Image Effects

Once upon a time, not too long ago, there was only film. When you shoot a scene with a film camera, you look through the lens and have to be concerned with what part of the frame would eventually be projected. Viewfinders show the Camera Aperture, which is everything the camera sees, and just inside that area is the Projector Aperture, which is what gets shown on-screen when the film is projected in a theatre.

Then came television. Now, many camera viewfinders have guide lines for the older and the new formats. In the Safe Viewing Area, you can see if the boom guy was letting the boom slip into frame ("CUT!"). If you are on a set, you might hear people talking about a prop or piece of a set that is, or isn't, "Out of Academy". That expression comes from 35mm film with an aspect ratio of 1.33:1 which is called Academy Aperture. You might also hear about an aspect ratio of 1.85:1, commonly refered to as Wide Screen.

Within those formats, you have TV Action Safe Area. What you don't see in the viewfinder is Title Safety. That's because camera operators, shooting live action, generally didn't create titles. In the "old days", this was done by specialists - like the Title department, because they were set up for it. People who did the opticals might also do the titles. Saul Bass was famous for doing titling for many of Hitchcocks' films.

Because any footage you shoot could end up anywhere: theatrical release, video stores, television/cable, or internet - it's necessary to have some sort of standard to consider when adding titles to a filmed product. First, you determine what the Full Screen is going to be. For television, that's the TV Transmitted Area. With film, that means any of the above mentioned (and some not mentioned) aspect ratios. You then need to calculate 90% of those heights and widths to establish the Action Safety Area. That's 5% on the left and right sides, and the top and bottom. This 90% of the inner frame is what mom and dad will see sitting in front of the tube. Title Safety is said, by the broadcast stations, to be 80% - 85% of the Full Screen. That's so titles don't end up hanging out along the edge of the family television set.

Most of the time, we only have to be concerned with 4x3 television aspect or the equivalant film ratio. However, we are now embarking on the next big thing: HDTV. HDTV has a television ratio of 16x9, with at least two resolution sizes that I know of - 1920x1080 and 1280x720. Translated to film-speak, that's an aspect of 1.78:1 which, if you were paying attention to what I said earlier, is close to Wide Screen. They are shooting sports events in HDTV right now - CBS broadcast the Buffalo Bills vs. New York game in November 1998 in Hi-Definition. Sony HDTV cameras will show the 16x9 safe area, as well as the 4x3 safe area. That's so you can frame the action so it can be useful and viewable in both aspects, obviously.

Those of you with LightWave, are probably already familiar with "Show Safe Areas" in the LightWave options. I've used it often myself. However, it's only accuate for 4x3 video output. Other programs, such as AfterEffects, have a built-in overlay, through which you can examine your frames and see what's in or out of Safety. Although the percentages are adjustable, most people keep them at the default of 90% and 80%. Well, now everyone's getting on the bandwagon. A few months ago, I saw an article explaining to Photoshop users how to make an image with Action and Title Safe Area markers. Recently, ImageFX users started shouting, "I want my "SafeTV"!!". (At least, that's what I thought that noise was!)

So, I set out to make scripts that would automatically create a buffer with the 80% and 90% guide lines, regardless of what format was to be used. Sure, most of you reading this are using Toasters so your resolution must be at 752x480 no matter what resolution your camera shoots - you probably don't need that flexibility. But some ImageFX users have DraCos or other motion video cards, which have different resolutions. Many of you may have become familiar with your cameras enough to not need the guidelines, but sometimes you get footage from an outside source and you might have to examine it's Safety, before converting it to video resolution.

Well, I did manage to create several crude scripts to do this, but what I really wanted was to have one all-encompassing elegant script that could be used by everyone. So dhomas came to the rescue - that much of an ARexx maven I ain't - and made exactly what we needed. It was as simple as cycling through some presets, or typing in the resolution you needed in the requestor and letting it generate a new buffer that you could overlay onto your footage. It was so useful, that I decided to write this article about it - which I did. What I wasn't expecting though, was that dhomas would then turn around and further enhance the script to make our lives even easier. He only does what I tell him (smile), at least until he sees where I really want to be going. When he started in on this article, he inferred what I really wanted and made all the necessary changes. So, the steps you see below are not what I started with - there's a lot less work now, thanks to a better script.

So now, when I get a digital camera - and some of these new consumer and prosumer models come with HDTV aspect just like the professional cameras - now I can film my epic with that aspect in mind. If I need to reformat to television aspect, I've got "Safety.ifx" to help me alter my footage as needed.

It's easy enough to do this. First, download this tutorial file archive here: safety.zip, unarchive it, and put the provided script into ImageFX's rexx/ directory. Then follow along with this short tutorial to find out how to use it.

1. Open "Sheep.iff" into a new buffer.
2. Click [ARexx] and select the "Safety.ifx" script.
3. If your image is already in the resolution that you want for final output; but you want to add some titling, you can just click [Okay]. "Safety.ifx" defaults to using the size of your Main buffer. Try it now.
You'll see three lines drawn into a layer called "* Safety *". The Red box represents the Action Safety area; the Green is the Title Safety Area. So, as long as your titles stay inside the Green box, you should be safe. If you can, adding titles in another layer makes it really easy to make changes if you need to.
If that's all you need or want to do, you can stop here - you're ready to go. But, you might want to read on to see what other perhaps useful things you can do. You might not ever think to do something like this, if you don't know you can.
4. Let's say that you shot your footage at the silly resolution of 640x360; but you want to reduce it to fit the display area of something strange like 500 x 300. Let's try it. [Delete Layer] the "* Safety *" layer we just made, and start the "Safety.ifx" script again.
Leave Size set to Custom; but change W to 500 and H to 300. Click [Okay]. This time, the White outermost box shows you the Full Screen size that you set or selected in the Safety options requester.
If you wanted to use one of the presets, you could instead select that, rather than Custom. For this tutorial, though, we have to use Custom, since 500x300 isn't a standard resolution.
5. Now you have guide lines to assist in the format changing process. At this point, you need to figure out how each scene needs to be adjusted to fit the smaller display. This usually means that you will have to Scale, Crop or maybe even use something like PanCanvas, to create a moving view across the larger source footage. You have a lot of options, and "Safety.ifx" can be a real help to examine footage and make the correct creative decision.
If Cropping is the way to go, use Move Layer in the Layers menu to reposition the "* Safety *" layer to the best section of the footage layer. Just select "Move Layer" and click in the buffer window and drag the mouse. When you let go you'll see the guide lines redraw. Keep adjusting until you get it where you want it. Then, make a note of the Offset X and Y values in Layer Settings.
Those values can then be used in the Crop module. When it comes time to Crop, set the resolution to the size you want your output video to be, that which you chose in the "Safety.ifx" requester, and use the Offset numbers you got previously to position the crop box to match. If you record the Crop settings to a macro, you can then resize all your footage in a batch process.
Notice in Figure 1, that looking through the television aspect layer doesn't even show the sheep it's so much smaller than HDTV. If you want to keep the sheep in view, you probably would want to scale your footage, rather than crop it. When doing that, you have to remember to first enlarge (scale up) the "* Safety *" layer first (be sure "Lock to Image's Aspect" is toggled on). Make it big enough to include whatever you'd like to keep in frame from your footage. I usually scale up in small steps to keep in control of the process.

There's a lot more that you can do, but I think you get the idea. In fact if you can come up with an interesting way to use this script, drop me a note, I'll write about it in a future article and we'll send you a (virtual) cookie. I have added to the script the HDTV sizes 1920x1080 and 1280x720, just because it's becoming alot more common. Children of DUNE was shot in HDTV.