About the editorial office

Purpose

MPEG Standards Lab is an independent scientific and engineering reference on the physics of the digital video signal. Its subject is the body of international standards that describe how a moving image is captured, sampled, compressed and carried, together with the colour science and the mathematics that those standards rest upon. The material is written for students, engineers and researchers who want a clear account of the principles before they turn to the specifications themselves.

The reference is deliberately theoretical. It explains how a coded bitstream is structured and why a given technique works; it does not distribute, host or endorse any particular product. Its diagrams are original vector drawings prepared for this edition, and its tables are compiled from the public specifications and from the published literature. Where a figure is representative rather than exact, the text says so.

Editorial method

Three rules govern the writing. The first is that every claim about a standard should be traceable to the standard or to a peer-reviewed account of it. The second is that a quantity should be stated with its unit and its convention; a bitrate without a unit, or a volume without a declared definition of the gigabyte, is not a useful figure. The third is that uncertainty should be shown rather than hidden. Where the literature reports a range, the reference reports the range and explains what causes it to vary.

Each sheet is prepared in a fixed order. The subject is first outlined in plain language, so that a reader meets the idea before the notation. The notation follows, and then a worked example. Diagrams are drawn at the smallest size that still shows the structure clearly, and each carries a caption that states what it does and does not demonstrate. Finally the text is checked against the source material by a second reader, and every figure is recalculated independently.

Independence

The editorial office is not affiliated with ISO/IEC MPEG, with the ITU-T, with any alliance of technology firms, or with any licensor of the standards it describes. It receives no sponsorship from manufacturers and accepts no payment for coverage. The reference does not host files, does not distribute products and does not operate any financial, brokerage or broadcasting service. Its only function is the written account of communication standards and the science that supports them.

Where a standard is named, the name is used to identify the specification, and the rights in the specification remain with its issuing body. Where a technique is attributed to a published paper, the reference summarises the argument and leaves the reader to consult the original. The goal is understanding, not substitution.

Corrections

Errors are corrected openly. A reader who finds a mistake is invited to write to the editorial office through the contact page, quoting the sheet and the paragraph. Where a correction changes a figure, the revision is recorded in the title block of the affected sheet, so that a reader can tell one edition from another. The revision convention is simple: the year of first publication, followed by a running number that advances with each correction.

Scope

The reference covers the general principles of the digital video signal, the frame architecture of compressed video, the comparison of coding standards, the mathematics of data rate, and the system layers that carry a coded stream. It does not attempt to reproduce the text of any specification, and it does not offer engineering drawings for any particular equipment. A reader who needs the formal syntax should go to the published standard; a reader who needs to understand why the syntax has the shape it does should begin here.