🐳PhotoShop prompt 📸

Here's the thing that makes photographic prompting work, and it's more mechanical than people expect: name a physical piece of equipment. Quality words and emotional modifiers barely move modern checkpoints, but "point-and-shoot film camera" or "golden hour lighting" drops the model out of beauty-default mode faster than anything. That's exactly the trick PhotographyParameters industrializes. It's a prompt-builder from 🐳 Pond Nodes (comfy_Pond_Nodes) that takes your base prompt and appends a properly-formed photography parameter string assembled from nine dropdown categories.

The display name is "🐳PhotoShop prompt 📸," which is a weird name - there's no Photoshop here, and no API. It's a text-builder that assembles camera language the way the community learned to write it.

How it works

Each dropdown is a list of English/中文 paired options - e.g. Golden Hour/黄金时刻, Bird's Eye View/鸟瞰, Canon EOS R5/佳能EOS R5. When you run it, the node splits each selection into its English and Chinese halves, and depending on output_language (English, 中文, or Both/双语) it builds a comma-joined parameter list: "Golden Hour lighting, Bird's Eye View angle, … shot with Canon EOS R5, skin texture: porcelain smooth." Your base input_text goes at the front, so the output reads like a hand-written prompt. There's real range here - 52 lighting options, 49 camera angles, 66 aesthetic styles, 61 moods, 49 camera models, 52 quality tags, and a monster 111-option perspective list covering mirror reflections, water reflections, and views through everything from telescopes to keyholes.

The inputs that matter

• input_text - your base prompt ("a portrait of a woman", "a neon-lit street at night"…).
• output_language - English, 中文, or both.
• The nine category dropdowns - lighting, camera angle, aesthetic style, emotional mood, camera model, image quality, perspective type, skin texture. Each defaults to None/无, so you only add what you want.

One output: text, the assembled prompt string. Wire it into your CLIP Text Encode or a prompt-input node. It's batch-friendly and needs nothing but a string.

Installing it

Standard for the pack. Manager search comfy_Pond_Nodes, or:

cd ComfyUI/custom_nodes
git clone https://github.com/Pondowner857/comfy_Pond_Nodes
cd comfy_Pond_Nodes
pip install -r requirements.txt

Restart, and it's under 🐳Pond/text. Pure Python, no deps beyond the standard library - safe to skip the pack's heavy requirements.txt if this is the only node you're after.

The gotcha

Big dropdown lists are tempting, and that's the trap: pile on eight categories and you get a prompt so dense the model loses the subject. The photographers' rule applies - a couple of well-chosen parameters beat a wall of them. Pick one lighting, one angle, one camera, and let the skin texture and perspective stay None unless the shot actually needs them. Also, this node appends; it won't clean your base text or balance weights, so feed it a clean subject line. And if a term comes out awkward ("Broad Light/宽光" → "Broad Light lighting"), it's because the option strings are written as display labels, not final grammar - a small price for not remembering the syntax yourself.