Car interiors are deceptively difficult to render. Most of the materials aren't especially dramatic on their own. We're usually working with dark polymers, textiles, leather or leather alternatives, painted surfaces, and relatively small amounts of metal. The realism comes from the differences between them.
If every dark surface has roughly the same roughness and texture frequency, the entire interior starts looking like it's made from one material. Getting those subtle differences right is one of the biggest things we can do to make an automotive interior feel convincing. The problem is that an interior can easily contain dozens of material variations, which is exactly the kind of work we built twirl to speed up.
Automotive plastic needs more than a flat shader
Plastic probably deserves more attention than almost any other interior material. Real automotive plastics aren't just gray shaders. Their grain changes how reflections break up across the dashboard, while differences in gloss can make two similarly colored polymers feel completely different.
Adobe's automotive material work included more than 50 procedural plastic grains based on physical references, including organic, fine matte, and geometric structures. Adobe's automotive interior material overview is a good reference for how much visual variation can exist inside what we casually call plastic.
For a convincing PBR version, much of that difference happens in roughness and normal information rather than base color.
In twirl, we can generate a complete seamless PBR material from that description, apply it to the IP, and decide whether the grain and finish work before spending time developing it further. We can then generate a slightly coarser grain for the lower IP or a softer matte version for the door and compare all three in the same scene.
Leather needs variation at several scales
Leather has a different set of problems. There is broad variation across the material, smaller grain and pore structure, and larger deformation created when the material is wrapped around upholstery. Stitching and seams add another level of detail again, and those features shouldn't all live at the same texture frequency.
Adobe approached its automotive leather materials by combining scanned physical skins with procedural systems for grain, seams, and other controls. For early visualization, we don't necessarily need that complexity. We do need the grain scale and reflection response to feel believable.
A useful twirl prompt could be warm graphite automotive leather, fine natural grain, subtle irregular pores, low sheen, clean new upholstery. We can generate the seamless surface in twirl and handle seams, stitching, and larger seat construction separately in the 3D application.
Textiles are about structure
A woven textile has direction. Fibers cross one another, the weave creates repeating structure, and those fibers interact with light differently depending on the viewing angle. At a distance, much of that detail collapses into a relatively simple surface. Up close, it becomes obvious.
This makes scale especially important. If we're creating a wide interior render, we may only need enough weave structure to break up the highlights. If we're making a CMF close-up, individual fibers and weave details become much more important.
This is another place where generating a few directions is useful. Instead of trying to force one generic fabric texture to work, we can describe the textile we're actually designing around, such as light gray recycled automotive textile, tight technical weave, subtle darker fibers, matte finish or charcoal woven seat fabric, large visible weave, alternating black and graphite fibers, soft surface.
Those are meaningfully different materials, and we can generate both without manually constructing two separate texture sets.
Metals need direction and restraint
Metal accents often occupy a relatively small amount of the interior, which makes them disproportionately important. Brushed aluminum needs a brushing direction that makes sense with the component. Bead-blasted metal needs a very different reflection response. Satin anodized surfaces shouldn't suddenly look like chrome.
For more complex automotive finishes, especially exterior paints and specialized coatings, we'd move toward more sophisticated material systems when accuracy matters. VRED supports workflows including MDL and measured materials for this kind of work. Autodesk's VRED material documentation covers those higher-fidelity options.
For concept work, though, a generated PBR material can get us surprisingly far because the first question is usually whether the direction belongs on the vehicle at all.
Stop judging materials only on spheres
A material sphere is useful for seeing whether a surface works technically. The vehicle is where we find out whether it works as CMF. Once we've generated the surfaces, we want to get them onto the actual interior quickly.
Can we distinguish the lower IP plastic from the upper IP? Does the textile insert feel softer than the surrounding polymer? Does the metal trim catch enough light to define the form without becoming distracting? These are the comparisons that matter more than whether a single material looks impressive in isolation.
This is why we built twirl around complete PBR materials rather than material reference images. The output is meant to leave twirl and go onto the actual design. Generate the surface, put it on the vehicle, and make the next decision from there.