[Phil Williams]

What was actually on the LCD? Can you cheat it by tracking in stuff shot with the A-cam?

[Phil Williams]

Assuming I’ve properly understood what you’re trying to do, I would suggest something like this: create a solid circle shape, set the opacity to 50%, position it over one of the dots on the tile, and then resize, feather, and colour the shape to taste, readjusting the opacity as necessary and probably trying some different blend modes to see what looks good.

Once you’re happy, simply duplicate this shape as many times as necessary, repositioning each one over a dot in the tile. Then you can animate the opacity of each one to fade it up in time with the counting.

[Phil Williams]

OK, I’m 3000 miles away from my Motion setup at the moment so can’t actually test this out for you myself, but off the top of my head, try this:

Use the freehand shape tool to draw a horizontal line across the screen. Don’t try and make it perfectly straight, it helps to have a “hand-drawn without a ruler” sort of feel. Then apply Behaviours>Shape>Wriggle Shape. Play around with the settings and you’ll see that this behaviour will animate the positions of the points that make up the line you’ve drawn, which will hopefully give it a sort of waveform look.

To make your audio track drive the animation, apply the Audio parameter behaviour to the amplitude parameter of the Wriggle Shape behaviour. You will probably want to add some randomisation as well, and try and find a way of limiting the movement of the points to the vertical plane only.

As I say, if I wasn’t 3000 miles away from my Motion setup I’d be able to offer more detailed suggestions, but hopefully you might find the above to be a useful starting point.

[Phil Williams]

I did try that earlier, but when you position or scale the caterpillar beyond the area covered by the Polaroid image you end up with this sort of thing:

BTW, am I right in thinking that the mask rotation problem is just an example of Motion being buggy? I guess a reinstall might fix it?

[Phil Williams]

If I just use a rectangle mask directly on the caterpillar clip, the mask moves and scales along with that clip. If I draw a rectangle mask around the picture area of the Polaroid and then apply it to the caterpillar as an image mask, I can move and scale the caterpillar clip independently of the mask, so that I can reposition and resize the caterpillar whilst always keeping it within the picture area of the Polaroid – I don’t have to keep repositioning the edges of the mask. Am I making sense there?

That side, surely if I rotate a group, everything within that group should rotate, including image masks? It’s just a bug, there’s nothing I’m overlooking?

EDIT – I think I’ve found a workaround. If I draw a solid rectangle, NOT a mask, and then apply that to the caterpillar clip as an image mask, it rotates perfectly.

[Phil Williams]

Thanks for that – I guess I’ll have to get Photoshop now 🙂

BTW, because I don’t have Photoshop there’s no way I would have found that out by experimenting by myself, and I did try to replicate the effect in Motion before asking. I do try and research things first, but sometimes you just gotta ask on a forum like this.

[Phil Williams]

I’m only working in SD at the moment, so 2K would be big enough for me. I hear what you’re saying about the 2D group or the solid background colour, but I want to achieve the effect of the TV frame being a window onto a sort of photocollagey/scrapbooky background that we can’t see all at once.

[Phil Williams]

Thanks for that, Mark. It should have been obvious, but I guess sometimes it’s easy to miss the obvious!

On a related note, I’d be grateful if you could give me your insight on the following:

I want to emulate the sort of style you see quite a lot these days, where an animation begins, and then moves off the edge of the frame. The camera then moves to “catch up” with it, typically using a motion-blurred whip-pan effect. Or maybe the camera will leave the animation to zoom off to reveal a different part of the background, and the animation will then “catch up” with the camera. The sense you get is of the camera moving and the animation happening over a very very large background.

Am I right in thinking that the best workflow for this sort of project – especially if your background is made up of lots of different textures and still images rather than using a single seamless texture – is first to use something like Photoshop to create the background as one very large graphic image (say, 2-3x bigger than the resolution of the project) to allow plenty of room for dramatic camera moves over it? Or is there a more RAM-efficient way?

[Phil Williams]

Thanks for the responses guys, although it’s not what I wanted to hear 🙁

I’m just about to start a project where I’d like foreground graphics to be motion-blurred while background graphics stay crisp, so I guess the only way to achieve that would be to render out the foreground graphics separately with motion blur and an alpha channel, and then composite them onto the backgrounds?

[Phil Williams]

Hi Zak,

Sorry about the long delay in replying but I’ve been on holiday and only just had a chance to try trashing the prefs in Motion.

Bottom line, it didn’t make any difference.

Typically what’s happening is that I’ll draw a mask of the exact dimensions of the TV screen, but when I apply that as an image mask, the mask will be in the right place but slightly too small – I have to extend it left and right to completely fill the TV screen.

Reinstalling Motion doesn’t make any difference either.

I can live with it, and hopefully I’ll be able to upgrade to Motion 4 soon.