Larry S. evans ii
Forum Replies Created
-
Ah, didn’t quite follow that from the original post, but yes I could see where that’d be a problem.
I don’t know that this would work, but have you tried putting the tracked clip inside it’s own comp and then making the comp a 3D layer? That might keep the tracking from diverging and still allow the footage to cast a shadow from the light. Just a wild idea, but if that works let me know so I can add it to my list.
Larry S. Evans II
Executive Producer
Digital I Productions -
You could duplicate the tracked layer, use scale to “squash” it down to where it appears to be a shadow on the ground, and then use that layer as an alpha matte for a solid. The solid can then be mixed in overlay or multiply mode, have it’s opacity adjusted, etc. to get a believable “shadow” look.
Larry S. Evans II
Executive Producer
Digital I Productions -
David:
You know what, you’re right. I was thinking of Cinepak. My bad.
That being the case, how would you account for the repeated frames being described here? If the compressor isn’t doing it, then wouldn’t you think it’s some odd sort of pull-down error?
Frankly I always get a head-ache when I try to puzzle out pull-down schemes, but this would seem to fall into that category more than any other. -R
Larry S. Evans II
Executive Producer
Digital I Productions -
I suspect the problem here is the “Animation” codec you are using for QT output. Animation is one of the worst compressors out there, designed years ago to create small clips for use as part of interfaces or displays, on machines that ran 15fps at best.
Try setting the compression settings to none and see if you get the same result. If so, then you may have some setting regarding video pulldown that is off. The nearest standard comp setting to the frame rate your are specifying is NTSC D1/DV Widescreen (23.976), or NTSC 24p.
If for some reason you’ve got a setting to remove pulldown or render fields it could be resulting in getting the two frames for every odd frame. The defaults are off for this, but your source video may have set the comp up differently. You can check by looking at the Render Settings when going to render out the clip.
Hope this helps.
Larry S. Evans II
Executive Producer
Digital I Productions -
Not to state the obvious, but have you checked the clips in some kind of player (i.e. Quicktime or Windows Media Player, etc.) to insure that the audio problem is not in the clip itself?
Failing that, you might try the following:
Save your file.
Go to the Edit menu and select Purge> All. This should clear your RAM of anything from the session.
In theory when you go to render the output module should find a nice clean spot of RAM to work with but that may not be physically possible.
If the purge doesn’t clear it, save the file, close out After Effects, and reboot your computer. Come back into After Affects, load the file and without performing any other operation, render it out.
If you are still getting the noise there is probably something else causing it besides a RAM issue, at least as far as RAM being in use.
Larry S. Evans II
Executive Producer
Digital I Productions -
Is it a “bad idea”? Anything that fits your creative vision is inherently a good idea, at least as far as the vision is concerned.
There are technical considerations involved. I think to get it to “stick” at 7.5 fps while the rest of the comp is running in 30fps requires placing it in a comp of it’s own and nesting that in the 30fps comp with a specific setting.
Truthfully I haven’t done that often enough to remember what the exact requirement is, but if you look around the forums here or in the tutorials at videocopilot.net you’ll run across the specifics. It’s probably in the manual/help files too, but good luck on finding it.
But bearing that it mind, if you can keep the “fixed” clip running as you want while the others run, you would be well on your way to getting it resolved.
Larry S. Evans II
Executive Producer
Digital I Productions -
It’s entirely possible that if this was something sent via e-mail that they altered the frame rate to make the file smaller. When you import that into a 30fps comp, then you get a sped up playback, and when you try to slow that down, you get a crappy slow-motion playback.
Always best to work with highest quality you can get, even if it means taking the time and bandwidth to FTP or paying for a courier service to get it on a disk.
Good luck.
Larry S. Evans II
Executive Producer
Digital I Productions -
Okay, knowing what you are trying to do gives me some better idea how to address the question. Unfortunately the answer is that you probably aren’t going to be able to get what you want.
As the images you are using already have motion blur in them (as you say -“you can see the blur (arms and legs) of movement”) changing the speed of playback is always going to look unnatural, because you will invariably get a “still” that shows that blur. There’s not really a way to remove it (there are tons of ways to add it). I honestly don’t know if frame-blending would address that or not.
Here is something that you might try, and I can’t guarantee that it would work, but it’s what I would try to do if I had a similar problem.
I would take the sequence you have a 1 second, and time stretch it to something like 1.5 seconds. then I would render that out to a new AVI file making sure I had turned Frame Blending on.
Once done, I would import that clip, and time stretch it by another 50% (make it 3 seconds long), again turn on Frame Blending and render it out.
You are still going to end up with a “slow motion” clip, but this might give you a more natural look to the motion blur. I really don’t know.
Changing the frame rate will have exactly the same effect, and there are a lot of complications to having a clip with one frame rate in a comp with another. But essentially, if it moves at 7.5 fps against a background of 30fps, it will be the same slow motion shot.
After a certain point you have to decide that the clip is just not going to work.
BTW, I posted to your thread regarding how to get the “printed” look of the baseball card. That one, at least, I think I might have been able to help with. In this case though, I’d really see if you can get a better piece of video to start with, or you try to stretch it less.
Larry S. Evans II
Executive Producer
Digital I Productions -
If you are talking about trying to project “white” onto a colored wall and have the wall turn white, you’ll have to change the laws of physics.
There’s currently no way, regardless of what you create in an animation, of removing the actual color of a painted backdrop using a projector (beamer). The best you will get is to have the areas turn brighter. Depending on your projector’s lumens rating this will be more or less. The more lumens, the brighter the light. A very bright light close to the backdrop might wash out the color enough to get the idea across, but I think it is impossible to make them white.
However, the area you want to be “white” in the design should be white in After Effects, and the background area should be black.
Hope that helps.
Larry S. Evans II
Executive Producer
Digital I Productions -
After Effects only recognizes the primary alpha channel, and I don’t think it’s a matter of what it’s named as much as if there is an additional channel.
However, there are ways of applying data from two alpha channels provided you render them separately.
You can bring in your clip with the basic alpha channel for transparency. And then you can import the z-depth alpha render and use it as an alpha matte to the imported video. This will then enable you to apply effects like fog, blur, color correction, etc. to the video track using the z-depth information.
I frequently use this approach as I have the capacity to output z-depth as a greyscale sequence from applications like Vue Infinite, so that I have just the “alpha” information to use with the rendered video plates. Of course in that instance, I would use luma matte instead of the alpha matte because I’m using the greyscale values of the z-depth render itself, and not the z-depth information that is embedded in an alpha channel of a second video clip. If Maya will render the z-depth out as a greyscale sequence (I would think it should), you can do the same thing, and have less file overhead for the same effect.
Hope that helps.
Larry S. Evans II
Executive Producer
Digital I Productions