In very general terms, this is what happens inside your camera:
- Light falling on your sensor is focused by tiny lenses on the photosite.
- This light passes through a color filter array (All the cameras in this guide have Bayer sensors) to separate Red, Green and Blue information.
- The sensor converts light into an analog electrical signal that is sampled to form a digital signal.
- The sampling might also include compression, conversion and other ‘data management’.
- The digital signal is passed to the circuitry for wrapping, transcoding, compression, and other proprietary activities. Being digital, all these channels of data should have the same source. On cameras that output raw files, the debayering step is omitted.
- The data is funneled through various ‘pipelines’ for different kinds of recording or ingest. Internal media records the signal meant specifically for it.
- External signals are usually encoded and packaged to a standard like SMPTE or HDMI. Some use proprietary technology like T-Link, SR Memory, etc, because existing standards don’t have provisions for these special data streams.
- External signals are transmitted via standard protocols so that the systems reading it can understand the signals.
- The signal that leaves the camera is ready to be read by an external recording device.
Why would you want an external recorder?
If the signal recorded internally by the camera is not of sufficient quality to satisfy the requirements of your production, you might be able to get a stronger signal with an external recorder.
Notice, I didn’t say ‘better’, but ‘stronger’. Very rarely have I seen an uncompressed HDMI or HD-SDI feed that is noticeably different visually from the internal recording on a camera. This is especially true of the lower-end prosumer camera models in our list.
The following is a rant. I feel it’s important enough to share but if you’re not in the mood, you can skip it.
Do you honestly think there is a hidden super-video in your camera that the manufacturer doesn’t want you to see? If that were the case, they wouldn’t put SDI or HDMI connectors on their cameras. They give you the option of uncompressed video to stream to monitors or to record to external recorders in a codec of your choice, not a ‘better’ codec.
Yes, this is a controversial statement. Let me explain:
Many people assume and propagate the notion that having an uncompressed or higher bit rate version of the original (like an intermediary codec) actually makes their footage ‘better’ in some vague sense.
If a sauce doesn’t taste good, adding more of it isn’t going to make it taste any better.
Your footage can never be better than the system that created it. A lot of compromises happen in the transduction, sampling and handling of data within the camera, and this is where the men are separated from the boys.
Most videographers don’t know anything about sampling and signal processing. Why should they? But these same people assume they know about these disciplines based on what manufacturers tell them. If somebody shouts out something long enough, people start believing in it.
Rule of thumb: The only way to know for sure whether you need an external recorder is to test its signal with the signal recorded in camera.
I wish I could tell you there’s an easier way, but there isn’t. I have seen too many ‘anomalies’ between claims and actual real-world performances to believe what is written in marketing brochures.
Be very conservative when you tread this ground. Don’t assume your camera and recorder will give you the best video just because somebody says it’s so. There is no free lunch.
There are situations where the stronger signal (not better, let me remind you again) is required for data manipulation down the workflow chain, like if you need to heavily color grade or ‘pull’ information (like chroma keying, match moving, etc) from your footage.
Rule of thumb: Only use an external recorder if you anticipate heavy (more than one level of) data manipulation. Most internal codecs are quite capable of at least one level of data manipulation.
What do I mean by this? Think of codecs as paper. Some kinds of paper can be folded only once. If folded again, they might tear. The tougher kinds of paper can be folded many times before they tear. Both papers are equally capable of being written on.
This is exactly how codecs work. Two codecs might look the same visually, but one may be ten times the data rate than the other. The codec with the higher data rate could be said (but this is not universally true) to have the resilience to be folded (manipulated) many times. An external recorder gives you this option. It does not offer you any ‘visual’ advantage. I hope I’ve made this clear.
If you want to know further about sampling and how digital signals are generated, please read the fictional love story Driving Miss Digital.