top of page

PHASE I

"Phase One"

 created in 1985, 0'56"
NICOGRAPH 1985 Amateur Award

phase_1_2.png

Work Explanation

  • This was the first installment of “CG animation using ray tracing method” by personal computer, aimed at technical verification and demonstration.

  • CG images displayed on the color CRT monitor were shot frame by frame with an 8mm film camera “Fujica Single 8 ZC1000” using motor control (*).

  • The work consists of four types of scenes. Modeling and lighting were focused on shadows and reflections to take advantage of ray tracing’s unique expressions.
     

Additional Information

  • Initially, the title track of the movie “Blade Runner” was used as BGM, but it has been replaced with a new original piece of music for internet release.

  • Music: Ishii’s Music Production Office, https://www.ishiimusicweb.com/
     

Trivia

  • At that time, there was a local amateur anime screening held at the Miyagi Prefecture Museum of Art, in Sendai. When this work was unveiled there, the venue immediately began to buzz (the audience felt a sense of incongruity with hand-drawn animation). When it was revealed during the final staff roll that it was CG footage generated by a computer, the audience’s buzz turned into a surprise murmur.
     

      (*) In 1985, the only way to shoot independent animation was with an 8mm camera, and the difficulty of this is referenced in “CHRONICLE: Historical Explanation Document” “001 ‘DoGA’ Establishment Background and Purpose” https://archives.cganime.jp/chronicle/
 

Image Data

  • CG part: 37sec x 12fps = 456 frames rendered (shot at 24fps)

  • Resolution: 640x400 with RGB each 8bit

  • Computing equipment: 6 units of friends’ PC-9801F2 (without i8087)

  • At that time, there were no HDDs as recording media, and image data was exchanged using hundreds of 720KB 5-inch 2DD floppy disks. Rendering was done by distributing 5-inch floppy disks to friends who owned PC-9801s every day and collecting the floppies with the rendered images. The rendering time was set to less than one hour per frame.
     

Software

  • We created our own ray tracing software using the N88-BASIC language.

  • The base was the source code published in “The 3 dimensional computer graphics” by Tsuyoshi Yamamoto. However, the following modifications were made for processing speed:
     

  1. For horizontal line processing on the screen, the first pass calculates the start and end positions of objects in a scanline manner, and the next pass performs ray tracing only between them for speed optimization.

  2. For vector product sum operations, calculations are processed by calling a machine language routine dedicated to operations without transferring memory internally.
     

  • We struggled to create random numbers to prevent periodicity from appearing in animations and textures. In Scene 2, the semi-circular frame through which the bike passes is distorted due to insufficient adjustment of computational accuracy.

  • We also created our own software in BASIC to animate objects. It internally processes the script-like definition of the movement and rotation of objects and cameras, converting them into coordinate information for each frame. In Scene 4, there is a mistake in the initial value of the rotor’s rotation, but it barely avoids collapse.

  • Since the PC-9801 can only display 8 colors on the screen, dithering is performed with 2x2 pixels, but in Scene 1, there was a mistake in fixing the pattern, and thereafter, it was changed for each frame.
     

Trivia

  • The software in charge (Mori) developed it on his own PC-8801 (8bit machine). The compatibility of N88-BASIC between the target PC-9801 (16bit machine) was helpful.
     

Photography

  • A stepping motor was connected to the film feed shaft of the 8mm film camera “ZC1000” owned by Tohoku University Animation Society(TAS), fixed on a wooden stand, and a homemade frame-by-frame shooting system was created. The stepping motor was controlled via the parallel I/F of the PC-9801.

  • The shooting control program allowed for automatic synchronization between the display on the PC-9801’s CRT monitor and frame-by-frame shooting of the screen. At this time, various tests were conducted to determine the optimal F-value and exposure time.
     

Trivia

  • The shooting took place in a room of an apartment in Yagiyama that Mori was renting at the time. To continue exchanging floppy disks day and night for camera shooting, a know-how was developed to darken the room by attaching aluminum foil to the window glass to prevent outside light from entering (!).

bottom of page