Lindenmayer

I thought setting up a basic L-System to counter the growing tubular shapes would be relatively simple. And I guess it is if you actually are a programmer. Finally got a basic version to work. Again written in Python, generated in Maya.

I did find some examples on-line but that code was rather dense and I need to mess with it. Now let's see how I can randomize this and make it grow. And preferably sway in the wind. So I am wondering again why I cannot be satisfied simply animating a cube...

It's Alive!

It's life Jim, but not as we know it:

See previous post (or posts) for details.

Update (4:44 PM): Here's another one, which required some improvements to the Python scripts and Maya Expressions:

That's more like it!

Yes, more of the same. But I am getting close to what I want. I switched to nParticles for the curve creation and luckily my scripts worked with just a little tweaking. They started failing when I limited the lifespan of the nParticles, but fortunately Python let's you "try:" stuff.

And tonight after another hour of intense screen staring I have my desktop render an animation of this shape, simply by animating the size of the particles (the radiusPP to be precise, with an expression using the particleID as variable)

So, YEAH! The real work can start: creating an interesting piece of animation using this technique.

Still Static

I got my curves-from-particles and put-particles-on-rebuild-curves script working. Now I need to take the next step: animating this to create a creepy crawly tubular shape.

More Curved Particles

Today I worked a bit on creating a more interesting particle system to convert into curves. The one to the right is quick result. The geometry is created by simply extruding a circle across the curve that traces the path the particle took over the course of 1000 frames.

The object on the right is created from the same curve, but this time I put particles back on the curve. To get the particles spaced more evenly I first had Maya "redraw" the curve. This is more in the direction of what I am looking for.

I want the particle system to split resulting in a branching of the curve. After yet another hour of tinkering (why are collisions events in Maya not object specific by default?) I think I may now have a particle system that could maybe yield an interesting result. So tomorrow (?) it is back to scripting. The script now traces a single particle, the new one needs to trace many.

Curved Particle

Still experimenting in Maya, I am trying to create an interesting shape (in the first very rough story and incomplete story outline named "Tubular Shape") by putting particles on curves and / or converting particles to curves. I came across this interesting scripting page by a recent SCAD grad, but unfortunately the scripts are not available (Access Forbidden: I did contact him). Those scripts could have saved me some time.

Today I wrote a script that draws a curve through the path a particle has taken. Then wrote it again after it got stuck in an infinite loop and finding out that Maya has no way to stop the execution of a script and yes, I always save but not this time. I think I may be getting somewhere, though as the image shows I need to put a bit more time into creating a more interesting particle system

Grow Up and Fall

In the process of trying to decide on the next big thing, I tinkered with Maya (and Python) and created this test. I woke up early the other day and came up with yet another project. May be pursuing that one... this could be a development test for it.

Node loss

I spent a few hours to assess where I am in porting the Eye Brain character from Lightwave to Maya. The resulting image is a tad dark, but I do like the painterly quality of it.

Note to self: deleting the materialInfo nodes in Maya is not very clever as the objects will loose all hardware shading. So that is what they are for! Ran createNode many, many times...

So where am I on Eye Brain? Many models done but no antagonist yet. Need to scan my hand for that.

The next big thing

What to do, what to do? Next to dealing with some uncertainty in my professional life, I need to decide what creative project to focus on next. I have at least three in the pipeline. Here is a look development test I created for ">Hello World":

Usually NPR cross hatching relies on properly laid out UVs on all objects. For "Seeing Red" I generated the UVs for the ground plane procedurally based on the world space coordinates and normal direction, but that is not an option for most objects. In this test I do not use UV maps at all but in stead project hand drawn cross hatching patterns onto the scene from the camera. Since the camera is moving, there are two projection position (the start and the end) and I had to add one extra to fill in a gap as some of the wall was obscured from both positions but visible inbetween.

But I am not sure I like this look for this project. I may use the technique (but not the look) on an older stalled project in stead: Eye Brain. That is the second project in waiting. A new project is brewing, involving integrating a CG version of this hooded character and some yet to shoot HD video. Or should I do something abstract again? Or maybe think up a project that does not take months to create? Make up my mind I must!