Video Basics

TechShop: Digital Video Basics

Want to work with digital video? Here's the essentials to get you up to speed with the medium.

What is Video?

  • Video, or moving image in general, is created from a sequence of still images called frames.
  • By recording and then playing back frames in quick succession, an illusion of movement is created.
  • Video can be edited by removing some frames and combining sequences of frames, called clips, together in a timeline.

What is a Video Format?

  • A video format defines the way in which video is recorded and stored. It normally specifies:
    • Codec/compressor
    • Frame rate
    • Frame size
    • Frame aspect ratio
    • Pixel aspect ratio
    • Scanning method (interlaced or progressive)
  • Common formats are DV, HDV and AVCHD.
  • Tape-based formats such as DV and HDV can be transferred to a computer for editing via Firewire.
  • File-based formats such as AVCHD are already stored as files and can be transferred to a computer for editing via USB.
  • File-based formats may need to be converted during or after transfer to be compatible with editing software.

What is a Codec (Compressor)?

  • Codec is short for coder-decoder and describes the method in which video data is encoded into a file and decoded when the file is played back.
  • Most video is compressed during encoding, and so the terms codec and compressor are often used interchangeably.
  • Transcoding is the process of converting from one codec to another.
  • Codecs can be Lossless, which means that they do not throw away any data, or Lossy, which means that data is lost during encoding.
  • Lossless codecs are higher quality than lossy codecs, but produce larger file sizes.
  • In a video workflow, you should avoid transcoding to a lossy codec until final output. This means that your final edit can be exported to H.264 for the web, but you should not use the H.264 file for further editing, DVD authoring, etc.
  • Examples of common codecs are:
Codec Lossless Recommended Use
DV-PAL No Capture/edit/output of DV-PAL video from camera; archive DV-PAL video
HDV No Capture/edit/output of HDV video from camera; archive HDV video
H.264 No Transcode video for web & computer playback (not suitable for editing)
MPEG-2 No Transcode video for DVD
Animation Yes Output/archive video (mainly motion graphics, animation, etc) from After Effects, Photoshop
Apple Intermediate Codec Yes Transcode AVCHD/H.264 video for editing; output/archive video
Apple ProRes 422 (Proxy) Yes Transcode AVCHD/H.264 video for editing; output/archive video
Apple ProRes 422 (LT) Yes Transcode AVCHD/H.264 video for editing; output/archive video

What is a Frame?

  • A frame is a single still image within a video clip.
  • As with any digital image, a frame consists of pixels (picture elements), with each pixel representing a colour within the image.
  • The higher the number of pixels, the more accurately an image can be represented. This is called resolution and is measured in megapixels.

Magnified portion of an image showing individual pixels:

pixels.png

What is Frame Size?

  • Frame Size describes the size of a single video frame: width x height, measured in pixels.
  • The width of the frame can vary depending on whether the pixels in the frame are square pixels or non-square pixels.
  • An example: a DV-PAL 4:3 frame is 768 pixels wide (square pixels) or 720 pixels wide (non-square pixels). This is because a smaller number of pixels are required to create the same size frame on screen if each individual pixel is wider.
  • Computer graphics use square pixels, whereas video cameras use non-square pixels. Therefore if creating graphics for video, use the square pixel frame size.
  • A square pixel aspect ratio can be written as 1.0 (width√∑height). All other values indicate non-square pixels.
  • Typical frame sizes are:
Format Frame Aspect Ratio Pixel Aspect Ratio Anamorphic Frame Size
DV-PAL 4:3 1.07 (PAL CCIR 601) No 720x576
DV-PAL 4:3 1.0 (Square) No 768x576
DV-PAL 16:9 1.42 (PAL CCIR 601) Yes 720x576
DV-PAL 16:9 1.0 (Square) No 1024x576
HD720 16:9 1.0 (Square) No 1280x720
HD1080 16:9 1.33 (HD 1440x1080) Yes 1440x1080
HD1080 16:9 1.0 (Square) No 1920x1080

What is Frame Aspect Ratio?

  • Frame Aspect Ratio describes the relationship between the width and height of a single video frame.
  • Video is landscape, so the width of a frame is greater than the height.
  • Typical Frame Aspect Ratios for video are 4:3 and 16:9.
  • Digital stills cameras often use 4:3 or 3:2.
  • 4:3 is refered to as standard.
  • 16:9 is referred to as widescreen, and is sometimes (inaccurately) called anamorphic. Some widescreen video is anamorphic, but not all.

Comparison of standard 4:3 and widescreen 16:9 frame aspect ratios:

4-3vs16-9.png

Mixing Frame Aspect Ratios

  • Ideally you should plan in advance and use a single frame aspect ratio throughout a video production.
  • If you use a mixture of cameras and/or sources of video, then you may end up with a mix of frame aspect ratios between your clips.
  • In this case you need to choose the frame aspect ratio for the video production that best matches the majority of your clips and then chose between cropping the other clips to fit, or accepting the letterboxing/pillarboxing that will occur from mixing aspect ratios:
    mixing_ratios.png

What is Anamorphic?

  • Anamorphic describes widescreen video that has been horizontally squeezed during recording to fit it into a 4:3 frame aspect ratio.
  • When anamorphic video is played back, it is horizontally stretched to a 16:9 frame aspect ratio so the image looks correct.
  • Anamorphic video uses 4:3 frame sizes with non-square pixels to achieve an end result of 16:9 widescreen video.
  • Most editing software will handle anamorphic video correctly, but check your captured clips prior to editing to ensure they look correct. If they don't, check the pixel aspect ratio setting and whether they are labelled (ticked) as anamorphic or not.

Comparison of anamorphic 16:9 video displayed correctly at 16:9 and incorrectly at 4:3:

anamorphic.png

What is Pixel Aspect Ratio?

  • Pixel Aspect Ratio describes the relationship between the width and height of a single pixel.
  • Different pixel aspect ratios are the reason why two video images with identical frame sizes can appear as different sizes on screen.
  • If width = height, then the pixel is square and the aspect ratio (width√∑height) = 1.0
  • If width > height, then the pixel is non-square and the aspect ratio is greater than 1.0 (e.g. DV-PAL = 1.07)
  • If width < height, then the pixel is non-square and the aspect ratio is less than 1.0 (e.g. DV-NTSC = 0.9)

Comparison of square and non-square pixels:

pixel_compare.png

What is Frame Rate?

  • Frame Rate tells you how many frames per second there are when recording or playing video.
  • Video cameras in Europe use 25 frames per second (fps). In USA & Japan 29.97fps or 30fps is used.
  • Animation works by recording each frame individually (e.g. with a stills camera) and then playing them back at a frame rate.
  • Animators often work with a lower frame rate (e.g. 12fps) so less frames are needed for the same length video clip.
  • If you change the frame rate of a 12fps video clip to 25fps, e.g. by adding it to a 25fps editing project, each frame will be repeated to keep the clip the same duration.

What is Timecode?

  • Timecode is how the duration of video is measured, and is divided into hours, minutes, seconds and frames, like so: HH:MM:SS:FF.
  • An example: 00:01:22:06 means 1 minute, 22 seconds and 6 frames.
  • It is recorded by the video camera as part of the video signal, and is used to navigate and locate video when editing.

What does Scanning Method mean?

  • A video frame is displayed on a screen by scanning each horizontal line from left to right to form the image.
  • Scanning method describes the way this is done, either by scanning first the odd lines, then the even lines (interlaced), or by scanning every line in order (progressive).

What does Interlaced mean?

  • Interlaced describes a method of recording video where each frame consists of two fields, which are combined (interlaced) together to form a complete frame.
  • SD televisions and DVD players use interlaced scanning to display video.
  • The Upper Field (field 1) contains the odd lines that form the image.
  • The Lower Field (field 2) contains the even lines that form the image.
interlaced_frame.png

What does Progressive mean?

  • Progressive or Progressive Scan describes a method of recording video where each frame is recorded as a whole image, with no fields.
  • This is similar to how a film or stills camera records images, so progressive video is often said to look more 'film-like'.
  • Progressive is often a recording option on HD video cameras.
  • It is a good choice for recording video that will be slowed down as it avoids any interlacing artifacts appearing in the slowed-down image.
  • Computer monitors and some HD televisions use progressive scanning to display video.
  • All video for playback on a computer (e.g. on the web) should be progressive scan, either by recording it as progressive scan, or by de-interlacing it if it is interlaced.

What is PAL?

  • PAL is the television standard used in the UK and most of Europe.
  • It specifies a frame rate of 25fps and a frame size of 720x576 pixels (non-square pixels).
  • For a 4:3 frame aspect ratio, a pixel aspect ratio of 1.07 is used.
  • For a 16:9 frame aspect ratio, a pixel aspect ratio of 1.42 is used.
  • DV-PAL is a DV video format that conforms to the PAL specification.

What is NTSC?

  • NTSC is the television standard used in USA & Japan.
  • It specifies a frame rate of 29.97fps or 30fps and a frame size of 720x480 pixels (non-square pixels).
  • For a 4:3 frame aspect ratio, a pixel aspect ratio of 0.9 is used.
  • For a 16:9 frame aspect ratio, a pixel aspect ratio of 1.2 is used.
  • DV-NTSC is a DV video format that conforms to the NTSC specification.

What is Standard Definition (SD)?

  • The term 'Standard Definition' (SD) describes the frame size of a video.
  • SD video can have either a 4:3 or 16:9 frame aspect ratio.
  • SD PAL 4:3 frame size is 720x576 pixels, or 768x576 if using square pixels.
  • A lot of SD web video uses a frame size of 640x480 pixels.

What is High Definition (HD)?

  • The term 'High Definition' (HD) describes a frame size that is larger than Standard Definition video.
  • HD video always uses a 16:9 frame aspect ratio.
  • HD refers to multiple video formats which use different frame sizes, frame rates and scanning methods.
  • Typical HD frame sizes are:
    • 1280x720 pixels (used for HD television and HD web video)
    • 1920x1080 pixels (referred to as full-HD or full-raster)
    • 1440x1080 pixels (full-HD with non-square pixels)
  • HD video formats are normally written like so: HDV1080p25 or HDV1080i50, which means:
Codec Frame Height Scan Method Frame/Field Rate Description
HDV 1080 p 25 HDV codec, frame height = 1080 pixels, progressive scan, 25fps
HDV 1080 i 50 HDV codec, frame height = 1080 pixels, interlaced scan, 25fps (50 fields per second)

Links

Vimeo Video School Tips
Useful sections of the Final Cut Pro on-line manual:
Video Formats
Frame Rate & Timecode
Anamorphic 16:9 Media

Unless otherwise stated, the content of this page is licensed under Creative Commons Attribution-ShareAlike 3.0 License