Loading…

Comparing image sliders

About 15 minutes · no sign-up · anonymous

An image slider takes one photo and one instruction — "make it winter", say — and turns that edit into something you can dial in by degrees. At the far left you get the original photo untouched; at the far right, the edit applied in full; and in between, partial versions of it.

You'll see two systems side by side with the original photo in the middle for reference, and one shared slider that moves both at once. Drag it back and forth, then answer two questions about the pair.

What makes a slider good

1. It makes the right edit. By the far right the instruction has actually been carried out, and carried out convincingly.
2. It keeps the photo. Everything the instruction didn't ask about — the place, the people, the objects, the framing — should still be recognisably the same photograph.
3. It looks good. No smearing, warping, strange textures or obvious artefacts along the way.
4. It only moves forwards. Dragging right should always add more of the edit, never partly undo it, and never wander off to something else and back.
5. It moves evenly. The change is spread across the whole range rather than arriving all at once — no dead zones, no jumps.

Question 1 — better overall

Taking all five together, which side is the better slider? There's no formula — weigh them however seems right to you.

Question 2 — more evenly spread

Then we ask about the last property on its own, because it's the easiest one to overlook.

A slider moves evenly when each step brings roughly the same amount of change as the step before. Two things spoil that. A dead zone is a stretch where you drag and almost nothing happens. A jump is the opposite — one small movement changes everything at once. They usually come together: a long flat stretch, then the whole edit landing in a single step.

Two different photos, each with its own instruction. In the first the fire builds a little at a time, so every step of the slider is worth about the same. In the second the first four frames are almost identical — you could drag three quarters of the way and see nothing happen — and then the entire edit arrives in the final step.

The two questions are related: a slider that jumps is usually a worse slider. But they can come apart, and a side can reach the better final result while moving far less evenly. Answer each question on its own terms.

Practical

Under each side you'll see five snapshots taken evenly across the slider's range, which makes dead zones and jumps easy to spot. You have 15 minutes; it stops on its own when time is up, and you always get to finish the pair you're on. Which system is on which side is randomised, and they're never named.

Time's up — thank you

Your comparisons are saved. You can close this tab.

Nothing identifying was collected — no name, no email, no account. Responses are stored under a random code generated in your browser.

Something went wrong

Could not load the study.

Please reload the page. If it keeps happening, let us know — nothing already answered is lost.