Guide
Conversion can re-encode pixels, change compression settings, and remove information the output format cannot keep.
01
JPG and common WebP settings use lossy compression. The encoder groups and approximates visual information, prioritizing patterns people are less likely to notice. At lower quality, this becomes visible as blockiness, ringing around edges, smeared texture, or banding in skies and gradients.
02
Conversion cannot restore texture or color detail removed by an earlier compression step. Saving a low-quality JPG as PNG only stores the already damaged pixels losslessly. Repeatedly opening and exporting JPG copies can cause generation loss, so return to the original source whenever possible.
03
Resizing requires the browser to calculate a new pixel grid. Reducing dimensions usually looks clean at the target size, but fine lines can soften. Enlarging cannot invent real detail and often makes blur more obvious. Resize once from the best available source instead of scaling through several intermediate copies.
04
Browser Canvas workflows may not preserve every embedded color profile or EXIF field. Images can therefore look slightly different in color-managed professional software, and orientation or camera information may disappear. For everyday web use this is often acceptable; for print and archival work, keep a color-managed master.
05
Start with the original, resize only when necessary, and use a moderate quality setting. Compare before and after at the final display size, especially around text, faces, and high-contrast edges. Stop lowering quality as soon as artifacts become distracting; a slightly larger clean file is often the better result.