Help
Subscribe to Genesis
1 Comments
I'm watching Mira playing Diablo II at the moment, and thinking quietly to myself about how many times we've played it over now, without getting terribly bored. One of the biggest points, I figure, is that the levels, monsters' attributes and a few other things are generated at random, ensuring the experience isn't the same every time. These are done via a formula that's been programmed into the game, providing parameters and rules which the game obeys to generate a unique experience with each playthrough, requiring no further input on the part of the developers. This is a kind of primitive example of what's called procedural generation; using a formula to get the computer to generate something.
For a really extreme example of this, you should check out a game called .kkrieger, a shitty game in and of itself but a fantastic proof-of-concept. The game is under a 1/2 megabyte to download but produces graphics comparable to your typical first person shooter circa 2008. We're starting to see this process used more in videogames, and I can see the reasoning for it; it takes the gruntwork off the designers' shoulders and lets them do more in the way of actual game design as opposed to concentrating on anal details that don't make much of an impact to the user's enjoyment of the product. That makes sense, as demand for bigger and better games is way outstripping the resources of a typical, honest development house.
What I was really wondering about is how this kind of procedural generation has the potential to be implemented in pretty much any sort of media, if an engineer were clever enough. On a primitive level, we could use it to make a pattern to use as a background for a website. Why not use it in the aural sense? Random sound effects to use in movies, or dare I say it, entire music that's randomly generated according to rules specified by the person wanting to listen? How about books that are written procedurally? That probably sounds retarded, but it's already been done in a crude manner. We've all seen 'random story generators' that piece together a fable by arranging pre-written events and outcomes. Employed on a more advanced level, it doesn't seem far-fetched to me that in the future, we might just have authors that write in a particular style, then see that style adopted by a computer to write hundreds of unique books in that style.
But of course, what makes a particular influential book or record to be so influential in the first place, is that it tried something new, that went against the formula.
Maybe this wouldn't be a great idea after all. Maybe we're already seeing it done anyway, in most of the movies we watch, the music we listen to and the books we read; assembly-line, cookie-cutter content that's pumped out by people, yes, but people thinking in a very computer-like manner. Why does so much music sound so similar on the radio?
For a really extreme example of this, you should check out a game called .kkrieger, a shitty game in and of itself but a fantastic proof-of-concept. The game is under a 1/2 megabyte to download but produces graphics comparable to your typical first person shooter circa 2008. We're starting to see this process used more in videogames, and I can see the reasoning for it; it takes the gruntwork off the designers' shoulders and lets them do more in the way of actual game design as opposed to concentrating on anal details that don't make much of an impact to the user's enjoyment of the product. That makes sense, as demand for bigger and better games is way outstripping the resources of a typical, honest development house.
What I was really wondering about is how this kind of procedural generation has the potential to be implemented in pretty much any sort of media, if an engineer were clever enough. On a primitive level, we could use it to make a pattern to use as a background for a website. Why not use it in the aural sense? Random sound effects to use in movies, or dare I say it, entire music that's randomly generated according to rules specified by the person wanting to listen? How about books that are written procedurally? That probably sounds retarded, but it's already been done in a crude manner. We've all seen 'random story generators' that piece together a fable by arranging pre-written events and outcomes. Employed on a more advanced level, it doesn't seem far-fetched to me that in the future, we might just have authors that write in a particular style, then see that style adopted by a computer to write hundreds of unique books in that style.
But of course, what makes a particular influential book or record to be so influential in the first place, is that it tried something new, that went against the formula.
Maybe this wouldn't be a great idea after all. Maybe we're already seeing it done anyway, in most of the movies we watch, the music we listen to and the books we read; assembly-line, cookie-cutter content that's pumped out by people, yes, but people thinking in a very computer-like manner. Why does so much music sound so similar on the radio?
Share This Entry:
1 Comments On This Entry
Page 1 of 1
Khorosho
19 June 2008 - 02:32 PM
It's an interesting thought! Maybe doesn't work for complicated intellectual work like books, songs and movies cause they require a level of creativity and sense of aesthetics which only humans can develop. But could indeed be used to create a unique web experiece.
Page 1 of 1
