In the video/film industry there are generally several ways to do things and they can get you a similar result. Likewise, with compressing video for web you can make a video a lot of different sizes (resolution), but in a technical and real world workflow there are right resolutions.
To make a long story short, mpeg (and most all modern compression codecs), use a shorthand to display image data (from 1′s, and 0′s) called macroblocks. The largest of these macroblocks is 16 pixels wide by 16 pixels tall. These macroblocks are very visible if you ever use your HDTV via an antenna to watch a broadcast that is not coming in all of the way. You don’t see scrambled snow like the old days, you see the actual macroblocks scrambled up.
Ok enough boring background, in real world scenarios the reason using the following resolution below are the right video resolutions are two-fold. This is because both the width and height are divisible by the size of a large macroblock (16×16). And here are those reasons:
- It is easiest for a processer to decode the compressed video
- It is easiest for the codec to create a clean encoded video
All of this theory and practice comes from the fact that there are 8 bits in a byte since the creation of the personal computer. You remember the time you bought a fancy 500GB hard drive and it wasn’t really 500GB? Same reason. OK, so now you know why these resolutions are ideal – here they are:
- 768×432 (my favorite for general use)
- 1280×720 (perfect for youtube uploads)
Here is an example of what I mean by divisible by 16, using 768×432 as an example. 768/16=48 and 432/16=27 – both 48 and 27 are whole numbers and not a fraction, which means you would not be compromising part of a large macroblock to create that resolution video.
Hopefully this helps some people down the road. There of course are times when you get silly requirements that you can’t do anything about, but when you can choose – always choose wisely.