Gear by Criteria, Not Model Number — The Gear Paradox, a Camera Checklist That Outlives Any Phone, and the Buying Rules for Light, Sound, Filters, Gimbals and Storage

The failure mode is analysis paralysis - comparing tools, assuming you need more of them than you do, and never starting. The industry name for it is GAS, and the diagnosis is uncomfortable: people with it enjoy buying gear more than using it. Waiting for perfect gear is described as the most dangerous habit a creator can develop. The prescription is a time allocation rather than a purchase decision, plus the release valve that makes acting easy - no gear decision is permanent, and you can switch once you actually know what you need. Keep a wish list; practise with what is in your hand.

The default recommendation is the phone already in your pocket - anything under about ten years old can produce content people succeed with. The decision to buy a dedicated body is framed by motivation rather than by an assumed quality ceiling: if you care about the output and about being efficient, stay on the phone; if you are a technology enthusiast who enjoys the process, spend the same money on a camera. The honest functional difference is interchangeable lenses, which buy apertures and looks a phone cannot reach. On phones he ranks iPhone and Samsung first, with Google, Huawei and Sony acceptable but not first picks - and tells viewers not to read their own phone's absence from the list as a reason to upgrade.

Six questions, applicable to a phone that does not exist yet. How many built-in lenses and how good - noting the primary lens is usually built to a higher spec than the ultra-wide or telephoto beside it. Sensor size, which drives low-light performance, editing latitude and background blur. The fixed aperture, where f1.2 to f2.0 is the desirable range and you cannot change it after purchase. Which resolution and frame-rate combinations actually exist. Which stabilisation - digital crops the image, optical moves lens elements, sensor-shift moves the sensor and reacts 'up to five times faster' than optical. And how good it looks straight out of camera, where HDR and a strong auto mode save real editing time.

Colour rendering index first, at 95 or above, because a light that renders colour badly cannot be corrected in post. Colour temperature next - 5000 to 5600K is daylight-neutral white, around 3200K is warm tungsten, and a bi-colour light buys the ability to move between them at the cost of maximum output, because it never fires all its bulbs at once. Output in lumens, scaled to the size of the scene rather than to ambition. Diffusion, which is what turns hard unflattering light into soft flattering light and is frequently not in the box. And portability. Underneath all of it he makes the argument that learning lighting beats upgrading cameras - and rules out ring lights for cinematic work, because flat even light removes the shadow that creates depth.

Proximity and room reflectivity decide the result before brand does. Get within six to twelve inches; hard surfaces echo, and rugs and acoustic panels absorb. Then match the microphone to the job. A lavalier is the versatile default because 'you can basically use lav mics in any situation'. A directional or shotgun rejects background noise well but degrades quickly as the subject moves away, so it suits vlogging or a fixed studio. Ambience and sound effects are a different problem entirely - record them on site if you can, use foley if you cannot, and fall back to a library. And the connector trap: TRRS is phones and computers, TRS is cameras and professional gear, and mixing them without an adapter produces no audio at all.

Filters: one variable ND covering roughly two to five stops handles most situations, budget $50 to $300 by brand tier, and always buy a thread size ABOVE your largest lens with step-up rings rather than one that is too small. A circular polariser is a different job - cutting reflections off skin, water, glass and metal rather than darkening uniformly - and many variable NDs double as one because of their two rotating elements. Gimbals: the rule that matters is matching payload capacity to body PLUS lens, not body alone, and price mostly buys payload rather than smoothness; for a talking-head YouTuber they are a nice-to-have, for commercial and wedding work they are not. Storage: SSD for anything being edited, cheap HDD for archive, because an HDD 'can become the editing bottleneck on an otherwise powerful computer'. Every brand fails eventually, so discount the one-star crash reviews.

A prompter solves two real problems - on-camera confidence, and batching, since you can script many videos and read them in a row. But he is unusually honest that reading well is itself a skill, and that some people get worse rather than better. The hardware advice is about mounting rather than price: adjustable height and a quick-release plate beat a screw-thread-only mount. The display advice is counterintuitive and good: use a phone in portrait rather than an iPad, because a narrower screen means less visible left-to-right eye movement. And the app that matters is the one that listens and scrolls at your pace rather than at a fixed speed.