[Tudor “ted” jelescu]

Precomp your AI artwork in a large comp and apply CC Sphere to it.
You can resize that precomp (in the settings) and the AI artwork (in the layer transform controls) to get more detail in your zoo. CC Sphere treats the layer it’s applied to as a bitmap so the bigger the layer the better the resolution zooming in.

Tudor “Ted” Jelescu
Senior VFX Artist

[Tudor “ted” jelescu]

Then it is best to vectorize that artwork in AI using live trace and then use the vector image in AE.

https://www.smashingmagazine.com/2010/11/15/illustrator-s-live-trace-sketch-to-vector/
https://www.wonderhowto.com/how-to-vectorize-graphics-adobe-illustrator-414330/

Tudor “Ted” Jelescu
Senior VFX Artist

[Tudor “ted” jelescu]

In your image there is a background- it is blurred and non descript, but there is a background.
To give the impression of water drops sliding down the glass you need that background to be clean, undistorted, without the water drops on it. You can paint that in Photoshop to clear out the path for the drops you want to animate. You can select then each drop you want animated and cut it out. Use that to composite on top of your plate. you need then to create a displacement mask in the shape of the path on which the drop will fall – a black and white image – black means no distortion, white is the maximum. Animate that displacement map to reveal the path as the drop goes down. Use it to displace the plate and to give the impression you’re after.
CC Mercury may be able to give you some good animations you can use as displacement maps as well.

Tudor “Ted” Jelescu
Senior VFX Artist

[Tudor “ted” jelescu]

“i think there would maybe be a benefit when rebuilding the rendered pieces back together since you would be going from the native codec to native codec rather than re encoding from jpg’s or tifs to pro res or animation.”

I am unclear what you mean by that- AE can render in pretty much any format you want it to render. What native format you are referring to?

The whole idea of a render farm is that it will render out individual frames and collect them in one location. I do not see any benefit in splitting your comp and rendering out whole parts of it.

Tudor “Ted” Jelescu
Senior VFX Artist

[Tudor “ted” jelescu]

Best to give more info than less when you ask a question- post the image, it matters a lot if there’s only a solid color beyond the glass or if it’s a window through which you can see a street for example.
Are you planning on compositing something behind or just animate that image?
Answer these for a specific answer.

Tudor “Ted” Jelescu
Senior VFX Artist

[Tudor “ted” jelescu]

It depends a lot on what kind of image- is it a photograph or artwork that can be vectorized? If it’s a photo the way you started is a good one- cutting the image in pieces and parenting them to a null.
If it’s artwork that can be converted to vector then it’s worth doing that and then you’ll be able to zoom in as close as you want and not worry about size.

Tudor “Ted” Jelescu
Senior VFX Artist

[Tudor “ted” jelescu]

In that case, a dslr is a good choice in my opinion. If you can get a few decent lenses then you can get a cinematographic look, and you can use that for stills as well. Just remember, if you want to shoot green screen or do heavy compositing or color grading you may not get the best results. For testing and school projects though, it is a good option on a low budget.

Tudor “Ted” Jelescu
Senior VFX Artist

[Tudor “ted” jelescu]

With a LOT of preparation and Hollywood size budgets…
For a less complex effect you can study these tutorials:
https://www.videocopilot.net/tutorials/advanced_time_freeze/
https://vimeo.com/11022627

Tudor “Ted” Jelescu
Senior VFX Artist

[Tudor “ted” jelescu]

CS5, 64 bit is the way to go. You can still get a good couple of years in my opinion on CS4, but it’s like when you switched from a single processor to multiple processors in a computer.

If you shoot for fx, camcorder for sure, especially if you can skip recording to a card/drive/tape and go straight to a hard disk.
A dslr will give you good looking picture but with a lot of compression which will take out most data from the image that you would use for chroma, grading and fx in general.

Tudor “Ted” Jelescu
Senior VFX Artist

[Tudor “ted” jelescu]

Stabilize the background, do your comp/fx, then reverse the stabilization to get back the camera shake on the finished comp. No need to worry about the buffers that way.

Tudor “Ted” Jelescu
Senior VFX Artist