THE NEED FOR COMPRESSION
THE PRIMARY REASON for any video codec—coder/decoder—is to enable the transmission of content without requiring lots of bandwidth. Take a standard movie with a 1080p resolution, running at 24fps: 1920 × 1080 × 24 (fps) × 3 (bytes per pixel) × 8 (bits per byte) = 1,194,393,600 bits of data for every second of uncompressed video. People with faster than gigabit connections might be able to handle that sort of data stream, but it’s inefficient.
But 1.2Gbps of data is only for 24fps content. A graphics card running at 1080p and 60Hz sends 3Gbps of data to the display. Modern high refresh rate displays can transmit at up to 54Gbps in the case of AMD’s DisplayPort 2.1 connection—4K at 240Hz for instance needs about 48Gbps. But sending gigabits of data over an HDMI or DisplayPort connection is one thing; sending it over the internet or storing content on a hard drive or SSD is another.
Using a video codec is like using JPG or PNG compression for images, rather than uncompressed bitmaps, but with video, there’s a third dimension available. Where a high-quality JPG file might be able to reach compression ratios of 12:1, video codecs can take uncompressed video and deliver good-quality alternatives with compression ratios of 250:1, 500:1, and even 1000:1 or more. And it’s a good thing because it’s what makes video streaming practical.
When you watch a 4K 60fps YouTube video, your PC sends around 12Gbps of data to your monitor. Your internet connection probably only has to deal with about 20Mbps. But not all video codecs are created equal, with differences in the algorithm, hardware requirements, and licensing costs. We’ve covered the three major contenders below.