Interview with Kim Goossens

My passion for computer games dates back from the 70's when my father showed me an amusement hall at the Belgian coast. I was completely hypnotized by the interactive mechanical games. It was already since that period that I wanted to do something with game development. But it was when I saw Tron for the first time that I knew for sure that I wanted to become a cg artist.

Yes, but I started my first 3D graphics on the Atari ST using antic's 3D software Cad 3D 2.0. It allowed me to render these flatshaded 16 color images out of a palette of 512 colors. Later on I changed to the Amiga when I saw the renderings of sculpt 3D in a magazine.

I was creating game graphics on my own using Deluxe Paint. And I was looking for a programmer who I could work with. I heard about a computer hobbyists' group called "Hobby Computer Club". And it was there that I first came into contact with Peter Verswyvelen. I showed him the graphics I've made and he really wanted to find an opportunity to work together. So he asked me to create the title screen for Ziriax. Peter is a very friendly and open guy and we soon became good friends.

I always created sketches on paper to get a clear idea about the design of what I wanted to do before I even touched the computer. The drawing is completely done by hand. I started drawing guidelines and curves to get the framework of the spacecraft ready. Then I started blocking out the image with colors and I started adding detail. During the detailing I was adjusting and testing different color palettes. I tested 32 colors, HAM and half-bright. Although I liked the amount of colors I could use with ham it was too difficult to get a sharp looking image. The pixels being placed were influencing their neighbors. So in the end I choose for using the half bright option which gave me 64 colors. It required some palette fine tuning to get the maximum out of the colors.

All the rest, antialiasing and gradient dithering is just manual mouse clicking labor.

Also in this case Peter contacted me for doing this game. It seemed like a great opportunity to me to work together with him so we went for it.

I was so enthusiastic to be able to create a game that I just looked at the basic concept and started creating images for it. When we showed the first demos, the guys of 21st Century Entertainment where happily surprised with the visuals and I basically got a "carte blanche" to continue.

I wasn't aware of that?

Yes absolutely, especially the first level. Later on I changed to other styles. But I was also influenced by Barbarian from Psygnosis

We wanted to make an impressive game. We were both inspired by the arcade graphics that were around that time, so we wanted to get to that quality as closely as possible.

I used Deluxe Paint and a scanner to scan in some drawings. Later I cleaned the images and colored them. The level tiles were drawn directly in Dpaint.

The palette contained 32 colors except for the background gradient. There we created an animated dithered gradient. It was an oscillating dithering pattern that animated at 50 frames per second. This allowed us to emulate a much more colors then the available 4096 color the Amiga had.

Some of the images were first drawn on paper and scanned. Other images I created directly in Dpaint. But I really don't remember, and those days I wasn't a hero in keeping track of all the hours I was spending on it, but some of those drawings where quite intense to create. :)

The global idea was that the barbarian would work his way out towards heaven saving all the fairies. So the last level became a sky level and therefore led to a shoot-'em up. Today I would probably have taken a different approach.

We wanted to aim for a graphically strong game. Actually we were young and a bit rebellious. 21st Century Entertainment (still being Hewson at the time) asked us to do a straight conversion of the C64 version to the Amiga, with some better graphics of course. But we wanted to take it way further. So we made a first level in which we've been putting in a lot of effort to show what we had in mind. And we've convinced them to take this route. Creating the game as it is right now required much more working hours and required much more memory. If I recall correctly, I think we couldn't afford to make it a 4/5 disk game.

Other than a few tries I made at home? No, not in a professional context.

It was way more complex. Instead of working as an individual in which you have full control, we'd have to work with a much bigger team. So you are suddenly dealing with all the aspects of management. Also the platform you are dealing with and all the assets that had to be created where that much more complex. But we had a great team. It is again sad to hear it was being released that late. The game was literally catching dust on the shelves for 2 years after it was finished. Although we were receiving some good feedback from the reviews, it was kind of bitter to hear then that the shooting controls where ripped off from geometry wars.

I'm a lecturer teaching procedural modeling, Digital sculpting and Compositing. I'm also supervising students that are doing their specialization, graduation and internships

What procedural modeling does, is allows game companies to create intelligent tools. This means that a lot of the manual labor is taken away, while still having a lot of control over the end result as an artist. The method allows parallelizing production flows.

For example, if one wants to create a race game in a traditional workflow. One would have first the cars behavior being programmed, simple road mockups being play tested and then the art levels being created. But this is a very rigid and destructive workflow. If for example an adjustment needs to be made on the behavior of the cars, then there is this whole chain above that will suffer the consequences.

In a procedural workflow one can start creating tools that would generate the environment based on the simple layout of the road being created. And so when adjustments have to be made on the layout of the game the impact will never be as severe.

Because the descriptive nature of procedural modelling, the entire content one is creating, is truly adding valuable assets for the company. For example, if you would have modeled a bridge the traditional way, you can throw that asset away once that game is being made. In a new game it will not fit anymore in the given context and it probably will be outdated in terms of polygons and texture quality. With the procedural method however, one has created a tool that can be reused in a different context and can be extended in terms of quality and flexibility.

If I play games I often play retro games. :) It is more out of nostalgia than anything else. I do like to see what is being created in new games. But my passion truly is the creation of games and my hobbies are still a continuation of the work I do.