Chapter 3 — Quiz

A self-check on the recording pipeline — codecs and containers, bitrate, bit depth and chroma, dynamic range, Log, and RAW. Answer from memory first, then check the key. This is the most technical chapter so far, and these questions are the vocabulary Chapters 5, 31, 32, and 36 assume you own cold. If you miss several in one cluster, that section is your re-read before Chapter 4.

Multiple choice

1. A codec is: - A. The file wrapper named by the extension (.MOV, .MP4). - B. The compression method used to encode video on record and decode it on playback (H.264, ProRes). - C. The memory card the footage is stored on. - D. The picture profile that sets contrast and color.

2. A container (like .MOV or .MP4) is: - A. Always the same as the codec. - B. The file wrapper that holds the video, audio, and metadata — named by the extension, and not a reliable indicator of the codec inside. - C. A type of memory card. - D. The bitrate of the file.

3. Two files both end in .MOV. One is 2 GB and edits smoothly; the other is 200 MB and stutters on the timeline. The best explanation is: - A. One file is corrupt. - B. They use different codecs inside the same container — likely ProRes (large, intraframe, easy to edit) vs H.264 (small, interframe, hard to edit). - C. The bigger file is always higher resolution. - D. .MOV files are random in size.

4. Bitrate measures: - A. The number of pixels in the image. - B. How many frames per second the camera captures. - C. The amount of data recorded per second of video (Mb/s) — higher generally means more detail and bigger files. - D. The physical size of the sensor.

5. Which is a compression artifact you'd expect from too low a bitrate on a hard-to-compress scene? - A. Rolling-shutter skew. - B. Macroblocking, banding, and smeared/mosquito noise in shadows, gradients, and busy motion. - C. Motion blur. - D. Lens flare.

6. Your camera records at 400 Mb/s. Roughly how fast must your card write (in MB/s)? - A. About 400 MB/s. - B. About 3,200 MB/s. - C. About 50 MB/s (400 ÷ 8). - D. Bitrate and card speed are unrelated.

7. Bit depth (8-bit vs 10-bit) controls: - A. The number of pixels. - B. The number of tonal steps per color channel — 8-bit = 256, 10-bit = 1,024 — which sets how much you can grade before banding. - C. The frame rate. - D. The lens aperture.

8. Why does 8-bit footage often band under a heavy grade when 10-bit doesn't? - A. 8-bit has fewer pixels. - B. Grading stretches the tonal steps apart; 8-bit's 256 steps pull into visible gaps, while 10-bit's 1,024 keep a fine step nearby. - C. 8-bit is always underexposed. - D. 10-bit uses a different lens.

9. In "4:2:0," the numbers describe: - A. The aspect ratio. - B. The frame rate. - C. Chroma subsampling — how much color detail is kept relative to full brightness detail (here, a quarter of the color). - D. The bitrate in Mb/s.

10. You can't usually see the effect of 4:2:0 chroma subsampling when simply watching because: - A. Screens can't display color anyway. - B. Human eyes are far more sensitive to brightness (luma) detail than to color (chroma) detail, so the discarded color is barely registered. - C. 4:2:0 keeps all the color. - D. The codec adds the color back on playback.

11. Dynamic range is: - A. The number of colors a file can store. - B. The span from darkest to brightest a camera can capture in one exposure while holding detail at both ends, measured in stops. - C. The range of frame rates a camera offers. - D. How fast the lens focuses.

12. Ungraded Log footage looks flat, grey, and desaturated because: - A. The camera is broken. - B. Log records the full captured range as a low-contrast curve so highlights and shadows survive to post; the contrast and color haven't been applied yet. - C. It was underexposed. - D. Log removes all the color permanently.

13. The most dangerous Log trap for a beginner is: - A. Shooting Log in 10-bit. - B. Shooting Log in 8-bit (stretching a flat 8-bit image to contrast bands badly) — and/or never grading it. - C. Exposing Log slightly bright. - D. Using Log outdoors.

14. RAW video is different from a normal codec (even Log) because it records: - A. Only the audio. - B. The sensor's near-unprocessed readings — before demosaicing and before a profile bakes in contrast, color, or white balance — so you build the image in post. - C. A smaller file than H.264. - D. Nothing that can be changed later.

15. The best one-line reason to choose 4K to deliver 1080 (Chapter 2) and the best reason to choose 10-bit/high-bitrate (Chapter 3) are the same underlying idea: - A. Higher numbers are always better. - B. Both are set-day decisions made to give the edit and grade room to work — "you shoot for the edit," paid for in storage. - C. Both make the camera focus faster. - D. Both reduce file size.

16. For a straightforward Project 1 talking-head you'll cut fast and correct lightly, the honest format choice is: - A. Always RAW, to be safe. - B. 8-bit Log, because Log is professional. - C. A good standard profile at the highest bitrate your card sustains (8-bit 4:2:0 is fine), white balance set in camera. - D. The lowest bitrate available, to save space.

17. Which capture situation most justifies 10-bit 4:2:2? - A. A quick vertical social clip. - B. A green-screen composite or a heavily-graded product on a smooth gradient backdrop. - C. A locked-off shot of a plain wall you won't grade. - D. A voice memo.

18. "This file won't import" is, nine times out of ten, a problem with the: - A. Container extension — just rename it. - B. Codec inside — transcode it or install the right decoder. - C. Resolution. - D. Frame rate.

19. The correct order of a safe card workflow is: - A. Shoot → format the card → hope the offload worked. - B. Shoot → offload to a drive → back up to a second place → then test import/playback → only reuse the card once two copies exist. - C. Shoot → reuse the card immediately → back up next month. - D. Shoot → delete the bad takes on the card → offload the rest.

20. Choosing a capture format should be driven by one question: - A. "What's the most expensive option?" - B. "What will happen to this footage in post?" — grade, VFX, deadline, storage, computer, and skill all follow from the answer. - C. "What do other people shoot?" - D. "What's the highest number the camera offers?"

True or false (justify your answer in one sentence)

21. The file extension (.MP4, .MOV) reliably tells you which codec is inside.

22. Compression damage like banding and macroblocking can be removed later in the grade.

23. 8-bit 4:2:0 footage is fine to watch but fragile to grade heavily.

24. Shooting Log always makes your video look more professional.

25. RAW makes changing white balance in post essentially free.

26. A single copy of your footage on one drive counts as a backup.

Short answer

27. Explain the difference between a codec and a container, and why "the file won't import" is usually a codec problem.

28. State what bit depth and chroma subsampling each control, and read the pair "10-bit 4:2:2" in one sentence.

29. Name the three strings attached to shooting Log, and say which makes 8-bit Log a trap.

30. Give the one question that should drive every capture-format decision, and apply it to a fast-turnaround talking-head versus a heavily-graded branded piece.

What's wrong here? (perception & diagnosis)

31. A creator shoots a sunset time-lapse in their camera's lowest-bitrate mode "to save card space." On the camera it looks gorgeous; in the edit, the moment they deepen the sky it breaks into visible stepped bands that worsen as they push. What went wrong, and what should they have shot?

32. A beginner flips their 8-bit phone to a downloaded "Log" profile because a tutorial said pros shoot Log, then delivers the ungraded footage to a client. The client says it looks "washed out and broken." Diagnose both mistakes and prescribe the fix.

33. An editor is handed 4K clips that play fine in a media player but stutter so badly in the timeline they can't cut. Exposure and framing are perfect. What's happening, and what's the fix (name the chapter)?

34. A shooter formats and reuses a card between two shoots to avoid buying a second card. Later, half a wedding is missing — the offload had silently failed and the card was already wiped. Name the two process failures and the one rule that prevents this forever.

35. A team shoots a green-screen product ad in 8-bit 4:2:0 at a low bitrate to keep files small, then can't get a clean key — the edges are blocky and the matte is soft. Name the two format decisions working against them and what they should have shot.

Answer Key

  1. B. A codec is the compression method (coder–decoder); the container is the wrapper (A), the card is storage, the profile sets contrast/color.
  2. B. The container is the wrapper named by the extension and does not reliably reveal the codec inside.
  3. B. Same container (.MOV), different codecs — likely ProRes (large, intraframe, smooth) vs H.264 (small, interframe, stuttery). Size difference is the codec, not corruption or resolution.
  4. C. Bitrate = data per second (Mb/s); higher → more detail and bigger files.
  5. B. Low bitrate on hard scenes gives macroblocking, banding, and smearing in shadows, gradients, and busy motion. The others are unrelated phenomena.
  6. C. 400 Mb/s ÷ 8 = 50 MB/s of actual writing — and you want a card rated comfortably above that.
  7. B. Bit depth = tonal steps per channel (8-bit = 256, 10-bit = 1,024); it sets grading headroom before banding.
  8. B. Grading stretches the steps apart; 8-bit's 256 open into visible gaps (banding), 10-bit's 1,024 keep a fine step nearby.
  9. C. 4:2:0 is chroma subsampling — full brightness detail but a quarter of the color detail.
  10. B. Human vision is far sharper at brightness than color, so discarded color detail barely registers when watching.
  11. B. Dynamic range = the darkest-to-brightest span held in one exposure, in stops.
  12. B. Log records a flat, low-contrast curve to preserve range for the grade; the contrast/color simply haven't been applied yet. It's not broken or necessarily underexposed, and the color isn't gone.
  13. B. 8-bit Log bands badly when graded, and un-graded Log looks broken — both are the classic beginner traps. 10-bit, careful exposure, and Log go together.
  14. B. RAW records the sensor's raw readings before demosaicing and before a profile — you build the image in post.
  15. B. Both are "you shoot for the edit": capturing extra pixels / tonal steps / data so post has room to reframe and grade, paid for in storage.
  16. C. A good standard profile at the highest sustainable bitrate, 8-bit 4:2:0 fine, white balance set in camera. RAW and 8-bit Log are the wrong tools here.
  17. B. Green screen and heavy grades on smooth gradients are exactly where the extra tonal steps (10-bit) and color detail (4:2:2) earn their keep.
  18. B. It's a codec problem — transcode or install the decoder. Renaming the container extension does nothing.
  19. B. Offload, back up to a second place, test import/playback, and only reuse the card once two copies exist.
  20. B. "What will happen to this footage in post?" drives everything; the highest number is rarely the right answer.
  21. False. The extension names the container; a .MOV or .MP4 can hold several different codecs — the extension doesn't reliably reveal which.
  22. False. Compression damage is baked in at capture and permanent; grading amplifies it rather than removing it.
  23. True. 4:2:0 8-bit looks fine to watch (your eye can't see the missing color or the coarse tonal steps) but bands and breaks under a heavy grade.
  24. False. Log only looks professional after a good grade, in enough bit depth, exposed correctly; mishandled (8-bit, ungraded, underexposed) it looks worse than a standard profile.
  25. True. RAW records sensor data before white balance is baked in, so white balance becomes a no-penalty post decision — a genuine safety net (not an excuse for sloppiness on set).
  26. False. One copy on one drive is a single point of failure; a real backup means the footage exists in at least two (ideally three) places before you erase the card.
  27. A codec is the compression method (how the picture is packed: H.264, ProRes); a container is the file wrapper named by the extension (the box: .MOV, .MP4). "Won't import" is usually a codec problem because the software can decode the container but not the codec inside — the fix is to transcode the file or install the decoder, not to rename the extension.
  28. Bit depth controls tonal steps per channel (grading headroom before banding); chroma subsampling controls how much color detail is kept relative to brightness detail (matters for keying and heavy color). "10-bit 4:2:2" = 1,024 tonal steps per channel and half-resolution color — the practical sweet spot for serious grading and green screen.
  29. (a) Log must be graded (ungraded it looks broken); (b) Log wants bit depth (stretching flat 8-bit Log to contrast bands — the trap); (c) Log needs careful, slightly-bright exposure to keep shadow noise down (Chapter 5). The bit-depth string is what makes 8-bit Log a trap.
  30. "What will happen to this footage in post?" For a fast-turnaround talking-head: a standard profile, high bitrate, 8-bit 4:2:0 — cut and post, no time or need for a heavy grade. For a heavily-graded branded piece: 10-bit and Log (or a flat profile), high bitrate, carefully exposed — because the grade needs the room, and you have the pipeline and time to use it.
  31. They mistook low bitrate (and likely 8-bit) for a harmless space-saver; the smooth sky is the hardest thing to compress and the easiest to band, so it broke under the grade. They should have shot the highest bitrate the card sustains and 10-bit if available — smooth gradients demand tonal and data headroom. (Damage is permanent; no post fix.)
  32. Two mistakes: shooting 8-bit Log (which bands when graded) and not grading it at all (so it looks flat and broken on delivery). Fix now: grade the footage back to a normal look (Chapters 31–32) if it holds up; fix for next time: if you can't shoot 10-bit and won't grade, shoot a good standard profile — it'll look better than mishandled Log.
  33. The clips are a heavy interframe/long-GOP codec (like 4K H.264/H.265): every frame must be rebuilt from neighbors, which a media player streams fine but a timeline chokes on. The fix is proxies / transcoding to an editing codec — lightweight stand-ins you cut with, swapped for originals on export — covered in Chapter 27.
  34. Two failures: (1) the offload wasn't verified (they assumed it worked), and (2) the card was reused before a second copy existed. The rule: never format or reuse a card until its footage lives in at least two other places, and confirm the copies actually opened (offload → back up → test → then wipe). Formalized as 3-2-1 backup in Chapter 37.
  35. Working against them: 8-bit (too few tonal steps for a clean matte) and 4:2:0 (quarter-resolution color gives soft, blocky key edges), made worse by a low bitrate. They should have shot 10-bit 4:2:2 at a high bitrate — clean color detail is what a key needs (green screen and keying: Chapter 35).

Topics to review by question

  • Codec vs container: 1–3, 18, 21, 27, 33.
  • Bitrate & compression artifacts: 4–6, 22, 31.
  • Bit depth & banding: 7–8, 23, 28, 31.
  • Chroma subsampling & keying: 9–10, 17, 28, 35.
  • Dynamic range: 11.
  • Log (standard vs Log, the three strings): 12–13, 16, 24, 29, 32.
  • RAW: 14, 25.
  • Choosing a format / "shoot for the edit": 15, 16, 20, 30, 36-thinking.
  • Card & offload workflow: 19, 26, 34.

Anything you missed, re-read that section before Chapter 4 — the recording format is the one Chapter 2–3 decision you can never take back, so it's worth owning cold.