Ordinary JPEG compression saves space by discarding image information.

Lossy compression trades numerical fidelity for storage and transmission efficiency. That trade can be valuable for photographs containing millions of pixels, where a substantial reduction in file size may produce little visible difference at the intended viewing size.

A JPEG quality control changes how aggressively the encoder simplifies image information. Lower settings usually reduce file size and increase visible damage, while higher settings usually retain more detail. Quality scales are not standardized across every application, so matching numbers do not guarantee matching results.

The trade becomes more obvious around sharp text, high contrast edges, and fine patterns. Compression can create ringing, block boundaries, or softened detail. These artifacts explain why a format that works well for a photograph may be a poor choice for a screenshot or logo.

The JPEG standard dates to 1992, and its name comes from the Joint Photographic Experts Group. It became widely used for digital photographs because comparatively small files were easier to store, transmit, and display across computers and networks with limited resources.

In the common baseline workflow, an encoder separates the image into components and processes blocks containing eight by eight samples. A mathematical transform describes each block in terms of spatial frequencies. Quantization then rounds those transformed values, discarding precision that the decoder cannot recover. Further coding compresses the resulting representation efficiently.

Many encoders also reduce color detail through chroma subsampling. Human vision often notices fine brightness changes more readily than equally fine color changes, so storing color at lower resolution can save space with a relatively modest perceptual effect. Nevertheless, colorful text and crisp graphics can reveal that reduction clearly.

Not every operation involving a JPEG causes another quality loss. Simply copying or renaming the file leaves its image data unchanged. Reopening an image to view it is also harmless. The risk arises when software decodes the image and then saves it through another lossy encoding step, especially after edits.

The wording "ordinary JPEG" matters because the wider JPEG standards family includes lossless methods and other coding systems. The familiar photographic files produced by common camera and web workflows usually use lossy encoding. Even a maximum quality setting should not be treated as a universal guarantee of pixel perfect reconstruction.

A practical workflow keeps an original or a lossless editing master and produces JPEG delivery copies as needed. Judge those copies at their actual display size, including detailed areas and smooth gradients. This approach avoids repeatedly editing an already compressed delivery file while still benefiting from efficient distribution.

JPEG’s usefulness comes from a tradeoff, not a promise of exact reconstruction. PNG preserves supplied image samples, while ordinary photographic JPEG aims for compact files with acceptable visual changes. The right choice therefore depends on the image and intended workflow, rather than a universal ranking of one format as better.