[Steve Bentley]

Re: the OP and the way-back-when post.
If you add an expression to the animated attribute but just let the default expression stay, then select all the keyframes and right click on one of them and choose “keyframe assistant”, then choose “convert expression to keyframes”, you will get keyframes for every frame, select them all (they are usually already selected at this point). Hit copy – because its all keyframes it will include the easing data.
Then that can be pasted to Word or Excel and parsed to columns of data.
Sometimes you have to let Word do some search and replace to convert double spaces to comma’s etc so you have a comma, or space, or tab delimited table of values that Excel will like.

[Steve Bentley]

What about just duplicating your “head” layer but first put a small wiggle expression on it. The layer number of each duplicate will change the random seed of the wiggle so each is slightly different.

[Steve Bentley]

That’s just animated masks, and the logo and the footage being linked (parented) so that they move together when animating the scale.
You can use an animated solid as a mask so when you move the footage you move the solid too, but use that solid as a track matte. That way they travel together and the masking layer is moving or changing at the same rate as the move.

[Steve Bentley]

As in another post recently I think there is also (adding to Dave’s post) a noise that is acting as a luma mask with the noise’s contrast being animated so that it fills in over time. The block of text seems to radiate outwards as it comes on and fills in unevenly which the noise will do.

[Steve Bentley]

K. is right, I don’t even think it’s got any depth – its just a flat layer that is “angled” by the 3D camera track. In fact, based on the accuracy of the track (or lack there of) I think the camera move (or layer angle and position) was just keyframed to match the source.
I would have been nice to see some of the leaves reflected in that glossy face. That would have helped sell it.

[Steve Bentley]

With AE 2017 and later you now have control of each node along a path or mask. You can now animate each one (ok technically you animate nulls that drag the points around with them).
It might be as simple as putting a sin equation that diminishes over time in amplitude and position, or simply animating by hand.
However, because AE is only 2.5D and not full 3D, one layer can overlap the next but then that lower layer cannot now cover the upper layer, even with 3D layers.
It’s messy, but you could do multiple copies of your strands and mask each one at the outer “curve” before it heads back in to overlap or underlap the one before. So get the animation right first for the whole braid, then duplicate your strands as many times as you have overlaps and mask each as it appears and is needed.

But even the lite version of C4D can do this with sweeps along splines.

[Steve Bentley]

I’m not so sure about Daves advice on that. I used to be a commercial illustrator. If I drew something based on a photograph, even if I only drew certain parts of the photo, I could be sued by the photographer, even though the illustration was 100% me (and in a different medium).
It comes down to: can the judge see enough of the original in your work? This goes double if the content is used in the same “marketing sphere”. So a Studio bumper that is used in a similar way to your sort-of-copy studio bumper (like at the start of a film) would be too close, and a target for drooling lawyers.
If it causes brand confusion (“Hey, I didn’t know that Brendan Films was a subsidiary of Fox! It must be, it feels like a Fox logo.”), then that’s a target for lawyers too. Remember they don’t have to have a winnable case, they will just bankrupt you until you stop – a defense costs money too, win or lose.

The above is all moot if its a parody – then, anything goes.

But these things should be iconic. Why have people think of Fox when you want them to be focused on your brand. Be bold, forge your own brand and take a different path. I know what kind of film I’m about to watch before the Studio logo has even resolved because that “style” becomes so ingrained in the zeitgeist. It becomes like the game “Name That Tune”: you know it in the first few frames.

And to add to Blaise’s post:

Good artists copy;
Great artists steal outright!
-Picasso

And my addition:
Great artists are rich enough to get away with it. (and have really good lawyers!)

[Steve Bentley]

If you have Particular you have full control over that kind of thing, but with just fractal noise you can try animating the contrast of the noise, this will give the effect of the “particles” scaling as each of the noise blobs gets smaller as it is crushed by the contrast.
The fractal whisps can be hard and that why programs like Particular and Form have come along. Combining what you had with Sabre might get you close – let Saber do the whispy stuff (you can turn off the core).
Don’t forget to use multiple layers (sometimes it takes 5 or 6 copies) and screen them over top of each other. If they are all slightly different and moving at different rates you get them adding up at different times as they pass through each other which gives a great ethereal “sparkle”.

[Steve Bentley]

There are a number of ways to do this:
You could pass the text through a mask that is made of noise (use luma matte) and then animate the contrast of the noise so it starts as all black and then it eventually “whites out”. This creates text with holes in it that fill in.
or
For more of an edge effect, premcompose a copy of the text, blur that text and animate the blur so that the text blurs so much you can’t see it and animates to sharp. Combine that with a noise in the precomp so you get a ratty edge- you may need to scale the noise to keep up with the blur of the text. Then change the contrast with an adjustment layer to have the precomp’s white areas expand or contract and use the precomp as the mask that bring on the text in the main comp. Then you can fade out the noise or use the contrast in the noise trick above to make the noise fade with the animation. This will bring the text on from the centre of each character out, like a water color bleed.

[Steve Bentley]

This is always a topic of heated debate around here. While a sphere should be the best shape because it doesn’t change shape or aspect no matter its orientation, it is affected by many things – size (distance from camera), shape, once motion blur is included, extended shape, when it starts casting shadows, and shading, when the sphere has only one side lit well. On top of that, when you incorporate the square grid of pixels that make up the image it starts to no longer resemble a perfect sphere.
For projects where tracking was used for highly accurate data tables, we’ve used small reflective spheres (scotchlite) or spheres coated in IR sensitive paint and photographed with that type of camera and light, so the only thing you can see is a perfectly round dot. Using a highspeed camera eliminates the motion blur and the near camera lighting eliminates shadows and shape distortion. If using film, use a radial shutter (leaf) in the lens (instead of a pull down) and if using digital use a super fast chip: both to remove image smear that happens during the exposure. For highspeed use a rotating prism.