Primer: Live-Switched Webcasting
Although it's possible to webcast with a webcam or smartphone and a streaming service provider, a professional live-switched webcast has more in common with a live TV broadcast than it does a kid with a smartphone. This article discusses the roles that must be filled in a live-switched webcast, the various features and types of video switchers, and a lot of the small details that are important considerations in the larger video switching and webcasting workflow.
Audio Inputs
Some video switchers have built-in, full-fledged audio mixing consoles with multiple audio inputs. Others allow a pair of audio inputs from a soundboard or single source, and some don’t even pass on embedded HDMI or HD-SDI audio and require the use of an external audio soundboard.
For streaming, the question of whether the webcast requires embedded audio or external audio is determined by the available inputs on the webcast encoder.
Embedded audio has the advantage of easy distribution amplification if additional video outputs are needed for the archive recording. The ability to monitor audio levels using volume unit (VU) meters and/or headphones is another consideration, and many video switchers, webcast encoders, and video converters that embed analog audio into a digital video output lack these capabilities. It is important to be able to visually monitor audio levels to ensure they are in the target broadcast range, and to listen to this output to ensure that no unwanted noise has entered into the workflow.
Frame Rates and Scaling
Especially when working with computer inputs, it is common to work with different frame and refresh rates, and some video switchers have the ability to internally convert frame rates and scale video inputs. When this isn’t available, webcast producers may need an external scaler or converter for noncompliant inputs.
Streaming, as with most modern video viewing, is done with progressive video, rather than interlaced video. Unfortunately, many video cameras and video switchers do not have a common progressive frame rate available from the video camera’s live video output. The video output is different from the available internal recording options on the video camera. This has resulted in 60i (interlaced) being the most common signal, rather than the 30p (progressive) that is typically webcast, although deinterlacing is not usually required because 60i on a native 30p or 60p CMOS sensor doesn’t show interlacing artifacts or scan lines the same way that a traditional 60i signal from a CCD sensor would.
Audio and Video Delay
The speed of light is much faster than the speed of sound, and this alone can account for audio that is out of sync with video, especially when the audio input is captured from microphones located away from the talent, which is the case for ambient microphones that record audience reactions and applause. Despite this, with live-switched video it is more common for the audio to be slightly ahead of the video. This is because audio delays that are introduced through audio equipment, especially with effects and processing, introduce less of a delay than does video processing. Video is more susceptible to delays because it is more common that converters, scalers, and signal processors are used. External scalers and converters, especially ones that convert an analog signal to a digital one, are the most likely to introduce a delay.
Using embedded audio on one of the camera inputs helps keep the video and audio in sync, but many producers prefer to embed the audio at the video switcher or video converter. Sometimes this requires an audio delay (and rarely a video delay). Some video switchers have built-in audio or video delay features; fewer video converters have this useful feature.
In terms of input formats and frame delays, HD-SDI is the lowest-latency video format with its 1 millisecond (ms) delay. HDMI lags a bit more with a 3ms delay, and some VGA to HDMI scaler/converters can introduce as much as 20ms of delay. The delay duration varies depending on manufacturer and equipment used, but in general, the minimal delay is one of the reasons that HD-SDI is the gold-standard video connection.
Preview Outputs
The technical director needs to see each video input as well as the preview and program outputs.This used to require a dedicated video monitor for each input that would be connected to a video loopthrough output from the video switcher.
But with the move to more on-location portable live video switches and large, inexpensive HDTVs, multiviewers have become more popular. A multiview output groups all the inputs and outputs on a single video display, usually on an HDMI output for compatibility with consumer HDTVs.
Auxiliary Outputs
The program output is the main mixdown video output from the video switcher, but similar to audio soundboards, video switchers can also have aux outputs that offer additional output options. An aux output is like having additional video switchers built into the to the main video switcher and can be used to provide an alternate video output for IMAG purposes or to provide additional critical video outputs from any of the inputs, preview, or program video feeds.
Keyers
Similar to layers in Photoshop, keyers allow the technical director to add layers and key out backgrounds. The upstream keyer or chroma key can be used for greenscreen work, where you key out the background of one image and overlay this image on a second background video or image. The downstream keyer is used for adding logos and text over the video.
Transitions
Similar to editing in an NLE, video switchers can offer a variety of transitions, but they can also be restricted to straight cuts. Common transitions include cut, mix or cross-dissolve, wipe, DVE, and stinger. It’s common to avoid using transitions when webcasting at low or aggressive bandwidths relative to the frame size because this might create too much change within the group of pictures (GOP) that delivery codecs use and cause excessive blocking or pixelation.