The era of VHS tapes and pellicule is but a melancholic memory for most of us. Nowadays it’s all video and the amount of different formats are numerous. So is the confusion about all those formats. In this blogpost we’ll give you an insight into codecs, compression and converting.
Let’s first take a short theoretical excursion to better understand how video is stored digitally. A video is basically a fast series of stills. In Europe we use the PAL standard, which defines 25fps - frames per second. So per second, 25 frames are shown consecutively, creating the feeling of motion. But if we save 25 frames per second on a hard drive, we would roughly need 410 Gigabyte for just one hour of video. Not only would that require tons of storage, heavy files also prevent an efficient workflow.
To make video more manageable, we use digital compression. Encoding is the process of putting data into a specialised format for efficient transmission or storage. Decoding is the opposite process - the conversion of an encoded format back into the original data. This is done by codecs. There are two kinds of compression techniques: Lossy and Lossless.
Compression: quality vs. file size
Lossless compression means that after the process of decoding, the data is exactly the same as before. There is not much loss of quality. But the actual compression is limited, so the files are still big.
Lossy compression on the other hand will produce a file that is not the same as the original. A certain amount of data has been eliminated in order to produce smaller files. Complex algorithms try to figure out how they can best reconstruct the file, with the limited data that they have. How they do this exactly, is beyond the scope of this blog.
Compression techniques of this kind can shrink huge files to a fraction of their original file size. One hour of video can now fit easily in a Gigabyte. What we give up in quality, we gain in usability. This the key consideration when compressing files: quality vs. file size.
Codecs and their use
There are roughly three groups of codecs: camera codecs, editing codecs and delivery codecs. Their names give you a clue about what they do.
Camera codecs like XDCAM, AVCHD, RAW,… are designed to be used by a camera, at the recording stage. Most of them are developed by the camera manufacturers.
Editing codec like ProRes and DNxHD are being used during the editing phase and to facilitate the exchange of footage between production companies. They try to sacrifice as little information as possible, while still enabling a smooth and flexible workflow. Quality trumps file size here, so Gigabytes gain quickly.
Does MP4, H.264, mpeg,… ring any bells? Chances are you’ve heard of them, cause these are some well know delivery codecs. They are aimed at consumer applications such as dvd, internet, blu-ray,… These codecs prioritise usability and sacrifice quality. They make sure your YouTube video plays smoothly, without a reloading stutter every now and then. To succeed in doing so, they have to give in quality. As a consequence, these videos are difficult to be re-used afterwards for editing purposes, because quality that is lost, cannot be restored.
Why does this .mov is playing smoothly and another one doesn’t play at all?
A common question! It’s because .mov is not really a format, but a container. Video files are often used in container formats such as .mov, avi, wmv, mkv. They act as containers of different media types. So beside a video file, they can also contain an audio file or even audio files in different languages, subtitles,… To know why one file plays and another file of the same type doesn’t, you have to look at the codec that is being used. If your computer doesn’t recognise the codec, you won’t be able to see the video.
Going back to our .mov example and bringing in our new knowledge of codecs: if you receive a .mov file that was made by a H.264 codec, which is a delivery codec that almost all browsers and video players can handle, you will be able to see the video. If on the other hand you receive a .mov file with a ProRes codec, and if your computer doesn’t have this codec installed, you won’t be able to play it back.
Sometimes clients ask us if they can get the rushes of the film shoot we did for them. No problem of course, but you understand now that our high quality rushes might be in a codec that cannot be played by a standard media player.
But we can convert files, right?
Indeed, files can be easily converted from one codec to another, using video encoding software. But bear in mind that information that is being lost, can’t be restored. So converting a file from a delivery codec to an editing codec, will not result in a higher quality file. Also, every time you encode a file, there will be a loss of quality. For this reason, to ensure the best result, it’s always advisable to keep at least one high quality file as your main ‘source’ for future conversions.
There you have it! We hope that the world of digital video is a little bit clearer now.