How does video compression work?
At its simplest, video compression uses codec algorithms to automatically remove unnecessary or repetitive sounds or images from a large video file, thus making the files smaller. There are three main factors that go into determining the size of a video – resolution, bitrate, and encoding – and they can all play a role in helping reduce video file size:
- Resolution is the number of pixels, typically represented by a horizontal x vertical measurement (1080p HD video, for example, has a resolution of 1920 x 1080)
- Bitrate measures how much information is transmitted for each second of video. It’s usually measured in megabits per second (Mbps)
- Encoding consists of both the codec (the code used to compress the video, including MPEG and ProRes), along with its container or file type (such as AVI or MP4); some file types feature more efficient codecs (MP4 files offer more efficient compression than FLV files, for example)
There are other factors that help determine video file size including video length and frame rate (measured by frames per second, or FPS).
The size of video files, like all data files, is typically measured in bytes or variations thereof: kilobytes (KB, 1024 bytes), megabytes (MB, 1024 kb), gigabytes (GB, 1024 megabytes), and terabytes (TB, 1024 GB).
Videos living in the gigabyte neighborhood and above traditionally have been considered for compression before transferring due to the enormity of this content. Without compression, these massive video files would choke networks and simply not get delivered without the use of more robust networking technology (such as UDP solutions like IBM Aspera) or the logistically complicated-if-tried-and-true method of saving files on a hard drive and shipping it by courier. The UDP option is fast but expensive. Hard drives are slow, and you actually run the risk of losing your content in the mail.
Because it involves shrinking or scaling video, it’s crucial for filmmakers or post-production houses using manual compression software to think about where their video will most likely be consumed.
If primarily on YouTube, then a resolution around 1080p will likely be adequate (YouTube offers a guide to its recommended upload encoding settings here), but a video meant for larger screens or sharp, 4k-and-higher displays, should have a higher resolution. (Technically, as of Fall 2020, YouTube is supposed to be able to support massive, 8k video content. We’re certain that November 2020’s YouTube outage is not because of MKBHD’s 8k uploads, as has been rumoured)
What are codecs and how do they compress video?
Codecs are the ubiquitous codes used by various file formats to compress large video (or audio). Several types of video codecs exist including MPEG4, Quicktime, ProRes, or WMV. These codecs compress data automatically when a file is saved as a certain format or file container (such as AVI or MP4).
Most video codecs perform “lossy” compression in that they always throw out some data in order to compress the file and make it smaller – hence the video quality issues related to compression that we mentioned earlier. Next-generation video codecs such as HEVC (also known as H.265) have also gained popularity in recent years, offering more efficient data compression with far less impact on video quality. Even newer next-gen codecs include Versatile Video Coding (VVC), Essential Video Coding (EVC), and Low Complexity Enhancement Video Coding (LCEVC), along with Google’s VP9 and the open-source AV1.
Most next-gen codecs such as HEVC are classified as “lossless” compression, in that they don’t lose any data and keep the same video quality after compressing (although there is mild debate about that in some quarters).
How To Compress a Video?
Two relatively easy ways to compress a large video file are to either make the video shorter by trimming footage, or to remove the audio completely (which probably isn’t a realistic option for most projects). Barring those methods, here are a few other ways to compress video files using either desktop or web software.
Method 1: How to Compress video via VLC
- Click Media > Convert/Save
- Click Add to select your video file (or multiple files)
- Click Convert/Save to bring up a list of conversion options. You can select your preferred type in the profile dropdown
- Select your conversion option (the software offers useful suggestions such as YouTube HD or Video for MPEG4 1080p TV/device)
- If you still need to reduce your file size, you can next lower the video resolution by going to Settings > Resolution
- Once you’re satisfied with your selections, simply hit Save, select your destination location (on your hard drive or cloud storage), and then click Start.
Method 2: How to Compress video via Shotcut
- Click Open File to open your video
- Click Export
- You’ll next see a large list of compression options – select your preferred option
- You can use the Resolution and Aspect ratio fields from this screen to further reduce video size and proportions (it won’t automatically adjust your video’s aspect ratio for you)
- Click export video
Method 3: How to Compress video via QuickTime
- Click File to open your video fileopen file to open your video
- Click Export As
- You’ll next get a list of options, but you’re limited to just four: 4k, 1080p, 720p, and 480p
- To compress your file, select a smaller file format than the original file
Other video compression services include VideoSmaller, Clipchamp, and HandBrake. Additionally, InVideo summarizes a few ways to compress videos on their blog. They all do the job reasonably well, but also suffer from the same drawback: They add several (often needless) steps to the workflow of video professionals, who in most cases have enough on their plates without having to compress large video files before sharing.
The downside of compressing video files
The addition of several steps is just one downside of having to compress video files. As we mentioned earlier, most codecs perform lossy compression, so you’ll almost inevitably lose some audio and video quality. In some cases this may be imperceptible, but it can be problematic in other scenarios, such as the color correction phase of post-production when minutely small visual details are vital (and can be lost forever after being compressed).
There’s also additional processing time required, especially for higher-quality codecs such as HEVC (which has been described as having “glacially slow encoding times”). File errors can crop up during the compression process. And your recipients may also not even be able to open your compressed file immediately if they don’t have the right software. Determining how to compress video properly and efficiently is a fine balance.