• Create BookmarkCreate Bookmark
  • Create Note or TagCreate Note or Tag
  • PrintPrint
Share this Page URL

Lesson 2. About Digital Video Editing > Understanding video data compression

Understanding video data compression

Editing digital video involves storing, moving, and calculating extremely large volumes of data compared to other kinds of computer files. Many personal computers, particularly older models, are not equipped to handle the high data rates (amount of video information processed each second) and large file sizes of uncompressed digital video. Use compression to lower the data rate of digital video to a range that your computer system can handle. Compression settings are most relevant when capturing source video, previewing edits, playing back the Timeline, and exporting the Timeline. In many cases, the settings you specify won't be the same for all situations:

  • It's a good idea to compress video coming into your computer. Your goal is to retain as much picture quality as you can for editing, while keeping the data rate well within your computer's limits.

  • You should also compress video going out of your computer. Try to achieve the best picture quality for playback. If you're creating a videotape, keep the data rate within the limits of the computer that will play back the video to videotape. If you're creating video to be played back on another computer, keep the data rate within the limits of the computer models you plan to support. It you're creating a video clip to be streamed from a Web server, keep an appropriate data rate for Internet distribution.

Applying the best compression settings can be tricky, and the best settings can vary with each project. If you apply too little compression, the data rate may be too high for the system, causing errors such as dropped frames. If you apply too much compression, lowering the data rate too far, you won't be taking advantage of the full capacity of the system and the picture quality may suffer unnecessarily.


DV has a fixed data rate of 3.5 megabytes per second, nominally 25 megabits per second; the DV standard compression ratio is 5:1.

Analyzing clip properties and data rate

Premiere includes clip analysis tools you can use to evaluate a file, in any supported format, stored inside or outside a project.

Locate and select the Sailby.mov clip from Lesson 1 and click Open.

From Premiere, choose File > Get Properties For > File.

The Properties window provides detailed information about any clip. For video files, the analyzed properties can include file size, number of video and audio tracks, duration, average frame, audio and data rates, and compression settings. You can also use the Properties window to alert you to the presence of any dropped frames in a clip you just captured.

Click Data Rate to view the data rate graph for the clip.

You can use the data rate graph to evaluate how well the output data rate matches the requirements of your delivery medium. It charts each frame of a video file to show you the render keyframe rate, the difference between compression keyframes and differenced frames (frames that exist between keyframes), and data rate levels at each frame.

The data rate graph includes:

  • Data rate: the white line represents the average data rate.

  • Sample size: the red bars represent the sample size of each keyframed frame.

If there are differenced frames, they appear as blue bars, representing the sample size of the differenced frames between compression keyframes. In this case, there are not any.

When you are finished, close the Data Rate Graph window and the Properties window.

For more information, see “Factors that affect video compression” in the Adobe Premiere Technical Guides found in the Support area on the Adobe Web site (www.adobe.com/products/premiere/community.html).

Choosing a video compression method

The goal of data compression is to represent the same content using less data. You can specify a compressor/decompressor, or codec, that manages compression. A codec may use one or more strategies for compression because no single method is best for all situations. The most common compression strategies used by codecs and the kinds of video they are intended to compress are described in this section.

Spatial compression

Spatial (space) compression looks for ways to compact a single frame by looking for pattern and repetition among pixels. For example, instead of describing each of several thousand pixels in a picture of a blue sky, spatial compression can record a much shorter description, such as “All the pixels in this area are light blue.” Run-length encoding is a version of this technique that is used by many codecs. Codecs that use spatial compression, such as QuickTime Animation or Microsoft RLE, work well with video containing large solid areas of color, such as cartoon animation.

Digital images are composed of pixels (A), which consume a lot of disk space when stored without compression (B). Applying run-length encoding stores the same frame data in much less space (C).

In general, as you increase spatial compression, the data rate and file size decrease, and the picture loses sharpness and definition. However, some forms of run-length encoding preserve picture quality completely, but require more processing power.

Temporal compression

Temporal (time) compression compacts the changes during a sequence of frames by looking for patterns and repetition over time. In some video clips, such as a clip of a television announcer, temporal compression will notice that the only pixels that change from frame to frame are those forming the face of the speaker. All the other pixels don't change (when the camera is motionless). Instead of describing every pixel in every frame, temporal compression describes all the pixels in the first frame, and then for each frame that follows, describes only the pixels that are different from the previous frame. This technique is called frame differencing. When most of the pixels in a frame are different from the previous frame, it's preferable to describe the entire frame again. Each whole frame is called a keyframe, which sets a new starting point for frame differencing. You can use Premiere to control how keyframes are created (see the Adobe Premiere 6.0 User Guide and Premiere 6.5 User Guide Supplement). Many codecs use temporal compression, including Cinepak.

If you can't set keyframes for a codec, chances are it doesn't use temporal compression. Temporal compression works best when large areas in the video don't change, and is less effective when the image constantly changes, such as in a music video.

In this animation clip, the only change is the circle moving around the ship.

A. Storing the clip without compression records all pixels in all frames. B. Applying temporal compression creates a keyframe from the first frame, and subsequent frames record only the changes.

Lossless compression

Some codecs use lossless compression, which ensures that all of the information—and thus all of the quality—in the original clip is preserved after compression. However, preserving the original level of quality limits the degree to which you can lower the data rate and file size, and the resulting data rate may be too high for smooth playback. Lossless codecs, such as Animation (at the Best quality setting), are used to preserve maximum quality during editing or for still images where data rate is not an issue.


To ensure smooth playback, full-frame, full-size video using lossless compression requires a very large defragmented hard disk and very fast computer system built for high data rate throughput.

Lossy compression

Most codecs use lossy compression, which discards some of the original data during compression. For example, if the pixels making up a sky actually contain 78 shades of blue, a lossy codec set for less-than-best quality may record 60 shades of blue. While lossy compression means some quality compromises, it results in much lower data rates and file sizes than lossless compression, so lossy codecs such as Cinepak or Sorenson Video are commonly used for final production of video delivered using CD-ROM or the Internet.

Asymmetrical and symmetrical compression

The codec you choose affects your production workflow, not just in file size or playback speed, but in the time required for a codec to compress a given number of frames. Fast compression speeds up video production, and fast decompression makes viewing easier; but many codecs take far more time to compress frames than to decompress them during playback. This is why a 30-second clip may take a few minutes to process before playback. A codec is considered symmetrical when it requires the same amount of time to compress as to decompress a clip. A codec is asymmetrical when the times required to compress and decompress a clip are significantly different.

Compressing video is like packing a suitcase—you can pack as fast as you unpack by simply throwing clothes into the suitcase, but if you spend more time to fold and organize the clothes in the suitcase, you can fit more clothes in the same space.

DV compression

DV is the format used by many digital video camcorders. DV also connotes the type of compression used by these camcorders, which compress video right inside the camera. The most common form of DV compression uses a fixed data rate of 25 megabits per second (3.5 megabytes per second) and a compression ratio of 5:1. This compression is called “DV25.” Adobe Premiere 6.5 includes native support for DV25 and other DV codecs, and can read digital source video without further conversion.

No single codec is the best for all situations. The codec you use for exporting your final program must be available to your entire audience. So, while a specialized codec that comes with a specific capture card might be the best choice for capturing source clips, it would not be a good choice for exporting clips, since it is unlikely that everyone in your audience would have that specific capture card and its specialized codec. This is a significant concern when exporting streaming media, since the three most popular streaming architectures (RealMedia, Windows Media, and QuickTime) use proprietary codecs in their players; a RealMedia stream, for example, cannot be played back through a Windows Media player, and vice versa. So, for the convenience of audiences with diverse players set as the default in their browsers, streaming media is usually encoded in multiple formats.

For more information, see “Video codec compression methods” in the Adobe Premiere 6.0 Technical Guides found in the Support area on the Adobe Web site (www.adobe.com/products/premiere/community.html).

  • Creative Edge
  • Create BookmarkCreate Bookmark
  • Create Note or TagCreate Note or Tag
  • PrintPrint