Monthly Archives: April 2012

Shall We Try This Again?

Hey Everyone! Err, this is a little awkward after all this time…

First, here’s something fun. Last year I worked with a couple of guys from Koelstof (some of you may know Jarno Cordia, who works with Phoenix Fly) to produce a prototype for an arcade wingsuit game. We put something together in a couple of days, and the result looked like this:

You can try out a webplayer version as well! (Requires Unity browser plugin)

This thing has been lying around for a year, and none of us were really doing anything with it. Recently some kind folks contacted me to ask if I would help them build their game, or if they could use parts of the source code for it. The game has a different theme, but shares some of the core mechanics. I’m working out a deal with them right now, so with any luck you will see this thing come to life soon.

(Just to be clear, this is not Volo, nor is it related to it. Volo is very much on the simulation side of the spectrum, while this game is distinctly on the arcade side.)


So what about Volo?

I have spent the last year setting up my freelancing business. I started out as a complete newbie, and have been stumbling my way towards a vague sense of competence ever since. I’ve been meeting a lot of interesting people, and have been learning a lot. Scoping projects, not not biting of much more than I can chew, and getting work done in time. I renamed my business to Ramjet Anvil (an almost-anagram of my name), and I am looking forward to working under that banner.

But crucially, I burnt out on Volo big time last year. It’s very ambitious for a first game, and to think I could pull it off was at least a little naive. Perhaps I should have done a 2D platformer first, heh.

Progress ground to a halt. I kept staring at the Unity project for days on end until I was banging my head against the monitor in frustration (no joke there). It got to the point where I figured I should cut my losses and move on to a completely different project; something a little more suited to my skill level. I thought I couldn’t realize this game, and I did not want to make promises to you guys that I would not be able to keep.

Meanwhile, every day I received youtube and facebook comments along these lines: “Holy shit! I’m throwing money at the screen but nothing is happening!“, as well as: “Hey man, are you still working on this game? It’s so sad to not see any updates…” These comments are very heartwarming, and they slowly made me realize that I was onto something too big to just throw away.

So now I want to try it again!

The biggest problem that I faced was that I wasn’t able to handle the sheer complexity of the physical simulation. With so many moving (body-)parts, and without adequate tools, it took weeks to understand the consequences of even the tiniest gameplay tweak. Not instantly seeing the effect of your changes has a devastating effect on your productivity. Imagine trying to draw a picture, but you only see the lines you draw appear half an hour after you’ve drawn them! It’s not impossible to work that way, but it is extremely difficult. I only really realized this after watching this wonderful talk by Bret Victor, a designer/engineer with a resume so impressive you’ll fall of your chair. If you’re at all interested in the fields of design and engineering I highly encourage you to check this talk out:

(And read his article on the Ladder of Abstraction; it applies so much to my Volo troubles I almost cried.)

So, I’m picking things up again, but taking a different approach.

First, the focus is not on realism, but on gameplay. This is to avoid the pitfall of adding lots of complex stuff even though it isn’t needed for the play experience. For example, most players think there is a very complex wind model in the game. There isn’t; so why would I focus a lot of time on putting one in, at least early on? It’s good enough as it is.) I’m also looking for a stylized look for the graphics. Something simpler to produce, better at communicating the simulation state, and something that will not look old and ugly the day the game comes out. A stylized and slightly fictional look would also enable me to get away with not simulating some things. Just boring things though. 😉

Second, I need to build adequate tools to deal with each bit of complexity I add to the simulation. Unity’s default tool set is great, but it will need significant additions if I am to tweak aerodynamics with it. I need wind tunnel testing with exquisite graphical readouts, and I need it now.

Third, multiplayer code needs to be in there right from the start.

With those things in mind I first plan to get some basic aircraft flyable on a multiplayer server. Once that is up and running add first/third person walking, getting in and out of vehicles, switching a character’s gear, and then finally wingsuit flight.


As always, let me know what you think!

And thank you for staying interested, it means much to me.