{"course":{"slug":"motion-and-audio","title":"Motion, Sound and Music","url":"https://addaly.com/learn/motion-and-audio","level":"intermediate","track":"create"},"lesson":{"slug":"frames-are-the-unit","title":"Frames, not seconds","url":"https://addaly.com/learn/motion-and-audio/frames-are-the-unit","position":3,"of":72,"minutes":9,"answer":"Frame rate is a commitment made before the first keyframe, because a duration in milliseconds only becomes a real animation once it has been rounded to whole frames.","keyIdea":"Frame rate is a commitment made before the first keyframe, because a duration in milliseconds only becomes a real animation once it has been rounded to whole frames, and changing the rate afterwards silently resamples everything you spaced by hand."},"body":{"format":"markdown","text":"## The software lies to you politely\n\nEvery animation tool shows a timeline in seconds, and almost everybody types in seconds. Underneath, nothing is stored in seconds. A frame is the unit, and a second is whatever number of frames you decided on before you started.\n\nThis matters the first time you specify a 300ms transition and hand it to somebody working at a different frame rate.\n\n- At 24 fps, 300ms is 7.2 frames. You get 7, and the real duration is 292ms.\n- At 25 fps, it is 7.5 frames. You get 7 or 8 depending on which way the software rounds.\n- At 60 fps, it is 18 frames exactly.\n\nNone of these are wrong. They are just not the same animation, and if you approved one and shipped another, the difference is real.\n\n## The rates, and what each one is for\n\n- **24 fps**: cinema, and anything that wants to read as film. The lowest rate at which camera motion is tolerable.\n- **25 fps**: broadcast in India, the UK, Europe, Australia. If your work touches television anywhere on 50 Hz mains, this is the number.\n- **30 fps (29.976)**: North American broadcast heritage, and the safest default for social video because most platforms re-encode towards it anyway.\n- **50 and 60 fps**: sport, screen recordings, and interface work where you want the motion to feel like the device rather than like a film.\n- **12 fps**: not a delivery rate, a working method. Hand-drawn animation holds each drawing for two frames inside a 24 fps timeline. Twelve distinct images a second is enough for drawn motion and nowhere near enough for a camera pan, which is why traditional animation works \"on twos\" and live action cannot.\n\n## 29.97, and where the missing time went\n\nWhen colour was added to American black-and-white television, the frame rate was slowed by exactly one part in a thousand to keep the colour subcarrier out of the audio. That 0.1% never went away. So 30 fps is usually 29.97, and 24 fps in a broadcast chain is usually 23.976.\n\n> Figure: What 300 milliseconds is, in frames. Bars (frames, before rounding): 24 fps, cinema 7.2; 25 fps, broadcast in India and Europe 7.5; 30 fps, social video 9; 60 fps, interface and screen capture 18. A frame is the unit and a second is however many frames you chose before you started. Each of these rounds to a whole frame, so the duration you actually get is 292 ms at 24, either 280 or 320 at 25 depending on which way the software rounds, and exactly 300 at 30 and 60. None of them is wrong and none of them is the same animation, which is why a brief says seven frames at 24 rather than about a third of a second.\n\nThe consequence is that a programme that runs for ten minutes of real time does not run for ten minutes of timecode. Over an hour, the gap is about 3.6 seconds. Drop-frame timecode fixes the *labels* by skipping two frame numbers every minute except every tenth minute: it does not drop any picture. Non-drop timecode counts honestly and drifts from the clock.\n\nIf you have ever wondered why an editor's delivery spec says `29.97 DF` and refuses anything else, that is why: a broadcast slot is measured by the clock, not by the frame count.\n\n## Set it first, and do not change it after\n\nPick the delivery frame rate before you place a single keyframe. Changing it later is not a neutral operation, and different tools disagree about what it means:\n\n- After Effects keeps keyframes at their *time* values, so they land on fractional frames and get sampled to whichever frame is nearest. A carefully spaced move loses its spacing.\n- Blender's frame-rate change keeps the *frame numbers*, so the animation changes duration instead.\n- Kdenlive will not let you change a project's profile frame rate after the fact without re-interpreting every clip.\n\nNone of these is a bug. They are two reasonable answers to a question you should never have to ask.\n\n## Higher is not better, it is different\n\nA common assumption is that 60 fps is a quality upgrade over 24. It is a different look, and the mechanism is exposure, not resolution.\n\nWith a conventional 180-degree shutter, the camera is open for half the frame interval. At 24 fps that is 1/48 of a second of blur per frame. At 60 fps it is 1/120: half as much. Less blur per frame means each frame is crisper and the strobing between them is more visible, which is exactly the \"looks like a soap opera\" reaction people report and cannot explain. Rendered animation has the same property: if you do not add motion blur in the render, a fast move at any frame rate will strobe.\n\nSo the honest statement is: 60 fps buys you temporal resolution and costs you the blur that hides the gaps. For interface motion, where elements are flat and travel short distances, that trade is usually good. For a camera move across a landscape, it usually is not.\n\n## What to write down\n\nOn every brief, before anything else: **frame rate, resolution, and duration in frames.** \"Seven frames at 24\" is a specification. \"About a third of a second\" is a conversation you will have again."},"check":{"question":"The same 300ms slide is built twice: once in a 24 fps composition and once at 60 fps, both travelling 600 pixels. Played back, the 24 fps version strobes during the fastest part of the move while the 60 fps version does not. What is the mechanism?","options":[{"text":"At 24 fps the object covers roughly 83 pixels between consecutive frames instead of 33, so the gaps between its positions become visible unless motion blur fills them.","correct":true},{"text":"The 24 fps file has a lower bitrate, so the encoder is discarding detail during fast motion.","correct":false,"tempting":"Confuses a temporal sampling artefact with a compression artefact; the strobing appears in an uncompressed render too."},{"text":"24 fps rounds 300ms down to 7 frames, and the shorter duration is what makes it look rougher.","correct":false,"tempting":"Notices the rounding, which is real, but 8ms of duration cannot account for visible strobing; the per-frame displacement can."},{"text":"The easing curve is evaluated once per frame, so a composition with fewer frames gives the software fewer opportunities to apply the ease, and the motion degrades towards linear as the frame rate drops.","correct":false,"tempting":"Treats easing as something sampled into existence rather than a continuous function evaluated at whatever frames exist; the curve is identical in both."}]},"prev":{"slug":"weight-comes-from-spacing","title":"Motion blur will not fix a floaty logo","url":"https://addaly.com/learn/motion-and-audio/weight-comes-from-spacing"},"next":{"slug":"the-graph-editor","title":"The panel most people never open","url":"https://addaly.com/learn/motion-and-audio/the-graph-editor"},"authorship":"Founder and series editor: Dr Mustafa Mun. Written by AI models to a written standard. Exam keys checked blind by a second model: 3,665 of 3,666 agreed. Academic review invited.","review":{"status":"Draft, open for academic review"},"cite":"Addaly (2026) 'Frames, not seconds', in Motion, Sound and Music. Edited by M. Mun. Addaly. Available at: https://addaly.com/learn/motion-and-audio/frames-are-the-unit (Accessed: 13 October 2026).","licence":{"name":"CC BY-NC-SA 4.0","url":"https://creativecommons.org/licenses/by-nc-sa/4.0/"},"reader":{"schema":1,"title":"Frames, not seconds","courseTitle":"Motion, Sound and Music","module":{"index":1,"slug":"how-motion-reads-as-real","title":"How motion reads as real"},"answer":null,"keyIdea":"Frame rate is a commitment made before the first keyframe, because a duration in milliseconds only becomes a real animation once it has been rounded to whole frames, and changing the rate afterwards silently resamples everything you spaced by hand.","hero":{"kind":"figure","figure":{"t":"figure","label":"Figure","figure":{"type":"bars","unit":"frames, before rounding","title":"What 300 milliseconds is, in frames","series":[{"label":"24 fps, cinema","value":7.2,"highlight":true},{"label":"25 fps, broadcast in India and Europe","value":7.5},{"label":"30 fps, social video","value":9},{"label":"60 fps, interface and screen capture","value":18}],"caption":"A frame is the unit and a second is however many frames you chose before you started. Each of these rounds to a whole frame, so the duration you actually get is 292 ms at 24, either 280 or 320 at 25 depending on which way the software rounds, and exactly 300 at 30 and 60. None of them is wrong and none of them is the same animation, which is why a brief says seven frames at 24 rather than about a third of a second."},"alt":"Figure: What 300 milliseconds is, in frames. Bars (frames, before rounding): 24 fps, cinema 7.2; 25 fps, broadcast in India and Europe 7.5; 30 fps, social video 9; 60 fps, interface and screen capture 18. A frame is the unit and a second is however many frames you chose before you started. Each of these rounds to a whole frame, so the duration you actually get is 292 ms at 24, either 280 or 320 at 25 depending on which way the software rounds, and exactly 300 at 30 and 60. None of them is wrong and none of them is the same animation, which is why a brief says seven frames at 24 rather than about a third of a second."}},"blocks":[{"t":"h2","id":"the-software-lies-to-you-politely","runs":[{"text":"The software lies to you politely"}]},{"t":"p","runs":[{"text":"Every animation tool shows a timeline in seconds, and almost everybody types in seconds. Underneath, nothing is stored in seconds. A frame is the unit, and a second is whatever number of frames you decided on before you started."}]},{"t":"p","runs":[{"text":"This matters the first time you specify a 300ms transition and hand it to somebody working at a different frame rate."}]},{"t":"ul","items":[[{"text":"At 24 fps, 300ms is 7.2 frames. You get 7, and the real duration is 292ms."}],[{"text":"At 25 fps, it is 7.5 frames. You get 7 or 8 depending on which way the software rounds."}],[{"text":"At 60 fps, it is 18 frames exactly."}]]},{"t":"p","runs":[{"text":"None of these are wrong. They are just not the same animation, and if you approved one and shipped another, the difference is real."}]},{"t":"h2","id":"the-rates-and-what-each-one-is-for","runs":[{"text":"The rates, and what each one is for"}]},{"t":"ul","items":[[{"text":"24 fps","b":true},{"text":": cinema, and anything that wants to read as film. The lowest rate at which camera motion is tolerable."}],[{"text":"25 fps","b":true},{"text":": broadcast in India, the UK, Europe, Australia. If your work touches television anywhere on 50 Hz mains, this is the number."}],[{"text":"30 fps (29.976)","b":true},{"text":": North American broadcast heritage, and the safest default for social video because most platforms re-encode towards it anyway."}],[{"text":"50 and 60 fps","b":true},{"text":": sport, screen recordings, and interface work where you want the motion to feel like the device rather than like a film."}],[{"text":"12 fps","b":true},{"text":": not a delivery rate, a working method. Hand-drawn animation holds each drawing for two frames inside a 24 fps timeline. Twelve distinct images a second is enough for drawn motion and nowhere near enough for a camera pan, which is why traditional animation works “on twos” and live action cannot."}]]},{"t":"h2","id":"29-97-and-where-the-missing-time-went","runs":[{"text":"29.97, and where the missing time went"}]},{"t":"p","runs":[{"text":"When colour was added to American black-and-white television, the frame rate was slowed by exactly one part in a thousand to keep the colour subcarrier out of the audio. That 0.1% never went away. So 30 fps is usually 29.97, and 24 fps in a broadcast chain is usually 23.976."}]},{"t":"p","runs":[{"text":"The consequence is that a programme that runs for ten minutes of real time does not run for ten minutes of timecode. Over an hour, the gap is about 3.6 seconds. Drop-frame timecode fixes the "},{"text":"labels","i":true},{"text":" by skipping two frame numbers every minute except every tenth minute: it does not drop any picture. Non-drop timecode counts honestly and drifts from the clock."}]},{"t":"p","runs":[{"text":"If you have ever wondered why an editor’s delivery spec says "},{"text":"29.97 DF","code":true},{"text":" and refuses anything else, that is why: a broadcast slot is measured by the clock, not by the frame count."}]},{"t":"h2","id":"set-it-first-and-do-not-change-it-after","runs":[{"text":"Set it first, and do not change it after"}]},{"t":"p","runs":[{"text":"Pick the delivery frame rate before you place a single keyframe. Changing it later is not a neutral operation, and different tools disagree about what it means:"}]},{"t":"ul","items":[[{"text":"After Effects keeps keyframes at their "},{"text":"time","i":true},{"text":" values, so they land on fractional frames and get sampled to whichever frame is nearest. A carefully spaced move loses its spacing."}],[{"text":"Blender’s frame-rate change keeps the "},{"text":"frame numbers","i":true},{"text":", so the animation changes duration instead."}],[{"text":"Kdenlive will not let you change a project’s profile frame rate after the fact without re-interpreting every clip."}]]},{"t":"p","runs":[{"text":"None of these is a bug. They are two reasonable answers to a question you should never have to ask."}]},{"t":"h2","id":"higher-is-not-better-it-is-different","runs":[{"text":"Higher is not better, it is different"}]},{"t":"p","runs":[{"text":"A common assumption is that 60 fps is a quality upgrade over 24. It is a different look, and the mechanism is exposure, not resolution."}]},{"t":"p","runs":[{"text":"With a conventional 180-degree shutter, the camera is open for half the frame interval. At 24 fps that is 1/48 of a second of blur per frame. At 60 fps it is 1/120: half as much. Less blur per frame means each frame is crisper and the strobing between them is more visible, which is exactly the “looks like a soap opera” reaction people report and cannot explain. Rendered animation has the same property: if you do not add motion blur in the render, a fast move at any frame rate will strobe."}]},{"t":"p","runs":[{"text":"So the honest statement is: 60 fps buys you temporal resolution and costs you the blur that hides the gaps. For interface motion, where elements are flat and travel short distances, that trade is usually good. For a camera move across a landscape, it usually is not."}]},{"t":"h2","id":"what-to-write-down","runs":[{"text":"What to write down"}]},{"t":"p","runs":[{"text":"On every brief, before anything else: "},{"text":"frame rate, resolution, and duration in frames.","b":true},{"text":" “Seven frames at 24” is a specification. “About a third of a second” is a conversation you will have again."}]}],"check":{"question":[{"text":"The same 300ms slide is built twice: once in a 24 fps composition and once at 60 fps, both travelling 600 pixels. Played back, the 24 fps version strobes during the fastest part of the move while the 60 fps version does not. What is the mechanism?"}],"options":[{"text":[{"text":"At 24 fps the object covers roughly 83 pixels between consecutive frames instead of 33, so the gaps between its positions become visible unless motion blur fills them."}],"correct":true,"tempting":null},{"text":[{"text":"The 24 fps file has a lower bitrate, so the encoder is discarding detail during fast motion."}],"correct":false,"tempting":[{"text":"Confuses a temporal sampling artefact with a compression artefact; the strobing appears in an uncompressed render too."}]},{"text":[{"text":"24 fps rounds 300ms down to 7 frames, and the shorter duration is what makes it look rougher."}],"correct":false,"tempting":[{"text":"Notices the rounding, which is real, but 8ms of duration cannot account for visible strobing; the per-frame displacement can."}]},{"text":[{"text":"The easing curve is evaluated once per frame, so a composition with fewer frames gives the software fewer opportunities to apply the ease, and the motion degrades towards linear as the frame rate drops."}],"correct":false,"tempting":[{"text":"Treats easing as something sampled into existence rather than a continuous function evaluated at whatever frames exist; the curve is identical in both."}]}],"order":[3,0,1,2],"why":null},"prev":{"slug":"weight-comes-from-spacing","title":"Motion blur will not fix a floaty logo","minutes":9},"next":{"slug":"the-graph-editor","title":"The panel most people never open","minutes":9}}}