A director asks for consistent weathering across forty props, by Friday

Look development on a project of any size runs on reference that has to stay consistent across a team, not just legible to whoever gathered it. A prop artist needs the same rust reference the environment lead used on a nearby surface, and the honest answer to "where did that come from" is usually a folder somewhere on a shared drive that nobody has fully mapped since pre-production.

Where look-dev reference actually lives on a real project

A working game art team's reference is rarely one library; it is photography gathered for specific surfaces, concept art handed down from an art director, and screenshots kept for lighting or composition, spread across however many folders pre-production produced. Each artist knows their own corner of it. Almost nobody can search someone else's corner, which is exactly the reference that matters when a director wants a prop's weathering to match a wall three biomes away.

A surface, found by what it is rather than who gathered it

Registering a shared drive's existing folders as one workspace does not require reorganising anything an art director already set up. in:Props scopes a search to prop reference only, while tag:any:rusted-metal reaches matching photography across Environments, Props and Characters at once, regardless of which artist saved it or which biome it was gathered for.

type:image tag:any:weathered-wood rating:>=4    rated wood surface studies
in:Environments tag:any:volumetric-fog          scoped lighting reference
added:>2026-01 -tag:archived                    this milestone's active set

A term the parser does not recognise renders as free text rather than an error, so a half-remembered material name from a review still returns something to narrow down from.

Rating what actually shipped

Concept art and photo reference accumulate fast during pre-production, and not all of it survives contact with the final art direction. Rating a reference once it is confirmed as the basis for something that shipped — four stars for the rust study that became the prop's final texture, one star for an exploration the director passed on — turns rating:>=4 into the set worth reaching for on the next prop that needs the same surface, rather than a re-scroll through everything gathered during exploration.

Keeping concept art and photo reference searchable as one set

Concept art handed down by an art director and photography gathered independently by individual artists are different in origin but often answer the same question — "what does this material look like under this light" — and do not need separate systems to be searched together. Tag both by surface and lighting condition rather than only by source, and tag:any:weathered-wood returns whichever one actually matches, concept painting or photograph alike.

A material taxonomy that survives a change of project

A studio's look-dev reference tends to outlive any single title, and a folder structure built entirely around one project's biomes does not travel to the next one. Tagging by material and condition — rusted metal, weathered wood, volumetric fog — rather than filing only by project or level name means a new title can reuse a rust study gathered years earlier, found by tag:any:rusted-metal regardless of which shipped or cancelled project it was first collected for.

What happens to reference sent by an outsourcing partner

Not all look-dev material originates inside the studio — an outsourcing partner's concept batch, a licensor's approved reference pack, a location scout's photography sent as a folder of files. None of that needs special handling beyond dropping it into the shared workspace already registered; the watcher picks it up on its own, ready to tag alongside anything gathered in-house.

Presenting a milestone review without building a separate deck

An art director reviewing a milestone wants to see the reference behind a surface choice, not just the final render, and assembling that reference into a slide deck for every review is a small project of its own each time. A .coii-canvas skips that step: arrange the photography and concept art that justify a material decision on one board and present straight from it. The file stays in the LookDev folder afterward, so the next artist working a similar surface finds both the decision and the reasoning behind it in the same place, rather than in a deck buried in someone's inbox.

Handling scanned or hand-painted concept sketches

Not every piece of concept reference starts digital — a director's thumbnail sketch photographed on a whiteboard, a scanned page from a physical notebook, a print marked up by hand during a review. None of it needs converting before it is useful: a scan or a phone photo is just an image file, indexed, tagged and searchable the same as anything produced digitally once it is dropped into a registered folder.

Keeping reference organized as a project moves from pre-production into production

Reference gathered loosely during pre-production tends to get abandoned once production ramps up and a team is heads-down on delivery, which is exactly when a director's original intent is most likely to get lost. Tagging reference by surface and biome as it is gathered, rather than leaving it unfiled until someone has time to sort it, means the same search that worked during pre-production keeps working once forty artists are actively pulling from the same shared drive under deadline.

Comparing this to a library tool that copies files into itself

Eagle is a capable option for the same job and copies whatever is imported into its own managed library, which suits some teams because everything then lives in one place under the app's control. Coii Ref indexes a studio's existing shared-drive structure exactly where it sits instead, so a folder an art director organised during pre-production never needs duplicating into a second, app-managed copy that could drift out of sync with the original. The round-up of Eagle alternatives covers the wider field if a tool that takes ownership of the files is the actual preference.

Working across a personal machine and a studio workstation

Contract and freelance game artists often keep a personal reference archive at home and work on a studio-issued machine during an engagement, and the two do not need to be one workspace. Register the personal archive on a personal Mac and, separately, the studio's shared drive on the studio machine — each builds its own local index, with nothing to sync and no account tying the two together. Licensing covers three Macs on one purchase, which in practice covers a home machine, a laptop and a studio seat without paying twice.

What it costs and what it runs on

$19 once, three Macs, updates included, nothing that renews. The trial runs 30 days with every feature enabled, no card and no account required. It needs macOS 11 Big Sur or later, on Apple Silicon or Intel, as two native builds.

The app's entire outbound network activity is a licence check on Activate and a launch-time read of a server's Date header only, so a wound-back clock cannot extend a trial. No in-app updater means no update check, and there is no analytics or crash reporting.

Past day thirty of an unpaid trial, one category of action stops: growing the library, meaning registering a new folder, running the importer, or starting an index by hand. Search, tags, ratings, notes and the folder watcher all keep working permanently on what is already indexed, so a milestone does not stall because a trial ran out mid-project.

Keeping reference searchable across engine and pipeline changes

A studio's underlying engine or pipeline tends to change more often than its actual look-development needs do, and a reference archive built around one engine's asset browser does not survive the switch to a different one. Keeping look-dev photography and concept art as ordinary files in an ordinary folder, tagged by material and lighting condition rather than by which engine a project eventually shipped on, means the archive survives every pipeline migration a studio goes through — registering the same shared drive again after a tooling change costs one re-index, not a rebuild of years of gathered reference.

Why this matters more the longer a studio has been open

A studio's look-dev archive is worth more the longer it has been accumulating, because a new title rarely starts its reference from zero — it starts by reusing whatever the studio has already gathered on a similar surface or biome. A ten-year archive spread across a dozen shipped and cancelled projects, searchable by material rather than by which of those projects gathered it, is exactly the asset a folder structure alone stops being able to deliver once it gets that large.

Keeping location-scouting photography usable beyond one project

A location scout's photography, gathered for one project's environments, is often specific enough to be reused on an unrelated title years later — a particular kind of rust, a specific weathering pattern on stone — provided it survives being tagged by what it actually shows rather than only by the project it was shot for. Registering the scouting archive as its own long-lived workspace, separate from any single title's production folder, keeps it searchable across however many projects come after the one it was originally gathered for.

What onboarding a new hire actually needs

A new artist joining a project mid-production needs fast access to what the team has already decided about a look, not a verbal walkthrough of folder conventions from whoever has time to give one. A shared workspace tagged consistently from the start means a new hire's first search — tag:any:rusted-metal rating:>=4 on their own machine, the day they get access to the shared drive — returns the team's actual, rated conclusions rather than an unfiltered pile of everything anyone has ever saved.

What a rating record protects against on a long project

A project that runs for several years generates far more reference than any one artist can hold in memory, and without a rating record, later milestones tend to re-research surfaces the team already solved. Rating consistently at each milestone — not only at final ship — builds a record a search can trust months later, so a returning artist or a new hire pulled onto the project inherits the team's actual conclusions rather than the same unfiltered pile of concept art and photography the project started with.

Who this actually suits

A team whose look-dev reference is genuinely spread across more folders and more artists than any one person can hold in their head, who needs a surface or a lighting condition found by property rather than by remembering who gathered it. If the actual need leans toward an on-device tagger that suggests tags automatically, refern is the more capable app on that specific front. The concept artist page covers a related but earlier-stage job built around fast exploration, the 3D artist page covers texture and reference setup for an individual seat, and the VFX page covers a plate-and-reference workflow built around a different pipeline entirely.

Thirty days, every feature, no card and no account: download it and point it at the shared drive that has been accumulating look-dev reference the longest. Tag the surfaces that actually shipped first — that is the search that pays off the next time a director wants a prop to match a wall three biomes away.