[Kalleheikki Kannisto]

Another option is to make something that looks like bad digital TV reception, digital glitches. Although a computer doesn’t do that, it is a more familiar-looking electronic glitch to most people. Along these lines:

https://www.youtube.com/watch?v=jaS37Z-DfPk

Some contents or functionalities here are not available due to your cookie preferences!

This happens because the functionality/content marked as “Google Youtube” uses cookies that you choosed to keep disabled. In order to view this content or use this functionality, please enable cookies: click here to open your cookie preferences.

[Kalleheikki Kannisto]

Sorry, I meant circular mask, not spherical. It is a sphere made from rotated circles. I tried this myself and it looked close to the example. Perhaps some more replications, I have just 10 here.

Distant view:

[Kalleheikki Kannisto]

You will get an eased animation if you add a null in between the two objects with the same expression. That is, use the expression to make the null follow the original object and then use the same (with a different target layer) to make the follower follow the null.

If you then make the settings

samples = 50
timeframe = 2
delay = 1

in both, you get something so smooth it looks like a rubber band!

For a bit more in-between, natural look, try:

samples = 25
timeframe = 1
delay = 1

[Kalleheikki Kannisto]

I mean take out the device from the first footage altogether (make a clean plate in Photoshop). Then cut out the device from the second piece of footage and put it into the the first one. Then they will match up perfectly when you zoom from one to the other. You’ll just have to fade the background from one to the next. Probably something more creative than just a plain opacity fade: maybe a burn film type of effect, like a track matte with soft fractal noise.

[Kalleheikki Kannisto]

Seems pretty straightforward:

1) Make a spherical mask

2) Use this as a Path Object, about 50 points

3) Replicate about 10-15 copies, set rotation to 0-360 or more (adjust until looks relatively even)

Use point and lines rendered with depth of field.

Animate.

[Kalleheikki Kannisto]

For follower object’s position something along the lines:

samples = 10;
timeframe = .25;
delay = 1;
smooth_position = 0;
for(n = 0; n<samples;n++){
smooth_position += thisComp.layer("Object").transform.position.valueAtTime(time-delay-timeframe/2+timeframe/samples*n)}
smooth_position = smooth_position/samples

That takes (with the current variable settings) 10 samples in the time frame of .25 seconds around the delay point of 1 second to smooth the motion. To smooth further, increase time frame (and samples). A custom smooth function, in essence.

[Kalleheikki Kannisto]

Depending on the length and complexity of the first piece of footage I would probably replace the entire device with the one from the second piece of footage. Then you could do it pretty seamlessly with animated color adjustments for the device.

I’d ask the client if this sounds like an acceptable solution before spending time on it.

[Kalleheikki Kannisto]

Per what you are saying, it should work. Are the ellipses 2D or 3D layers? For 2D layers the path should be exactly the same size as you copied, for 3D layers the camera distance will affect the perceived size.

[Kalleheikki Kannisto]

32768 is the normal white value for a 16-bpc project. As opposed to 255 for 8-bpc and 1 for 32 bpc.

[Kalleheikki Kannisto]

As for your original question, don’t select “relative”, it adds your given size to the existing size.

—

More importantly, doing a brochure layout in Photoshop is not something I would recommend, InDesign is the tool for that. I’d go directly to InDesign for anything that has more text than a couple of large headlines.

But if you’re going to do it this way, just about the only sensible output format is PDF. The main being is that Photoshop is raster (bitmap) based including text, so setting type to print out properly from Photoshop is a real hassle. You’d need to have the file in CMYK format and text in 100 black only (0C, 0M, 0Y, 100K) for black text, then export to PDF to retain text in vector format (embedded fonts in PDF). Doing anything else, you’re going to wind up with the double nightmare of rasterized text that is printed in all four colors.

If you’re overprinting text onto an image, it gets even more challenging…