A Quick Video Primer Flash

If you’re a neophyte to digital video, then you need to know some basic terms and procedures involved with digital video. The following section will be useful if you’ve never used digital video or used it without really knowing what you were doing.

A brief history of digital video
In the past, digital video on the desktop computer was almost impossible. It required expensive hardware such as superfast processors, huge hard drives, video-capture boards, and professional-quality video decks and cameras. Beginning at $15,000, such systems were out of reach for most users. But like most technology after it has been around for a while, digital video equipment has become much more affordable for the average user. Although digital video still requires fast and efficient computers to work well, it isn’t nearly as expensive anymore. You can get 30GB hard drives for under$500! Since the advent of the DV (Digital Video) format (a.k.a. DVCAM or miniDV), consumer-level video cameras and decks are almost as good as their professionallevel counterparts.

The need for space
Why does digital video require so many resources? To begin with, digital video is entirely raster-based. This means that, unlike Flash and other vector file formats, each frame of digital video requires that almost every pixel on the screen is remembered and stored. Vector formats, on the other hand, use mathematical descriptions of objects on the screen and compute their movement very efficiently. The resolution of an average television set is roughly equivalent to a 640 × 480 resolution at 24-bit color depth on your computer monitor. Mathematically speaking, one frame of digital video at this resolution is nearly 1MB!640 × 480 × 3* = 921,600 bytes = 900KB = 0.88MB

If that isn’t bad enough, consider that 1 second of video contains 30 frames. That’s 26MB for just 1 second of video! Only the fastest systems and hard drives on the market could deliver such performance. One solution to this performance bottleneck was to compress the data. Thus, most digital video now employs some form of compression (for storage) and decompression (for playback). The short form of this expression is codec (compression and decompression). You may have already heard of many codecs in use today, but what you probably don’t know is that there are three kinds of codecs: software, hardware, and hybrid.

DVD, or digital versatile/video disc, is a new storage medium that can handle feature- length movies in a snap. DVD should not be confused with DV. DV refers to true Digital Video, in which the source video originates as binary (zeros and ones) data. Furthermore, the general term digital video should not be confused with DV. Digital video usually refers to the any video that has been stored as binary data, although it most likely originated from an analog source such as a regular VHS or BetaCam video camera. DV refers to video that originated from a digital (a.k.a. binary) source and that remains digital through any number of edits on a digital system.

With the current implementation of DV, using IEEE-1394 (a.k.a. FireWire or iLink) technology, video is transferred from digital tape to your computer hard drive with virtually no loss of quality. The DV footage is not recompressed unless the image in the footage is changed during editing by adding effects or transitions. But like any digital video, DV still requires a lot of hard drive space about 2GB for every 9 minutes.

Codec, frame size, and frame rate: The keys to manageable video
Before you begin any digital video project you should have a clear understanding of codecs. Most software-based codecs are intended for computer playback and distribution, while hardware-based codecs are intended for capturing and editing original footage to be used for television broadcast or feature films. You can repurpose hardware-based codec video by compressing it with a software-based codec. Most video developers take high-quality video and shrink it, in both frame and file size, to fit onto multimedia CD-ROMs or the Web.

Three variables can be applied to digital video to make it more manageable for most consumer computer systems: frame size, frame rate, and compression. Developers often use all three variables to shrink huge 9GB video projects down to 3 to 5MB, which may lead to undesirable results.

First, let’s talk about frame size. Although most professional video uses a 640 × 480 or greater frame size, you may have noticed that most video on multimedia CD-ROMs only takes up a quarter or less of your entire computer monitor. Most video on the Web or CD-ROMs is rendered at 320 × 240 resolution, half the resolution of broadcast video. Actually, this is only slightly less than the horizontal-line resolution of your VHS recordings.

What about frame rate? You may have also noticed that video on multimedia CD-ROMs often looks a little jerky or choppy. Although this may be due to a slow processor, it’s more likely that in order to cut the file size the frame rate of the video was reduced. It’s not uncommon to find CD-ROM frame rates as low as 12 or 15 fps (frames per second) about half of the original frame rate of broadcast video. This slower frame rate is also the default frame rate of a new Flash movie, to ensure consistent playback on slower machines. Despite the drop in video quality, the lower frame rates result in much smaller file sizes with fewer frames for the processor to play within each second, which delivers better CD-ROM performance.

Finally, how does compression affect video? You’ve probably noticed that Web and multimedia CD-ROM video is often blocky looking. This is due to the software-based compression that has been used on the video. Codecs look for areas of the frame that stay consistent over many frames, and then log those areas and drop them from subsequent frames. The result is that no unnecessary repetition of data exists that needs to be continually decompressed. But, depending on the level of compression used, the properties of the codec itself, and the settings used in running that codec, the video varies in quality.

Keeping with the trend of better and faster, digital video continues to improve dramatically. This is well illustrated by the fact that many popular Web sites, such as Apple’s QuickTime Web site , now enables visitors to download larger, higher-quality videos (upwards of 15MB) for playback on newer, faster systems.

Playback bottlenecks
Digital video needs to be kept small for two reasons: storage and playback. So far, we have largely discussed storage issues. But playback (or transfer rate) further complicates the creation of digital video. Despite the relatively large capacity of CD-ROMs (650MB), most CD-ROM readers have limited transfer rates of about 600KB/second. It’s important to note that each second of video cannot exceed the transfer rate, otherwise the video will drop frames to keep up with the audio. So if the video is distributed via CD-ROM, this factor results in serious limitations.

Let’s look at some of the math involved under ideal (choppy) playback conditions: If you use 15 fps for compressed video, you are limited to a maximum of roughly 40KB per frame. (Remember, though, that the playback stream usually includes an audio track as well, which means that less than 40KB is available for the video component of each individual frame.)

Unfortunately, the Web still affords less than ideal playback conditions for video. On the Web, transfer rates can be as slow as 500 bytes/second. On average, a 56KB modem downloads around 4KB/second. The ideal Web video streams to the user while loading the page. If you intend to stream video quickly, you have to keep this very small transfer rate in mind. Large videos simply will not stream! This is why most Web sites offer larger videos as a download file. But you do have an alternative.

Later in this section, you can learn how to extract a minimal number of frames in order to simulate digital video motion with Flash, yet keep your Flash files streaming quickly. As modem technologies get faster, though, we’ll most likely see bigger and better video delivered across the Web. The ADSL (Asymmetrical Digital Subscriber Line) modem was developed with the MPEG-1 and -2 standards in mind.