recreate a 3d led array in flash

Hey guys im a basic flash user not used it in like a year but im a fast learner

i need to recreate this or something like this in flash

http://www.youtube.com/watch?v=Aj3_v7xCyJ0

anyone have any ideas how to do this

Dave

Does it need to be real 3d or can you fake it?

ooo spiffy :B

That is basically just a 3D array with each element filled with a hexadecimal data. Then each element is plotted into a 3d world. The 3D world renders it properly to location. The problem you will have with this concept is it will require a LOT of particles to accomplish something that large...you'd prolly have to make something with less elements than that cube if you did it in flash.

Another big thing is, if you wanted to make it real 3D where you can change angle of how you view the device, you even have to make it smaller because 3D elements in flash take a toll on CPU.

So figure out what you want exactly then let us know :D

16x16x16 is 4096 elements, which is quite a bit to try to do in "real" 3d. But, you don't really have to, since you can optimize for your cube model. If you don't need to rotate the cube or move the camera, then you can just precompute the screen positions for each element. You could even use pv3d for this purpose. If you do need to rotate the cube, you'd probably be better off using the new flash10 2.5d transform features.

I'd agree with neznein9 that you're probably going to want to fake as much 3d as you can get away with.

i only need to make a small faked version of that just to show a concept so what is the best way to go about it

any example would be nice that i can hack to bits :)

oh and it doesnt have to rotate just do a few simple things like "sparkle" like it does in the video and be able to scroll some text flash up a smiley face stuff like that

Here's how I'd approach it:
Model your cube as a 3d array of objects placed appropriately in 3d space. Use pv3d or pure math to project those 3d coordinates to 2d coordinates. Hereafter, all graphic manipulation will be 2d, but I'd keep my 3d array in memory for ease of computation.

For whatever animation you want to do, figure out which 3d leds are lit, get their 2d projected coordinates, and draw into a common bitmapData. Repeat as necessary.

For ease of programming (although, it would be hard on the proc), I'd make a square bitmap and give it an isometric distortion (rotate 45° and squash 50%), then stack up a bunch of those vertically to make the cube. Store each layer to an array and youve automatically got z-depth by the array, and then x,y is still relative to each bitmap.

Couldnt find this link earlier today but it floated up in a related story on infosthetics today.

If it's just a design concept rather than programming: could you mock up some masks and solid overlays in illustrator using a prospective grid and layer them? Then just tween skewed twinkling colour using the masks.