[Joey Foreman]

Try selecting all the keyframes and adding Smoother (Window > Smoother), playing with different values. Undo if you don’t like the results, and tweak to taste.
There is also a Smooth expression, but I haven’t used it.

[Joey Foreman]

I don’t really understand the question. What is a “pre-animated clip?” Pre-rendered, maybe?

And what do you mean by “frames that won’t animate properly”?

If you’re talking about Prerenders, possibly you’re rendering to a non-editable codec such as H.264?

I refer you to Dave LaRonde

Dave’s Stock Answer #1:

If the footage you imported into AE is any kind of the following — footage in an HDV acquisition codec, MPEG1, MPEG2, mp4, m2t, H.261 or H.264 — you need to convert it to a different codec.

These kinds of footage use temporal, or interframe compression. They have keyframes at regular intervals, containing complete frame information. However, the frames in between do NOT have complete information. Interframe codecs toss out duplicated information.

In order to maintain peak rendering efficiency, AE needs complete information for each and every frame. But because these kinds of footage contain only partial information, AE freaks out, resulting in a wide variety of problems.

Also, I’ve seen Render Multiple Frames Simultaneously cause problems on some systems, so if that’s on try turning it off.

Finally, disable OpenGL if it’s on. I think Dave might have a stock answer for that as well.

[Joey Foreman]

This isn’t a very tough key.

I duplicated the layer and applied Keylight to both. One for a Core Matte and one for an Edge Matte. Both are set to Combined Matte, with the top layer (core) set to Add. These two are precomposed and set as a Luma Matte for the unkeyed original layer.

For the Core Matte, in the Screen Matte controls, crank down the Clip Black and Clip White controls until all the green is gone from around the subject, and the body is fully matted with no holes. In Combined Matte this will be a fully white matte for the body surrounded by full black. There will be no edge detail. Then use Screen Shrink to choke this in.

For the Edge Matte, I just cranked down the Clip Black until the green around the subject was gone. The matte is full of holes, but that’s what the underlying Core Matte takes care of.

This method bypasses Keylight’s Spill Suppresion, so you might need to add AE’s Spill Supressor afterwards. Your subject was at a good distance from the greenscreen, so it doesn’t look like you have much spill anyway.

Follow this up with a simple rectangular garbage matte (mask) around the subject on the base layer.

I suggest just keying out the shadow completely, because his is so faint.
With some work you can add a realistic shadow, or use a 3rd party plugin like Red Giant Warp.

[Joey Foreman]

He mentioned he used some 3rd Party plugs. Maybe Mocha Shape? Along with the Stroke Effect?

https://www.imagineersystems.com/products/mocha_shape_for_After_Effects/

[Joey Foreman]

Is your y position a whole number or a does it contain decimals?
If so, try rounding off.

[Joey Foreman]

What do you mean by “noticeably different quality?”
Edge detail, color/gamma fidelity, lack of visible artifacting?
All of these I suppose.

Animation at full quality is widely considered visually lossless.

Arguments aside, you’ve got several choices. Why not run some tests?

8 or 10 bit Uncompressed
ProRes 4444 or HQ
None
Tiff Sequence

[john lynn] The footage comes out uncompressed and as if it was straight from the camera.

Very few cameras have uncompressed output, e.g. Sony F950, Viper, CineAlta, and certain ‘lower end’ cameras that have been hacked to bypass compression through HDMI and a capable capture card. And of course film, via scanning.

Color or Gamma shifts may occur with any of these without a comprehensive color management strategy that takes into account platform, application, broadcast spec, external monitoring, bit depth, chroma subsampling, etc., etc., not to mention acquisition format, in-camera compression, and, well, that just about scratches the surface.

And if your footage is interlaced, the deinterlacing algorithm used is yet another factor in determining quality.

Oh yea, don’t forget about de-artifacting.

[Joey Foreman]

I think you’re overcomplicating this. In FCP your Sequence settings should match your final delivery settings. It’s ok to render the AE file with less compression, i.e. Animation Codec. In FCP it will render down to your Sequence Settings’s compressor. What this is should have been specified by your client.
If your AE render is non-interlaced which it is by default, then it will look jaggy in FCP if your Sequence Field Dominance is set to Upper or Lower.
And as far as settings themselves, there are only a hand few:

Resolution
Pixel Aspect Ratio
Frame Rate
Interlaced or Non.
Compression Codec

The reason I suggested changing your Sequence compressor to Intermediate Codec is that it has some of the advantages of the new ProRes codecs, and is good for maintaining quality of items rendered with higher quality codecs like Animation.
But that’s immaterial if your client needs output back to tape or something. Otherwise Apple Intermediate should serve you well for a final Export from FCP, whether you’re going to web or DVD.
But then you’ll then to make another, differently compressed version for those, and that will need to be done in an application other than AE or FCP.

[Joey Foreman]

Hopefully will solve all of it. I know I’ve got my fingers crossed:)

[Joey Foreman]

I keep forgetting you’re on 5. Change the Sequence Compressor to Apple Intermediate Codec. And do set the Field Dominance to None.

You’ll need to render your sequence afterwards.

Those After Effects settings are for the Composition dimensions and frame rate, but they don’t have anything to do with which codec you choose to render to.

[Joey Foreman]

Oh,and change the compressor to ProRes 422.