When working in audio I feel pretty comfortable working with WAV, raw, PCM, MP3 or whatever compressed/uncompressed file gets thrown my way, and life was relatively neat rowing down the bitstream of audio format. But when I dared dip my toe into the perplex and competitive world of video postproduction it was time to buckle up and ride out the storm for a new, ever changing course of elements.
Here are some basic explanations that I hope can get you through the confusing world of digital video and codec:
What is codec?
The basic explanation of codec is in the name:
- Co- compress
- Dec- to decompress.
You’re probably more familiar with container formats then you think:
AVI (audio, video interleave) is a container known to many. It can contain many formats, but is commonly known for containing J-MPEG or DIVX.
QuickTime: A popular container format that works across many platforms including Linux, it was developed by apple, as opposed to the AVI Microsoft framework. This also carries many formats including the two above.
Basically used to represent the amount of colour used in an image, therefore the lower the bit depth the less colour-rage to chose from, with a higher bit depth meaning better resolution. This can make a difference between and clear and grainy piece of wok.
The example above shows 16 bit and 8 bit, although you can visibly see the difference it is mindful to note:
8 bit codec has of Red, Green, Blue information = 256
16 Bit codec has of Red, Green, Blue information = 65, 536
It is important to capture in a high bit depth to represent the wide colour range, and detail, for production.
Basically used to define the amount of colour and luminance within and sample, usually for every four pixels.
Some examples are:
4:2:2 = meaning for every 4 pixels 2 chroma (colour) values are used and 4 luminance values are used.
4:1:1 = uses 1 chroma sample, with 4 luminance.
4:4:4= no subsampling at all
To put this into perspective, if we have a red and blue image the transition from one colour to the other will be repentant on the subsample:
Progressive Vs Interleaving
We all see 1080p or 1080i when we are recording or rendering our final product but do you know the difference? Firstly let’s take a look at the numbers, 1080 is the numbers represented in resolution, currently digital high definition TV has a resolution of 1,920 x 1,080 pixels, however how these pixels are displayed depends on the scanning of the images, this is where the letters come to play. P, stands for progressive, this form of scanning displays both the even and odd lines of information on screen at the same time. Interlaced video even and odd lines as separate fields therefore for a smoother capture of an action shot 1080p is always better.