[Kalleheikki Kannisto]

By “infinite”, do you mean repeating from one end of the line to the other?

The Illustrator dotted line is a repeating pattern, which repeats to infinity. You don’t input every dot separately, jut the length of dots and spaces to define the shortest repeating pattern.

Thus, if you make a dotted line with the value of “2 points”, you’ll get a repeating pattern of two points of stroke follow by a two-point gap, repeating ad infinitum. The additional slots are for more complex patterns, such as a repeating short and long strokes.

[Kalleheikki Kannisto]

If you can determine the line breaks beforehand — or do them manually — you can use point text instead.

[Kalleheikki Kannisto]

A simple handling is to add a curves adjustment and pull in the highlight and shadow ends of the alpha channel about 15%.

[Kalleheikki Kannisto]

Markers might be a good way to do this. Check out marker.nearestKey(time) in the expression reference.

[Kalleheikki Kannisto]

Essentially you’ll need an array with your numbers

my_array = [48, 50, 57, 32, 44];

Then the source text for your test layer will need to pull out the number by time:

my_index = Math.floor(time);

Math.floor() makes it so that only complete seconds are used as index numbers, not partial. Now you can get the corresponding number from your array:

my_number = my_array[my_index]

This will work as long as your comp is not more seconds than you have items in the list.

If you want the last number to stay on longer, you can fix the second line of code to

my_index = Math.min(Math.floor(time),my_array.length-1);

This takes the smaller of the two numbers, either the time in full seconds OR the number of items in the array. Thus the last number will stay on when time passes beyond the last item.

So, all in all:

my_array = [48, 50, 57, 32, 44];
my_index = Math.min(Math.floor(time),my_array.length-1);
my_number = my_array[my_index]

[Kalleheikki Kannisto]

There is no comp open, nor is there one visible in your project window. I suspect that either

A) you have it tucked away in one of the project folders, in which case you need to locate it in them and double-click on the comp icon to open it, or

B) this is not the latest version of the project, in which case you could check “recent project” in your file menu or

C) you closed the project but didn’t save it after making a comp.

[Kalleheikki Kannisto]

It’s a bit complicated, but it is possible. Here’s how:

1) Make your text box the width you want it.

2) Select the text or place cursor in the text and select from menu Type>Area Type Options.

3) Turn on “Auto Size”. This will make the text box height fit vertically to text.

4) Create a separate box within which you want to vertically center the type. Doesn’t have to have any fill or stroke.

5) Select both the text box and the “alignment box”.

6) Click on the “alignment box” again, it will highlight with a darker line (it becomes the key object for aligning.

7) Align the objects vertically centered. The text box will move to the center, the “alignment box” will stay in place.

Once set up this way, you can replace text and do 5-7 to re-align vertically.

[Kalleheikki Kannisto]

“doesn’t it look kind of odd when the presenter takes steps but doesn’t move?”

Yes, then there is that. Stabilizing the shot too perfectly on the person will give a similar effect to GoPro footage where the camera is attached to something, wherein the “something” (in this case, the person) looks static while the background moves in a dizzying fashion.

[Kalleheikki Kannisto]

Ok then. There are a whole bunch of factors that determine the speed of tracking, not the least of which is the computer you are using, and obviously the length of the footage. But one important thing is you want to have as small a tracking target and, even more importantly a small search area, the latter just large enough to account for the largest motion from one one frame to the next. That’s where most of the processing power goes, searching through those pixels, so you don’t want any extra there.

Then you could consider alternate strategies. It is not necessarily a good thing that the camera is moving constantly, so using you manual method to to re-position the footage but only doing it when it is absolutely necessary might even be faster than tracking the shot.

Alternate strategy #2: Shrink the footage so that it fits completely in that half of the screen, then take occasional close-up views when the person is not moving. Maybe there is a reason you could do this, such as showing other visuals in the “extra” space that you will create.

[Kalleheikki Kannisto]

I assume they are actually falling straight down but due to the camera movement they appear to be falling diagonally. The easiest solution would be to use horizontal wind to push them over a the same speed as the camera moves. You’ll have to experiment to find the right wind speed.