[Andrew Somers]

Newton looks pretty clever – though this can be done easily with expressions.

Here are a few tutorials:

https://www.videocopilot.net/blog/2007/08/gravity-in-after-effects/

https://library.creativecow.net/articles/toula_jesse/Gravity-in-After-Effects/video-tutorial

Here’s an expression you can cut/paste:

https://www.motion-graphics-exchange.com/after-effects/Better-Bouncing/4d620fac97e8e

[Andrew Somers]

Red or film footage it is preferable to work in 32 bit linear for all compositing. You *do not* want to work in “Log” space.

You will only convert to LOG DPX sequence if that is what the DI house wants – but that is an export issue, not a working space issue.

FIRST: let’s clear up a misconception – what is “LOG” space really? It is actually a LINEAR encoding of FILM DENSITY – but guess what? Film density is log – therefore, we are used to talking about film footage in 10 bit DPX as “log”., or put another way, the gamma curve is “log” – interestingly, the gamma curve for regular video is actually quite similar to log – and it is technically incorrect to call regular video with a gamma curve “linear”.

“Lin” or Linear means a “straight line”, also known as “Gamma 1.0”. “Vid” or Video typically means with a particular gamma curve applied – and that gamma curve will vary depending on the color space used.

Stu discusses this at length:

https://prolost.com/blog/2009/9/30/passing-the-linear-torch.html

https://www.fxguide.com/featured/After_Effects_32-bit_Workflow_In-Depth/

But to sum:

Vid: Implies a gamma encoded image file.
Lin: implies an image file with no gamma (aka Gamma 1.0)
Log: Implies an image file encoded with film densities, i.e. a gamma that evaluates as film densities.

Your workflow:

So what you want to do is bring in your RED footage to an AE project that is set to 32-bit linear, and pick a colorspace for working – in most cases I suggest sRGB when working in linear space, though there are reasons to work in a larger space at times, in *most cases*, a working space of Linear (gamma 1.0) sRGB is preferable, particularly if you are going to end up in Rec709 (HD) space.

If you are going to output to film or to Digital Cinema, a case can be made for working in a larger space like Adobe 98 or DCI/P3 (particularly if you have a true wide gamut display that is properly calibrated). I do suggest that you avoid working in any of the really large spaces like “ProPhoto” however.

When you do your compositing in AE (or Nuke) using linear space, then when you ADD, you will be adding images the same way that lights add in real life (that is, real light is linear).

Set your comps to *Use Display Color Management” Under the “VIEW” menu.

When you EXPORT, you can use the Cineon converter method, or use the “Universal FIlm Profile”. You should discuss with your DI house what specs they want for the show. If you use the Cineon converter method, one common setting when *coming from a linearized workspace* is:

“Gamma 1.0” is the same as saying “Linear working space”.

The “Log Convert” checkbox will then convert to log space.

Use your working space for your color profile in most cases here.

If you use the “Universal Film Profile”, it takes care of the Cineon/log conversion for you.

[Andrew Somers]

You can do *nearly everything* that the 3rd party plug ins do, right in After Effects, using the built in plug ins.

The 3rd party plug ins just “make workflows faster/more efficient”.

Personally, I like to design/create my own effects using the “standards” (levels/exposure/curves/blur/displacement), and expressions and scripts I write for the function. Since I’ve been fortunate enough to work mainly on features, I’m afforded the time to do the R&D to get these custom looks. For Advertising and Television though, deadlines are so tight that a good set of plug-ins means the difference between making deadlines vs not sleeping.

But sometimes there simply is no time, so then I may use a 3rd party plug-in.

AE ships with a version of Mocha – for tracking you can do *almost* everything you’d be able to with a $$$$ tracker (see the Copilot tutorial on using Mocha to place objects in 3D space).

But there are a few cases where a 3rd party plug in really makes a difference in terms of final quality. I tend to think of the following plug-ins as “Hugely Helpful” and well worth the money – mainly for making AE’s 3D space actually a useable feature:

For working in 3D:

Trapcode Particular – the best particle generator for AE, and it works in 3D space, with the AE 3D camera and lights. Related to this, Trapcode FORM (which I actually don’t use) also allows for some really spectacular abstract 3D designs.

Zaxwerks 3D Invigorator Pro- for down and dirty 3D inside AE, this is pretty amazing. I still haven’t upgraded to v 5 though as I am now using external 3D for the CGI stuff. But for $500, and the easy integration with AE, it really can’t be beat.

Some “generally useful” ones that are often worth having:

Trapcode SHINE does the light ray ting batter then the built in CC Light Rays, and it renders very fast even in 32 bit mode. Trapcode LUX is a nice trick for volumetric lighting in AE 3D space.

RE:Vision effects have some very useful plug ins, especially Reel Smart Motion Blur – this is indispensable for working with CGI. RE: Fill and Twixtor are also very useful when you need them.

If you can afford them, the Sapphire lug ins can do wonders for your workflow speed – I don’t have them but I have used them and they are super-fast and versatile to get a lot of those “looks” that some clients demand.

For compositing CGI, the Walker effects are quite useful – though he sold them to Digieffects, and they are included in the DELIRIUM collection.

Also, upgrading from the bundled Mocha to Mocha AE v 2.6 gets you more features and better integration at a pretty reasonable price.

But otherwise, for someone starting out with AE, I recommend learning how to use expressions and the built in plug ins *first* instead of relying on the off the shelf “cool thing” – your skils in AE will greatly improve as a result.

[Andrew Somers]

When frames go behind other frames. you’ll need to use masks – Mocha is great here too, as you can make roto masks that auto-track the moving objects.

Currently you can get Mocha 2.6 for AE, which comes with a free plugin for AE that allows for easy import of Mocha data.

[Andrew Somers]

Yes, you may need additional plug ins if the distortion is complex – if you have Sapphire plug ins they have one called “WarpMagnify” that has a several additional parameters.

[Andrew Somers]

There is a built in lens flare under the GENERATE list of effects.

[Andrew Somers]

I agree – the PS 3D to AE is not a particularly useable solution, but Zaxwerks 3D Invigorator is quite reasonable in price, and very useable.

The one caveat is that 3D Invigorator does not integrate will with other 3D layers, and effects like Trapcode Particular (you need to use a few work arounds), and it does not do ray tracing. But otherwise is a joy to use and a very fast way to knock out basic CGI.

[Andrew Somers]

Steps:

1) Out of PF track, export a frame (or a few) of the image after the UN DISTORT node. Ideally, choose a frame with a lot of sharp details, and vertical and horizontal lines.

2) In after effects, create a comp with the camera original footage in it, and add the frame you exported from PF track so that it lines up with the same frame(s) as a layer over the original footage.

3) Add the OPTICS COMPENSATION plug in to the undistorted frame from PF track.

4) Set the transfer mode of the undistorted frame so that you can see it overlaid on the camera original – I usually use the DIFFERENCE transfer mode, as once you have both images exactly aligned, the total image becomes black (or close to it).

5) Now adjust the optics compensation effect until both images align exactly (in difference mode, the entire screen should be nearly black.

NEXT:

6) Create a pre-comp for your CGI elements.

7) Place this precomp in the comp with your original footage.

8) Add the optics compensation effect to the precomp using the exact same settings that you came up with in step 5 above.

[Andrew Somers]

That is more commonly referred to as Moiré

https://en.wikipedia.org/wiki/Moiré_pattern

The term “Zebra” is used to describe the overlay that is sometimes used to indicate “out of legal” brightness or colors in a camera viewfinder (or similar places).

To avoid moire in the future, test wardrobe to make sure that it does not have a pattern to cause the moire’

In the meantime, it can be corrected using a low pass filter,, that removes the higher frequency detail from the image.

In after effects the “low pass” filter is also called FAST BLUR or GAUSSIAN BLUR. 🙂

To fix the moire, create an adjustment layers and add FAST BLUR, and adjust just high enough that the moire disappears. THEN use a mask on the adjustment layer to constrain the blur to the area needed.

[Andrew Somers]

Here’s an idea off the top of my head:

For “continuous motion” of the foreground man, I think he needs to be in a shirt that starts with stripes and is then animated into a solid.

Shoot the actor in a shirt that is striped with your final blue but with pure green stripes (for keying purposes).

Have him step to the right.

Shoot him in the red shirt, stepping left.

The BLue/Green shirt will be the foreground man, and the red shirt will be comped behind.

Then Key the green and make it match the rest of the blue in the shirt, using a mask to wipe into the final solid blue.

The big difficulty here is getting the density of the green and blue stripe to be close. But probably easier to getting the geometry correct of stripes onto a pure green screen shirt. By putting the green stripes on the blue shirt, you can use the blue with an additional keyer to help match density.

(note: if you can’t reshoot, then take the element of the red/blue striped shirt, and see if you can key the blue into a red that matches for a full red shirt – but without planning beforehand, this may not work well.

If you could post some key frames of your footage, it would be helpful.