
Original document
GPT Image 2 · en
Scene:
Create a square reference image for Three-Dimensional Product UI Depth System. Build a reference anchor establishing the core system with a premium, production-ready finish. Use a polished modular board or campaign surface with clean spacing, premium game-UI finish, and clear focal hierarchy. The composition should feel intentional, clean, and reusable for gallery, research, campaign, or internal handoff contexts. Lighting must support readability first, with the main forms clearly separated from the background and enough breathing room for the system to stay legible at both hero scale and thumbnail scale.
Subject:
Center a coherent original game-art system built for premium production and marketing surfaces. For this image, focus on reference anchor. Keep the same system logic stable across screenshots, cards, banners, and overview boards. This image is the canonical anchor that later series images must follow. The viewer should immediately understand what role this surface plays inside a cohesive game-production asset pack.
Important details:
Lock framing, icon grammar, palette blocking, and surface hierarchy so the kit feels production-ready and consistent. Preserve the core silhouette, framing language, and material logic from the reference image while each variation changes function. Keep materials believable, edge hierarchy disciplined, and the visual language consistent with 3D UI Depth. Use abstract interface blocks or diagram accents when helpful, but avoid any readable language. The final result should feel like a deliberate production asset rather than a random illustration.
Use case:
Three-Dimensional Product UI Depth System series study 1/9, designed as reference anchor within a cohesive asset pack.
Constraints:
No readable text, no watermark, no logos, preserve exact 1:1 aspect ratio, no unrelated objects, no broken visual hierarchy, no malformed anatomy or broken perspective, no random decorative clutter.How to use
- Copy the complete original prompt, including any parameters.
- Replace the subject, text and composition with your own requirements.
- Use a compatible external tool; check its current access conditions and review the result.
Before you start
- Check whether a reference image is required. A compatible model label does not mean this example was generated with that model.
- Keep names, figures and visible text accurate; results can vary between model versions.





