
原文
GPT Image 2 · en
Scene:
A square documentary image composed like a premium archaeology magazine feature. Use rose-red sandstone, warm desert amber, cool canyon shadow, and fine airborne dust. Frame the Petra Treasury facade emerging at the end of a narrow siq, with sand drifts at the base and high canyon walls framing the reveal with crisp depth layering, restrained contrast, and subtle atmospheric depth so the scene feels excavated, observed, and historically grounded rather than theatrical or game-like. Keep the light directional but believable, as if captured during a careful field visit.
Subject:
Center the image on the Petra Treasury facade emerging at the end of a narrow siq, with sand drifts at the base and high canyon walls framing the reveal. Include culturally accurate surface logic, region-specific setting, and believable human-scale cues in stairs, doorways, cut blocks, relief edges, or water depth. Shapes, carvings, masonry joints, and site wear should feel studied from real archaeology, with no generic ancient-world mashup and no ornamental filler borrowed from another civilization.
Important details:
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. Emphasize rose sandstone, wind-shaped sand, chisel-cut columns, shadowed canyon walls, and softly weathered Nabataean ornament. Keep inscriptions, painted registers, symbolic motifs, or carved bands visually plausible but not readable. Preserve a museum-grade balance between beauty and evidence: light should reveal relief depth, edge wear, repairs, sediment traces, and excavation context without turning the site into fantasy spectacle. The composition should feel like a definitive reference frame that later variations can inherit exactly.
Use case:
Nabatean Petra Rediscovered series study 1/9.
Constraints:
No readable text, no watermark, no logos, preserve exact 1:1 aspect ratio, no modern barriers, no souvenir stalls, no fantasy caravans, no electric lighting, no tourists, no anachronistic props, no collage layout, no exaggerated fantasy lighting.使い方
- パラメータを含む元のプロンプト全体をコピーします。
- 被写体、文字、構図を自分の要件に置き換えます。
- 対応する外部ツールで使用し、利用条件と結果を確認します。
開始前の確認
- 参考画像の要否を確認してください。互換モデルの表示は、そのモデルでの生成実績を意味しません。
- 名前、数値、画像内の文字を確認します。モデルの版で結果は変わります。





