AI in games. My initial angle was actually to talk about how we can design better and more empathic NPCs on the whole, but when I saw Critical Distance's Blogs of the Round Table request ('Blogger Jam' would be a way better title, no need to thank me), I set out to take my article concept out of the realm of Notepad musings, and into a properly published format.

And wow. This essay is way too long, hardly ideal for putting on the web. But I felt that to explore the idea in a way that doesn't make people frantically sprint after my own leaps in logic, the first half was necessary. Plus, since this is written with other media bloggers in mind, hopefully those that stumble across this will make it to the end (clearly you have, or you wouldn't be reading this!)

Honestly, I struggled a little reworking this to fit the theme in a way I felt was complete. Writing about AI and good storytelling are both simple enough, but how do you link the two without sounding like an idiot, vague, or a vague idiot? Then I realised the potential was in AI characters telling their own story separate to that of the protagonist's, intersecting when and where the player wants, and there being consequences for those decisions. Story telling and character design need not act at odds to developing AI, but can work with it.

Forgive the large number of Non-AI examples used in the first half; I feel like examples of good design and dialogue without an emphasis on AI would be an important contrast to AI-based solutions. I won't apologise for making fun of Tropers, Power Ranger fans and Katawa Shoujo fans though; y'all are just asking for it.


  1. Good read- a lot of good points, many of which I agree with. I do think it will be quite a while until we see good AI in games, though- not only does it take a lot of processing power, it's hard to do well. Given that video games are now a multi-billion dollar industry, they're largely produced by large companies, and any businessperson at Activision will tell you that it'd be foolish to devote the massive resources required for good AI into a game when many players may not even care or experience it. In a way, we get the games we deserve. While I'd like to say that everyone wants well-written convincing games, even if they don't know it, a part of me knows that a lot of people just want a game that blows off some stress, the art be damned.

  2. I don't know - we may reach the point where the ability to differentiate yourself from your competition matters more, and with the way the Mass Effect and Borderlands series are going, it seems that developers might be finally catching on that characters matter, even if they don't necessarily hit it right all the time. Good NPC AI could be one way for a company to stand apart in a time when "Super High-Res graphics!" or "10,000 quests!" no longer draws audiences.