Compare the actual sizes in a batch video compression run
Admin ·
A shared compression setup can produce different results for clips with different durations, motion and audio. Run two practice clips together and inspect the measured size for each output.
Set two output budgets in one MP4 queue
Choose Load example queue. Video 1 starts at 744,408 bytes with a 960 × 540 picture. Video 2 starts at 104,317 bytes with a 320 × 180 picture and a mono tone.
Keep MP4 and Target file size. Set Default target size in MB to 0.20 and the target for video 2 to 0.10. Set Maximum video long edge to 480 px and Maximum frame rate to 15 fps. Leave the first audio track enabled.
MP4, 480 px long edge, 15 fpsFFmpeg documentation on transcoding, filtering and stream selection
Compare byte counts, dimensions and duration
Choose Compress queue. Video 1 is reduced to 480 × 270, while video 2 stays 320 × 180. The frame-rate cap keeps up to 15 frames per second. The source durations are four and two seconds. Preview both outputs separately to check the picture and audio.
Read the actual byte count and target result in each row. The encoder uses a bitrate budget plus the selected frame size, frame rate and audio settings. When the first attempt exceeds the target, the workspace can try one smaller budget. Container overhead and scene complexity still affect each result.
FFmpeg documentation on transcoding, filtering and stream selection
Continue when one target is too small
Set the target for video 1 to 0.05 MB with audio enabled. Its four-second timeline leaves too little budget for the minimum video bitrate. The row reports the problem and video 2 stays ready.
Download results ZIP includes completed clips. Compare the encoded byte counts for the completed rows before packing them. Duplicate source names receive distinct row numbers in the download names.
Common questions
Does the tool increase the frame size of a smaller clip?
A maximum long edge caps large dimensions. The workspace keeps smaller source dimensions, rounded to supported even sizes when encoding requires it.
Why might a result be larger than its original?
A source can already be efficiently encoded. Changing codec, container, audio or quality may add bytes even if the result remains within its target.