
Oryginalny dokument
GPT Image 2 · en
Scene:
A square close study framed within the same site established by the reference anchor. Use a tight crop, shallow but believable depth of field, and raking light that reveals tactile evidence in surface relief, pigment, tool marks, moisture, or marine wear. The image should feel like a museum conservation photograph elevated to gallery quality.
Subject:
Focus on a materially rich portion of the same site: carved edges, painted plaster, fitted stone, root-wrapped masonry, submerged paving, or quarry-cut facial planes depending on the civilization. The composition should remain anchored to the original location and show enough surrounding context that viewers still recognize the parent site rather than mistaking it for an unrelated artifact.
Important details:
Preserve the exact site identity from the reference model while shifting emphasis to forensic surface evidence. Keep the monument recognizably Petra, with faceted rock-cut massing, canyon compression, and carved architectural order that feels cut directly from stone rather than assembled from blocks. Make micro-texture, fracture lines, pigment retention, water wear, plant interaction, dust, or chisel traces visually persuasive. Avoid decorative clutter. Every mark should feel like evidence of making, weathering, excavation, or conservation. Keep any symbolic marks non-legible and historically appropriate, and make the light reveal depth without flattening the material richness.
Use case:
Nabatean Petra Rediscovered series study 3/9.
Constraints:
No readable text, no watermark, no logos, preserve exact 1:1 aspect ratio, preserve exact site identity from the reference image, no modern barriers, no souvenir stalls, no fantasy caravans, no electric lighting, no isolated floating artifact, no studio backdrop, no fake macro abstraction.Jak używać
- Skopiuj cały oryginalny prompt wraz z parametrami.
- Dostosuj temat, tekst i kompozycję do swoich potrzeb.
- Użyj zgodnego narzędzia zewnętrznego; sprawdź warunki dostępu i wynik.
Przed rozpoczęciem
- Sprawdź, czy potrzebny jest obraz referencyjny. Zgodność nie potwierdza modelu użytego do przykładu.
- Sprawdź nazwy, liczby i napisy; wyniki zależą od wersji modelu.





