31 July 2012

In the SOuP

After struggling for a while to understand how to use the new nodes offered in the SOuP plugin for Maya, developed by Peter Shipkov, I finally cracked it a couple of weeks ago..... it's simple when you know how!!
I wanted to use the PointCloudField, a field that is emitted through the vertices of any attached geometry, to affect the nParticle debris from a DMM demolition, sounds technical but really all of the work has been done by coders and all I needed to do was apply it.

The thing is, there's very little documentation to support many of the free third party plug-ins but the community for SOuP is very friendly and, I was pleased to discover, is happy to offer support to those who ask.

So, why did I want to  use the PCF? Simple really..... ordinarily Maya fields eminate from a single point in the scene and the force is set to specific directions dictated through the attribute editor but the PCF allows the field direction to be dictated by the normal direction of the emmiting object.... Very useful in demolition simulations where turbulence is created through the scattering of the debris.

This illustrates the difference between PCF and an regular radial field....

The radial field is to the left and the PCF is on the right. The blue debris scatter from the PCF is far more pleasing to the eye as the radial field has little to no effect. After the test I adjusted the magnitude value of the radial field to see whether I could obtain a result similar to that of the PCF and the result at 5 times the original magnitude yielded similar scatter but at a far more even and, as it's name would suggest, radial spread.

25 July 2012

200 Frames Per Second Torrential Outpour Blues

So, I wanted to make a slow motion (200fps) shatter effect and mix it with a fluid simulation.... ahoy, is that a steep learning curve that I see ahead?

Things I learned along the way....
1: It's not as straight forward as it would first appear to use DMM simulations in Realflow.
2: PCF doesn't seem to work so well at high speeds.
3: My computer is Murtaugh from Lethal Weapon (too old for this sh*t).

Originally I had three glasses all rendered nice, refractions, raytrace shadows, photon emission etc.. (see still image below) but I felt that 16 minutes per frame was a little excessive for a technical excercise so I reigned it in to my usual plasticine style render, that's another thing to add to the list of stuff I learned...

4: Don't bother making things all shiny if it's just a technical exercise!*

* unless the technical exercise is rendering shiny stuff!

11 July 2012

Balls to the Wall...

Further developing my understanding of the DMM plugin I built a cavity wall to destroy. The idea here was to work with multiple objects and different materials so I have two layers of brick, iron wall ties, wooden planks and a plaster render.

I've used nParticle instancing again and this time have two emitters, the front wall and the plaster layers. I need to spend a bit more time refining the particle motion and my understanding will likely develop the more I use it.
My workflow is definitely improving and I have found that I am able to "sculpt" a simulation rather than just apply the plugin and go with the default outcome. I feel that I need to spend more time working on the applied DMM materials and perhaps push my processor a bit harder to create smaller chunks.

Note to self: get some dust debris going and address the sinking blocks in future simulations.

9 July 2012

This is Cracking Me Up!

I've spent some of my free time over the past couple of weeks investigating, experimenting and getting o know the DMM plugin for Maya and I have to say, it's great!

The plugin is intuitive and, once a basic understanding is developed, easy to use. One thing that appears to be missing from DMM is the ability to create debris but this has been dealt with by Olivier over at Pixleux who has writen a handy Python script that will emit particles from the faces that are created by the fracture.... handy!

I found a really nice video by Sylvain Nouveaux on Vimeo ( Click To View ) that I thought would be a good piece to replicate in order to best develop my understanding of DMM. I feel that my attempt is pretty successful, there are areas that could be developed further but for an early learning exercise it's not too bad.
I still didn't get my head around the SOuP plugin and had to resort to using the standard radial field to affect the particles so this is something that I can further investigate another time.