Channels split. Signal lost.

No. The pixel manipulation runs entirely in your browser. Your photo never touches our servers — we don't run an image pipeline at all.

We split the source photo into red, green, and blue channels. Then we shift the red channel by some pixels in one direction, blue by the opposite, and keep green steady. Where the photo is colorful, the channels separate visibly — the chromatic-aberration look. Where it's neutral grey, no visible shift.

Subtle uses a 4-pixel shift — it reads as a high-end-camera lens flaw, almost invisible until you look closely. Anaglyph uses a 22-pixel shift — like the red/blue 3D-glasses effect, blatantly visible. Lens sits in between.

Adds horizontal random offsets to bands of pixels, simulating tracking errors on a VHS tape. Click 'Reshuffle bands' for a fresh random pattern. Pair with CRT scanlines and the result reads as a frame from a busted CRT TV.

Channel-shifting moves R and B channels, leaving G in place — so colored edges sharpen but neutral edges can soften slightly. This is the same softness real cheap-lens chromatic aberration produces. For sharper results, use less intensity.

Not in v1 — R goes one way, B goes the opposite, G stays still. Direction-angle slider rotates the R/B axis. For asymmetric shifts (R one way, B another, G a third), you'd need to run the photo through this tool multiple times.