Take another look at Figure 5.2. Remember that I’ve trimmed all the clips of the doctor to remove unwanted frames from the beginning and end, so the clips don’t flow smoothly from one to the other. For this reason, when the video moves from one clip to another, it’s likely the image will jump from the frame shown on the left in Figure 5.4 to the frame shown on the right.

Figure 5.4. A “jump cut” showing the last frame of the fifth video and the first frame of the sixth.

This is called a “jump cut,” because the video looks like it’s jumping around. One moment viewers are watching the doctor with his hands on the chair, and the next moment they’re seeing his arms waving in the air. Jump cuts are very disconcerting to the viewer because they make no visual sense.

And, they create a problem any time you edit video of a relatively static scene and subject shot with a fixed camera. Fortunately, jump cuts are easy to avoid, if you take the time to shoot cutaways or noddies. This is illustrated in Figure 5.5.

Figure 5.5. Use noddies or cutaways to hide jump cuts.

The screenshot in Figure 5.5 shows an editor that offers multiple video tracks, which makes the concept of avoiding jump cuts easier to visualize. Basically, you have three choices:

Inserting cutaways or noddies using editors with multiple video tracks is simple. You drag the video to the target location; sever the link between the audio and the video in the inserted clip (usually a right-click command that “unlinks” the audio and video) and then delete the audio. Or, you can simply mute the audio if that is easier.

With most multitrack editors, after deleting the audio, you can click and drag the newly inserted cutaway down to the main video track to free the space and avoid inadvertently shifting the cutaway clip during subsequent edits. This is shown in Figure 5.6, where the cutaway and noddie are now part of the main track.

Figure 5.6. Drag the inserted clips down into the main track to free the space.

If you’re working with an editor that doesn’t have multiple video tracks, inserting cutaways is a bit more complicated, though most editors are up to the task. Technically, the procedure is called “insert editing,” so check the program’s help files or manual for that term and instructions on how to perform this edit in your tool of choice.

What if you don’t have noddies or cutaways to hide the jump cut? This happens frequently when you’re editing group discussions or executive brief-ings and you’re trying to consolidate the footage. Typically, I’ll insert a short transition, such as a dissolve, between the two clips to let the viewer know that there was a change in time, which is the basic purpose of a transition.

Briefly, transitions are effects that smooth the flow from clip to clip and can range from simple dissolves, where the first clip smoothly blends into the second, to more visual effects such as the ’60s-era flower-power effects used in the Austin Powers movies. Most of the time, I’m pretty conservative with transitions and almost exclusively, stick with dissolves—especially for business videos.

Unfortunately, in jump-cut scenarios, where there’s minimal change between the first and second scene, a dissolve can be practically invisible, as are most static two-dimensional effects such as wipes. In these instances, I’ll either use a push transition (this pushes the first scene off the screen and pulls the second on screen), or an unobtrusive three-dimensional transition such as the cube spin shown in Figure 5.7.

Figure 5.7. If you don’t have cutaways, use a transition that makes it evident that a change in time is occurring, such as this cube spin (don’t worry, the effect is more evident when you watch the video).

A split edit is any edit where the video and audio move from one scene to the next at different times. If you watch television interviews, you’ll notice that the sound and video seldom cut to the next scene simultaneously, as this is less interesting to the viewer. For example, if Larry King was asking O.J. Simpson, “Did you do it?” what you really want to see is O.J.’s face as the question is asked, not Larry King’s.

Of course, Larry has the benefit of using multiple cameras so the producer can cut from camera to camera at any time. We don’t have that luxury. More often than not, usually, you would ask the question in the field, get the response, then come back and record asking the question again while shooting the noddies. Showing O.J.’s face while asking the question is a classic split edit, in this case called an “L-cut” for reasons that will become obvious momentarily.

Figure 5.8 shows two clips on the timeline. O.J. declined to be interviewed for this book, so we’ll have to use our doctor instead. The first clip shows me prompting, “Tell me about the quality of the small town practice.” Then, both audio and video cut simultaneously to the physician as he responds, “You know, that was one of my chief concerns ...”

Figure 5.8. Audio and video cut over from one clip to another simultaneously.

In Figure 5.9, I’ve extended the video from the second clip over the audio from the first, allowing the viewer to watch the doctor as I ask the question. The L-cut gets its name from the shape the audio makes in the timeline when it’s extended beneath the second track. You get this whenever you cut from one scene to another on the video track, but let the audio from the first scene continue.

Figure 5.9. An “L-cut” edit is where the audio from the first clip continues under the second clip, forming an L shape.

Sometimes you want the audio from the second clip to start while the video from the first clip is still showing, generally to presage the content of the second video. In Figure 5.10, for example the first video is Dr. Sumie commenting, “Patients really make the practice.” And the second video clip is an examination where he’s saying, “Breathe deeply, now.”

Figure 5.10. Creating a “J-cut,” where the audio from the second video creeps under the video from the first.

Here, I’ve cut the audio to the second clip before the video, so while the viewer is still watching Dr. Sumie nod with affirmation after commenting, “Patients really make the practice,” they’re listening to the examination under way. The audio presages the video, which is a much more interesting presentation to the viewer. Since the audio from the second clip creeps under the first clip roughly in the shape of a J, this is called a J-cut.

When I’m editing, I use L-cuts and J-cuts as frequently as possible to avoid cutting audio and video simultaneously. Producing these split edits varies from program to program, but, as with insert edits, “unlink” the audio from both video files so that you can move the audio independently from the video.

To produce an L-cut, you adjust only the video track, as shown in Figure 5.11. Regardless of what editing program you use, creating an L-cut always includes these steps:

Figure 5.11. Completing the L-cut.

With a J-cut, you adjust only the audio track, as shown in Figure 5.12, exactly the same way you manipulate the video to create an L-cut. Specifically:

Figure 5.12. Completing the J-cut.

As I mentioned earlier, the purpose of a transition is to clue the viewer of a change in scene or time. Unless your video includes changes in scene or time, you shouldn’t insert transitions between the individual video sequences with one notable exception: to avoid a jump cut. Even then, it’s better to use a cutaway or noddie rather than a transition.

The best way to understand how transitions are used in business type footage is to watch interviews on the evening news, or programs such as 60 Minutes. You’ll notice that transitions are rarely used; when they are used, they’re nearly always-dissolves or fades to black.

Inserting transitions is generally very simple. Typically they sit in their own library or collection bin, and you simply drag them into the timeline or story-board between the two clips.

If you’re producing titles for display on a television set, either via DVD or by writing to VHS or other analog formats, note that television sets don’t display the outer edges of the video—an area called “overscan.” If you place your titles or graphics in the overscan area, they won’t be seen by your viewers.

To help avoid this, most titling utilities can display a “safe zone” comprised of one or two boxes such as those shown in Figure 5.13. The innermost square is called the Title Safe region, which represents the 80 percent of the screen that all televisions should display. Position your title within this title safe area and your television viewers will be able to see it; go beyond the region, and it may be cut off.

Figure 5.13. Title Safe and Action Safe regions in the title.

The outer square represents the Action Safe region. Video within this outer box will most likely display without being cut off by the television set, but action outside of this box will most likely be obscured as overscan. Note also that these boxes do not actually appear onscreen after you apply the title.

Note also that when a computer plays a video file, it displays all pixels in the video file. Accordingly, the concepts of overscan, title safe, and action safe are not applicaple to streaming video or other video that will never appear on a television set. While you shouldn’t position your titles at the very edges of the video, a few pixels in from any side should work well.

Simply stated, computers can display more colors than a television set, particularly the brighter spectrum of colors like red, yellow, and green. For this reason, when producing your titles on a computer, you want the colors to be safe for television, in other words or “broadcast safe.” Otherwise, the color may appear distorted or may pulse visibly on the TV screen.

Many prosumer programs display warnings if you select a color that isn’t broadcast safe. Such a warning is shown on the left in Figure 5.14. If your program doesn’t provide a warning, you can ensure that your colors are broadcast safe by making sure that all three Red, Green, and Blue values have at least 15 bits of color. As shown on the right in Figure 5.14, adding color bits corrected the problem with the overly red title color in the example.

Figure 5.14. To make your colors broadcast safe, make sure all RGB values have at least five bits of color.

As with the title safe and action safe zones, title color isn’t an issue for videos displayed solely on a computer; in this case, achieving the right aesthetics and contrast are the primary factors in choosing a color for your titles.

Your first consideration when choosing a font should be legibility, or the viewer’s ability to read onscreen text. Note that this metric varies according to screen size and whether the video is compressed and to what extent. This is shown in Figure 5.15.

Figure 5.15. Fonts compressed to 128Kbps, on the left. and 32Kbps on the right.

To produce Figure 5.15, I created a title using the named fonts at 36 points for the larger text, and 18 points for the smaller text. After creating the title, I rendered twice—once at 128Kbps (shown on the left), and then at 32Kbps (shown on the right), both times using Microsoft’s Windows Media Format with output at 320×240 resolution.

On top is Times Roman, an “Old Style” font with serifs, the little pointy edges you can see at the ends of many of the letters. If you scanned a book or magazine, you would notice that most text is printed in Old Style fonts such as Times Roman because it is highly readable.

However, Old Style fonts have two characteristics that don’t work well at high compression. First, as you can see on the right, the serifs tend to disappear at high compression, degrading the appearance of the font. In addition, if you closely analyze the font, you’ll notice a variance in the thickness of some of the letters. Compare, for example, the upper left regions of lower case o’s and e’s with the right side of letters like lower case m’s and d’s. This is important, because at high compression, these thinner lines tend to disappear, degrading the appearance even further.

Beneath Times Roman is Arial, a sans serif font. If you study books and magazines, you’ll notice that most chapter and heading titles are in sans serif fonts such as Arial. That’s because, as you can see in Figure 5.15, sans serif fonts stand out. In this regard, they’re easier to browse than serif fonts, and better for quick hits of information, but not as readable over the long term.

For our purposes, note also that Arial, like most (but not all) sans serif fonts, is “monoweight,” which means all lines are the same thickness. This makes Arial the most easily compressed font of the group, with great readability even at 32Kbps.

Next time you’re watching a television program, make a mental note to determine whether they’re using serif or sans serif fonts. In my experience, when the issue is readability (news, advertisements, public service announcements), sans serif fonts are used almost exclusively, as they are in most high-volume streaming media.

Back to our font comparison. I included Palatino, an Old Style font, because I like its appearance and use it frequently. It’s slightly thinner than Times Roman, but otherwise shares the same characteristics regarding readability and lack of compressibility.

The final is Park Avenue, a “script” font, which means it’s designed to look as if handwritten. It shares the serifs and nonuniform line size of Times Roman, but carries much more detail. Predictably, it quickly degrades at high compression ratio.

Interestingly, at 128Kbps, all fonts retain close to original quality, so with most video produced for DVD, television, or desktop viewing, post-compression readability shouldn’t dictate the font. It’s worth noting here that EIA-708, the FCC regulation that delineates the requirements for closed captioning on U.S. television, prescribes the use of eight font types, including serif, sans serif, and script fonts, meaning that all the fonts shown in Figure 5.15 would be “legal” for a closed-captioned feed. But at high compression rates, Arial or other sans serif fonts are the obvious choice. (See Chapter 11 for a full explanation of how to produce closed captions.)

Clearly there’s is no absolute right or wrong way to go in font selection, but over the years I’ve adopted the following conventions for producing my titles. (Assume that I’m not producing for high compression (in which case I’d simply use Arial.)

Obviously, being consistent with font usage—within a single project, and among similar or related projects—is as important as your choice of fonts. For this reason, most organizations create standards regarding fonts, font sizes, colors, duration, and other title attributes.

In the next few sections, I’ll show you the standards I use for most business-oriented productions. Note that many of these are my own personal preferences; as I said, there is no true right or wrong. What is important for you is to recognize the need to create and apply similar standards and consistencies in your own videos.

Many parameters like font, font size, color, and positioning need no further explanation. Before introducing my standards, let me briefly describe some parameters that may not be so obvious.

With this as prologue, let’s move to the opening title sequence.

The opening title is used at the very start of the video (shown on the left in Figure 5.16). With educational or training usage, be sure to identify both the speaker and the topic, and any other relevant information to let the viewer know the subject of the video (they’ll want to know what they’re watching the correct video).

Figure 5.16. A video title screen on the left; section headings on the right.

If your video has discrete sections you’d like separate by titles, use similar parameters, but a smaller font size, around 35 to 50 points. These are the titles shown on the right in Figure 5.16. I also use the same approach and font for the “The End” screen that I include in many productions, generally using a point size of around 50.

These titles identify speakers when they appear onscreen. Use a consistent background on all name titles; I prefer to make the title background semi-transparent (see the title shown on the left in Figure 5.17); this creates sufficient contrast to make the text readable without appearing obtrusive, though not all programs support this.

Figure 5.17. This title identifies the speaker when he or she appears on screen.

Borders around titles (such as those shown on the upper right of Figure 5.17) look good in video produced for television, but the lines degrade severely at high compression. For this reason, I also avoid drop shadows and inner and outer strokes on titles used in videos destined for high compression. Keep it simple is the best advice at high compression ratios.

After the opening titles comes an information screen; this is text intended for the viewers to read (Figure 5.18). The small caps font shown on the right is a popular technique for many information screens.

Figure 5.18. Information screen, shown in a normal font and in small caps.

Using upper and lower case (shown on left), is considered easier to read, because readers recognize the words as words, rather than having to read the individual letters and decipher the word. Experiment with both to see which you feel is more readable.

Bullet points are titles introduced to highlight certain audio portions of the video. For example, while the physician in our video was outlining the benefits of the small town practice, you might produce a title with the bullet points shown in Figure 5.19.

Figure 5.19. Bullet-point titles shown against two backgrounds. The image on the right with the background darkened, is much more legible than the one on the left.

While you can certainly build these titles against a flat black or blue screen, I like to use a thematic background, when available, as shown in Figure 5.19. When the image is too bright to provide the necessary contrast (as shown on the left), darken the background image, either within the video editing application or using an image editor, to improve readability (on the right).

Closing credits generally don’t contain mission-critical information, so you’re free to get a bit more creative. A range of font and background options are shown in Figure 5.20. The options in the style guide are those used in the larger title on the left.

Figure 5.20. Closing credits are less formal than other titles, allowing greater creative leeway.

Technically, all video is copyrighted whether or not it contains a notice like that shown in Figure 5.21. Nonetheless adding copyright notices “discourage” copyright infringement, or the use of your video without your written permission. It also keeps legal types off your back, and insulates you from barbs from bosses or co-workers asking, “Where’s the copyright notice?”

Figure 5.21. Various versions of the copyright notice; the notice on the upper right contains the minimum required information.

To be complete, the copyright notice must contain the copyright “bug” © or the word copyright, plus the year of creation and the name of the copyright holder. I go beyond that with the large copyright notice shown on the left in Figure 5.21, moving towards a more minimalist approach on the upper right.

When producing video and titles for streaming video, keep these points in mind.

The good news is that streaming video is almost always watched on a computer screen, which displays much greater detail than a television set. However, when producing for low-bitrate streaming at reduced resolutions, such as 320×240, you’re essentially cutting the point size of your fonts in half. That’s because when you shrink the video from 640×480 to 320×240, you shrink everything in the video by 50 percent. This means your highly readable 20-point font becomes a tiny, smudged-looking 10-point font.

This won’t matter for larger titles like the opening title and bullet points, but will affect many smaller titles, particularly name/affiliation tags. When producing for a 320×240 window, boost the point sizes for these titles by about five points for improved legibility. In addition, consider bolding the text, because bold text survives compression with much greater legibility. This is shown in Figure 5.22, where both video files were compressed to 32Kbps.

Figure 5.22. Bold fonts compress with much greater legibility than normal fonts.

Though it will seem like the larger font size is consuming gobs of space, remember that you can ignore the safe zone when you’re working with streaming video (since it isn’t displayed on a television set). For this reason, you can reclaim the space by moving the title much closer to the bottom or left edge of the video.

Virtually all video editors can import digital images and convert them to video. This includes not only digital pictures and logos, but also the output of programs such as PowerPoint, Visio, AutoCAD, and other electronic design programs. Adding images to video is a great way to explain your work or product and can help break up otherwise monotonous talking-head footage.

Adding still images to your editor is also fairly simple. Generally, you drag them down to a video track and grab the right edge to set duration. Most programs let you add transitions between still images and adjust color, brightness, and similar values with the same filters available for video.

However, when adding line art and other finely detailed images, you must keep several critical points in mind. First, the resolution of video is limited. Even a high-quality format like MPEG-2 (used in DVD) has a resolution of 720×480 pixels, while streaming files are often produced at 320×240 pixels. Suppose you included a high-resolution image such as an architectural drawing produced in AutoCAD 2004 shown in Figure 5.26.

Figure 5.26. Output from AutoCAD 2004.

The image has a resolution of 1,340×960. If you’re encoding for DVD at 720×480 resolution, there aren’t enough lines of pixels in the video file to display all the pixels in the image. To convert from the 1,276×820 source to the 720×480 output during encoding, the video editor will discard image detail, degrading image quality.

Plus a television set’s inability to display fine details with accuracy only compounds the problem. Basically, it’s almost impossible to display high-resolution images in a video file without losing detail.

To workaround, cut your image into smaller chunks as shown in Figure 5.27. Then, import the files into your video editor and display them in the desired order.

Figure 5.27. To show a high-resolution image, display it in chunks.

You can do this in two ways. First, divide the image up into separate images—each containing the desired views—using an image editor like Ulead’s PhotoImpact or Adobe Photoshop. When using this approach, keep the following rules in mind:

As you become more familiar with your video editor, you’ll find the easiest way to show portions of an image is to insert a high-resolution version into the video editor and use the editor’s zoom and positioning controls to show only the desired region of the image. Take a mental break for a second, and I’ll explain the procedure.

Imagine you were at Mount Rushmore with your camcorder. You stand in one spot and zoom in to shoot George Washington, where it all began. Then you zoom out to show President Washington with President Jefferson. Then you zoom back in and pan across to President Lincoln, then President Roosevelt.

Obviously, the magnificent carved mountain is there the entire time; you’re just using the camera’s zoom controls to zoom in and out and your own two-dimensional positioning controls (hands and arms) to move the camera around the mountain. Most video editors have controls that work the same way, allowing you to pan and zoom around high-resolution images to show only low-resolution sections.

Once you’ve imported the image into the editor, you can zoom into and out of the image with magnification controls, and pan around the image using positioning controls. The only segment of the image the editor will include in the rendered video appears in the viewport; the other regions of the image are still there, but out of camera.

This is illustrated in Figure 5.28; the black area is the viewport, and the gray regions are out of the camera’s current scope. Basically, you drag that portion of the image to display into the viewport, and use zoom controls to zoom into or away from the image.

Figure 5.28. Two-dimensional (2D) editing controls allow you to show different regions of an image at different magnification levels.

The only caveat is to keep the viewport smaller than the output resolution of the video, just as you did for each subset of the image you isolated in your image editor. The image editor will automatically set the viewport to the correct output resolution, so you don’t even have to worry about choosing a 4:3 ratio. This is faster and simpler, and while the interface may vary from what’s shown in Figure 5.28, virtually all editors offer these controls.