Differences

This shows you the differences between two versions of the page.

--- general:common_myths [2010/04/16 23:31]
j.m
+++ general:common_myths [2012/03/14 23:28]
j.m [Software vs. format, format vs. codec]
@@ Line 22: / Line 22: @@
 ===== Software vs. format, format vs. codec =====
-Contrary to popular myth, codec is not the same thing as format. Audio or video does not have to be decoded with the same codec it was encoded with. For example, when you encode an MP3 file with the LAME codec, it is not "LAME file", "LAME format", "LAME audio" or just "LAME" – you don't encode "to LAME". It is an MP3 file and it can be decoded with other MP3 decoders, too.
+Contrary to popular myths, a codec is not the same thing as a format. A compression format is //a way// audio or video data are encoded, which is described in its specification (if it is standardized, it can be called a standard). An audio or video codec (a codec means en**co**der/**dec**oder) is a software or hardware //tool// that encodes or decodes audio or video data. Tools used for encoding are called encoders, tools used for decoding are called decoders. Software codecs, encoders and decoders are typically implemented as libraries.
-However, the common practice in digital video is using these two terms interchangeably, which is caused by years of industry obfuscations, greedy proprietarism (pretending that their codec is a format is a common business model for codec vendors and a significant source of their income, as it ties the customers to their products), intentional lying in various magazines, websites etc. (in order to make it "easy" for the readers who could be "confused" by correct terms) and nonsensical use of codec [[http://en.wikipedia.org/wiki/FourCC|FourCC]] in AVI (that is, codecs have their own FourCC in AVI files, so if the codec uses a common standard format, users are supposed to install multiple decoders for the same format). Historically, when a company released their own proprietary codec, it was automatically supposed to be the only existing codec for that proprietary format.
+==== Examples of compression formats (standards, specifiations) ====
-But there are standards such as MPEG-1, MPEG-2 and MPEG-4. These are specifications and developers can develop software or hardware implementations that encode or decode audio and video according to the specification. There can be multiple codecs, made by different people, using the same compression format such as MPEG-2 (used e.g. on DVDs) or MPEG-4 ASP. Codec is a software (or hardware) implementation that encodes/decodes (hence the term "codec") data to/from some format which can be compatible with other codecs that follow the same specification. Software is not a format.
+  * MP3
+  * Vorbis
+  * MPEG-4 Part 2, ASP
+  * H.264 (MPEG-4 Part 10, AVC)
+  * VP8
+None of the items listed above is a codec, even though all of them are often erroneously referred to as such. A codec is a tool that actually does something (encodes/decodes, as the term encoder/decoder suggests), while a specification does not do (encode or decode) anything. It only describes how to make tools that do the actual work.
+==== Examples of codecs ====
+  * LAME
+  * libvorbis
+  * libavcodec MPEG-4
+  * Xvid
+  * DivX Codec
+  * libvpx
+None of the items listed above is a format, even though some of them are often erroneously referred to as such (see below). A software tool is not a specification. It implements the format described in the specification.
+==== A single standard, multiple implementations ====
+This all means that audio or video does not have to be decoded with the same codec it was encoded with. For example, when you encode an MP3 file with the LAME codec, it is not a "LAME file", "LAME format", "LAME audio" or just "LAME" – you don't encode "to LAME". It is an MP3 file and it can be decoded with other MP3 decoders, too.
+However, the common practice in digital video is using these two terms interchangeably, which is caused by years of industry obfuscations, greedy proprietarism (pretending that their codec is a format is a common business model for codec vendors and a significant source of their income, as it ties the customers to their products), intentional lying in various magazines, websites etc. (in order to make it "easy" for the readers who could be "confused" by correct terms), and the resulting ubiquitous ignorance and lack of knowledge. The nonsensical use of codec [[http://en.wikipedia.org/wiki/FourCC|FourCC]] in AVI files was a major factor in this massive confusion (video codecs have their own four-character identification in AVI files that is used by the player software for choosing a video decoder, so if the encoder uses a common standard format, users are supposed to install multiple decoders for the same format). Historically, when a company released their own proprietary codec, it was automatically supposed to be the only existing codec for that proprietary format.
+But there are standards such as MPEG-1, MPEG-2 and MPEG-4. These are specifications. Developers can develop software or hardware implementations that encode or decode audio and video according to the specification. There can be multiple codecs, made by different people, using the same compression format such as MPEG-2 (used e.g. on DVDs) or MPEG-4 ASP. Video encoded with one of them can be decoded with another one, if they use the same compression format, which means they are compatible. So the essential thing for describing a video file is its format, not which codec was used for encoding.
 ==== DivX ====

Avidemux

User Tools

Site Tools

Differences

Page Tools