Image Compression Calculator: File Size Estimator for JPEG, PNG and RAW

Lossy compression permanently discards some image data to achieve smaller file sizes. JPEG and WebP lossy are lossy formats: once saved, the removed data cannot be recovered and repeated re-saving compounds the quality loss. Lossless compression (PNG, WebP lossless, TIFF) uses mathematical algorithms to reduce file size without discarding any pixel data, so the image can be perfectly reconstructed from the compressed file. The trade-off is that lossless files are larger than lossy files at comparable apparent quality.

For web use, JPEG quality 80 to 85 offers a good balance of visual quality and file size, with most viewers unable to distinguish it from quality 95 at normal viewing distances and screen sizes. Quality 90 to 95 is suitable for professional output and images that may be further processed or re-saved. Quality 70 and below is generally only appropriate for thumbnails or cases where bandwidth is severely constrained. Quality 100 is rarely useful: it produces very large files with minimal quality benefit over quality 90 to 95.

JPEG is almost always preferable for photographs, as its lossy compression reduces file sizes dramatically (often 10 to 30 times) with minimal visible quality loss on typical photographic content. PNG is better for images with sharp edges, flat colour areas, transparency, or text, such as logos, screenshots, diagrams, and illustrations. PNG lossless compression achieves much better ratios on these types of images than on photographic content. For photographs that need transparency, consider WebP lossy instead of PNG.

WebP is an image format developed by Google, offering both lossy and lossless compression. WebP lossy typically produces files about 25 to 35 percent smaller than JPEG at equivalent visual quality, while WebP lossless is typically about 26 percent smaller than PNG. WebP also supports transparency (unlike JPEG) and animation. It is supported in all modern browsers and is widely recommended for web use. The main limitation is that some older image editing software and operating systems do not yet support WebP natively.

Bit depth determines how many tonal levels each colour channel can record. An 8-bit channel has 256 levels per channel, while a 16-bit channel has 65,536. Moving from 8-bit to 16-bit doubles the raw (uncompressed) data size, since each pixel requires twice as many bytes. Professional cameras often capture 12-bit or 14-bit RAW files, which fall between these extremes. Higher bit depth is important for post-processing flexibility (avoiding banding in gradients during heavy editing) but the extra data is largely discarded when exporting to 8-bit JPEG for web or print distribution.