Wave Candy

Displays the wave form created by the audio signal.

• Pinch - Restricts the movement of the ends of the waveform (as if they are being 'pinched').
• Interpolate - Smooths zoomed waveforms.
• Update - Waveform synchronization. Right-click to sync to a specific note/frequency. Turn to the right to match higher frequencies, left to match lower frequencies.
• Window - Zoom, turn to the left to zoom in on the waveform. Turn to the right to zoom out.
• Stereo separation - Turn right to separate the L and R channels of a stereo signal.

Displays a moving trace of the audio signal. Color hue represents intensity (level), vertical position represents frequency (20 Hz to 20,000 Hz, bottom to top is represented) and horizontal position represents time.

• Update - Scroll speed, right is slower.
• Max res - Increases the resolution of the analysis at the expense of CPU load.
• Scale - Right increases the low frequency detail, left increases the high frequency detail. Piano keyboard - Rotate the knob fully right to see a Piano keyboard to aid in note identification from audio files. When looking at a single note, the fundamental/root pitch (note name) will generally be the lowest horizontal line in the group. The lines above this are likely to be harmonics. With practice you will be able to recognize the root notes in chords and other complex music.
• dB range - Changes the intensity scaling.
• Natural weighting - Sets the color intensity gradient to better represent the way we hear frequencies. Specifically, it changes the brightness weighting from white noise, a linear relationship between frequency and display brightness to pink noise. Pink noise decreases the intensity by 3dB/Octave, making it more suitable for musical uses as the higher frequencies no longer dominate the spectral display brightness, which is closer to how we hear those same frequencies.

Displays the audio level as a pair of vertical bars (left and right channels) on a dB scale.

• Scale - Zooms the scale range.
• Hold time - Amount of time the peak indicator remains at its peak. NOTE: The peak shown will be the peak of the Mode type selected. This won't be the absolute peaks, unless you are in 'Peak' mode.
• Mode - Choose a metering algorithm. Apart from 'Peak' all the other metering algorithms apply some 'average', 'weighted average' or 'secret squirrel mathematics' to the signal to better represent 'loudness' as perceived by humans. The problem is that the ear responds in a complex way to sound pressure level, depending on frequency and duration of the sound, and so various arguments about what method best represents 'loudness' have broken out among audio engineers. Welcome to the 'metering wars' -
  • Peak - Displays the precise and instantaneous level of the audio. Most useful for spotting clipping problems and seeing absolute maximum peak levels.
  • PPM (Peak Program Meter) - Similar to the 'Peak' meter with a slightly slower response time (~5 ms averaging window) so that transient peaks are ignored and levels are easier to read.
  • RMS (Root Mean Square) - The RMS calculation is performed over a 300 ms 'time window'. This moving average evens out transient peaks, to better represent loudness.
  • Vu (Volume Unit) - Originally developed in 1939 by Bell Labs, CBS and NBC for measuring and standardizing the speech levels in telephone lines, Vu was one of the first methods used to represent perceived loudness of audio signals by metering. Rather than showing peaks Vu has an intentionally slow and logarithmic response to changes in input to better represent loudness. The main shortcoming of Vu is that the fluctuations in the metering are greater than the changes in volume the ear perceives. This problem is addressed by the next two methods:
  • Leq(A) - Another attempt to 'more accurately' represent loudness. The Leq(A) metering algorithm is essentially an A-weighted average of the audio signal. The A-weighting takes into account the changing sensitivity of the human ear as a function of audio frequency. Leq(A) is reputed to be useful for matching speech levels across different tracks.
  • ITU-R BS.1770 - Yet another algorithm (developed by the International Telecommunications Union, published 2006) to 'more accurately' measure perceived loudness of lengthy sections of program material (particularly for multi-channel 'surround' audio). This algorithm measures a frequency-weighted' average over a long time-window (several seconds) and is probably best suited to the 'Mastering' stage when you need to match the perceived loudness of a number of different songs.

  Advice: Use a Peak mode to watch for clipping and your ears to make adjustments to loudness, not so difficult after all. The art of metering is really most important for audio engineers in the broadcast industry, where they need to match the volume of a wide-range of source material so that we don't have to keep adjusting the volume on our TVs & Radios. If only they would agree to use Leq(A) or ITU-R BS.1770 to match the volume of advertising to the surrounding program material! By the way, Leq(A) or ITU-R BS.1770 are not your new weapons in the 'Loudness War' , we're watching you!

• Decay speed - Rate at which the metering bar falls.

The Vectorscope measures the differences between the left and right channels of a stereo input and can provide information about stereo content, panning and phase. The display requires some additional explanation:

The flying spot - follows the positive and negative amplitude oscillations of the left and right input waveforms. The stereo channels share the spot by having their amplitude axes orthogonal to each other, so 'vectors' are created between the two channels, hence the name Vectorscope. It sounds a bit complex but is really quite simple, think of the display as an oscilloscope-style graph where one axis (+R/-R) represents the right channel amplitude and the other axis (+L/-L) is the left channel amplitude. The technical name for the visual display on a Vectorscope is a Lissajous figure (as shown left).

Left, Right & Mono signals - When fed a lone left or right channel input, a straight line is generated as the spot bounces up and down along the input axis (the dotted +L/-L or +R/-R grid). A left channel input will generate a line sloping to the left, while a right channel input generates a line to sloping to the right. A mono input, panned to center, will generate a vertical line. This is the 'vector' between the left and right channels. By observing the orientation of the line OR general orientation of a 'Lissajous figure', the display will give a visual indication of the average pan position. NOTE: The background grid shows center pan, 50% pan and 100% pan for left and right).

Stereo - Stereo is created by amplitude differences between the left and right channels, so stereo signals cause the display to deviate from the vertical line, creating a 'Lissajous figure '. The more 'stereo' content the signal has, the more complex and 'fatter' the Lissajous figure will be. Thin figures mean there is low stereo content (lines mean there is no 'stereo' effect and indicate pan position).

Phase - If a straight line, or Lissajous figure, appears oriented along the horizontal axis this indicates that left and right channels are 180° out of phase. Note that the more stereo content the signal has the more difficult it will be to see the orientation of the 'Lissajous figure'.

Options

• Update - Display update speed, right is slower.
• Thickness - Thickness of the Lissajous figure's flying-spot.
• Mode - Menu with several 'flying-spot' options.
• Add - Additive blending (improves visibility).

NOTE: If you want to use the plugin to create visual effects, try sine-wave sounds (Sytrus 'Default' for example) with phasing or flanging added. Two note chords will add another layer of complexity. Generally, keep the input simple and you will be rewarded with the most coherent and beautiful Lissajous figures. But it's not a toy, get back to work and improve your mix!