Digital video
Digital Video (DV) is the generic term for the DV standard that was introduced in 1994. It includes the cassette formats DV, MiniDV, DVCAM , Digital8 , HDV , DVCPro , DVCPro50 and DVCProHD .
Development and dissemination
After starting in 1996, digital video made rapid development. As early as 1999 there were products from a large number of manufacturers, with prices falling steadily. The system originally intended for the consumer market quickly secured a niche in the (semi-) professional sector because of its acceptable quality and low price.
Technical specifications
The images are recorded digitally and reduced to around 10 percent of the uncompressed storage space. In the DV standard, the individual images are individually encoded using a JPEG- like method ( intra-frame encoding ). First, the RGB signal generated by the camera's image sensor is converted into a YCbCr signal , in which the three light primary colors are not transmitted separately, but a brightness signal ("black and white image") is supplemented by two color difference channels from which the color can be reconstructed. As with the analog signal , chroma downsampling takes place : a reduction in the resolution of the color information, since the human eye has less visual acuity for color vision than for differences in brightness. This takes place in the ratio 4: 1: 1 ( NTSC ) or 4: 2: 0 ( PAL ), which results in a data reduction of half compared to the original signal (4: 4: 4). The ratio means that the two color component signals are each stored with a quarter of the resolution of the brightness signal: Four pixels therefore always have the same color value and only differ in their brightness. The designation 4: 2: 0 does not mean halving a component signal and completely omitting the other, but only serves to distinguish 4: 1: 1. With the latter, four neighboring pixels are always grouped in color in one line, while with 4: 2: 0 only two neighboring pixels have the same color, but also the pixels of the line below (subsequent field ) in the same place. The four color-coded pixels are arranged here as a 2x2 square instead of in a horizontal row. The chroma subsampling is followed by a lossy, digital image compression based on the discrete cosine transformation (DCT) with subsequent quantization and entropy coding , which reduces the data rate again to around a fifth, but at the expense of the formation of compression artifacts . The process is thus similar to MJPEG and, in contrast to MPEG, offers the advantage of being able to cut videos at any point without having to calculate intermediate images first. Reverse playback is also possible without any problems; in the case of inter-frame coding (such as is used in MPEG codecs, for example), this is associated with a high computational effort and requires significantly more working memory . However, since DV works in SD resolution that is low according to today's standards , this advantage has become largely insignificant on a modern PC with a modern CPU and several gigabytes of RAM .
The data rate of a DV stream is 3.125 MByte / s (25 megabits per second), i.e. around 1 GByte per 5 minutes. The image quality is due to the lack of noise and analog image disorders as well as a slightly higher resolution compared to older Hi8 - or SVHS - cameras subjectively better, but can - especially in MiniDV tapes - digital artifacts (blocks) caused by faulty tape recording, similar to you know them from disturbed digital television reception. With Digital8 , the problem is much less common due to the wider tapes and higher quality cassettes. This also applies to the professionalized DV formats (DVCAM, DVCPro), which work with a higher tape speed instead of wider tapes. Digital recording enables lossless transferring or capturing (importing the video onto a computer, for example for cutting or further editing), especially in connection with the iLink interface . Compared to professional, digital video formats such as D-1 or Digibeta , DV has a significantly lower image quality and a higher susceptibility to interference, but it is a far more cost-effective alternative, with more compact and lighter cassettes (and consequently also devices), which are usually per tape have a longer playing time. This led to an increasing acceptance of the format in the broadcast sector too , especially for electronic reporting .
DV uses 720 × 576 pixels ( PAL ) or 720 × 480 pixels ( NTSC ) as the image resolution . In the PAL format, 50 fields per second (50i) are recorded, with NTSC 59.94 (60i). Some DV cameras also allow full-frame recording at frame rates of 24p or 25p (also known as progressive scan ), whereby for technical reasons each frame is split into two fields that are recorded sequentially (at 25p, only on PAL cameras), or in alternating 3 and 2 fields ( 3: 2 pull-down ) on NTSC cameras.
The audio format used in DV is uncompressed PCM sound with a sampling rate of 48 kHz and a resolution of 16 bits, as is the case with contemporary digital cinema sound formats (such as Dolby Digital or DTS ), but which (to enable multi-channel sound at a feasible data rate) lossy compression ( AC -3 algorithm ). With DV there are two channels (stereo) available. At the expense of the sound quality, 32 kHz with 12-bit sampling on four channels were also possible. The latter option is limited to consumer devices and is generally not found with semi-professional DV cameras or with DVCAM and DVCPro . It was probably intended to provide bilingual stereo sound for films released on DV retail cassettes. Since the DVD and not (as Sony hoped) DV replaced the VHS cassette as the standard in this area of application , this never happened. With the DVCAM format, picture and sound are synchronous (locked audio); With consumer DV format, a sound-image offset of ± 1/3 frame (13.33 ms) is permitted (unlocked audio)
The width of the tape material is 6.35 mm for all variants (except Digital8).
MiniDV
In the amateur field, the MiniDV has established itself, which records in the same data format as DV on smaller cassettes. MiniDV cassettes are offered in capacities of 60/90 (60 minutes standard play / 90 minutes long play) and 80/120 (80 minutes standard play / 120 minutes long play), although the latter is generally used due to the thinner and therefore more sensitive tape advised against. The cassettes are far cheaper (also in terms of playing time) than the standard DV cassettes, which are roughly twice as large and are only used in the semi-professional and professional sector.
Compatibility with DVCAM
A MiniDV cassette can also be recorded in professional DVCAM format using the appropriate equipment . However, since these devices move the tape faster when recording, the recording time indicated on the cassette is reduced by 1/3. Nevertheless, the use of DVCAM certified tapes is recommended for recording in DVCAM format, as these i. d. Usually have a higher quality tape.
Longplay
Most consumer DV camcorders, but also some semi-professional devices, have a long play mode that increases the running time per cassette by 50%. A DV cassette with a regular playing time of 60 minutes fits 90 minutes. The special thing about it is that the use of the long play mode with DV does not lead to any loss of picture or sound quality, in contrast to most other video systems. This fact leads many users to frequent, if not permanent, use of longplay, as this saves tape and money without any disadvantage that is directly apparent to the user. In fact, long play has a number of disadvantages and dangers.
The DV long play mode does not correspond to any set standard. This means that virtually every manufacturer of DV equipment is free to design the LP mode at its own discretion. In practice, however, the manufacturers have agreed on certain tolerances in order to achieve at least some of the compatibility. However, the exchange of DV tapes that were recorded in long play mode is extremely problematic. Often, error-free playback is only possible on the device on which the tape was recorded. Playback may be difficult even on other models from the same manufacturer.
With DV, exactly the same data is stored on the tape in long play mode as in SP mode (hence the same quality). However, the belt runs at a speed reduced by 33%, resulting in a narrower track width. Due to the greatly increased data density, however, there is hardly any leeway to compensate for possible tape errors, which leads to frequent dropouts . The signal strength on the tape is also considerably lower than in the SP mode and is only slightly above the legibility limit even with a newly recorded tape. All of this means that DV tapes recorded in LP mode generally have a very short lifespan and are often no longer readable after a few years.
Related formats
Usually only the cassette formats DV and MiniDV, which are recorded with the usual DV codec, are referred to with the term DV in the narrower sense. There is also the HDV codec, which enables HD data to be recorded on these two cassette formats.
The professional formats DVCAM (Sony) and DVCPro (Panasonic) use a similar technique , whereby the tape is transported faster, which means less data loss is to be expected. With DVCPro50 , double the data rate (50 Mbit / s) is used in order to achieve a higher quality. There is also the DVCProHD format, also from Panasonic, which is used in the VariCam camera system, for example . This format has a higher resolution than the normal DV format and is used for recording in HD resolution. In the future, high-resolution videos will be recorded in the AVCHD format, which is supported by Panasonic , Sony , Samsung and Canon , among others .
Digital8 is a digital further development of the Video 8 / Hi8 product line , uses the same cassette format as its analog predecessors but uses the usual DV codec.
transmission
Among other things, the DV devices have the same analog interfaces ( composite , S-Video ) as video devices in VHS or Hi8 format. Therefore you don't need a PC to display the data on the television. In order to be able to copy them digitally and thus without loss, the FireWire interface developed parallel to DV is used. This connection, usually referred to as DV-Out, is also used for data transfer from and to PCs. Since the data packets are transmitted without loss and without conversion, the load on the processor is very low.
In addition to a DV output, some camcorders also have an input, the so-called DV-In. In addition to 1: 1 copies from one camcorder to another, it is also possible with devices equipped in this way to play back videos edited on the PC without loss on MiniDV. From a purely technical point of view, all DV camcorders would have a DV-In (which is also the case in the USA or Japan, for example), since camcorders with DV input are considered to be video recorders for tax purposes in Europe, which means that higher import duties apply , the DV-In has been deliberately deactivated by the manufacturer on many models. Often there are also two versions of the same model that only differ in terms of DV-In and the resulting higher price. In addition to the DV-In, some camcorders also have an analog input. B. VHS tapes can be dubbed on MiniDV.
With older MiniDV camcorders, the DV-In was blocked purely on the software level and could therefore often be reactivated with little effort. Instructions for this can still be found on many websites today, as can the cables and software required. "Unlocking" the DV-In was not legally prohibited, but the warranty was often lost with devices modified in this way. The manufacturers reacted gradually by deactivating the DV input on the hardware side, so that subsequent activation of the DV-In is no longer possible with current models. Professional camcorders and DV / DVCAM video recorders have i. d. Usually via a DV input as standard.
When transferring the video stream from the camcorder to the PC, the DV material is usually stored on the hard drive in AVI container format. The video file then has the file extension ".avi". Apple's Macs mostly use “ QuickTime Movie” as the container format and the file extension “.mov”. DV is the standard format for video editing at Apple (e.g. iMovie , Final Cut Pro ).
File size
Storage space requirement per cartridge
Although most DV camcorders have a (mostly hardware-based) compression codec, the result is still fairly high bit rates and the associated large amounts of data. The DV codec, which is identical for MiniDV, DV and Digital8 (in contrast to the cassettes), delivers a data rate of approx. 28,000 kBit / sec with stereo sound. This corresponds to about 0.2 GB per DV minute with stereo sound. It follows:
- 60 minutes DV: 13 gigabytes
- 80 minutes DV: 17 gigabytes
Automatic scene recognition
However, this does not necessarily have to be identical to the size of the individual files obtained in this way; so you can z. For example, with many capture programs, automatic scene recognition based on the recorded data code - which contains information about various camera settings as well as the date and the exact time of the recording - provided that the camera time has been set before the recording, so that this data is always interrupted Codes of at least one second a new file is started. Some editing programs also support automatic, chronological sorting of the recordings, even if they are on different tapes or were even recorded with different cameras. With some devices in the consumer sector, the date and time are only stored in a volatile memory so that they have to be reset if the battery is removed or the battery is completely discharged.
The data code should not be confused with the time code , which is also recorded by all DV devices.
System-based limits on the possible file size
As a rule, however, very large files are created when transferring the contents of a DV cassette 1: 1 to a PC. You must therefore ensure that the file system of the hard drive supports this,
- z. B. the NTFS used by Windows , which has an upper limit per file of 16 TiB .
- FAT16 , however, has a max. File size of only 2 gigabytes (i.e. only about 10 minutes of DV data), FAT32 4 gigabytes.
- The standard HFS + file system on Apple computers also supports large files of up to 16 terabytes
- Linux and Unix file systems have no significant restrictions here either (e.g. ext3: 2 terabytes, ReiserFS: 1 exabyte , XFS: 8 exabytes, btrfs: 16 exabytes).
In addition, the file format used used to be important. The size of an AVI file was limited to 4 GiB due to its internal structure. This restriction has now been removed. With MOV files, there was and is no significant limitation in this regard.
To avoid this problem, practically all capturing programs support automatic file splitting, so that a new file is automatically started after the maximum size permitted by the respective file system has been reached. Some manufacturers, e.g. B. Canopus , also support so-called reference AVI. A virtual .avi file is created, which only contains information about where the actual video data is located on the hard drive - split into small, individual files (often with extensions such as .001, .002 etc.). Editing programs then process this reference AVI like a large video file of unlimited length.
DVD authoring: From DV file to DVD
The data can be processed well with different video editing programs, nowadays you can easily achieve very effective effects like dissolves, titles, etc. with most standard PCs.
In order to finally get the produced video in a DVD-compatible format, for example, further compression is required, since a standard DVD only has 4.7 GB. The standard codec for DVDs is therefore the more highly compressed MPEG-2 format to which the DV data must be converted. For the conversion on the PC you need DVD authoring software that ensures that all common video and DVD standards are complied with when writing to the DVD, so that it can be played in any commercially available DVD player. Some video editing programs are already equipped with such a function.
DV camera vs. Dvd camera
In contrast to camcorders with integrated DVD burners, DV cameras still have a decisive quality advantage, as the raw material is only slightly compressed and stored on the tape, which is particularly important during post-processing (editing, etc.). In the case of a “DVD video camera”, however, an integrated real-time MPEG2 codec takes care of the compression; depending on the type of camera and the quality set, this can show greater losses, whereby MPEG-2 generally suffers from greater quality losses than DV.
Another advantage of the MiniDV devices is that the recordings on the MiniDV tape - as far as we know today - are retained in their original quality for considerably longer than on DVDs; on the other hand, there were also camcorders that record on DVD-RAM .