An old army joke about a compression bucket that the “experienced” senior comrades are instructed to bring to a young soldier, checking his quick wits, is more relevant than ever for beginners in video editing on a PC. At first, they are simply shocked by the enormous amount of unfamiliar terms. Especially those who before this had no business with computers. Among our acquaintances there is a girl – an artist, who works perfectly in the complex compositing program Adobe After Effects, but falls into a state close to panic, if necessary, create a directory on the disk or make any kind of file manipulation. This lack of education was eliminated within 10 minutes by assurances that “it’s not difficult” and a little practice. The case is anecdotal, but revealing.
We venture to say that among the active “installers” on the PC there are many who do not know about video compression. Allow me a few words on this topic, which is quite important for those planning to get hold of a couple of video editing complexes.
The main problem that arises during video processing is the huge amount of data and, more importantly, a very large stream of this data when recording. You need a very powerful computer (and, first of all, with a high-performance disk system) only in order to “drive” the film from the VCR to the computer’s hard drive. To reduce the data flow, special data encoding and data compression are used (Data compression is compression, you won’t put it in a bucket).
Coding is simple:
To display the color of each point, 24 binary bits are required, 8 bits for each of the three primary colors – red, green and blue (RGB color palette). However, the eye is very sensitive to the luminance component and less sensitive to color. Therefore, the luminance component of each point of each frame is transmitted in 8 bits. But color difference signals (R-Y and B-Y) are not transmitted for each frame. For example, a 4: 2: 2 coding scheme means that for four samples of the luminance component, only two color difference samples are transmitted.
Such coding (more precisely, a conversion scheme) significantly reduces the data stream during recording. But still, the digital stream remains very large. For example, for a 4: 2: 2 scheme, the digital stream is about 177 Mbps.
To record such a continuous stream of data, you also need not only a PC with a high-performance processor, but also a powerful disk subsystem capable of managing to record this data without gaps and distortions. And the amount of disk space for recording at least one hour of video should be quite impressive. (Continuing the above calculations, you can calculate how much disk space is needed to record, for example, 10 minutes, one or several hours of video). Naturally, the cost of such an integrated solution is quite high and can not afford the average user. To reduce data flow, and save disk space, various methods of signal compression are used. Today, the most common compression standard is MJPEG. It is generally accepted that compression up to five times, that is, with a factor of 5: 1, of this standard allows you to maintain the quality of the Betacam SP format without errors (artifacts) that are not visible to the eye. Accordingly, the data flow and hardware requirements of the installation complex are reduced. As for the bottleneck – the disk subsystem, in this case it is possible to use even IDE disks with a rotation speed of 7200 r / s, which significantly reduces the cost of the solution. Currently, an even more progressive MPEG2 compression format is gaining ground.
The data stream of compressed digitized video can be 25 or 50 Mbps. This is also a lot, but is already available for most modern computers.
The main thing to remember is the higher the compression ratio of any format, the more artifacts that are noticeable to the eye, especially in multi-layer editing. But then less PC resources are required. And vice versa. Everyone must find the necessary compromise for himself.