Choosing Cameras and Lighting for Single-Speaker Webinars

While most video producers are sold on professional video cameras for concerts, large seminars, and similar events, are they overkill for single-speaker webinars? What type of camera is best for these productions, which are becoming ever more popular? And what kind and how much lighting do you need to pull them off?

These are the issues that I’ll address in this article.

By way of background, since I’ve been producing a lot of webinars lately, I decided to take a good hard look at this subject, comparing the quality of three webcams and two camcorders. Once I finished my research, it seemed natural to host a webinar on the subject. But you can skim through the results much faster in this article, so I recommend you start here.

Cameras and Tests

The three webcams I compared were the embedded webcam on my HP EliteBook 8760w; the Logitech HD Webcam C615, an external unit you can buy for under $40; and the Logitech Broadcaster, an external unit that costs around $200 (Figure 1, below). The C615 runs on both Mac and Windows computers, while the Broadcaster is Mac-only.

Figure 1. The Logitech Broadcaster on the left; Logitech HD Webcam C615 on the right. Click the image to see it at full size.

As you can see in Figure 1, the Broadcaster is a much larger unit, with a length of about 4.6”, which is plenty of room for a longer lens and larger sensor than the C615. The unit also had a 1/8” microphone input and a small video light on the front, the latter of which came in handy during my tests. Though rated at 1080p, compared to the Broadcaster’s 720p, the C615’s optics are obviously smaller, which seemed to be a factor in overall quality.

Capture connections varied by webcam, with the embedded unit on my HP notebook captured internally, while the C615 attached via USB. The Broadcaster sent the signal to my Mac Pro via WiFi, a cool feature that lets you use the unit to stream directly to Ustream without a computer.

The two camcorders I tested were the Canon Vixia HFS10, a consumer-class camcorder that cost around $950 when new, and the Panasonic AG-HMC150, an older prosumer AVCHD camcorder still available for around $2,600. The Vixia has a video light up front which I used during some tests, while the Panasonic is a 3CCD model that I use frequently in my production work (Figure 2, below). I captured a component signal from both camcorders because the BlackMagic Decklink capture card I was using refused to recognize HDMI from either camcorder, and captured audio using the C615 webcam. I obviously ignored audio quality in this review, and just used the audio on the C615 in the test videos that I posted to the internet.

Figure 2. The Panasonic HG HMC150 on the left, Vixia on the right. Click the image to see it at full size.

I ran two series of two tests each, all in Telestream Wirecast 5.0, outputting a 720p 5Mbps H.264 stream in all tests. The first test simulated a webinar with the video within a small picture-in-picture window in a Powerpoint slide (Figure 3, below), the other a full-screen video. I tested under two lighting conditions: optimal, which had two softboxes on me as dual key lights, and two on the back wall eliminating shadows; and typical, which simply used the overhead fluorescent lights in my office.

Figure 3. This was the picture-in-picture webinar configuration. Click the image to see it at full size.

In all tests, I adjusted the available configuration options to optimize video quality. For the webcams, this meant boosting gain significantly, because the bright wall caused a backlight condition that darkened my face when running in auto mode. With the Vixia, I switched into backlight mode for the optimal tests, and switched back to full auto for the typical tests. With the AG HMC150, I optimized exposure and focus manually with optimal lighting, and ran in full auto mode for the typical tests.

In the webinar, I presented the results from worst to best, first under optimal lighting conditions, then under typical lighting. I’ll do the same here.