Understanding different types of bit-rate encoding

In Microsoft Expression Encoder, you have a choice of three methods of encoding: Constant Bit Rate (CBR), Variable Bit Rate (VBR), and Multiple Bit Rate (MBR) encoding modes, each of which is optimal for particular playback or streaming scenarios.

CBR encoding is designed to work optimally in a variety of streaming scenarios. You can constrict the bit rate to guarantee consistent playback across a wide range of systems. The bit rate remains fairly constant and close to the target bit rate over the course of the stream. The disadvantage of CBR encoding is that the quality of the encoded content is not constant. Because some pieces of content are more difficult to compress than others, some parts of a CBR stream are of lower quality than others. In addition, CBR encoding gives you inconsistent quality from one stream to the next. In general, quality variations are more pronounced at lower bit rates.

VBR encoding is designed to work optimally in high-bandwidth scenarios and is especially suited for encoding content that is a mixture of simple and complex data. The encoder allocates fewer bits to the simple parts of the content, leaving enough bits available to produce good quality for the more complex portions. Content with consistent data (for example, a "talking head" news story) would not benefit from VBR encoding. However, when it is used on content with varying complexity, VBR encoding produces a much better output than CBR encoding, even if both methods produce files of identical size. In some cases, a VBR-encoded file might have the same quality as a CBR-encoded file that is twice as large because the VBR encoding compressed the less complex portions much more than the CBR method did. However, VBR encoding would give more bandwidth to complex portions, unlike CBR encoding.

MBR encoding is designed to adjust to fluctuations in bandwidth. MBR encoding combines several bit rates into a single encoded file. Alternatively, you can choose to create a separate file for each bit rate. When the file is accessed, the server determines the appropriate bit rate based on the available bandwidth, and then serves the encoded file at the optimal bit rate. If the available bandwidth decreases for any reason, a stream at a lower bit rate can be served. With a lower bit rate, there is also a decrease in quality.

For all encoding, whether or not you use a template, Expression Encoder will still create a file that is optimized for playback in the scenario that you choose. For example, you can create a video that is highly compressed so that it has a low bit rate, thereby enabling high-quality playback for users ranging from those who have dial-up connections to those who are using high-speed LANs. Conversely, you can choose to encode a video with a high bit rate, thereby enabling only users who have high-speed connections to successfully play back the file.

When compressing the file, the encoder can make either one or two encoding passes through the content, depending on the encoding method that you choose. VBR encoding offers only two-pass encoding. In two-pass encoding, the encoder analyzes the content during the first pass, and then encodes it in the second pass, based on the data gathered in the first pass. This method produces the highest-quality result, but requires more encoding time. The CBR encoding method offers both one-pass and two-pass encoding. In one-pass encoding, Expression Encoder uses only one pass through the content to analyze and compress it. Note that two-pass encoding does not take twice as long to encode as one-pass encoding. The second pass is faster than the first, and so total encoding time increases only by roughly 20 percent.