How Should You Compress an Online Course Video?

Why course videos end up "fine but unreadable"

Viewers forgive compression artifacts in natural scenes like faces and landscapes, but notice them immediately on regular text, code indentation, and thin chart lines. At the same strength, a landscape looks normal while small code in a screen recording can blur a comma into a full stop.

The good news: screen recordings are mostly static, and video encoders handle that well — frames that barely change need almost no extra data. So even on High quality, a screen recording is often smaller than live footage of the same length.

Settings by content

Frame-rate options only lower, never raise. A 60fps screen recording at 30fps still shows smooth cursor movement and saves a good amount; slide narration that holds each slide for a long time looks the same at 15fps.

For text-heavy content, use compatibility mode

tinyclip uses the browser's hardware encoder first, which is much faster. The trade-off is slightly lower picture quality than software x264 at the same file size (in the tool's own testing, roughly 1 to 1.5 dB lower PSNR), and small text and thin lines show it most.

Checking "Compatibility mode" in the advanced settings switches to ffmpeg's software x264, where CRF and encoder-speed settings fully apply and text edges come out cleaner — at the cost of being much slower. Compress one lesson the default way first; if the smallest text is not clear, use compatibility mode for that kind of course.

Batch-processing a whole course

1. Put every video in the course into one folder (chapter subfolders are fine) and drag the folder in.
2. Videos queue and process one at a time, so several large files never compete for memory.
3. Each video shows live progress; you can do other things, but keep the tab open.
4. When everything is done, download a ZIP; for very large courses, downloading files one by one is more reliable.