Let the delivery brief set the boundaries
Before comparing camera bodies, establish the destination. A short intended for theatrical projection, a vertical social release and a visual-effects test have different framing and finishing needs. Ask about resolution, aspect ratio, frame rate, colour space, sound requirements and any required versioning. These decisions shape storage, monitoring and post-production long before a camera is powered on.
More recorded pixels can support reframing or stabilization, but they also increase processing and storage demands. High frame rates need more light at a comparable shutter angle and may restrict available recording modes. A useful camera is therefore one that meets the shot requirements reliably, not simply one with the largest number on its specification sheet.
Pair the sensor with the lens and the movement
Sensor size changes the field of view produced by a particular focal length. It does not, by itself, determine whether an image looks cinematic. Lens coverage must encompass the selected sensor area, while the mount must provide the correct physical and electronic connection. Confirm these details for the exact camera mode; some modes use a smaller sensor region.
Compare focus behavior, minimum focus distance, distortion and flare in a recorded test. A lens that breathes noticeably during a focus pull may complicate a composite. An extremely shallow depth of field can separate a performer from the background, but also makes focus less forgiving. For an effects-heavy shot, a little more depth may preserve tracking detail and reduce difficult edge work.
Camera movement introduces another set of dependencies. Rolling-shutter sensors read the image progressively, so fast pans or vibration can bend vertical lines. A gimbal does not remove that readout behavior. Test the intended movement with the intended lens and support, especially when the shot will require a digital camera match.
Expose for a usable recording, not a reassuring screen
A bright monitor is not an exposure measurement. Use the camera's waveform, false-colour tool or other documented exposure aids, and understand whether the display includes a viewing transform. Log recordings deliberately look low in contrast before a suitable transform is applied; that appearance is not a reason to add arbitrary contrast or underexpose the source.
Shutter angle links exposure duration to frame rate. At a 180-degree shutter, each frame is exposed for half of its frame interval: at 24 frames per second, that is approximately 1/48 second. Different settings alter motion blur and may interact with practical lights. Test mains-powered and dimmed fixtures at the actual shooting settings rather than assuming a familiar shutter value eliminates flicker.
ARRI LogC, Sony S-Log3 and Canon Log 3 are examples of camera-specific log encodings, not interchangeable looks. Their recommended exposure practices and transforms differ. Read the documentation for the exact model, recording mode and firmware. A generic lookup table cannot recover highlights that were clipped during capture.
Choose a recording format the whole team can handle
Resolution, bit depth, chroma subsampling and compression describe different properties. Greater bit depth can make substantial tonal adjustments more resilient. Higher chroma resolution can help fine colour edges in keying. Neither guarantees a clean result if the source is noisy, poorly lit or heavily compressed.
Apple ProRes and Avid DNxHR are common intermediate codec families; H.264 and H.265 often prioritize smaller files but can demand more decoding work, depending on the profile and available hardware. RAW formats preserve a different level of sensor-related flexibility, with support and controls that vary by implementation. Test a sample in the actual editing and finishing applications, including audio and metadata.
End the shoot with verified media
Use recording media approved for the selected data rate, not merely a card with a promising headline speed. Estimate capacity from the recording mode and expected shooting time. Allow room for additional takes and the time required to offload them. A production that can record quickly but cannot copy safely has only moved its bottleneck.
Preserve original folder structures and create independent, verified copies before reusing a card. Checksum verification compares data rather than relying on a successful-looking drag-and-drop operation. Keep a camera report with clip identifiers, frame rate, lens information, sound references and any unusual settings. These notes become useful evidence when an editor or effects artist encounters an unexpected shot.
Bench test: record one representative shot, copy it, synchronize sound, make a proxy, relink the original and export a short finished file. A successful round trip is better evidence than a compatible logo on a box.
For model-specific guidance, consult the ARRI learning and support resources and the documentation supplied by your camera manufacturer. Manufacturer guidance explains operating limits; your test establishes whether those limits suit the production.