Posts Tagged ‘ninja’

• By Rich

  Tags

  3D, fp10, ND3D, ninja

  Posted in

  3D

  6 comments

  3D Ninja Test 2 – Fists of 30 fps Fury

  8th Jun 2010

  6

  

  A few weeks ago I posted my first demo of an animated 3D ninja. At the time I had high hopes of creating a primitive Virtua Fighter styled game. But there were several obstacles to overcome. The poly count was a bit too high, the scene only had one model in it, and it took Away3D a fair time to parse the MD2 data for the key frames.

  So last night I decided to revisit the code and see what could be improved. Here is the result – I dropped Away3D and decided to run a test with ND3D instead. It’s a much more light-weight 3D library, and doesn’t include features like lights or shaders. But what it does do, it does very well, and very fast. The MD2 parser in particular kicked several bails out of the Away3D one.

  This, combined with an optimised MD2 model (many thanks to Adam Biles for help) allowed me to get two fighters in the scene, each independantly animated and textured, with a ground plane and free roaming camera. I did have a skybox in as well, but the camera perspective didn’t look right and made the fighters seem as if they were floating in space. So I’ll save that for a different project.

  Feel free to have a play with the demo.

• By Rich

  Tags

  3D, away 3d, ninja, tekken

  Posted in

  3D

  10 comments

  Ninja 3D Test 1

  27th May 2010

  10

  I had a desire to see if it was possible to create a simplistic “Virtua Fighter” or Tekken style game in Flash 10. The biggest initial hurdle was getting a convincing model displayed and animating it. And then seeing if the frame rate shot through the floor like a dead weight, or was actually playable.

  After a lot of messing around with Milkshape, md2.qc custom files and the Away3D MD2 parser doing some really weird shit, I finally managed to assemble this tech test (click the picture to launch, FP10 required):

  

  Ok so it’s not going to set the world on fire, but I was genuinely surprised at the speed, even when running a full animation sequence. And the model could be optimised significantly too.

  More tests will follow as time permits, but this is very encouraging at least.