Digital Video Formats

Serial Digital Interfaces - SDI         Digital Component Video         Windows Video File Formats



DV

Developed in 1994, tape based format, consumer and semipro level, recording on the DV cassette, entry level, low cost operation, JPEG-similar compression image by image, data rate 25 Mbps
NTSC: chroma subsampling 4:1:1, 720 x 480 pixels
PAL: chroma subsampling 4:2:0, 720 x 576 pixels
DV is always an interlaced video signal, starting with the second (!) line ("second" field) first.
For the 16:9 format the image is anamorphically stretched.
See: Anamorphic.

Audio: DV1 with 32kHz sampling and 12bit quantization, DV2 with 48kHz sampling and 16bit quantization.



HDV

Developed from Sony and JVC, tape based format on standard DV cassette, entry level, low cost operation, MPEG-2 compression, data rates 19Mbps (720p), 25Mbps (1080i), 4:2:0
HDV1: 1280 x 720p, frame rates 25p, 50p, 30p, 60p, 24p, MPEG-2, long GOP (six frames), the data rate is 19Mbps
HDV2: 1440 x 1080i, frame rates 25i, 50i, 30i, 60i, MPEG-2, long GOP (12 or 15 frames), interlaced format with the first, uneven, line first (first field first). The 1440 horizontal pixels are stretched in the ratio 1:1,33 (non-quadratic pixels), to generate a 16:9 image format.



XDCAM EX

Developed by Sony, tapeless format, MPEG-2 data compression, entry level for tapeless recording, media: SxS PRO memory card,1080 or 720. Recording medium: professional optical disc recording media is a 12cm (5 inch) single-sided, rewritable optical disc with a capacity of 50GB or SxS memory card. A 16GB SxS records approximately one hour of video in the format XDCAM EX, card data transfer up to 800 Mbps.



XDCAM HD mpeg HD

Developed by Sony, tapeless format, MPEG-2 data compression, 1080, recording rates: 18Mbps, 25Mbps, 35Mbps, up to 50Gb dual layer professional disk, all-digital, file-based workflow from video acquisition through editing. Recording medium: professional optical disc recording media is a 12cm (5 inch) single-sided, rewritable optical disc with a capacity of 50GB or SxS memory card.



XDCAM HD mpeg HD422

Developed by Sony, tapeless format, MPEG-2 data compression, top range of XDCAM, up to 50Mbps, 1920x1080 resolution, 14 bit AD conversion, 4:2:2 recording



DVCPRO HD

Format developed by Panasonic, tape based, also called D7-HD, same tape as used for DVCAM and DVCPRO, formats: 720/60p, 1080/60i, 1080/24p in development, compression rate 6.7:1, 10-bit, 8 audio channels



D5 HD

Format developed by Panasonic, tape based, 720/60p, 1080/24p, 1080/60i and 1080/30p, compression rates 4:1 in 8-bit mode and 5:1 in 10-bit mode, 8 audio channels



AVCHD

'Advanced Video Codec High Definition', developed by Panasonic and Sony for consumer camcorders using highly efficient codec technologies, formats 1080i and 720p, recording on SD Memory Card or Harddisk, MPEG4-AVC (H.264) video compression at a maximum of 24 Mbps, audio stream as AC-3, no uncompressed linear PCM audio is supported, video, audio, subtitles and ancillary streams are multiplexed together into an MPEG-2 transport stream that is stored as binary files.
pixel formats: 1920x1080i, 1440x1080i, 1280x720p, 720x480i, 720x576i, compression format: MPAG-4 AVC / H.264. The recorded format of 1440×1080 is upsampled to 1920×1080.



HDCAM

Developed by Sony and introduced in 1997, tape based format, high quality professional broadcast recording format, HDCAM is the HD version of Digital Betacam, 24PsF, 25PsF, 50i and 60i. The recorded format of 1440×1080 is upsampled to 1920×1080. The bit rate is 144 Mbps



HDCAM SR

Developed by Sony and introduced in 2003, tape based format, top professional broadcast recording format used by most top broadcast facilities, defined in the standard SMPTE 409M-2005, 4:4:4 recording, bit rate of up to 880Mbps, 12 uncompressed audio tracks uncompressed at 24bit 48kHz.



MPEG

stands for 'Moving Picture Experts Group', the organization that developed a set of standards for video and audio compression and multimedia delivery.



MPEG-1   H.261

was designed in 1990 for coding progressive video at a transmission rate of about 1.5 Mbps for videoconferencing, Video-CD and CD-i media. The audio standard MP3 has also evolved from the MPEG-1 (audio layer-3).
Files with the .mlv extension are usually elementary streams that contain only video information. Files with the mpeg and mpg extension usually contain MPEG-1 encoded video and MP2 (MPEG-1 Layer II) encoded audio.
(The file extensions mpeg and mpg are not exclusively MPEG-1, but used for MPEG-2 content too.)
Image formats 176 x 144 or 352 x 288 pixels at a data rate of 128 and 384 kbps, color subsampling 4:2:0, YCbCr.



MPEG-2   H.262

was designed in the mid 1990s for coding interlaced images at transmission rates above 4 Mbps and is/was the general format for SD digital TV broadcast and DVD. The compression rate for MPEG-2 video can be low as 2 Mbps, but the recommendation of the European Broadcast Union (EBU) is a minimum compression rate of 10 Mbps. To achieve lower compression rates with acceptable results the newer compression format MPEG-4 Part 10 should be used (see below).
Still now (2010) the overwhelming majority of on-air tv channels uses MPEG-2 as the compression method of choice.
To move from MPEG-2 to MPEG-4 all STBs (set top boxes) of the tv watching clients must be changed. This makes sense with the introduction of new channels but is often hard to achieve with an otherwise not changed tv channel (audience acceptance).
Best pragmatic approach is to use multiplexers which can handle both MPEG-2 and MPEG-4 streams at the same time.



H.263

enhancement of H.261, standardized in 1995/96, mostly for videoconferencing
Image format 352 x 288 pixels



MPEG-3

was intended for High Definition TV (HDTV), but it became apparent that the MPEG-2 standard met the HDTV requirements.



MPEG-4

MPEG-4 does not define specific transport layers. The adaptation to existing transport layers has been defined as transport over MPEG-2 Transport Stream and transport over IP.
The MPEG-4 format allows the hybrid coding of pixel-based natural images and video together with computer generated synthetic scenes. Video can be progressive or interlaced. Resolution can be up studio resolution of 4000 x 4000 pixels.

MPEG-4 Part 1
Version 1 was approved by MPEG in December 1998.

MPEG-4 Part 2   H.263
Version 1 was approved by MPEG in December 1999. MPEG-4 Part 2 is the standard for the Sony HDCAM SR recording format.



MPEG-4 / Part 10, ISO/IEC 14496-10, AVC (Advanced Video Coding)   /   H.264

The MPEG-4/Part 10, AVC video compression format is identical with the ITU-I standard H.264. Both (MPEG and ITU) worked together on this standard in the so-called "joined video team". The standard was finally published in 2003.
Special features of this compression standard in comparison to MPEG-2 are:
- the ability to use multiple reference pictures, up to 16 frames or 32 fields,
- implementation of multiple motion vectors,
- improved entropy encoding with statistical data to predict picture details
- up to 50% data reduction can be achieved in comparison to the former MPEG-2 standard

H.264 is used for BluRay and HD TV via DVB-S2.



MPEG-4 Part 12
On the basis of the MPEG-4 format, in 2004 the ISO Base Media File format was defined, which is the basis for file formats like 3GP and Motion JPEG 2000.
The ISO Base Media File Format was published as the MP4-Family by Apple, named in Annex D in MPEG-4 Part 12.

MPEG-4 Part 14
Revised standard, published in 2003 and 2004 (ISO/IEC 14496-14:2003, ISO/IEC 14496-12:2004). MPEG-4 Part 14 is based on the MOV Quicktime container format (developed by Apple). It is basically identical with the MOV format with additional features.
MPEG-4 Part 14 has replaced the initial MPEG-4 Part 1 standard, published in 2001 (ISO/IEC 14496-1:2001) based on the very first MPEG-4 standard developed in 1999 (ISO/IEC 14496-1:1999).
The MPEG-4 Part 14 format is a container format that can contain any number of video, audio and/or subtitle streams. The file extension will not determine the types of streams in the container.
The official file extension was introduces as .mp4, but the extension .m4a (introduced by Apple) can be used in exchange. These could be video and/or audio files.
The file extension .m4v was originally introduces for raw video files but is used also for general video files in the MP4 container.

MPEG-4, Network Abstraction Layer
Introduced in 2007, enables scalable video coding. The program is coded into different layers within one bitstream and different layers of the stream can be used by different receiving devices (base and enhancement layers). Low resolution decoders only need to decode the elements needed for the desired image resolution. Higher resolution decoders choose the appropriate parts of the stream. Basic services can be updated without downloading the entire stream again by just download the needed additional parts. In a scalable codec each layer is coded with standard MPEG-4 AVC tools.



MPEG Transport Streams (TS)

The Transport Stream is a container for multiple video, audio and meta data streams. The concept of transport streams was developed for less reliable transmission in comparison to Program Streams for more robust formats like DVD etc.
The Program Stream results from combining one or more Packetised Elementary Streams (PES) with a common time base, the Transport Stream combines one or more Packetised Elementary Streams (PES) with independent time bases into a single stream.
The transport stream concept is used for all transmission of video with more than one audio stream but can be used for multiple video streams at the same time too. These streams are multiplexed and can be re-played simultaneously.

The transport stream is identified by Packet IDs to communicate with the multiplexer what streams are included in the multiplexed signal. A PMT (Program Map Table) lists all elementary streams with their Packet IDs.



Formats for Mobile Video Devices

ATSC M/H 416 x 240 WQVGA
DVB-H (MPEG-4/AVC Level 1.2) 288 x 352 CIF
832 x 480 WVGA
624 x 360 3/4 WVGA (nHD)


3GP

3GP is a variation of the MPEG-4 Part 12 format used for mobile phones etc. The file extension is normally .3gp or .3g2.



WebM

WebM is Google's implementation of the VP8 format, developed by On2. WebM should be comparable to the codec H.264 both in quality and in performance, with a slightly higher visible quality of movies coded with H.264.
WebM is a royalty-free media file format, especially developed for web use.