Hi,

Can anyone suggest an efficient way of reproducing the Digital Domain effect used in the pin block on the Nine inch Nail vid ‘only’? I’m not talking about driving it with video footage but just being able to manipulate it with rigid bodies/fields/whatever colliding with the back of the pin block.

I have set up a rigid body solution, with the pins as active rigid bodies forced into an enclosed space (ie a passive rigid body with holes cut out for each pin) but it interpenetrates all over the place. When i attach individual contraints to each pin it takes for ever and is really unpredictable. Is there a way to constrain the pins so they can only move along one axis but still use rigid bodies to generate organic movement to that axis only- if that makes sense? Is it possible to rig it so you don’t have to use rigid body constraints but just normal (less unpredictable) ones?

any help would be v helpful!

bert

	Attachment

I’d probably still use the same technique you would use if you want to do a video. Render a depth map of the scene from the position of the board and use that to drive the position of each pin with an expression that uses colorAtPoint to find the color value at the UV that correspeonds to the relative position of the pin in the grid.

But, you can of course set up a pin to be constrained in one direction only simply by breaking the connections on two of the rotation axes.