Triple-Camera Interview Setup: X-H2S, X-T5 & Canon R5C

Corporate interviews are often treated as production afterthoughts - a talking head, a bland backdrop, something to cut around. Done well, they're some of the most demanding single-day shoots you can do: fixed lighting, no second takes, long recording times, and a premium on getting it right in-camera when there's no time to grade.

This is the setup I used for a recent triple-camera interview, with the Fujifilm X-H2S as A-camera, the X-T5 as B-camera, and a Canon R5C as additional coverage. The subjects were executives; the deliverable was broadcast-quality content. Here's how I approached it.

Filtration On close-ups, I used 1/8 Black Mist filters to soften skin tones without losing detail in the eyes. I'd avoid this on wide shots - the diffusion affects the whole frame and tends to look odd on wider compositions. Neither Fujifilm nor the R5C has an internal ND, so I also had variable NDs to hand for aperture control, keeping the close-ups at f/1.4 for shallow depth of field without blowing the exposure.

Capture settings Knowing this would be a long interview - 20 minutes plus - I shot H.265 Long GOP 4:2:2 at 200Mbps rather than chasing the highest possible codec. It's a sensible compromise: good image quality, manageable file sizes, and no overheating issues. At those settings, 30 minutes of footage sits around 45GB. The X-H2S recorded to CFExpress Type B; the X-T5 to V60 II SD cards.

Colour I was tempted to shoot FLOG2 for maximum latitude, but with three cameras running and a tight turnaround, I needed a look I could deliver with minimal grading. I settled on Nostalgic Negative with custom tweaks - Highlights +1, Shadows -1, Colour -2, Sharpness -4 - dialled in to approximate a REC709 look comparable to Sony Cinetone or Canon Cinestyle. White balance was locked to a single Kelvin value matched to the lights. No auto, no drift between cameras.

I've never fully warmed to Eterna for this kind of work - it's too flat as a starting point when you need to move quickly in post. Nostalgic Negative gives you something that already has shape to it.

Autofocus Face/Eye detection on AUTO, AF+MF enabled for manual override. I run AF speed high (+5) and tracking sensitivity at +1 as a baseline, though in environments with multiple faces in frame it's worth checking for AF hunting before you commit - sometimes AREA mode is more stable than full face detection. Peaking on, red high, for any moment where I need to step in manually.

Monitoring When running an external monitor, make sure HDMI Output > Info Display is set to ON - without it, you'll get a compatibility error that wastes time you don't have. On-camera I keep the histogram live (more useful than zebras for judging exposure), plus Rec Frame Indicator so there's no ambiguity about which cameras are rolling.

Audio Record locally to the wireless pack if your system allows it - on-camera preamps are rarely good enough to be your primary source. I use RØDE microphones that record to the device itself, and treat the camera input as redundancy. If you're plugging directly into the camera, double-check you're in the MIC input rather than the headphone jack. It sounds obvious. It still catches people out. Set a limiter at -9 or -6dB and leave it.

The broader point A multi-camera interview setup is less about having the best gear and more about removing variables. Locked white balance, matched looks, clean audio, continuous recording - every decision you make on the day is one fewer problem to solve in the edit.