[Doyle Rockwell]

Hey Brian,

I just played with this a bit, and it’s quite simple and runs fast:

1. Draw a large circle shape (representing one of the binocular viewports).
2. Duplicate the circle shape and set it to overlap, giving you the basic binocular shape.
3. Group the two shapes into a new layer, name the new layer ‘Binocular Mask’.
4. Apply the Gaussian Blur filter to the Binocular Mask layer. Set the Amount to 30.
5. Add an image mask to the footage you want to binocularize (my own new word!). Found in Object->Add Image Mask menu.
6. In the layers list, drag the Binocular Mask layer into the image mask. You should see a masked image.
7. Add the Compound Blur filter to your masked footage.
8. In the layers list, drag the Binocular Mask layer into the Compound Blur filter. This tells the blur to blur only where the binoculars are.
9. In the Inspector, view the Compound Blur filter and set the Amount to 10 and toggle ON the Invert Map checkbox.

You should now have a nice, soft-edged binocular effect, with blurs on both the dark edges AND the edges of the unmasked area. Plays back fast, too. If you want to simulate coming-into-focus, just add a Defocus filter to the masked footage and keyframe it. You can even make a Library preset of this whole dealie and jsut replace the masked fotoage as needed. Plugin, shmugin, I say!

[Doyle Rockwell]

Hey Sarah,

Take a look at the short replicator tutorial in the Help menu, or even the replicator section of the manual (also found in the Help menu). You’ll have a much easier time of it than slowly beating the steps out of us 🙂

[Doyle Rockwell]

Hey Brian,

Isn’t this effect usually achieved with a two-overlapping-circles matte effect? I would just draw two circle masks on the layer containing the footage, overlap ’em, feather ’em, and call it a day 🙂

[Doyle Rockwell]

Heya,

As Noah mentioned, you’ll need a pixel shaders-capable graphics card, which I believe all G5s have shipped with. If memory serves (isn’t that what the host of Iron Chef says before biting that onion?), the 1.6GHz G5 shipped with the Nvidia GeForce FX 5200, which does support Motion, but it’s the lowest-performance card available. You’d see a night-and-day difference if you were to upgrade the card. Good luck, and have fun!

[Doyle Rockwell]

Hey Sebastian,

Some behaviors stop affecting their target parameters after the behavior ends…and some don’t. Throw, for example, affects the Position parameter of an object, but the object doesn’t jump back to its original location after the Throw ends. Grow/Shrink, as you’ve noticed, does stop affecting the Scale parameter when the behavior ends.

There are a couple of ways to get around this. As Noah suggested, simply extending the lifetime of the behavior will keep its affect in place. In the case of Grow/Shrink, though, the behavior is constantly driving Scale, so extending the behavior keeps the animation going, and you want it to stop growing/shrinking at a certain point, right? Your new best friend: the Stop parameter behavior. Stop freezes a parameter in place. Simply add a Stop to the Position parameter of the object, having the Stop begin at the last frame of the Grow/Shrink. To add the Stop, just right-click on the Position parameter name in the Inspector, and select it from the list. Stretch the Stop’s length out for as long as you want the scale to hold.

Hope this helps. Good luck!

[Doyle Rockwell]

Hey Les,

Motion does not currently offer any tracking functionality.

[Doyle Rockwell]

Hey Mike,

They may be artifacts from running into the memory limits of your card. There are a few factors involved: the memory dedicated to the display, the size of the images being processed and the bit-depth of the project. If you have two displays, for example, the system divides the card memory between them, so in your case it would be 64MB for each (assuming your card has 128MB). It sounds like your project is HD-rez (if you’re doing HDV), so those are bigger images. Then, when you go into 16-bit, each image now takes up twice as much memory on the GPU as 8-bit. Between any and all of that, you may be toeing the line as to what the card can handle, so it could be producing funky, corrupted images. Or maybe it’s an ATI driver bug…who knows?

I usually refrain from the constant cry of “Buy a new card!”, but you may find your problems gone if you were to get a 256MB GPU. Also, the Nvidia cards, like the 6800GT, have accelerated 16-bit rendering that is MUCH faster than the ATI 16-bit. You may also want to see if optimizing your project helps at all. Are the smoke particles running way off-screen and living for a long time? Remember: Motion renders everything, not just what is inside the project size. If you can adjust the life of the particles, they die right after moving off-screen, which results in a smaller render image, which means you might squeeze by without artifacts. Give it a shot.

Hope this helps. Good luck!

[Doyle Rockwell]

LOL, don’t know why I said “Hey, Brian”…. sorry about that, Mike. 🙂

[Doyle Rockwell]

Hey Brian,

The color patches you describe sound like OpenGL freak-out. What graphics card do you have in your system? Also, what operations were you doing that required 16-bit?

[Doyle Rockwell]

Hey Rene,

The Animation codec is essentially lossless, but that means that it doesn’t compress the file size much, so you can end up with big files (such is the price for quality). I think the PP book may have meant that the codec is more efficient (i.e. gets smaller file sizes) on images that don’t change much. Regardless, it is, for all intents and purposes, lossless, just not quite as huge as Uncompressed or None.