MIXING & EFFECTS

Emphasis

Emphasis is a multi-stage, single-band mastering limiter designed for sound and loudness enhancement while preserving the original dynamics profile and spectral balance. At the heart of the processor, are two interdependent limiting stages: one for crystal-clear gain reduction, the other providing wave-shaping with arbitrary characteristics. The limiter provides only very few high-level parameters, but it can be customized fully and quickly to get the best out of any genres or styles, from classical music to harsh noise. Alternatives to Emphasis are Maximus, Soundgoodizer, Fruity Limiter and Fruity Compressor.

Visualizations & Controls

The left side of the display shows a rolling history of the plugin activity, including output peaks, gain reduction functions, and LUFS plots. The right side of the display holds the peak meters that are the source of the rolling display. Controls are located under the Options menu, as shown above.

  • Peak Metering - GR, L/R and (I, S, M):

    Video Tutorial

    Controls

    • Gain Reduction (GR) - Gain Reduction relative to the input caused by the Emphasis effect. The display illustrates the cumulative effect of the first (clean limiting) and second (clipping limiting) reduction stages, with the display color saturation indicating the intensity of each stage. The upper orange segment shows the first clean limiting stage, while the red segment below it reflects the second stage clipping limiting. The extent of the second stage reduction is controlled by the Hardness setting. After those two stages, there is another stage with 'Brickwall' and True Peak limiter, combined with the first two stages.
    • Output Peaks (L/R) - Output peak levels. Dark gray shows peak levels and light gray RMS (an averaged level).
    • LUFS metering - This meter can show Momentary (M), Short-term (ST), or Integrated (I), depending on the settings. Direct access (Left-Click) the drop-down menu to the right of the S control in the screenshot above OR use the Options menu to change the setting.
      • Integrated (I) - It provides the loudness measure averaged over the entire signal length being analyzed. Music production normally targets track-length integrated LUFS values between -9 and -12 LUFS.
      • Short Term (S) - 3 second integration (averaging) period.
      • Momentary (M) - 0.4 second integration (averaging) period. Values are represented as negative compared to full scale zero (0). As the LUFS value approaches zero your track will have less dynamic range and punchiness. There are no right and wrong answers here, as music levels are part of the 'art'. However, if you aim for somewhere between -12 and -10 dB (for the loudest parts of the track) you should retain good dynamics and with enough loudness.

      LUFS Measures - LUFS (Loudness Unit Full Scale) is a synonym for LKFS (Loudness, K-weighted, relative to full scale), so you may see these terms used interchangeably. LUFS is 'simply' a modified peak level measurement. The input signal is equalized (filtered) to better represent human frequency sensitivity. That is, frequencies humans are most sensitive to, will contribute more to the measurement. Then, LUFS peak measurements are averaged over the track. In this way LUFS attempts to simulate the time and frequency characteristics of how the ear responds to sound pressure level. Of course, various arguments about what method best represents 'loudness' have broken out among audio and broadcast engineers, and LUFS was a response to these arguments designed to help broadcast entities deal with loudness variations that can upset listeners. LUFS is not magic, it is just an agreed industry standard for the apparent loudness of audio.

      How to use LUFS? - LUFS is widely used by 'broadcast' entities such as Television, YouTube, Spotify and iTunes. If everyone uses the same LUFS scale, and agrees on a target level, then it's easier to balance the apparent loudness of music and other audio, from a wide range of sources. Practically, audio levels are re-scaled to match LUFS targets, so successive tracks don't sound dramatically different in volume. For each platform using LUFS, if your track is above their LUFS target, they will turn it down. If it is below the target, they will turn it up. Since LUFS targets differ between broadcasters, we recommend Googling your specific interest, 'YouTube LUFS', for example. If you master your tracks with a given broadcast target in mind, it should be served untouched by the broadcaster during delivery.

      NOTE: LUFS measures are reset when the Stop button is pressed.

    • Range - Difference in LUFS units between the highest and lowest parts measured.

Parameters

Input Gain

  • Input - Input level in decibels. Input levels will have a significant impact on the processing of the effect.

Enhancement

  • Emphasis mode - There are two modes with manual emphasis control, and two modes with adaptive mechanisms for automatic emphasis adjustment. For the adaptive modes, once choosing a Hardness setting that is adequate for the music style (see below), turn on RMS Compensation and increase the input gain until you hear any artifacts – keep switching the bypass button on/off to assess that. Then, slightly dial back the input gain until there is no apparent difference between bypassed and processed signals. Then, turn RMS Compensation off to hear the results. You can proceed in a similar fashion with the manual modes, except you can play with both the input gain and the emphasis knob to find your sweet spot.
    • Versatile (adaptive) - This algorithm is the best compromise between loudness and transparency. Even when using high Hardness settings, it prevents distortion in signals where shaping creates the most audible artifacts, while it allows for saturation whenever possible to maximize loudness and presence. All of this happening with the focus of preserving the original dynamics of the input.
    • Loud (adaptive) - This algorithm is designed for maximum loudness. While it also minimizes distortion where artifacts are most apparent, this mode can be pushed somewhat extremely, applying considerable saturation and reducing the overall dynamics by pushing everything towards the ceiling.
    • Transient (manual) - This mode emphasizes transients while maintaining clean steady-state signals. Used gently with moderate Emphasis values, it sharpens hits; when pushing the Emphasis parameter up, it boosts loudness and body by shaping a wider portion of the signal after the transient. NOTE: The Hardness parameter is crucial, balancing distortion levels and signal sharpness.
    • Steady (manual) - This mode enhances signals above a threshold regardless of their transient or steady-state nature. When the Emphasis parameter is all the way up, this algorithm is nothing but a saturator whose shape can be adjusted via the Hardness setting; hence, it can be driven arbitrarily via the input gain. With smaller Emphasis settings, the saturation amount is reduced. NOTE: The Hardness parameter is crucial, balancing distortion levels and signal sharpness.
  • Emphasis Amount - This multi-parameter control adjusts both the level of emphasis and dynamic boost applied to the signal. Lower values retain more of the source dynamics while subtly shaping louder peaks. Higher values increase both loudness and the extent of shaping through compression or limiting. Note: The quality of the shaping also depends on the Hardness setting (see below) and whether Emphasis is automatically controlled - such as when using the Versatile or Loud adaptive modes.
  • Envelope - Affects more than just the shape of the envelope curves (gain reduction curves) due to the algorithm's self-regulating mechanisms, which respond dynamically to the behavior of these curves. Higher numbers (orders) increase the S-shaped curvature of both attack and release, resulting in more pronounced movement in the gain reduction curves. They emphasize a smaller portion of the transients, creating stronger dynamic shifts. Lower numbers produce a more natural dynamic response, prioritizing louder results by emphasizing a broader portion of the transients.
  • Hardness - Controls the transparency of the emphasis limiting stage, functioning similarly to the knee parameter in a compressor. It offers 10 steps, ranging from: 0 - Completely clean emphasis, with no clipping limiting applied to 9 - Maximum hardness, where the emphasis stage effectively hard-clips the signal.

    Recommended settings:

    • 0 to 2 - Ideal for classical or instrumental music.
    • 3 to 6 - Suitable for most genres of music.
    • 7 to 9 - Best for electronic music or tracks with dense, complex frequency spectra.

Routing

  • Linking - Controls how the gain reduction curves are calculated independently or not for the stereo input channels. At 0% they are based on the separate Left and Right channel peaks, while at 100%, they are based on a combination of both channels. Higher linking values help maintain the original stereo image. When the Mid/Side (M/S) processing switch is enabled, the linking knob applies the same mechanism to the Mid and Side signals (see below). To maximize the stereo widening effect in M/S mode, use lower linking values.
  • Mid/Side - Mid/Side processing switch. When this switch is enabled, the limiter is more likely to be driven heavily. In most cases, you will be able to compensate this by decreasing the input gain by about 3 to 6 dB.

Output

  • Output - Output ceiling in dB.

Lower Controls

  • Lock - Locks the controls that are highlighted when you mouse-over the control. This prevents most controls other than the Emphasis section being changed.
  • True Peak - True-peak limiting in compliance with the EBU R 128 standard. Peak analysis is over sampled to use True Peak as the metric for calculations. This means that instead of just measuring the peaks at standard sampling rates, the system looks at extra points between samples to detect the actual highest level of the audio. This helps avoid clipping and distortion that might go unnoticed at sample-rate calculations.
  • Oversampling - Choose Off, 2x, 4x, 8x, or 16x oversampling. NOTE: If oversampling is desired, 192 kHz will be more than adequate. This corresponds to ~ 4x. If FL Studio is set to 192 kHz, oversampling in the plugin should be disabled.
  • Bypass - Seamless bypass switch for A/B testing of the effected and original signal. This is most useful when combined with the Level Matching switch.
  • Delta - The Delta switch allows to monitor difference between the input and output signals. When the Level Matching is on, the residual heard is based on the compensated output; when Level Matching is off and Delta is on, the output is automatically scaled down by the negative input gain. When Delta is on, the Output knob is disabled.
  • Level Matching - RMS-based output loudness compensation ensures that input and output audio maintain an equivalent perceived loudness. Since louder often sounds better - can you handle the truth?

Signal Flow

Plugin Credits

Code: Dario Sanfilippo: DSP algorithm and audio programming.

GUI: Miroslav Krajcovic (implementation and design) & Nenad Milosevic (design concepts).