ControlRoom Channel and Master Plugins

  • Compressor
    • Peak or RMS detection
    • Feed-forward, Feed-back, and Opto compression styles
    • Mojo – analog transformer modeled distortion
    • Crush – to really squash your tracks
    • Continuously Variable controls
    • Ratios from 1:1 up to 10:1 (or much much further with Crush mode ON)
    • +/- 18dB of makeup gain
    • Adjustable wet/dry mix
  • EQ
    • Dedicated Lo Cut and Hi Cut filters with Slope up to 96 dB/Octave
    • 8 fully adjustable bands with Hi Shelf, Lo Shelf, Bell, or Notch Shape
    • Continuously Variable Controls for Slope, Freq, Gain, and Q
  • Spatial Panner w Proximity DSP
    • Push towards far to put sounds further back in your mix
    • Pull towards near to place sounds front and center in your mix
    • Custom Proximity DSP combines selective filtering and harmonic saturation enhancement. Designed to be subtle but effective.
    • 3 Modes to choose from different levels of Bass Preservation
    • Unique 2D interface lets you smoothly control panning and proximity by clicking and dragging the sphere interface
    • Revolutionary 3D interface from ControlRoom VR lets you control proximity, panning, and level simultaneously in 3D space.

 

Below, some words from Persp3ctive DSP engineer Jatin Chowdhury:

ControlRoom EQ

The ControlRoom Channel EQ is built on the idea of doing exactly what you want to do and nothing more– no artifacts, no distortion. Just a smoothly and precisely shaped frequency spectrum. To that end, the channel EQ includes variable-slope filters, bell, notch, and shelf shapes, each with variable frequency, gain, and Q.

Filters

In designing the filters for ControlRoom Channel and ControlRoom Master, our goal was to make filters that sound clean and transparent, that allow you to shape the spectrum of your sound without adding anything extra. This idea led us to the electrical engineering concept of a Butterworth Filter, an analog filter notable for its maximally flat response in the pass-band and smooth rolloff in the stop-band.

Now as a mixing engineer, what does this mean? The pass-band refers to the part of the spectrum that is “passed through” by the filter. In other words– the sound you hear coming out of the filter. Having a flat response in this part of the spectrum means that the output of the filter is optimally identical to the input except for the parts of the spectrum being filtered out.

Similarly, having a smooth roll-off in the stop-band helps the filter to sound natural, and changing the slope of the roll-off allows us to control just how much we want the frequencies of the stop-band to be attenuated. Musically, this slope control appears to us as the “sharpness” of the filter. Do we want a “smooth” transition to the stop-band of our filter, or do we want to terminate everything outside of a certain frequency range, with more of a “brick wall” type response?

A true Butterworth Filter can only exist in analog. This is due to the fact that filter slope can only be evaluated as frequency approaches infinity, and a digital signal can only represent a finite set of frequencies. So the challenge for us was to model the Butterworth filter as closely as possible in the digital realm, retaining all of its useful and great-sounding characteristics. With regards to filter slope, we match the analog response as closely as possible up to the high frequency limit of the digital system. An important thing to note is that as the sampling rate of the digital system increases, our accuracy in approximating the analog filter in the audio band increases as well.

Bell

The bell shape is super useful for sculpting out a small peak or null in the frequency spectrum. In our opinion, the most important parameter for the bell is the Q. Originally for the “quality factor” of a filter, Q refers to the bandwidth of the filter.In other words, for high Q the filter will only affect the frequencies very close to center frequency, while for low Q, the filter will affect a wide band around the center frequency. Pro tip: automate the center frequency of the bell filter for a cool “wah pedal” type of effect.

Notch

Similar to the bell, the notch filter cuts out the center frequency of the filter and leaves the rest of the spectrum alone. Notches are great for cleaning up a sound: maybe you left your fridge running while recording your vocals, and now the 60 Hz hum of your fridge is all up in your sound (hey, it’s happened before).The good news is that a notch filter at 60 Hz can take care of that noise, while leaving the rest of your track perfectly intact. Similar to the bell filter, the Q parameter adjusts the size of the filter band being notched out.

 

High/Low Shelf

For a shelving filter, the idea of “bandwidth” doesn’t really have a meaning.Instead, the Q parameter or “quality factor” adjusts the resonance of the filter at the center frequency. A Q of 0.7071 (1/sqrt(2)) gives a maximally flat filter with no resonance; higher values increase the resonance, while lower values decrease the resonance (or increase the damping) of the filter. With that in mind, the resonant shelving filter of the channel EQ are perfect for shaping the extreme ends of your spectrum. Need a little more shine on your synth track? Try boosting a high shelf with a low Q. Need some more size out of your bass? A low shelf with some low frequency resonance can work wonders.

ControlRoom Compressor

For ControlRoom Channel we wanted a compressor that was light and fast, while also being extremely versatile. A compressor that can be clean and precise when you need to clean up a vocal track, smooth and subtle when you need a synth to blend into your mix, or give some attitude and crunch to your drum bus. And do all of this so efficiently that you can run it on every channel in your mix, with your CPU hardly breaking a sweat.

In order to maximize versatility, we came up with some options to supplement the standard compressor parameters (threshold, ratio, knee, etc). First, we provide you with two level-detection schemes– Peak and RMS, both with adjustable attack and release times. Peak detection tracks the level with near-instantaneous detail, and can be surgical and precise when needed, while RMS detection uses more averaging (similar to how the human ear detects volume) resulting in a smoother compressor response.

We also offer three distinct compressor architectures: Feed-Forward, Feed-Back,and Opto. Again, feed-forward offers a little extra precision and speed in knocking down transients, while feed-back sounds slightly smoother and is more similar to how the human ear compresses sound. Finally, Opto mode is a model of an optical compressor circuit, similar to that used in the famed LA-2A analog compressor. This mode gives the smoothest response of all, along with a little bit of old-school charm for your mix.

If you want a little extra something from your compressor, we also offer switches for “Crush” and “Mojo” modes. Crush mode takes no prisoners and will squash your track’s dreams of ever having some dynamic range. Mojo mode uses a model of a particularly unique output transformer, which saturates warmly as you drive the output of the compressor. Using some cutting edge analog modeling techniques and a mild understanding of electromagnetic physics, we were able to digitally recreate the magnetic hysteresis characteristic that makes this particular transformer sound as warm and crunchy as it does.

The downside of implementing a compressor with so many options is that often creates a lot of overhead on the programming side. Traditional signal processing structures often don’t support smooth transitions between different architectures or different components, so a common solution is to run both architectures simultaneously and fade between them as necessary. While that solution works okay, it makes the compressor half as efficient, plus making the programmers’ lives exponentially more complicated when it comes time to add more options or make changes to the existing ones.

Inspired by functional programming philosophies, namely the Faust programming language, and facilitated by modern C++ concepts, we’ve developed an architecture where the computer views the functions doing the signal processing as a variable no different from the parameters being used by the processor, meaning they can be changed smoothly with minimal overhead and zero redundancy. In other words, the computer sees no difference between you changing the compressor threshold, and you changing the compressor architecture, it all gets handled quickly and smoothly.

All of these elements combined allow us to make a compressor with unprecedented versatility. One that’s light on your CPU, but provides some hefty compression power. As a part of the ControlRoom audio production environment, the possibilities with this compressor are truly boundless.

Happy mixing!

A comparison of peak and RMS level detecors. Note the slower andsmoother response of the RMS curve.
Feed-Forward compressor architecture.
Feed-Back compressor architecture.
Optical compressor architecture.
A comparison of feed-forward (FF), feed-back (FB), and Opto architectures.
Our model of the transformer hysteresis curve compared to the measured
response.

Below, some words on the Proximity DSP from Persp3ctive DSP engineer Saul Laufer:

ControlRoom Proximity

ControlRoom Proximity is a psychoacoustic effect designed to both emulate how humans perceive sound changing at distance as well as enhance those changes in a virtual environment.  At Proximity setting 75, the effect is a pass-through – there is no adjustment happening to the sound and you can think of it as the normal distance of your speakers to your ears. The “near” and “far” end of the spectrum effectively move you and the sound source further and closer from each other respectively, so you can change where various sound sources are in proximity to you while mixing. This is all done without any delay or reverb effect, and will keep everything phase accurate.

A good way to think of this effect is to think of how different sounds are more or less apparent depending on your distance. The length of a low frequency wave is longer than that of a higher frequency. Some low frequency waves can reach up to miles before the first crest. This is part of the reason why you can hear thunder from a much greater length than you can hear a car alarm.

 

Controls

Proximity:

The amount of the effect. From 75 to 0, the sound source will behave as though it is further from the listener, being processed through a series of filters to lessen the presence of some frequencies.

From 75 to 100, the effect will mimic the changes of sound sources moving closer to the listener by saturating or emphasizing select frequencies.

Preserve Bass:

While A or B is selected, two different methods of preserving bass over the “far” end of the spectrum (0 – 75) are employed, similar to a bass management system. While the effect is set to Off, no bass correction is employed and you’re hearing the filters as designed.

 

Design

The harmonic saturation portion of the effect was designed to replicate certain subjective harmonic examples selected by the team. Upon investigation, we discovered that the examples and “sounds” in question resembled idealized and modified versions of harmonic distortion created by select types of push-pull amplifiers popularly used in the 1970s. Specifically, a small portion of the output stage of  what is known as a “Totem Pole” push-pull amplifier was analyzed for it’s driver circuit’s asymmetry.

An important byproduct of these designs is odd-order harmonic distortion and the respective volumes of harmonics generated. Harmonics are multiples of the fundamental frequency (e.g. if 1kHz is the fundamental frequency, 5kHz is the 5th order and is a higher order than 3kHz, or the 3rd order). Contrary to an old belief, odd-order harmonics are quite musical but the higher the order of the generated harmonic, the less pleasant the perception according to subjective listening tests1. For this purpose, the cascading gain of the harmonics created by these old amplifiers was used as an audible reference point for the saturator.

1E. Geddes; L. Lee, “Auditory Perception of Nonlinear Distortion,” AES 115th Convention (October 2003).

Spectrum analysis of a 1kHz sine wave at -20dB  through the saturator algorithm.

 

During processing, our algorithm first creates all audible odd-order harmonics of the fundamental frequency while fitting all harmonics to a “satisfying” non-linear curve so no harmonic is too loud or too quiet, creating a proportion of harmonic distortion totally unique to ControlRoom.

 

The resulting algorithm creates an idealized, isolated and modified version of the harmonic distortion unintentionally created by 1970s tube amplifiers to colorize the sound at correct increments regardless of what is passing through.