Gemini (2002)

I designed this fluoro flyer for a small trance party in Edinburgh in February 2002. The flying ‘logo’ and spiral were built and rendered in 3D Studio Max.


Elements (1995)

Elements was a piece of eye candy software I wrote way back when I was in university, in around 1995. The sole point of eye candy programs is to produce trippy imagery. Usually you’d watch them while listening to music. The imagery was often generated using simple mathematical formulae chosen to produce spirals and infinitely receding perspective effects.

Many of these kinds of programs used what was called colour cycling. Computer displays at the time were often palette-based, meaning that the pixel values contained in the image were only 8-bit values and, instead of containing actual colours, were just indexes into a look-up table, or palette. The palette could contain any colours you liked, but was of limited size. Colour-cycling exploited this by cycling the palette: shifting all the entries one place to the left or right in each animation step, with wrap-around.

In the case of Elements, I didn’t actually cycle the palette directly. Instead I would blit in a completely new palette in each frame, which I would cycle manually myself. This allowed me to blend between two different palettes, cycled in opposite directions — something that strict colour cycling didn’t allow. The palette was blended over time between successive palettes picked at random from a pre-defined set chosen for their pleasing colours. The blending between different palettes cycled in opposite directions helped to hide the use of colour cycling and resulted in new blended colours that weren’t in any of the predefined palettes. Usually, this was a good thing.

As well as colour-cycling, Elements also used a few more sophisticated tricks. While the palette was cycled and blended, the image on the screen blended slowly between successive pairs of key frames. A third key frame would be generated line-by-line in the background while blending between the current pair in the foreground, so that the next key frame was ready when the current blend was complete.

The equations that generated the key frames were structured carefully using a small number of well-defined parameters controlling the number of axes of symmetry, the spacing between concentric rings, and so on. The parameters of successive key frames were kept consistent (with only occasional random changes) to ensure that the structures of successive key frames were similar. This helped ensure that successive pairs of key frame images ‘fitted’ together. You can see the results in the video above.

Elements was quite well received in what were the early days of the web. I used to show it at parties where I mixed visuals with a mate, under the name Psynaptics. I also have fond memories of seeing Elements played at parties where I wasn’t doing the graphics, or walking into clubs to find it playing on a big screen above the dancefloor.

Binary (1999)

This image was made in 1999, and is a variation on one that I created for the January/February round of the Internet Ray Tracing Competition, a geeky graphics competition originally for users of POV-Ray, the free ray-tracer. My image was modelled in 3D Studio Max, not POV-Ray, but it still came 22nd out of 143 entries.

The theme for this round was Imaginary Worlds, and my image was an obscure maths pun about how integers, with their boring, humdrum lives, dream of the exciting lives of imaginary numbers. Yes.

Mandala (1999)

I made these psychedelic mandala images back in 1999, at a time when I was into the psy-trance party scene in Cape Town. Psy-trance, also known as Goa trance, was a style of dance music with a steady 4/4 beat (at around 140 BPM) and a danceable rhythm constructed from layers of melody. At the time I was mixing graphics at parties under the name Psynaptics. As you can tell, I like purple and yellow.

Pulse (1998)

This design was for a ‘teaser’ flyer for a trance party called Pulse in 1998. At the time I was into the psy-trance scene in Cape Town. A friend of mine, Laz, was organizing this party in Joburg, and I got the job of making the flyer. Happily I also got to go: I was part-timing as a VJ, mixing graphics live at parties with another friend called Freddie Bell under the name Psynaptics, and we played at the party. We drove up for the weekend since Joburg is a 13 hour drive from Cape Town.

The flying logo was modelled in 3D Studio Max. The SoundStream logo was my creation and was designed in Macromedia Freehand. The end result was composed in Photoshop.

Fast, Cheap and out of Control (1999)

Album art that I made in June 1999 for a DJ friend, Richard Human, for one of his trance mixes. It ended up on the front cover of a ‘mix-tape’ style CD; I also made a back cover.

The flying robot model was authored by hand in 3D Studio Max using polygon-modelling. The photo in the background is mine, and shows The Sentinel viewed from Chapman’s Peak, a well-known piece of Cape Town scenery.

The name, Fast, cheap and out of control, is coincidentally also the name of a documentary film made in 1997. In our case the name refers to a concept put forward in a 1989 paper whose full title is Fast, Cheap and Out of Control: A Robot Invasion of the Solar System, by Rodney A. Brooks and Anita M. Flynn. Hence the flying robots.

Milkshake (1999)

I modelled this milkshake and burger in 3D Studio Max in 1999. The point of the image was actually the glass, which was created using polygon modelling and was the subject of a tutorial I wrote. I added the burger later to make a scene out of it. Pity I didn’t think to add a tiled counter-top!

Quetzalcoatl (1999)

A 3D model I made in 1999 of a statue of Quetzalcoatl, the feathered serpent god of Central America. My version, modelled in 3D Studio Max, was based on the real statues that still adorn the Temple of Quetzalcoatl at Teotihuacan, Mexico.

The main image with the pyramids in the background, and an actual featured serpent flying in the sky, was my entry into the March-April 1999 round of the Internet Ray Tracing Competition. It placed 10th out of 67 entries, which I was pretty pleased about. Still, I regretted that the statue model actually looks way better in the starkly-lit test images than in the final picture, which is less well lit and let down by some frankly half-assed work on the pyramids and ground.