Variable Frame Rate Check

How a variable frame rate check saves an afternoon

Drop the clip and read one line. Variable, constant, or the container did not say.

The reason to do it now rather than later is simple. Discovering the problem after you have cut forty clips means re-syncing forty clips. Discovering it before you start means converting once and forgetting about it.

Nothing is uploaded. We read the container in your browser, so a fifty-gigabyte project file never leaves your drive to answer a one-word question.

Is my video VFR or CFR, and why does it matter?

A constant frame rate file gives every frame the same duration. A variable one does not — the rate rises and falls while recording, so some frames sit on screen longer than others.

Playback handles that fine, because a player reads each frame’s own timestamp. Editing software does not. It assumes even spacing and builds a timeline on that assumption, and every frame that broke it adds a small error.

Those errors accumulate in one direction, which is the part people find counter-intuitive. Nothing looks wrong for the first minute. By minute ten the dialogue no longer matches the mouths.

Fixing audio out of sync in Premiere, properly

The symptom is distinctive. Audio that lines up at the start and slides progressively out over a long clip is a timing problem, not a sync-point problem, and nudging the track will not fix it because the offset keeps growing.

Premiere has a preserve-audio-sync path that adds or drops frames to compensate, which helps and is not a cure. Resolve tends to hold the audio steady and let the picture slip, which shows up as pops and stutters instead. They fail differently rather than one being safe.

The real fix is to transcode to a constant rate before the footage enters your project. Our video format converter will do it, and so will any encoder you already trust. Pick a generous bitrate and the quality cost is negligible next to the time saved.

Why your screen recording frame rate problem is not a bug

Screen recorders write variable footage on purpose, and so do phones.

A recorder skips frames when nothing on screen changed, because storing an identical frame is wasted space. A phone drops the rate when the sensor struggles in low light so it can hold the exposure longer. Both decisions produce smaller files that play perfectly.

OBS, QuickTime screen capture, phone cameras and most browser-based recorders all do this. It only becomes a problem when an editor arrives with a different set of assumptions. Worth knowing before you blame the software.

Converting VFR to constant frame rate without making it worse

Order matters more than settings here.

Convert first, before trimming, compressing or colour work. Every later step inherits the timing of whatever it was handed, so a compressor fed variable footage normally hands back variable footage with a new name on it.

Choose the target rate deliberately. Matching your timeline is usually right; taking the maximum from the source is usually safer than the average. And be careful with 29.97 against 30 — mixing those causes its own slow drift that people mistake for a variable-rate problem, which is covered alongside the other numbers in checking whether a file is H.264 or H.265.

Trimming is not a fix. A stream copy preserves the original timing exactly, so a trim without re-encoding gives you a shorter variable file rather than a constant one.

When the container refuses to answer

Sometimes there is no mode recorded at all, and we say so rather than picking the reassuring option.

Some muxers write neither a mode nor a minimum and maximum, which leaves nothing to infer from. Browser-recorded WebM does this routinely, and so do a few phone recorders. An empty field means the file did not say, not that the file is constant.

It is worth remembering what these fields are. A muxer writes what it believes, and nothing verifies it afterwards, which is why whether video metadata can be faked is a fair question to ask of any field on this page.

Our advice in that case is unglamorous: if you are already seeing drift, treat the clip as variable and convert it. The conversion is cheap and the alternative is a second afternoon lost to the same symptom. The full container field list is in what video metadata is, and the video metadata viewer shows every field this page deliberately hides.

Run the check before you build the timeline

A variable frame rate check costs two seconds and belongs at the top of an import routine, next to backing the card up. Drop each new clip in, read the line, convert the ones that need it.

Do that and the drift never happens. Skip it and you will meet it at the worst possible moment, which is the reason this page exists at all. While you are looking at the file, our codec and bitrate checker covers the other fields that decide whether footage plays cleanly on your machine.