Image file size estimator
Model pixel payload, packing, alignment, padding, headers and multiple frames; estimate storage only from your supplied compression assumption.
A QUICK WALKTHROUGH
How to use this tool
- Name scenarios and enter integer dimensions; choose channels/bits or explicit bytes/pixel; declare packing, alignment, padding, frames and headers; optionally enter a ratio or observed bytes/pixel, calculate and export.
Model and scope
Pixel bits = width×height×channels×bits, or width×height×bytes/pixel×8. Row packing rounds each row upward, adds row padding then aligns each row. Whole packing rounds the frame upward, adds height×row padding then aligns the frame. Raw bytes = frames×stored payload + frames×frame header + file header. Ratio estimate compresses stored payload only; observed bytes/pixel applies to pixels only. Both estimates round the combined payload upward and add unchanged headers. Zero dimensions give zero pixel payload but retain declared headers. For a zero-area frame, pixel packing and row padding are omitted; declared headers remain.
Result
Original decimal strings become exact fractions. Display decimals stop after 12 places (…); all byte boundaries use exact integer rounding.
How to use
Name scenarios and enter integer dimensions; choose channels/bits or explicit bytes/pixel; declare packing, alignment, padding, frames and headers; optionally enter a ratio or observed bytes/pixel, calculate and export.
GOOD TO KNOW
Common questions
Does this predict an actual encoded file?
No. There are no universal PNG/JPEG/WebP presets. A user ratio or observed bytes/pixel is an assumption, not a prediction of a codec. All processing is local; no image is read or uploaded.