[Steve Bentley]

Re: color not matching what you made in illustrator now that you have it set to linear in AE: There might be some color management in use in a few places – AE, Illustrator, the file format you save to, or even your monitor depending on the brand. So choosing a color by eye or by the numbers in Illustrator won’t necessarily match by eye alone in another package where the same color management is not in use.
I’m not sure if Illustrator can embed the color profile like pshop can but keep in mind that embedded profile might be coming in to play as well.
And finally, sRGB is a pretty old color space and very limited (there are so few blues it is literally impossible to do an HD sized gradient from one side of the screen to the other without a band appearing). So when possible choose a color space with a little more… um … space.

[Steve Bentley]

There’s a couple of ways:
The mask way is simple but if you have overlapping splines you have to additionally mask where the overpass leaks through (like a script “e or a”). or in your case where the head catches up with the tail.

You can also use the gradient Wipe effect. Make a stoke or path in either pshop or AE than has a gradient that travels over the length of the path. So an S would be dark at the top and gradiate to white at the bottom. Then turn this layer off and take the layer you want to animate and apply the Gradient Wipe to it. In the effects dialog select the hidden layer and use this as your “Source” and then animate the “Transition complete” paramater. You can make the stroked gradient larger/wider than the shape you want to animate on and then choose a layer to be the masking layer to keep the original shape.

There is a script on aescripts.com called Boa that is stellar at this and has tons of other features too.

You can use the mask idea again but use a paint stroke that you animate on (stroke it in real time and then adjust the recorded stroke to your liking). Use this element as your revealing mask.

Depending on the shape you can use particular to acheive this as well.

[Steve Bentley]

There are all sorts of scripts available (aescripts.com) that mimic the functionality of the cloner in mograph in C4d. They can all do a radial object that you then populate with whatever you want. Then use an effector to change the size or density to match the pattern you have shown.
You could also write a fairly simple expression that would populate objects in concnetric rings.

[Steve Bentley]

Keep in mind too, when using transfer/blend styles other than “normal”, you will get a different result when there is no background layer. Just having your comp have a black background in the comp preferences will have a different look to using a black solid at the bottom of the layer stack. With no true layer background there is nothing to do the math on.

[Steve Bentley]

You can add an expression to the fill color parameter (also the outline) and then pickwhip another shape layer’s color. When you change the one that all the layers are pickwhipped to, they will all change.

[Steve Bentley]

We’ll if you are going to say with CS6 keep in mind that it won’t saturate either the processor or the ram like newer versions will. I don’t remember if it was finally in AE2017 or 2018 but the first time I saw all my procs pegged and the ram filled to the brim I almost cried – finally after all these years! Now if they could just get their memory management system sorted, and y’know a proper paint.
If only adobe owned a preeminent paint package they could maybe borrow from.

[Steve Bentley]

Anybody remember hand animating something with frame by frame drawings? The same element over multiple frames is a dead giveaway. Most cell animation was/is done at 12 or 15 fps and then double printed to get up to 24 or 30 fps.

[Steve Bentley]

It could have been rendererd out of AE with interlace or there could have been a dot pattern from CMYK printing on the original scanned art that is creating a moire with the screen pixels of the video.

[Steve Bentley]

The better fog machines can use Mineral oil and you can tweek the nozzle to reduce particle size. Water’s surface tension is such that even when you generate small droplets, the minute you let them loose they glob together and you are back to a big droplet. One trick we have used in our water foggers is to add some camphor. This reduces the surface tension and may get you what you want.
Such is the nature of natural elements (fire too – trying to blow up a model is just as challenging due to flame sizes – the brain just knows something wrong).
If you use oil just remember to bag your camera and use a disposable UV filter to keep the mess out of the camera and use appropriate breathing protection.
You could also try dry ice (and water). Its the water particles you are seeing but because it wants to sink to the ground it tends to hold together a bit better and be more dense – but it does have a look.
The biggest issue with any of the above is the time between shots and the lighting. Once you start you build in this base fog level to the studio and then you have to wait for a while for it to dissipate or it shows in the shot.
Shoot wider than you think you will need but with a long lens.
The minute the fog leaves the frame you suddenly have an edge that limits where you can place the element. When shooting erupting fire we always shoot sideways so the fire leaps in the long direction of the frame and we tell the fire union that it can’t leave the frame..
And shoot at the highest true rez you can. This has the advantage of letting you use the footage anywhere in the comp and when you scale it down you are also scaling the water droplet size.
If you want to generate some footage of some fairly even fog moving about you can always use a turbulent displace to give it a few more “knots” once you pull it into AE. Then you can manually match the scale of the stuff one notices to the shot.
Shoot at the highest color depth you can. You can then crush levels in AE (work in a 16 or 32 bit comp and use 32 bit clean filters) and keep the fidelity but that crushing will bring out whisps and knots you didn’t even know were there.

I think you could get there with Turbulence FD ($$) or Blender (free – I’m not a blender guy but the sim engine looks eerily similar to TFD) but it can take a while to get the look you like and you will look up from the computer and the day is shot.

[Steve Bentley]

I’ve got both Mac and PC’s sitting in front of me and I’m a die hard Mac guy from way back…. but. If you look at cost/performace its hard to stay in the Mac camp. We recently added a new big box to our arsenal with specs like 24 cores, running overclocked at nearly 4 ghz, 128gb of Ram and a not-too-big Nvidia. On the PC side that came in around $4500. On the Mac side it was pushing $12k!
As for the graphics card, while there are more and more features in AE taking advantage of the myriad Cuda cores on Nvidia, it’s really the Ray tracing in Ae’s 3D world that uses them. I think for Particular it’s just the preview in the new Designer area that uses the cores. I haven’t seen a huge difference in render times in the comp. For the kind of work we do we’ve found more physical cores have been better than more Cuda cores.
Re stuttering – We did have shredding problems on the big box at the beginning with Nvidia – and that was a first for us after dozens and dozens of Nvidia cards. It turned out to be a monitor incompatibility in one case and a digital vs analong DVI input in another. The Nvidia guys are stupid smart so you shouldn’t see any issues. If you do it’s probably not the card. Make sure to download the latest drivers from Nvidia and not from your card manufacturer.