Find what consumes the bytes
Measure the source and identify its largest components. Choose lossless optimization or a deliberate lossy target.
Compression & Quality
A working understanding of lossless and lossy compression, quality metrics, and delivery budgets.
4 guides in this collection
START HERE
Lossy vs. lossless compression in plain English
Lossless coding removes repetition; lossy coding also removes information judged less important to perception.
9 min read →02
Smaller, not worse
03How to choose an image quality setting
A quality slider is not a percentage of retained information and is not standardized across encoders.
10 min read →Smaller, not worse
04A practical file-size budget for faster websites
Performance improves when each asset has a budget before design and export decisions become fixed.
10 min read →Smaller, not worse
PSNR, SSIM, and VMAF: what quality scores miss
Metrics help compare encodes, but no single score understands every texture, viewing condition, or business need.
8 min read →How to approach compression & quality
Compression decisions should compare perceptual quality, structural fidelity, decode support, and transfer cost - Not size alone.
Measure the source before moving a quality slider. Lossless compression removes repetition while preserving the original information; lossy compression trades some detail for a smaller output. Compare at least two settings against a real file-size or transfer budget, inspect the most difficult image, audio, or video passage, and keep a high-quality master so later delivery files are not repeatedly re-encoded.
Set a quality budget before a size target
Encode two representative settings. Compare artifacts and size in the real delivery context.
Compression comparison protocol
- The result meets the actual transfer or storage budget.
- Artifacts are checked in the hardest visual or audible passage.
- Dimensions, duration, and structure were not reduced accidentally.
- The uncompressed or highest-quality master remains available.