Working with digital video can be a pain. The challenge is that almost every piece of video is expressed through a different codec (which is just short for “compression/decompression”). The codec is the mathematical code that shrinks (compresses) all the full frames of video data into something small enough to fit onto a piece of magnetic tape or a data card, or to swim through the internet without clogging up the pipes; then, on the other end, it unshrinks (decompresses) it in a quick way so it can be watched.
Each codec is designed for a different use: Blu-Ray, cell phones,internet video, camcorders, etc. Most people have heard of Flash – it’s one codec; others may have heard of H.264 – another new and rather popular codec. All said, getting source video from one codec to another (also known as transcoding) can cause a lot of headaches trying to make everything as good as possible in as small a file as possible.
Recently, we found ourselves back in the encoding world finishing up the digital files for use in the Seek Social Justice. We have also been putting together a new participant download series for Modern Parables, which has brought on its own set of issues. After doing this on-and-off for an number of years, we’ve learned a few things that may help someone new to the world of encoding:
- Always start with an uncompressed file. Yes, this is very big, but the problem with going from one codec to another is that sometimes things get messed up when making the translation. By starting with no codec, you know you have all the data necessary for the codec to shrink it down. This will normally give you the best image.
- Not all software is the same, so buy something that will give you good results. To be completely fair, hardware encoders such as the Sony Vizaro are far better than software solutions; but most people doing digital video today don’t have access to hardware solutions and have to rely on software. Some of the better known brands in this field are Telestream’s Episode , Sorenson Squeeze, and Apple’s Compressor (which comes in the Final Cut Studio). There’s a useful article in the Jan 09 edition of Millimeter that compares most of the major sub-$1000 encoders and gives results. Both Telestream and Sorenson do well across the board. But what happens if Compressor came with your Final Cut Suite and you want to use it?
- Learn to tweak your own system through trial and error: take time to experiment. We use Apple’s Compressor simply because we’ve worked at it enough to know how to tweak it. The newest Compressor 3.5 looks promising, but the reality is that if you know something about what you have (and it is good enough: see above point), you can get nice-looking videos. Take a minute (ok, an hour or two) to read the manual. It will explain a lot. Don’t just use presets; tweak and compare side to side. For instance, today we wanted to see what a 854×480 H.264 MOV looked like at a bitrate of 1250kbs vs 1000kbs, as well as at a 120 keyframe vs. 24 keyframe. So we put all the settings on one file (uncompressed) and then looked at them side by side playing at the same time. The results were pretty obvious (and sometimes surprising: 120 keyframes looked better than 24 at the same bitrate when the frame was doubled in size – which happens when people make it full-screen to watch).
- Read forums if you have the stomach for it. You can learn a lot on forums like CreativeCOW and DVXUser (there are tons more out there), you just have to wade through a bunch of comments that have nothing to do with your problem. Then suddenly you’ll hit someone with the exact issue and, voila: you have your answer. This is a fairly non-linear way of learning, but in a world that’s constantly shifting, it’s one of the only ways to keep up.
Encoding and transcoding are the necessary evils of the digital video world and will continue to change as technology moves forward; if you find you have to live in that world, the best thing you can do is roll up your sleeves and starting learning it.
