DV
From Free net encyclopedia
Image:MiniDVMpegMan.jpg Template:Alternateuses
Digital Video (DV) is a video format launched in 1996, and, in its smaller tape form factor MiniDV, has since become one of the standards for consumer and semiprofessional video production. The DV specification (originally known as the Blue Book, current official name IEC 61834) defines both the codec and the tape format. Features include intraframe compression for uncomplicated editing, a standard interface for transfer to non-linear editing systems (FireWire also known as IEEE 1394), and good video quality, especially compared to earlier consumer analog formats such as 8 mm, Hi-8 and VHS-C. DV now enables filmmakers to produce movies inexpensively, associated with no-budget cinema.
There have been some variants on the DV standard, most notably the more professional DVCAM and DVCPRO standards by Sony and Panasonic, respectively. Also, there is a recent high-definition version called HDV, which is rather different on a technical level since it only uses the DV and MiniDV tape form factor, but MPEG-2 for compression (and, as a result, records higher resolution but more highly compressed video).
Contents |
Technical standards
Before compression
Before arriving at the codec compression stage, light entering the camcorder is converted to analog signals, then digital code (binary). Incoming light first makes contact with the CCD(s), which transforms the light into analog electrical signals, varying depending on the levels of light.
The RGB channels from the CCD are then sampled and quantized by an analog to digital converter (ADC or A/D) into a binary stream. The stream is processed by a digital signal processor (DSP) or custom ASIC.
Chroma Subsampling: The analog signal is now sampled 13.5 million times per second (13.5 MHz) for the Y (luminance) signal and 3.375 million times per second (3.375 MHz) for the (B-Y) and (R-Y) (chrominance) signals. This produces an accurate rendition of the activity of the video image. The Y channel is sampled 4 times more than the (B-Y) and (R-Y) channel; this ratio is commonly commonly denoted 4:1:1. In the PAL standard, the (B-Y) and (R-Y) channels are sampled at a frequency of 6.675 MHz. However, the chrominance samples from alternate lines are discarded, reducing the vertical chrominance resolution by a factor of two. This sampling format is denoted 4:2:0.
Quantization: The quantization process represents each sampled signal with an array of numbers; one for Y, one for U, and one for V. The Y channel has numbers appearing 13.5 million times per second, while the other two have 3.375 million numbers per second each. Each number in these arrays is between 0 and 255, which is known as 8-bit quantization.
Binary Notation: In the final stage of A/D conversion, the decimal strings are transformed into binary strings of 0's and 1's. These strings form the digital data stream to be recorded to tape. However, the process creates a huge amount of data, about 162 million bits per second (162 Mbit/s). In order to reduce the amount of tape required, the data is first compressed, using one of several algorithms including Discrete Cosine Transform (DCT), Adaptive Quantization (AQ), and Variable Length Coding (VLC).
Video compression
DV uses DCT intraframe compression at a fixed bitrate of 25 megabits per second (25.146 Mbit/s), which, when added to the sound data (1.536 Mbit/s), the subcode data, error detection, and error correction (approx 8.7 Mbit/s) amounts in all to roughly 36 megabits per second (approx 35.382 Mbit/s) or one Gigabyte every four minutes. At equal bitrates, DV performs somewhat better than the older MJPEG codec, and is comparable to intraframe MPEG-2. (Note that many MPEG-2 encoders for real-time acquisition applications do not use intraframe compression.)
Chroma subsampling
The chroma subsampling is 4:1:1 for NTSC or 4:2:0 for PAL, which reduces the amount of color resolution stored. Therefore, not all analog formats are outperformed by DV. The Betacam SP format, for example, can still be desirable because it has similar color fidelity and no digital artifacts. The lower sampling of the color space is also a reason why DV is sometimes avoided in applications where chroma-key will be used. However, a large contingent feel the benefits (no generation loss, small format, digital audio) are an acceptable tradeoff given the compromise in color sampling rate.
Audio
DV allows either 2 digital audio channels (usually stereo) at 16 bit resolution and 48 kHz sampling rate, or 4 digital audio channels at 12 bit resolution and 32 kHz sampling rate. For professional or broadcast applications, 48 kHz is used almost exclusively. In addition, the DV spec includes the ability to record audio at 44.1 kHz (the same sampling rate used for CD audio), although in practice this option is rarely used. DVCAM and DVCPRO both use locked audio while standard DV does not. This means that at any one point on a DV tape the audio may be +/- 1/3 frame out of sync with the video. This is the maximum drift of the audio/video sync though it is not compounded throughout the recording. In DVCAM and DVCPRO recordings the audio sync is permanently linked to the video sync.
Connectivity
The FireWire (aka IEEE 1394) serial data transfer bus is not a part of the DV specification, but co-evolved with it. Nearly all DV cameras have an IEEE 1394 interface and analog composite video and Y/C outputs. High end DV VCRs may have additional professional outputs such as SDI, SDTI or analog component video. All DV variants have a timecode, but some older or consumer computer applications fail to take advantage of it. Some camcorders also feature a USB2 port for computer connection, but these are sometimes not capable of capturing the DV stream in full detail, and are instead used primarily for transferring certain digital data from the camcorder such as still pictures and computer-format video files (such as MPEG4-encoded video).
On computers, DV streams are usually stored in container formats such as AVI or QuickTime, but sometimes raw DV data are recorded directly.
Physical format
Image:DV tape sizes 2.jpg The DV format uses "L-size" cassettes, while MiniDV cassettes are called "S-size". Both MiniDV and DV tapes can come with a low capacity embedded memory chip (most commonly, a scant 4 Kbits for MiniDV cassettes, but the system supports up to 16 kbits). This embedded memory can be used to quickly sample stills from edit points (for example, each time the record button on the camcorder is pressed when filming, a new "scene" timecode is entered into memory). DVCPRO has no "S-size", but an additional "M-size" as well as an "XL-size" for use with DVCPRO HD VTRs. All DV variants use a tape that is 1/4 inch (6.35 mm) wide.
MiniDV / DV
The "L" cassette is about 120 x 90 x 12 mm and can record up to 4.6 hours of video (6.9 hours in EP/LP). The better known MiniDV "S" cassettes are 65 x 48 x 12 mm and hold either 60 or 90 minutes of video depending on whether the video is recorded at Standard Play (SP) or Extended Play (sometimes called Long Play) (EP/LP). 80 minute tapes are also available and can record 120 minutes of video in EP/LP mode. The tapes sell for less than USD 4 each as of 2005. DV on SP has a helical scan track width of 10 micrometres, while EP uses a track width of only 6.7 micrometres.
Software is currently available for ordinary home computers which allows users to record any sort of computer data on MiniDV cassettes using common DV decks or camcorders. A 60-minute MiniDV tape will hold approximately 13 Gigabytes of data in this form of usage as the DV video format has a constant data rate of 3.6 Megabytes per second (3.6 MB/s x 60 seconds x 60 minutes = 12,960 MB per hour, divided by 1024 = 12.66 GB per hour).
DVCAM
Sony's DVCAM is a semiprofessional variant of the DV standard that uses the same cassettes as DV and MiniDV, but transports the tape 50% faster, leading to a higher track width of 15 micrometres. The codec used is the same as DV, but because of the greater track width available to the recorder the data are much more robust, producing 50% less errors known as dropouts. The LP mode of DV is not supported. All DVCAM recorders and cameras can play back DV material, but DVCPRO support was only recently added to some models. DVCAM tapes (or DV tapes recorded in DVCAM mode) have their recording time reduced by one third.
DVCAM is now also available in HD mode.
DVCPRO
Panasonic specifically created the DVCPRO family for electronic news gathering (ENG) use (NBC's newsgathering division was a major customer), with better linear editing capabilities and robustness. It has an even greater track width of 18 micrometres and uses another tape type (Metal Particle instead of Metal Evaporated). Additionally, the tape has a longitudinal analog audio cue track. Audio is only available in the 16 bit/48 kHz variant, there is no EP mode, and DVCPRO always uses 4:1:1 color subsampling (even in PAL mode). Apart from that, standard DVCPRO (also known as DVCPRO25) is otherwise identical to DV at a bitstream level. However, unlike Sony, Panasonic chose to promote its DV variant for professional high-end applications.
DVCPRO50 is often described as two DV-codecs in parallel. The DVCPRO50 standard doubles the coded video bitrate from 25 Mbit/s to 50 Mbit/s, and improves color-sampling resolution by using a 4:2:2 structure. DVCPRO50 was created for high-value ENG compatibility. The higher datarate cuts recording-time in half (compared to DVCPRO25), but the resulting picture-quality is reputed to rival Digital Betacam, a more expensive studio format.
DVCPRO HD, also known as DVCPRO100, uses four parallel codecs and a coded video bitrate of 100 Mbit/s. Despite HD in its name, DVCPROHD downsamples native 720p/1080i signals to a lower resolution. 720p is downsampled from 1280x720 to 960x720, and 1080i is downsampled from 1920x1080 to 1280x1080 for 59.94i and 1440x1080 for 50i. Compression ratio is approximately 7:1. To maintain compatibility with HDSDI, DVCPRO100 equipment internally downsamples video during recording, and subsequently upsamples video during playback. A camcorder using as special variable-framerate (from 4 to 60 frame/s) variant of DVCPRO HD called VariCam is also available. All these variants are backward compatible but not forward compatible.
DVCPRO cassettes are always labeled with a pair of run times, the smaller of the two being the capacity for DVCPRO50. A "M" tape can hold up to 66/33 minutes of video. The color of the lid indicates the format: DVCPRO tapes have a yellow lid, longer "L" tapes made specially for DVCPRO50 have a blue lid and DVCPRO HD tapes have a red lid.
DVCPRO VCRs can play (but not record) DV and DVCAM tapes. MiniDV tapes can be played with a cassette-adaptor.
Other Digital Formats
Sony's XDCAM format allows recording of MPEG IMX, DVCAM and low resolution streams in an MXF wrapper on an optical medium similar to a Blu-Ray Disc, while Panasonic's P2 system uses recording of DV/ DVCPRO/ DVCPRO50/ DVCPROHD streams in an MXF wrapper on PCMCIA-compatible flash memory cards. Ikegami's Editcam System can record in DVCPRO or DVCPRO50 format on a removable hard disk. Note that most of these distinctions are for marketing purposes only - since DVCPRO and DVCAM only differ in the method in which they write the DV stream to tape, all these non-tape formats are virtually identical in regard to the video data.
JVC's D-9 format (also known as Digital-S) is very similar to DVCPRO50, but records on videocassettes in the S-VHS form factor. (NOTE: D-9 is not to be confused with D-VHS, which uses MPEG-2 compression at a significantly lower bitrate)
The Digital8 standard uses the DV codec, but replaces the recording medium with the venerable Hi8 videocassette. Digital8 offers DV's digital quality, without sacrificing playback of existing analog Video8/Hi8 recordings.
See also
External links
- Adam Wilt's DV page with in-depth technical information
- Digital Video Information Network messageboards
- DV/DVCAM Video production at FilmDailies.com - a filmmaker's blogde:Digital Video
fr:Digital Video ko:DV nl:Digital Video it:DV ja:DV pl:Digital video sk:DV fi:DV sv:DV