Video Looks Washed Out After Export? Check the Colour Range

You finish editing a short film for a ceramics studio. On the timeline, the potter’s charcoal apron still shows its weave and a pale glaze catches the window light. In the exported file, the apron looks grey and the glaze looks dull. It is tempting to turn up the contrast and export again. But if the export or player is interpreting the video’s levels differently from the editor, that extra contrast can damage a picture that was right in the first place.

A range mismatch is one possible explanation. It is worth testing because it changes where you look for a fix. A useful review separates the picture you intended, the file you delivered and the version people actually see. Start with a short section that makes the difference easy to spot, then follow it through all three.

Why the same numbers can make different pictures

Video files encode picture information as numbers. The range tells the playback system which numbers correspond to reference black and white. In the 8-bit example in section 4 of ITU-R Recommendation BT.709-6, reference black is 16 and nominal peak white is 235 for luma, the signal that represents picture brightness. In a full-range 8-bit interpretation, the corresponding endpoints are 0 and 255. The BT.709 values here describe luma; the standard specifies separate values for the colour-difference channels.

Imagine a file with reference black at 16 and white at 235. If a playback path treats those values as full range without remapping them, 16 sits above its black and 235 below its white. The result can look flatter than intended. The reverse mismatch can make a picture look too harsh: a path expecting limited range may map dark and bright full-range values beyond the display endpoints, losing shadow or highlight detail when they are clipped.

That does not make one range better than the other. A limited-range file interpreted correctly can still show deep blacks and bright whites. What matters is that the encoding, its signalling and the playback interpretation agree. And range is only one part of colour handling. A difference in transfer function, an HDR-to-SDR conversion or a display setting can also change how a picture looks. The appearance alone is a clue, not a diagnosis.

Compare the same frame at each handover

Pick a short section with a deep shadow that should retain detail, a light surface with visible texture and something in the middle, such as a face or a neutral wall. In the ceramics example, use the apron, a bright patch on the glaze and the potter’s hands. Mark the time and write down what should remain visible. “I can see two folds in the apron” is a better check than “the blacks look good”.

• Timeline: Check the marked frame in the editor. If the apron is already grey there, inspect how the original footage is interpreted and whether the grade produced the effect. The export may simply be carrying it through.
• Local export: Compare the rendered file with the timeline on the same display. If you can, bring the export into a clean comparison timeline with matching colour settings and no extra grade. Compare the same frame and use a waveform, if available, to check whether the dark and light areas moved. Treat the waveform as evidence of how that application decoded the file.
• Delivery copy: Once processing is complete, check the same moment in a private preview or test upload if the destination offers one. If the local file is right and the delivery copy is different, investigate that handover before changing the master.

If two players on the same display disagree about the local file, record which applications you used. The disagreement gives you a lead about playback or interpretation, but it does not tell you which version is correct. Testing the same material at each handover is also useful for timing problems, as our guide to checking video audio drift shows.

Change one thing, then make a short export

Start with the destination’s delivery requirements if they are available. In your editor or encoding tool, check how the source is being read and what the export preset is expected to produce. The FFmpeg scale filter documentation, for example, distinguishes the input sample range from the output sample range. That distinction helps even if you use a different editor: what the tool believes it received and what it is producing are two separate settings.

Also separate changing the range label from converting the sample values. If the values are already correct and the label is wrong, correcting the interpretation or label may solve the problem. If the destination needs a different range, the samples may need to be remapped and the result labelled accordingly. A new label on its own will not restore texture that was clipped in an earlier conversion. If you do not know which happened, keep the original and test on copies.

Export a few seconds with one deliberate change. Keep the grade, source, frame and viewing conditions the same. Review the shadows and highlights again, and check the destination copy if that is where the problem appeared. Keep a note of the setting you changed and the result you observed. That record makes the next video easier to troubleshoot and gives a collaborator something more specific than “try more contrast”.

A good export review ends with a creative decision you can defend. If the apron still has the weave you wanted and the glaze still holds its texture, the handover has preserved the details that mattered. If it has not, you know which stage to investigate before touching the grade.

Sources

• ITU-R, Recommendation BT.709-6, June 2015, section 4.
• FFmpeg, Scale filter documentation, input and output range options.