Saturday, March 6, 2010

The FX Process

Here's the thing. Its not about how did I or why did I do something in a scene. Its a process. I start with 1 fluid 1 emitter kinda low rez (50x50x50'ish) I turn everything on, b/c I know what I'm doing and what I want to make ;) I set only 3 attrs damp 0.01 swirl 10 temp turb 5, the second 2 are just the sliders all the way up. Playblast. Hmm looks like the fuel isn't burning fast enough, turn up reaction speed and turn down ignition temp. playblast. Hmm looks like temp isn't fading out now. Turn up dissipation. playblast. Nope not enough, turn up more. playblast. Hmm looks too regular.

make hypothesis.
I wonder if I separated the temp into a smaller emitter it would react differently with the fuel?

test, playblast, nope settings off. tweek, playblast.

ok thats kinda better not much but it's better. hmm uprez. playblast. ok now it needs more bouancy or something, tweek. playblast. Still looks kinda regular.

what if I make the fuel emitter separate and give it a really high value and strong turbulence?

test. ok all my settings are now off. Tweek playblast, tweek playblast, tweek playblast. Hmm ok that looks better.

the point is, why did I do something and is it "the way to do it?" is hard to say and very case by case. Thats why I think of it as "playing" its not a real shot, its kinda general R&D. No rules, try things. sometimes they work other times epic fail.

I can't talk and play at the same time b/c my mind is totally focused on, what settings have I changed? What settings are the animating settings/strongest influences to velocity? Where is the balance of fuel to temp? What are the last 20-50 changes I made? (not THAT much of an exaggeration) if I make 3-4 changes then playblast what were they, and what were the last good numbers? Remember you can't really undo too much with fluids b/c the timeline change will jack your computer.
If I'm at work doing tests and someone talks to me, it could be the end of that test. I can't remember what I was doing. Now I just see a bunch of settings, but I don't understand why I did that, hah.

So play, have fun, test out you ideas. Only experience will make you better. Keep testing and playing. Guess at how it works, test that guess and see if you are right. A failed test doesn't mean you are wrong it just means the test failed. I find a lot of times what I thought was wrong was right later, I just didn't do the right test.