As you evaluate cameras for shooting DSLR video, one of the most important factors is the format or codec the camera records in. Different manufacturers use different file formats (and may offer different choices within their own product lines). It is essential that you understand the impact of codec, let’s compare how video compression relates to still compression.
When shooting stills, you can often choose to shoot raw, TIFF, or JPEG. Raw files allow for superior latitude during post-processing since raw images are not permanently processed by the camera. At the other end of the spectrum, JPEG files “bake” a lot of information into the file, making it necessary to get the shot close to perfect at the time of capture.
Your approach to exposure and white balance when shooting video should be similar to shooting JPEGs. Accuracy in-camera matters. It’s most important to avoid overexposure, since there’s no way to rescue blown highlights. But it’s also important not to underexpose, since making a video brighter also introduces noise.
How about raw?
In the world of high-end video, raw video exists, but is often cost-prohibitive. It also adds significant increases in budget due to additional processing time and equipment requirements. When shooting video on video-enabled DSLRs, the video signal is heavily compressed in order to write the data to the compact flash or secure digital memory card. Some of these formats are more heavily compressed than others, so you need to think about the effect of each on overall image quality.
H.264 and AVCHD
H.264 is a video compression standard that’s based on the MPEG-4 standard. This is a widely popular video format used for a variety of devices including portable media players, Blu-ray Discs, and even television sets. H.264 is a modern codec that offers remarkable footage with little loss of quality and minimal compression artifacts.
H.264 has become a standard for high-quality web compression, but recently it’s also being utilized for high-quality acquisition. Currently, Canon cameras like the 5D MKII, 7D, and the Rebel T1i all record H.264 formatted video. Newer Nikon cameras like the D7000 and D5100 are also using the newer format.
AVCHD is a very close cousin of the H.264 codec. Essentially, AVCHD is a brand name for H.264 encoding. Offering the same benefits as H.264, AVCHD is marketed by Panasonic and Sony, and video DSLRs like the Panasonic Lumix GH1 can use AVCHD to record HD video.
Note that AVCHD has a complex folder structure for the video file and for associated metadata, as shown in Figure 1. While the video file itself can play without all the other stuff, some software applications will reject an AVCHD file if the directory structure has been altered.
For years Motion JPEG has been a stalwart compression scheme. Motion JPEG is the oldest compression scheme that modern video-enabled DSLRs use. It has the lowest overall image quality but also allows for the most recording time since it is so heavily compressed.
Besides small file size, another benefit of shooting Motion JPEG is that it is natively supported by several video-editing applications. Unlike H.264 and AVCHD, it is not very processor intensive, making it less demanding on your computer during postproduction. Currently, older Nikon video-enabled DSLRs record using the Motion JPEG codec.
This is an older format that has quickly fallen out of favor. If your camera only supports this recording format, you are required to use it, and it may provide suitable quality. If purchasing a new camera, this format should be avoided due to its quality limitations.
Other format characteristics
Your capture format has a number of other important characteristics that you need to understand when specifying and setting up for video capture. These include frame size, frame rate, and compression. We have written a thorough description of these characteristics in the Format section of the website.
I originally wrote this article originally for DPBestflow.