Video BasicsAug 9, 2026

Video editing terms, explained in plain English

Gain is not volume, a LUT is not a grade, and a container is not a codec. Here are the words that stop beginners, grouped by the part of the job where you actually hit them.

8 minute read

Satvik ShrivasFounder, FrameOS

The timeline, and the words for where everything sits

Everything happens on a timeline: a strip where time runs left to right and your media stacks up in tracks, sometimes called layers. A clip is any piece of media sitting on it, which is the first confusing thing you meet, because clip is also what everyone calls the finished short you post. The playhead is the vertical line marking the frame you are currently looking at. In and out points mark the section of a source file you want to use, and trimming drags a clip's edge to change where it starts or stops. The distinction that catches people is delete versus ripple delete: a plain delete leaves a gap where the clip was, while a ripple delete closes it and pulls everything after it earlier. Snapping makes edges click together. Nesting bundles a run of clips into one object you can move as a unit.

Cuts and transitions, which are not the same thing

A cut is simply the join between two shots, and most of an edit is cuts with nothing on them at all. A jump cut leaves the camera where it is and removes time from the middle, so the subject appears to hop, which is the standard move for tightening talking-head footage. A match cut hides the join by lining up shape or movement across it, a different idea people regularly mix up with the jump cut. A cutaway sends the viewer somewhere else briefly, and B-roll is the supporting footage you cut away to. Adobe describes J-cuts and L-cuts as split edits, where picture and sound change at different moments: in a J-cut the next scene's audio arrives first, in an L-cut the outgoing audio hangs over the new shot. Dissolves, fades and whip pans are transitions, and beginners reach for them far more than working editors do.

Audio: level, tone, and repair

Audio vocabulary splits three ways. Gain sets how hot a signal is going in, before anything else touches it; volume sets how loud the clip plays back. Reaching for volume when you needed gain is the most common early mistake, because it leaves a weak, noisy signal turned up. Normalising lifts a whole clip to a target level. Compression narrows the gap between the loudest and quietest parts, and a limiter is a wall peaks cannot pass. Perceived loudness gets measured in LUFS rather than decibels: the European Broadcasting Union's R 128 recommendation, which most loudness meters implement, sets broadcast programmes at minus 23 LUFS with true peaks no higher than minus 1 dBTP, while consumer platforms normalise playback to their own, louder targets. Room tone is a recording of your room doing nothing, used to patch gaps. Ducking dips music automatically under speech.

Colour: correction, grading, and the words in between

Colour work has two halves that beginners collapse into one. Correction is the technical pass: setting white balance so white reads as white, fixing exposure, and matching shots from different cameras so a cut does not flash. Grading is the creative pass that comes after, and it is what people mean by a look. Log is a flat, washed-out picture profile some cameras record in to protect detail in highlights and shadows, and it is meant to be graded rather than posted as-is. A LUT is a lookup table, a saved colour transformation you apply as a starting point, and it is not a grade in itself, which is why dropping a film-look LUT onto ungraded footage usually disappoints. Rec. 709 is the standard colour space for ordinary HD delivery. Scopes are the graphs that tell the truth when your screen lies.

Frames, resolution and speed

Frame rate is how many still frames make up a second, written as fps. YouTube's own encoding guidance lists 24, 25, 30, 48, 50 and 60 as the common rates and tells you to upload at the rate you recorded rather than converting first. Resolution is the pixel count, so 1080p means 1920 by 1080 and 4K is roughly four times that, and it is not the same thing as aspect ratio, which describes only the shape of the frame. Progressive means each frame is whole; interlaced, the older broadcast method, splits every frame into two half-images and is worth avoiding entirely. Real slow motion needs a high frame rate captured at the camera and then conformed to a slower timeline, so you cannot invent it afterwards without smearing. Time remapping is the tool for speed ramps within a single clip.

Aspect ratio, cropping and framing

Aspect ratio is width against height. 16:9 is the landscape default, and YouTube's help pages still describe it as the standard shape for watching on a computer, while 9:16 is the full-screen vertical used by TikTok, Reels and Shorts, 1:1 is square, and 4:5 is the taller portrait shape feeds allow in-line. Letterboxing means black bars above and below, pillarboxing means bars at the sides, and you get the wrong one when a landscape file is dropped into a vertical frame without reframing. Cropping throws pixels away and changes what is in shot; scaling only makes what is already there bigger or smaller. The safe zone is the middle of the frame you keep clear of the app's own buttons, username and caption. Headroom is the space above someone's head, and too much of it is the most common vertical framing error.

Export: containers, codecs and bitrate

Export is where the words get genuinely technical, and two of them cause most of the trouble. As Mozilla's developer documentation explains, a container is the file wrapper, such as MP4, MOV or MKV, while a codec is the compression method used on the streams inside it. H.264 is the safe default, H.265 and AV1 compress harder for the same picture, and ProRes is a large, lightly compressed format meant for editing rather than posting. Bitrate is how much data you spend per second: YouTube recommends roughly 8 Mbps for 1080p at standard frame rates and 12 Mbps at 48 to 60 fps, in an MP4 with H.264 High profile and AAC audio. Rendering is your editor building previews as you work, exporting is producing the finished file, and transcoding is converting between formats, which costs a little quality each time.

Captions and the AI-era vocabulary

The newest words are the ones nobody stops to explain. Burned-in captions are drawn permanently into the pixels; closed captions travel as a separate file, usually SRT or VTT, which a viewer can restyle or switch off. Subtitles assume you can hear and only carry the dialogue, while captions add speaker labels and sound. ASR is automatic speech recognition, the transcription engine behind every auto-caption button. Diarisation is the separate step that works out who spoke. Auto-reframe tracks the subject so a landscape shot can be rebuilt as vertical. Text-based editing means cutting the video by deleting words from its transcript. A virality or clip score is a model's guess at which moments will travel, and a guess is exactly what it is. Tools like FrameOS chain those steps together, which is mostly worth knowing so the labels on the buttons stop being mysterious.

FAQ

What is the difference between a codec and a container?

The container is the file type you can see, such as MP4 or MOV, and the codec is the compression method applied to the video and audio inside it. Mozilla's developer documentation treats them as separate layers for exactly this reason: two MP4 files can hold completely different codecs, and a player has to support both the wrapper and the contents.

Is rendering the same as exporting?

No. Rendering is your editor pre-building preview files so effects and stacked layers play back smoothly while you work, and those files stay inside the project. Exporting produces the finished, shareable video from your timeline. Some editors use the word render for both, which is where the confusion starts, but the outputs are different things.

What is the difference between a jump cut and a match cut?

A jump cut removes time from the middle of a continuous shot, so the subject visibly jumps, and it is the everyday tool for tightening talking-head footage. A match cut does the opposite: it hides the join by aligning shape, movement or composition across two different shots so the change feels smooth.

What frame rate should I edit at?

Match what you recorded. YouTube's encoding guidance names 24, 25, 30, 48, 50 and 60 fps as the common rates and asks you to upload in the rate the footage was shot at, because converting between rates introduces judder or duplicated frames. The exception is slow motion, where you deliberately record fast and conform the clip to a slower timeline.

Sources

YouTube recommended upload encoding settings · YouTube Help
Media container formats (file types) · MDN Web Docs
L cut: What is an L cut and J cut in film? · Adobe
Loudness · European Broadcasting Union

Turn one long video into a week of posts

300 credits for 3 days · no card.

Start free