🌀 JPEG to AVIF Converter

Frequently Asked Technical Questions

Question 1: What is AVIF, and why is it more efficient than JPEG?

Answer: AVIF (AV1 Image File Format) stores a single still frame using the AV1 video codec's compression tools — the same codec used by modern video streaming. AV1's intra-frame prediction, larger transform blocks, and more sophisticated entropy coding allow it to represent photographic detail with meaningfully fewer bytes than JPEG's much older DCT-based approach from 1992, typically 30-50% smaller at visually comparable quality.

Question 2: How does the native browser encoding path work?

Answer: This tool calls canvas.toBlob(callback, 'image/avif', quality). Per the Canvas specification, browsers are only required to support PNG — additional formats like AVIF are optional. Chromium-based browsers from version 124 onward implement genuine AVIF encoding here. Because unsupported formats silently fall back to PNG, this tool explicitly checks that the returned Blob's type is actually "image/avif" before trusting the native path, rather than assuming success.

Question 3: What happens when native encoding isn't available?

Answer: The tool dynamically imports a WebAssembly build of libavif (the reference AV1 image codec, via the open-source jSquash project) directly from a CDN, extracts the canvas's raw pixel data with getImageData(), and passes it to the WASM encoder's encode() function, which returns a genuine AVIF binary as an ArrayBuffer. This all happens locally in your browser — the WASM module never phones home with your image data.

Question 4: Does the WASM encoder require special server configuration?

Answer: No. Modern libavif WASM builds detect whether the page is running in a cross-origin-isolated context (which enables multi-threaded encoding via SharedArrayBuffer) and automatically fall back to a single-threaded build when it isn't — which is the case on typical static hosting without special COOP/COEP headers. Encoding is simply somewhat slower in single-threaded mode; it still works correctly.

Question 5: Why does the tool offer a "Force WASM" option if native encoding is faster?

Answer: The native and WASM encoders are different implementations and can occasionally produce very slightly different output for the same quality setting. Forcing the WASM path guarantees byte-for-byte consistent behavior across every browser and machine, which is useful for testing or when reproducibility matters more than speed.

Question 6: How is the quality percentage mapped for each engine?

Answer: For the native path, the percentage is passed directly as the Canvas API's 0–1 quality parameter. For the WASM path, the same percentage is passed to libavif's encoder options; internally, libavif translates this into its own quantizer scale. Both aim for equivalent perceptual quality at a given percentage, though exact output byte sizes can differ slightly between the two engines at identical settings.

Question 7: How does the resize feature choose the output dimensions?

Answer: The tool compares the image's longest side against your selected limit. If the image is already smaller, it's left untouched. Otherwise, both dimensions are scaled down by the same factor to fit within the limit, preserving the original aspect ratio exactly, and the result is drawn onto a canvas sized to the new dimensions with high-quality smoothing enabled before encoding.

Question 8: Why is AVIF encoding slower than JPEG or WebP encoding?

Answer: AV1's compression tools — including multiple prediction modes, larger and more varied transform block sizes, and computationally heavier entropy coding — trade encoding speed for better compression efficiency. This is a deliberate, well-known property of the AV1 codec across both video and still-image use, not specific to this tool.

Question 9: Does converting to AVIF preserve transparency?

Answer: AVIF fully supports an alpha channel, but source JPEG files never contain transparency in the first place — JPEG has no alpha channel — so there is nothing to preserve when converting from JPEG specifically. If you need to convert a transparent PNG to AVIF, that would go through a separate PNG-based pipeline.

Question 10: Can this tool encode 10-bit or 12-bit AVIF for HDR content?

Answer: No. The HTML5 Canvas 2D API works with standard 8-bit-per-channel image data, so pixel data extracted from a canvas is always 8-bit, regardless of the encoder used. True high-bit-depth AVIF encoding would require a specialized HDR-aware pipeline outside the scope of a Canvas-based converter.

Question 11: Why batch multiple files into a ZIP instead of individual downloads?

Answer: Browsers can prompt for confirmation or block rapid successive downloads when many files are triggered at once. Bundling everything into a single ZIP with JSZip avoids that friction and gives you one clean download containing every converted AVIF file.

Question 12: Is the AVIF output identical every time I convert the same image?

Answer: With the same engine, settings, and browser version, output is deterministic and reproducible. Switching between the native and WASM engines, or using a different browser or browser version, can produce byte-different (though visually equivalent) results, since each is an independent AV1 encoder implementation.