Cily.it - Simone Ciliani vfx and TD - [project] The First Locomotive

[project] The First Locomotive - Puffing Billy

Details: Category: Projects

Why Puffing Billy?During my visit at Science museum London, I was fascinated by it. An incredible invention that was 200yrs ago, AMAZING... amazing "retrò" design, you could see his engine, everyting works "unskinned"... WOW... I went home and Immediatly I researched how it could works... the steam machine, the plungers, the breakes, the whells... COOL... it was love at first look! I had to making in 3d! My TD shot was there... and Voilà...

Storic Note:
Puffing Billy is an early railway steam locomotive, constructed in 1813-1814 by engineer William Hedley.

During my Td corus i Made an "early" version of it...

I used Ncloth system to simulate the explosion but... slow calculation and so many time ingestible things (a lot of bug like: with paint on glue of costraint... and an animated ramp, iI solved after 2 days of works)... I used a Layerd Simulation with ncloth and works with layer for the collidings things but like Peter Kyme (from Dneg) said: "good simulation is when you have:

1) Stability
2) Speed
3) Flexibility"

Whell ok... my result was RUBBISH!!
Dneg and a lot of other company have a RBS selfmad (like Dynamite...) based on Bullet libraries... why I didn't use that? I want Stability and Speed! Bullet can made me happy! I'm started to used maya 2013 for that project with Bullet insted by Autodesk (maya bullet)... but let me said... when I tried to use some Field to try my RBS... wtf is that... maya, after you make a field, convert your bullet nucleus into a nomarl maya nuclus!!! "ARE YOU KIDDING ME?" I unistalled immediantly maya 2013 and I continued to work in 2012 with Bulllet DYNAMICA (free from Disney)... well is not perfect but is free and works! to help me in my projects I made this simple Script on MEL.

OK I lost some funny and interesting stuff, I lost flexibility.... but i kept Speed and Stability.... and DONE, in only 2 days I baked my projects!
meCSYQddu7M

Some people said me the gravity is to "slow", there isn't an error about scale I done that becouse i like more the "coreograpich" result.

In the old projects I had a stupid problem into rig about the whell and the rail... iI fixed with simple scripts... now I've a dynamics velocity and correct grip with rail.

Finally I had the 3 Fluids, i worked to have less trasparency, big Density ( with a significatly cutoff into the graph) and not small proxy grid to improve the speed in simulation and rendering (about 1 hour of simulating for each fluid)