Joey Foreman
Forum Replies Created
-
The “black bars” are just the underlying background color of your comp being revealed as the edges of the footage shift position to compensate for the stabilization.
If you’re working with a comp size that’s the same as your footage, yes, you will have to scale up your footage some, which will soften it. This is the main reason I don’t use stabilization.
One alternative would be to scale your footage up in a program such as Apple Compressor, using Frame Controls, Better or Best settings for scaling.
Another alternative would be Red Giant’s Instant HD, to scale up to a larger res of HD.
I’ve also gotten pretty good results from Magic Bullet, using its antialiasing controls.
None of these methods is perfect, though. You will lose some quality from upscaling, but depending on your footage, it might be imperceptible.
Just go ahead and increase the scale of the layer past 100% until you can’t see the background, just to see if the results are acceptable to your eye.
-
I linked to a tutorial in my post.
-
I think you’re talking about Stabilization via the standard AE tracker. I’ve never used Boris, but I’ll guess that it outperforms AE’s.
However, AE CS4 ships with Mocha AE, a planar tracker, that basically performs miracles, and, unless Boris utilizes planar tracking as well, I’d say it will beat the pants off Boris.
Check it out:
https://www.imagineersystems.com/products/mocha4ae/Mocha_image_stabilization.mp4
-
AE doesn’t have light falloff functionality. But you’re right, you can fake it with an expression:
https://www.motionscript.com/design-guide/falloff.html
-
It’s important to understand the difference between vector and raster imagery:
https://help.adobe.com/en_US/AfterEffects/9.0/WSFD576BBB-B11B-472b-971F-25CCD5CE25CBa.html
Once you’ve scaled down those items in AE and rendered out a movie, you can’t get back the original resolution of those clips. Their scaled-down size is now “baked-in.”
Why not do all your zooming and panning in AE? That’s what it’s for.
But keep in mind that you can’t do any zooming in on an image such that the zoomed-in position exceeds its original dimensions (over 100% scale), in the comp window, without losing quality.
-
Check your Final Cut Pro Sequence Settings and make sure that they match the footage exactly.
Same Resolution.
Same Codec – Animation – Not NTSC/DV
Same Frame Rate
Same Pixel Aspect Ratio
Field Dominance Set to None.Starting with FCP 6, the sequence creation process became automated. With 5 you have to be more vigilant and make sure all this stuff matches up.
-
To try and answer your questions in reverse… The light on your object was set in photoshop, either in Scene or Material options.
To put your object in a new comp of different dimensions, you have to nest the comp it’s already in, into the new comp, and make it a 3d layer.
Now you can add new 3d layers in this comp. They can intersect, but only as a 2d object intersecting a 2d object in 3d space. Good old 2.5d as it’s known.
And as far as I understand, the 3-button mouse is used to manipulate the Unified Camera Tool, not the “3-d” objects themselves.
Photoshop 3d in AE is very limited, and, of course it’s not really 3d at all. It’s sounds like you would really appreciate the flexibility and near-endless possibilities of a dedicated 3d app. Why not download Blender and give that it go?
-
I’d go with Kevin’s first option. I used to do the 1280 X 720 Square comp thing, but it was pointed out that this causes a loss of quality by stretching the footage horizontally.
Just toggle Pixel Aspect Correction while you work, then render with the same settings as the comp.
Quicktime player will only display the actual size, so it will appear squished, but FCP knows how to display it, and will export it properly. -
Yep, Todd’s right. You can learn to work well within your limitations, rather than just throwing money at the problem.
I started working with AE in 2000 on my trusty ole G4, 1.5 Gb RAM, on projects with tons of layers and effects. You figure out pretty quickly how to generate longer previews by switching off what’s inessential, lowering resolution and/or quality, prerendering, using proxies, region of interest, etc., etc.
I work on much more powerful hardware these days, but none of that’s changed.
It just has to do with the limitations of today’s hardware. Eight core, 64 bit, whatever, you’ll still have to employ the same workarounds. $500.00 worth of new RAM might get you another 15 or 20 seconds, unless you add another effect or layer, then you’re right back where you were.