PNG

PNG stores raster images with lossless compression and several color and transparency representations.

A pixel format with several useful representations

A PNG may use a palette of indexed colors or store grayscale or truecolor samples directly. Transparency can be expressed through alpha samples or a transparency chunk. Encoding an existing pixel grid as PNG does not itself need a lossy quality setting.

PNG is useful for screenshots, interface graphics and transparent artwork. Photographic copies can be considerably larger than a lossy delivery format. APNG extends the container with animation data; a converter that processes one decoded image does not automatically preserve that sequence.

Text chunks and color information

Text can be stored in tEXt, zTXt and iTXt chunks. PNG can carry color information, an ICC profile and an eXIf chunk. XMP is commonly carried as an iTXt packet under the Adobe XMP keyword convention.

Still PNG support in PRVKIT

Convert, Compress and Resize work with supported still PNG. PNG compression uses the browser encoder; a smaller file or a requested byte target is not guaranteed at unchanged dimensions. Current image workflows reject detected animations rather than exporting every frame.

SVG to PNG uses a restricted image-context renderer. Redact Image offers solid regions and verifies the encoded result. Image Inspector can inspect supported text, EXIF, XMP and ICC groups; some metadata types and thumbnails remain outside its coverage.

A new PNG is not an unchanged original

Changing size or rasterizing vectors creates a new pixel grid even if the PNG encoding is lossless. For SVG input, choose the output resolution deliberately; the PNG no longer contains editable SVG geometry.

A PNG created by prvkit is a browser-encoded copy. Source EXIF, XMP and ICC are not copied by these conversion workflows. Inspect the generated file when its embedded information matters.

Lossless describes one stage of the workflow

A PNG encoder preserves the pixel samples it is given without a lossy quality tradeoff. That does not guarantee that those samples are identical to every earlier source. Resizing, color conversion, compositing and rasterizing vectors all happen before or around encoding. A JPEG converted to PNG keeps its decoded appearance, including earlier JPEG damage.

This distinction matters for screenshots and redaction. A clean pixel grid can be useful for small lettering and exact opaque covers. It does not mean that a PNG is an unchanged copy of a camera original or that every private structure has been eliminated. Keep the intended property specific: retained alpha, readable edges or verified marked pixels.

Transparency has more than two states

Alpha can represent partly transparent edges as well as fully visible and fully empty pixels. Anti-aliased lettering, soft shadows and hair can contain many intermediate values. A cutout that looks fine on a white background may reveal a pale fringe on black, so inspect it against contrasting colors.

A checkerboard is a viewer's explanation of transparency, not a requirement to paint squares into the output. An image can also contain an opaque white rectangle despite being stored in PNG. Choosing PNG does not automatically remove that rectangle or identify which part of the artwork you intended as background.

Flattening to JPEG replaces alpha with a selected background. That can be appropriate for a fixed layout, but it removes the ability to reuse transparent pixels over another color. Keep an alpha-supporting master when the same graphic will appear in several designs.

File size depends strongly on the picture

A screenshot containing broad flat areas can compress quite differently from a noisy full-color photograph. PNG has no photographic quality slider in PRVKIT's focused workflow. If the file is too large, ask whether it needs all its pixels or whether another format is acceptable, rather than expecting a lower quality percentage to solve it.

For a strict upload ceiling, the compressor measures output and reports when the chosen constraints cannot meet it. A transparent graphic may legitimately need fewer dimensions or a different accepted representation. Do not sacrifice a required alpha channel merely because an opaque copy happens to be smaller.

PNG's specification also permits representations that a particular tool does not preserve fully. High-depth samples, color information or animation-related structures require explicit support. A still-image browser workflow should not claim universal preservation simply because PNG can express those properties in another application.

Use the source that matches the next task

For an editable logo, retain vector artwork when it is available and permitted. A large PNG remains a raster grid; enlarging it does not restore paths or independent text. SVG to PNG is useful when a recipient needs pixels, but its output should not replace the vector source for future resizing or editing.

For a layered design, save an editable project alongside the flattened PNG. A transparent export can be useful in another composition without retaining the original layer hierarchy. The destination determines whether the needed asset is an image, a vector document or a project with separate objects.

Verify the actual delivery copy

Open the downloaded file and check width, height, orientation and alpha behavior. Inspect fine edges at the intended use size and review any sensitive visible content. Source metadata may not have been copied during re-encoding, but filenames, logos, faces and text can still disclose information.

Use Image Inspector for supported embedded properties and Compare Images for decoded visual differences. A hash comparison answers byte equality instead. These observations are complementary; none turns an ordinary PNG into a universal certificate of privacy, authenticity or suitability for every screen and print workflow.