Compressing video for the iPod: don’t sacrifice quality

Many people who produce online media have been socialized to keep file size in mind above all else. Concerns about download time and storage space have left many people I work with perpetually wary of creating files much larger than what once fit on a floppy disk — even if it means sacrificing image and sound quality.

You don’t need to do that.

Compression is still important – the iPod will only play video compressed into one of two specific formats (H.264 and MPEG-4, both described later in this posting) and only the most powerful computers can play uncompressed video without significant glitches. However, these playback limitations are not necessarily due to the amount of memory a file requires (the new iPods come in 30GB and 60GB models, so they can hold some hefty files) instead, limitations come from the processor power required to play video in a digital environment.

As a result, compressing video, or converting it into a form which can be more easily played back by your audience, means adjusting the data rate or bit rate (the amount of information that needs to be processed during each second of playback) as opposed to thinking only about the total amount of memory in the file. This holds true both for podcasting and for online streaming (it also applies to DVD production). Files with a higher bit rate contain more information and are harder for a device (such as an iPod, computer, or set-top DVD player) to process.

Apple and a number of other companies offer a variety of tools to help you compress video into the iPod compatible H.264 and MPEG-4 codecs. (Codec stands for compression/decompression. As a producer you use a codec to compress your media into a more efficient delivery format; your audience then uses the same type of codec, for example H.264 or MPEG-4, to decompress it and play it back.) Apple also offers some great online tutorials for both Mac and PC users.

More important than your ability to use a particular tool, however, is your knowledge of the underlying process and how you can make it work for you as a digital filmmaker. Not all framings and compositions compress equally well. Shots with lots camera movement and detailed backgrounds are often harder to compress than stable shots against a solid color background or a background that contains less fine detail. Close up shots of people’s faces, which I wrote about in my last posting, not only hold up well at small screen sizes but they generally compress well too. (For more information, I cover compression and codecs extensively in DV Filmmaking: From Start to Finish).

Because shots compress differently, and different parts of a video file may require different bit rate settings to play back smoothly, it’s important to use the manual settings in your compression tool, and to experiment: compress your media at a few different compression levels, and then test each on your target playback platform to see which settings work best. If you’re distributing your work as a podcast, test your compression by playing the video back on an iPod to be sure it looks the way you want it to before you make it public for the whole world to see. It takes some work, but the results are worth it when your video looks like a million bucks.