Photo Trends

Verified against ChatGPT · 2026-09-10

1986 Diwali night family photo

Rows of oil lamps, a sparkler and warm light coming from below. The one prompt built entirely around uplight.

ChatGPT (GPT Image)Nano Banana (Gemini 3.1 Flash Image)Gemini app

The prompt

Ready to copy — highlighted parts are example details you can swap.

A family on the front steps of a house on Diwali night in 1986, lit from below by a row of clay oil lamps with one sparkler burning
Use my uploaded photo. Count the people in it and render exactly that many — no more, no fewer. Anchor each face to its counterpart by left-to-right order. Do not merge, average, swap or blend any two identities. Keep each face exactly as it is — same features, skin tone, age and natural asymmetry — and preserve the age gaps. Warm lamp light may fall on our faces from below, but every face must stay clearly readable and no feature may be reshaped, narrowed, enlarged or lightened by the lighting. No slimming, smoothing or beautifying. Change only hair, clothes, jewellery, pose and surroundings. Put us on the front steps and doorway of an Indian house on Diwali night in 1986: neatly done festival styling for each age, oiled and combed, crown volume on the younger adults, a flower pinned for one. New festival clothes in deep reds, golds and magentas — silk sarees with zari borders, kurta-pyjama, a child in a stiff new outfit that does not quite fit — with gold jewellery, bangles and a bindi. One of us holds a lit sparkler at arm's length, another a brass plate with a small lamp. We stand and sit together on the steps, some crouched beside a row of small clay diyas along the step edge, one child mid-laugh, most looking toward the camera. Around us rows of lit clay oil lamps along the steps and parapet, a marigold toran across the doorway, a rangoli in coloured powder on the ground and dark night beyond. The dominant light comes from below: the row of lamp flames lighting the undersides of chins, the lower cheeks and the fronts of the clothes, and throwing soft shadows upward. The sparkler adds a small hard white-hot point that blows out slightly and casts a second set of shadows, and a dim warm bulb in the doorway lifts the background just enough to read. Beyond the lamps the night falls to near-black. 35mm film at night without flash, 1986: coarse grain, heavy warm-orange cast, strong halation around every flame, motion blur in the sparkler trail and one moving hand, dense blacks. No LED or electric fairy lights. Horizontal 4:3.

Why this works

Uplight is the rarest lighting direction in any image model’s training data, which is why asking for it produces something that does not look like other AI output. The instruction has to be specific about consequences — light on the undersides of chins and lower cheeks, shadows thrown upward — because "lit by diyas" alone returns a normally-lit scene with some lamps in it. Uplight also distorts faces more than any other direction, hence the explicit clause that it may not reshape a feature.

What you get back

A dark, warm, coarse-grained night frame lit from below by lamp flames, with a blown-out sparkler and upward shadows. Verify faces are not distorted by the uplight.

Verified against

ChatGPT GPT Image · 2026-09-10

Nano Banana / Gemini 3.1 Flash Image Gemini app · 2026-09-10

Changelog

  • 2026-09-09 Written for the September 2026 80s photo trend.
  • 2026-09-10 Rewritten as a single copy-and-go paragraph with no fill-ins, and paired with an example image.

Need this built into your business?

If a prompt isn't enough — what Scult builds, built and maintained for you — that's Scult's day job.

EXPLORE WHAT SCULT BUILDS
All Photo Trends prompts

Check your AI visibility

One URL in, a 0–100 score and the exact fixes out.

RUN THE CHECK

Browse all the tools

15 tools across six categories
13 of them never send your data anywhere

Free · No signup · No trial clock

SEE THE DIRECTORY