Can I Do This In Final Cut Express?

[gvdv]

Hi,
I'm about to buy FCE HD, and I'm relatively new to editing digital video in general (edited film briefly about 20 years ago and have just done a couple of amateur videos to teach myself the ropes).

I have an effect that I'd like to achieve, and don't really have the language to describe it or the awareness of how to go about it, although my research may have pointed me in the right direction. So, I'm hoping that people here can help me and that you can tell me whether or not FCEHD is capable of doing this.

I'm thinking that this is an example of compositing - so let me know if I'm wrong.

Here is what I would like to do.

1. Start with a black screen.
2. Slowly, as if we were zooming out and moving backwards, the letters of a word begin to appear on screen (they don't fade in - this is difficult to describe - they appear at the edges of the screen, and then, as the illusion of the viewer moving backwards continues, more and more of the word is revealed. It's kind of like the text has been 'outside' of the screen, and moves into the screen, becoming smaller as the screen encompasses it. The only illustration that I can think of that does this is the THX 'The Audience Is Listening' text that precedes some movies in theatres).
3. The letters of the word at first are white (actually transparent), but as we see the whole of the word on screen, we see live action video (and hear audio of their voices) of different people 'under' the letters.

So, is this compositing? (Not the blue screen type). If not, what is it, and how do I accomplish this with FCE?

Thanks for any help,
GVDV

 

[Multimedia]

LiveType 2 Page Link • Understand FCEHD 3.5 Is Going To 4 Soon

Start your research from this FCEHD LiveType2 web page plus Google.

You do understand Final Cut Express HD 3.5 is probably about to chnage versions to Final Cut Express HD 4 in April right?

 

[gvdv]

Hi Multimedia,
Thanks very much for the web page address, and yes, I think you mentioned elsewhere that FCEHD4 was about to come out. Thanks for the reminder, though.

All the best,
GVDV.

 

[seniorstinky]

Is this kinda what you are thinking of?

http://carsonfamily.name/movies/demo.html

 

[gvdv]

Is this kinda what you are thinking of?

http://carsonfamily.name/movies/demo.html

seniorstinky,
That's almost exactly what I was thinking of!!!!! How did you do it?

The only difference that I have in mind, and I didn't say this in my original description, is that I would have many different, smaller, video images going on within each letter, rather than one video image with the letters overlaying it.

Having said that, the zoom is right, and the principle of the letters overlaying the video is right on.

Can you let me know what the technical 'terminology' is for this - are we talking overlays, alpha channels or what? And did you do this in FCEHD?

Thanks for taking the time to do this,
GVDV

 

[AviationFan]

I would have many different, smaller, video images going on within each letter, rather than one video image with the letters overlaying it.

So as the letters change size, will the many different video images associated with each letter also resize and move around?

Can you let me know what the technical 'terminology' is for this - are we talking overlays, alpha channels or what? And did you do this in FCEHD?

For the example that seniorstinky pointed you to, you'd create two video tracks: the bottom one has the video image, the top one has a mask, in this case a plain black still image with transparent "cutouts" for the letters. The use keyframes in Final Cut's "Motion" tab to scale (and move, if necessary) the mask layer.

As usual, there are different ways to achieve this result, but this is how I'd do it.

- Martin

 

[seniorstinky]

I did this mainly in LiveType, with the background of the lettering being a movie.

Once imported into FCE, use the motion tab to change the scale to the biggest size possible. Keyframe the initial frame and the frame where you want the size of the text to be "normal"

You'll want to change the scale back to 100 (or something smaller... whatever your "normal" size will be) on the end keyframe.

In your particular need, you'll want to follow AviationFan's advice. The only difference I see is that you need multiple video tracks, one for each letter that you want to "animate" with live video. You'll want to scale these down and perhaps mask them to fit the letter. You'll also want to move them around to fit into the letter using the wireframe.

Hope this helps.

Let me know if you need more clarification.

 

[seniorstinky]

The only difference that I have in mind, and I didn't say this in my original description, is that I would have many different, smaller, video images going on within each letter, rather than one video image with the letters overlaying it.
GVDV

I also was curious about how many different video images you are going to have within each letter. One video image is pretty simple, getting more complex may be a bit problematic.

And did you do this in FCEHD?
GVDV

All of what I did and from what I understand, all of what you need to do, can be done with FCEHD.

 

[huntercr]

Start your research from this FCEHD LiveType2 web page plus Google.

You do understand Final Cut Express HD 3.5 is probably about to chnage versions to Final Cut Express HD 4 in April right?

Where did you hear this? I know new FCP is coming out, but I didn't think it was time for a new Express. What features would there be new? The only thing I could think of would be shifting to the XML file format. There's very little about "express" that is crippled anymore for SD video anyway. The only thing I can think of off the top of my head is the 3 way color corrector, and support for 24fps, and of course the bundled tools ( DVDStudio Pro, Compressor, etc )

I'm curious, please let me know everyhting you know!!

( I'm a little sad becuase I just bought 3.5 2 weeks ago )

 

[jelloshotsrule]

everyone did a great job of explaining that 1. this can be done in fce, and 2. how to go about doing it...

as for the terminology, i'll take a stab at a couple of the things.

first, it is compositing. basically compositing is any sort of use of multiple layers to create some particular effect. whether it's blue screen stuff, or even titles, it all involves some level of compositing... putting together different components into one final product.

second, yes, in this case the letters would not only be white text, but you'd actually be using the layers as an alpha channel to reveal the video levels beneath it.

hope that helps.

 

[gvdv]

hi seniorstinky, aviation fan, and jelloshotrule,
Many thanks for the wonderful help that you have given me here.

Can't believe I didn't call my original idea a mask - I've only been using graphics programs for several years, and that adjective would have helped my original desription.

I do have two more questions: when you say this was all done in FCEHD, you're including LiveType as part of that description, right? Because it is designed to work with FCE? Or can this all be done within FCEHD itself?

And my final question, to do with your questions about whether or not the many different video images associated with each letter will also resize and move around. The short answer to this is that is maybe. There will be many (let's say 6 or 7 different video images) within each letter. I may want to have these fly in to each letter (in which case they will be like shards of glass flying in to the letters, and decreasing in size until they settle in to their final size). Or they may just fade up as the word takes its final place on screen.

So, from your previous answers, I presume that the latter case (many images fading up) would simply mean the use of multiple video tracks. Is that right? And what about the latter? That sounds more difficult, and like a mix of animation and ????? (I've no idea what I'm talking about now).

I'm actually really enjoying learning about this stuff, so many thanks for all of your help.

GVDV

 

[seniorstinky]

I do have two more questions: when you say this was all done in FCEHD, you're including LiveType as part of that description, right? Because it is designed to work with FCE? Or can this all be done within FCEHD itself?
GVDV

It can be done just in FCEHD itself, although I prefere LiveType for intro title scenes. The more I think about, the more it might be better, btw, to have a big font and shrink it down (rather than regular font and blow it up). Here's my thinking:

- Shrinking a font should show little to no pixelation compared to blowing up a small font

And my final question, to do with your questions about whether or not the many different video images associated with each letter will also resize and move around. The short answer to this is that is maybe. There will be many (let's say 6 or 7 different video images) within each letter. I may want to have these fly in to each letter (in which case they will be like shards of glass flying in to the letters, and decreasing in size until they settle in to their final size). Or they may just fade up as the word takes its final place on screen.

So, from your previous answers, I presume that the latter case (many images fading up) would simply mean the use of multiple video tracks. Is that right? And what about the latter? That sounds more difficult, and like a mix of animation and ????? (I've no idea what I'm talking about now).

I'm actually really enjoying learning about this stuff, so many thanks for all of your help.

GVDV

The way I see it is multiple video tracks as well using motion and wireframe to move the "shards". Getting that right is tricker than the fade up as the shards will be in the foreground.

 

[jelloshotsrule]

honestly, i, myself, would use after effects for this, and have barely used fce (only pro really)... but for the multiple clips within each letter, it's really just a matter of time and detail.

you will essentially have to resize and move (and keyframe both attributes) for all the video clips. now, what i don't know is if fce has a way for you to do a sort of nested clip.

it will be tough to make sure the video clips all move exactly the same way so that they don't slide in relation to each other. what might be worth experimenting with is making a video with just the word as it is going to be in the end, and just line up all your background videos under it as desired. export that as a video file (or perhaps just nest that sequence within another, if fce would allow that?). then take that, and scale it up/move it so that it's all black screen (essentially zoom in on the black space), and then animate it shrinking/moving back to regular position. while that is probably the least time consuming way, it will not give the greatest quality results as you will get pixelation as you see the letters/video zooming,since you are basically doing a digital zoom.

how long will the overall zoom/movement take? if it's quick, it might be worth doing it that way, if you will see the letters closer up on screen for a longer time, then you would probably need to do it the longer, more tedious way to have higher image quality.

 

[gvdv]

seniorstinky and jelloshotrule
thanks for these replies.

The pixelation that you're talking about sounds like it is the equivalent of noise in the still image world that can result from resampling a .jpg. Is this right?.It can provide a graininess with lossy formats like .jpg. However, I thought that the letters in programs like livetype were actually like vectors, i.e. infinitely (in theory anyway) scalable without any loss in quality. I hate anything that is in anyway representative of a digital zoom, and am a resolution freak in all contexts (hate jaggies, and anything other than crisply defined edges of letters).

jelloshotrule, the letters will probably be quite big on screen and the zoom will be on the slower side, so the tedious route is probably safest.

Thanks once again for your help in this.
GVDV

 

[jelloshotsrule]

The pixelation that you're talking about sounds like it is the equivalent of noise in the still image world that can result from resampling a .jpg. Is this right?.It can provide a graininess with lossy formats like .jpg. However, I thought that the letters in programs like livetype were actually like vectors, i.e. infinitely (in theory anyway) scalable without any loss in quality. I hate anything that is in anyway representative of a digital zoom, and am a resolution freak in all contexts (hate jaggies, and anything other than crisply defined edges of letters).

well, sort of... same concept. but it'll be basically the same effect as if you were to take a jpeg and zoom in beyond 100%. so, the software is trying to guess as to what the pixels should be between the resolution it was originally taken in, and what you have zoomed into... try a digital zoom on your cell phone or dv cam, etc... same thing

you are right that in livetype, fcp, etc, the text you create is vector based and IS scalable without loss in quality... so the issue would only come about if you were to use my theoretical approach... i'll explain a bit more

jelloshotrule, the letters will probably be quite big on screen and the zoom will be on the slower side, so the tedious route is probably safest.

the letters will come to rest quite big? that would actually make my proposed plan a bit better, although with a slow zoom i'm still not sure it's worth the time saved. what my suggestion would do is essentially take the final size... let's say the text fills up much of the screen, you have clips within the text, etc... and that clip is 720x486 or whatever. well, you would render that as a quicktime movie (or might just be able to use it as a nested sequence, not sure)... and then from there you would take that result, which is now a fixed 720x486 resolution, bring it into the final sequence, and zoom way into the black space to start, and slowly zoom/move out to get the final result... the text would not longer be vector, but now based on pixels, and thus would be more jagged. again, this is really not the best quality solution, just maybe the fastest.


i shouldn't talk without knowing more about how the nested sequences of final cut work... so maybe someone can chime in about that.

the problem i see with just sorta throwing some text out, arranging clips in their end size/position (within that text), and then scaling the text/clips up to be off screen is that there would most likely be slippage... maybe i'll try this when i get home just to quickly test it and see if i can more accurately recommend the method or advise against it.

 

[gvdv]

jelloshotsrule
I really appreciate the time you're taking to go through this with me.

I've been thinking about this quite hard and actually think that the quality of the text/letters within which the video clips will appear is more crucial than the quality of the clips within the letters (in fact, might even put some kind of opaque mask effect over them so that there is a kind of uniform element uniting them, and so that they are a subordinate element because there are other elements to this, too).

So, given that - that the quality of the text is highly important to maintain credibility as a solid, quality presentation and appearance, what would you suggest?

GVDV