The last of the Noise parameters; Position Scale!

Noise parameters, part 2:  Noise Strength!

Now, for the Noise Series.  Upres Factor creates a higher resolution grid for the noise, essentially multiplying the domain resolution by the Upres Factor.  Just like cranking up resolution, going crazy with the resolution/upres factor combo platter can quickly swamp your computer!

Dissolve is fairly straightforward--you tell blender what age (in frames) you want the smoke to start dissipating.

For the last of the Gas Parameters--Vorticity!  This one will get out of control if you aren't careful!

Next up in the Gas Parameters section; Heat!  Not the most exciting of parameters, but that just keeps the excitement building that much more, right??

Now, on to the Gas Parameters section!  Buoyancy Density seems simple enough; the less dense something is, it should rise faster through a container of constant density.  How hard could that be?

Resolution always leads me down the Primrose Path.  I find myself thinking that if I just had a little better resolution, I could see what I needed to change.  That quickly leads to running the simulation overnight, waking up to find it didn't do what I wanted, and wasting another day waiting for an overnight run.  The question is always how much resolution is enough?

This is the first in a series of short posts where I examine various parameters of Blender's smoke simulator one at a time, while holding the rest constant.  I started this because I was tired of the amount of time I spent going around the tweaking-simulating-tweaking loop.  I hope this will provide me the background understanding to quickly get close to the look I want before I spend a lot of time on high-resolution simulations that take time to process.

This happens to me frequently; I start a project and things go smoothly.  I can make a widget--a leaf or a column or a fish.  But it doesn't occur to me until too late that when I duplicate that nice widget, especially if I duplicate it hundreds of times, my system might complain.  So I've learned a few lessons that, if I listen to myself, will make future projects finish much more smoothly. 

 I have been experimenting with the SwiftBlock addons that simplify the process of taking a mesh in Blender and defining that mesh in OpenFOAM.  There are two versions and both are about at the same level of development.  While they both work as advertised, I have come to the conclusion that neither will be appropriate for complex models.  It was at this point, the little voice in the back of my head said, 'Of course, dummy!  A wise man once said, "Always use the right tool for the job!"' (That little voice loves to quote himself!)  While Blender is great for visualizing the results of numerical analysis, it was not made for creating the meshes used in said numerical analysis.  Luckily, I found something that is the right tool for the job...