The demoscene group Conspiracy took 1st place last April at Revision 2015 with this demo.
The demoscene has its own culture of computationally enabled art, frequently overlapping with procedural techniques.
Like most modern demos, it’s available to download and view on your own computer, if you have the hardware to handle it.
The official deadline is past and there are 96 entries this year. Go check them out! The talks have been posted on YouTube (and embedded at the top of this post.)
Timestamps for talks:
4:10 - Michael Cook - Welcome To PROCJAM!
15:58 - Kate Compton - Let It Grow: Practical Procedural Generation From The Ground Up
49:54 - Sean Oxspring and Kieran Hicks: Stuff And Things (Weird APIs)
1:22:53 - Dan Stubbs - The Hit
2:17:25 - Allison Parrish - Speaking Terms: A Few Thoughts On Verbalising The Semantic Web (YT Version Forthcoming)
2:46:15 - Tommy Thompson - First Foot Forward: Procedural Generation Research In Sure Footing
3:23:00 - Tom Betts - The Wider World Of Generation (With Musical Intro!)
4:28:00 - Complicated Intro To Thrice’s Talk (there’s some minor audio issues in the first minute also - sorry!)
4:30:55 - thricedotted - Found Lexicons, Generated Grammars: The Dynamic Possibilities Of Twitterbots
5:02:10 - Michael Cook - Closing
(If you’re still working on your submission, ProcJam has always accepted late submissions, because it’s not the kind of jam where that matters. Go ahead and email Michael Cook or tweet at @mtrc for a submission link!)
As 2015′s ProcJam is wrapping up (don’t forget to submit your projects!) let’s take a look back at another ProcJam project from last year.
Diversitizer, by runevision, is “a simple study in environment diversity”.
I think it’s a clear example of what I’m talking about when I mention the value of structure and order in procedural generation. Just throwing down plants could quickly devolve into boring noise. Instead, the generator looks at the hidden parameters underlying the landscape, resulting in clumps of plants that feel right and hold your interest for much longer than a completely chaotic process would.
Because there’s an order to it, you can start to discern landmarks and anticipate what you might find next. And that anticipation is important, because without a pattern to set the players’ expectations, how can you hope to surprise them?
Realtime Sound Visualization - PROCESSING from Jonas Wedelstädt on Vimeo.
Here’s a cool sound visualization project. Done in Processing, it apparently runs in real time, which is the part that interests me. Going by his site, Jonas Wedelstaedt is a Visual Artist based in Berlin. He’s done some other real-time music visualizers, like this one, and this one.
Since this is ProcJam week, there are a lot of ProcJam projects I’d love to talk about…but they aren’t finished yet. So instead, I’m going to talk about one from last year’s ProcJam.
The Inquisitor, by Malcolm Brown, is a murder mystery with a procedurally-generated murder to solve. You get to look for clues and interview suspects, trying to work out who committed the crime and why.
Simulating a murder like this is, I think, one of the best approaches to implementing a video game mystery game. The classic whodunnit already has enough elements to be approached as a system, and it avoids some of the problems with interactive detective stories, where you have to second-guess to author to figure out the clues. Here, the clues are spread liberally around and the puzzle is figuring out how they all fit together and who has been lying to you.
I happen to think that a more sophisticated version of the murder mystery system here could easily turn into a genre by itself. Though, of course, it helps that it’s abstracted: the clues are allowed to fit into the game’s toy world, rather than trying to force them to be too realistic. Likewise, the framing of the game, where you are an inquisitor rather than a modern police detective, lets the game pleasantly get away with the charmingly intrusive prying that keeps the mechanics simple.
You don’t have to know how to program to make things with procedural generation.
Want to participate in ProcJam, but don’t know how to program? Here’s an example of procedurally generated music from over a hundred years before the first computer. A Musikalisches Würfelspiel is a musical dice game, letting the player assemble a piece of music (frequently a minuet) by rolling dice and using the result to pick the next section of music.
The piece in the video above is generated by Johann Philipp Kirnberger’s Der allezeit fertige Menuetten- und Polonaisencomponist (German for “The Ever-Ready Minuet and Polonaise Composer”). There were many other generators created by different musicians, including two allegedly composed by Mozart.
There have been a few online implementations of Mozart’s version, but most of them are apparently out of commission (including one from 1995, which I can’t get to output midi files at all). If someone’s looking for a programming project, implementing Kirnberger’s, C. P. E. Bach’s, or Maximilian Stadler’s music generators is a worthy goal.
These generators are demonstrations of how it is often just as useful to have a good aesthetic sense of the thing you are generating. Creating a music generator can be more dependant on your knowledge of music, rather than just your skill in programming.
But what if you can’t program at all? Well, these musical games show one way to make a procedural generator that doesn’t need programming. Just a bunch of source material and some rules for randomly combining the parts. Maybe your good eye for drawing or painting, or your ear for music, or your interest in something else can suggest some way for you to invent a system for generating new things.
(video via https://www.youtube.com/watch?v=3SQYWsfL_Fo)
Both ProcJam and NaNoGenMo are in full swing, so this is the perfect time to highlight useful programming resources!
Instead, I’m going to talk about git-man-page-generator. Git is an open-source version control system for developing software. Version control is important when you’re working on a large programming project, because it lets you keep track of what changes have been made. There’s nothing worse than having a working system, making a small change that breaks it, and being unable to fix it to the way it used to be. With version control, you can just revert to your earlier version. It’s even more important when you’re working in a team, because good version control can prevent you from overwriting your teammate’s work.
The catch is that understanding git can be tricky. While the basics are relatively simple and there are tools to help, the arcane intricacies of how it operates can baffle most people.
Created by Lokaltog, git man page generator is absolutely no help for that problem. Instead of displaying a useful page from the manual, it creates a totally new one, which probably isn’t even real.
Oh, and for an actually useful resource for those of you doing your own procedural generation projects right now, the git-man-page-generator is based on Baba, which is a designer for text generators. Try it out!
Generate your own git man pages: http://git-man-page-generator.lokaltog.net/
I’m looking forward to seeing what you all make!
Tamperdrome Collection (by G. P. Lackey, who was also involved in GenerateACat) is a suite of generators that make cats, islands, goblins, dogs, inhabitants, and ghosts in response to your input. It, or parts of it, have apparently been exhibited in physical form.
I like projects like like this that aren’t pressuring themselves to be more than they are. While I can think of all kinds of uses for these tools as a part of a larger interactive sim (and will probably borrow some ideas from here and from Strangethink in the future) I admire the way that the Tamperdrome Collection is more interested in being what is it rather than trying to be something that it isn’t. Generative art can be an end in itself.
Continuing the theme of procedural generation examples in Unity, here’s a student project that combines four different generation processes to create a village. The generators are available in a Unity project if you want to play with the generator and see how it looks in action.
The generator was originally created as a student project to explore some of the different ways that procedural content generation is used. The research writeup that accompanies the project is interesting in itself, especially if you’re looking for inspiration or ideas for your next procgen project.
You can find TOWN here: http://delca.itch.io/town