I didn’t realize when I was taking trigonometry in high school that I’d really be using it in life. Turns out that there’s a lot of math to be done when it comes to sophisticated Actionscript animation. Here’s a clever use of a multi-dimensional array to create a polygon by Flepstudio.
I must master the art of modeling glass and in different colors! I exported a png of this out of Maya and brought it into Photoshop onto its own layer. I was so happy to see how well Maya’s alpha channel translates into Photoshop. On this layer, I added a layer mask which perfectly masked the varying levels of opacity in the alpha channel by command-clicking on the layer to make a precise selection, inverted the selection, created a layer mask, then filled the selection with black. By having this layer mask on the image from Maya, I can now add adjustment layers without affecting anything underneath it. This is something I’ve been trying to figure out how to do forever! Embarrassingly, layer masks is something quite new to me.
| On1 Steps | NY On2 Steps | ||
| 1 | <-> | 6 | |
| 2 | <-> | 7 | |
| 3 | <-> | 8 (pause) | |
| Neutral | 4 (pause) | <-> | 1 |
| 5 | <-> | 2 | |
| 6 | <-> | 3 | |
| 7 | <-> | 4 (pause) | |
| Neutral | 8 (pause) | <-> | 5 |
Snagged from:
http://addicted2salsa.com/2006/09/19/basic-salsa-steps-table-mapping-on1-to-on2-timing-steps/
My favorite part of Iron Man was the ending credits. Danny Yount of Prologue Films is the man behind it. It turns out that Maya was used to create it and contour rendered with Mental Ray.
Also playing with the depth-of-field in this iteration. The way to figure out the focus distance is so old-school and unintuitive. I hope the day will come soon when you can focus interactively like a real camera lens.
What seemed like an easy task turned into a real challenge. I first attempted to model it as polygons which evolved into the mess you see above. Too many polygons too soon resulting in rectangular connections.
Then I tried modeling it with a NURBS sphere and cylinder. This would’ve been a possible solution if it was only two spheres.
Then with a revolved NURBS curve. This would’ve been an even better solution for two spheres. The third sphere throws all NURBS solutions out the door.
Returned to polygon technique, but this time starting with the least amount of spans. Bridged between two of the spheres, then while using the last NURBS attempt as a guide, moved the vertices one by one into place. Used the smooth tool and got the parabola shape (pictured above).
I realized around this point that this whole model was about mirrored halves and merging the center points between the halves. So I cut the parabola shape to end up with 1.5 spheres (pictured above). Before cutting, however, I straightened out the points where the cut was going to be so that I can ensure a straight cut. I mirrored the ice cream scooper shaped parabola and got the three sphere shape. I merged the centerpoints between the mirrored halves. (I couldn’t get the mirror command to work properly so I ended up just duplicating and moving/rotating into place.) Then a final run through the smooth command.
Final model.
This simple engraving of text was a major pain in the ass. I started with two spline primitives, one for the text and another for the rectangle. Made them both editable and then connected them via the Connect command. I parented the resulting combined spline to an Extrude NURBS object. Lastly, I applied a subtle motion blur just to the spin to spice it up.
I realized from making this that the order of the objects I animate is crucial to getting a clean animation. This is especially clear when working with the F-Curve editor. First animate the main subject as completely as you can with the F-Curves and then move onto the camera and its F-Curves.