You spend all afternoon in an empty concert hall perfecting the house EQ for the concert that night. But after the audience fills the hall, everything changes. Or how about an outdoor concert? The temperature drops between sound check and the concert, and the EQ settings are all wrong. Or worse, after spending three days perfecting a large installation, you finally lock down the system. Then two days after you fly home, your client calls with problems.

No matter how long you spend at setup, it's impossible to anticipate every change in the acoustics of the performance space. Wouldn't it be great if someone came up with a sub-audible noise generator so you could continue the real-time analysis during the concert? Then you could measure and compensate for changes in the sound field during the performance. While we're dreaming, why not hook the RTA directly to the house EQ and let it push the sliders automatically to maintain your EQ curve in real time?

Impossible? Not any more!

Introducing the Sabine REAL-Q2 Real-Time Adaptive Equalizer, the first adaptive DSP tool that uses ten times more processing power than your PC to automatically analyze and equalize your sound space during the program - even when you're not there.

The REAL-Q2 only adjusts for changes in the room acoustics, not changes in the mix. You keep full control of the mixing console, program EQ, FBX-Feedback Exterminators, and all other signal processors that precede it in the signal path. If you decide to add bass to the mix, the REAL-Q will not compensate. However, if the audience absorbs the bass, the REAL-Q automatically boosts the equivalent amount.

REAL-Q2 installation and operation is about as easy as using your CD player. Just patch it in and in a few keystrokes, the REAL-Q2 automatically analyzes and flattens the room, or matches any curve you create. Then it monitors and compensates for any changes in the auditorium - changes that will otherwise ruin your sound.

On the other hand, if you want full control, the REAL-Q2 provides a full-featured RTA and noise generator, digital graphic EQ, graphic display of the room's response curve, and storage for 100 different response curves.

A Fundamental Breakthrough For Audio Engineers

The REAL-Q2 represents a fundamental breakthrough that goes far beyond any other audio engineering tool. A REAL-Q2 frequency response curve is not just a picture of the REAL-Q2 filters alone, as is the case with all other programmable devices that claim to "learn" or "compensate." A REAL-Q2 curve is the actual system response curve as heard at the reference microphone in the audience. That means acoustical architects, engineers, and designers can now specify the room frequency response curve _ not just at setup, but indefinitely, under all conditions.

Use the REAL-Q2 in any installation or live sound situation to make sure your audience hears the same high quality sound you work so hard to achieve during system setup. Call Sabine for a demonstration of the REAL-Q2 Real-Time Adaptive Equalizer.

How the REAL-Q2 sets up the room and maintains your EQ settings

1. SETUP MODE: Normally, it takes a good engineer hours to properly shape the response curve of a room using the best equipment. The REAL-Q2 does it in about eight minutes per channel. You select the desired response curve from Preset Curves, or define your own curve, and then select Initial Room Mapping. The REAL-Q2 sends a series of reference tones through the sound system where they are received by the reference microphone. The REAL-Q2 analyzes the relative loudness of each signal and automatically adjusts its Master EQ to match your desired response curve.

2. Using similar reference tones, the REAL-Q2 then "maps" the response of your room as heard at the reference microphone. The REAL-Q2 now has a complete picture of the entire audible spectrum of the room.

3. ADAPTIVE MODE: Once the program begins, the REAL-Q2 automatically utilizes sub-audible reference tones to continue the real-time analysis.

4. The REAL-Q2 is constantly analyzing the program and the reference tones via the reference microphone. These reference tones are completely inaudible to humans; only the REAL-Q can detect them. The relative loudness of the reference tones is compared with the mapped response curve. If there is a change, the adaptive EQ adjusts to maintain the original curve.

5. You can manually edit the REAL-Q2's curve at any time during the program, and you may "freeze" any of the REAL-Q2's adaptive EQ sliders if you don't want those frequencies updated. You can even recall and insert other pre-programmed response curves during the program.

6. Only the sub-audible reference signals factor into the analysis; changing the program level or equalization does not affect the signals. You keep full control over your mix, while the REAL-Q2 maintains your specified base-line room equalization.

The Secret of the REAL-Q2'S Sub-Audible Reference Signals

The REAL-Q2's patent pending algorithm employs a three-step process - Masking, Filtering, and Adapting.

MASKING: The first order of business is to hide the reference signal in the program. The REAL-Q2 constantly analyzes the program for an opportunity to check system frequency response. For example, in order to check at 400 Hz, the REAL-Q2 waits until the program material is rich in content near that frequency in order to mask the reference signal.

FILTERING: Once the masking frequencies are detected, the REAL-Q2 momentarily inserts an extremely narrow filter at 400 Hz, typically 80 dB deep. A 400 Hz signal is momentarily injected into the center of the filter, typically 60 dB below the average program level. Meanwhile, the reciprocal filter is placed in the reference microphone circuit, removing all program material. Now only the reference signal remains. This algorithm is so powerful, it accurately measures the reference signal even if it is buried 6 to 12 dB below the room's average noise level.

ADAPTING: The REAL-Q2 compares the level of the 400 Hz signal to the level stored at setup. If the level has changed, the REAL-Q2 makes an appropriate adjustment to the 400 Hz slider, in 1 dB steps. This EQ adjustment increment can be set by the user, up to 3 dB. This whole process typically takes about 300 milliseconds per measurement, and is repeated across the entire audio spectrum every 4 minutes. Typically there are three measurements per EQ slider to average out local anomalies like room resonance and reverberant field reflections.

The strength of the REAL-Q2 algorithm is its combination of simplicity and robustness. Since it generates and measures its own reference tones, it is independent of changes in the mix; in fact, the REAL-Q2 cannot hear the program material during analysis. Program level, signal processing, and even reference microphone placement do not affect the measurements. The REAL-Q2 operates in the background to constantly provide the correct room equalization - indefinitely.

REAL-Q2 Displays

1. Initializing the Real-Q2

Initializing the REAL-Q2 is a simple, automatic process that doesn't require a separate graphic equalizer or RTA. When you select Initial Real-Q Setup from the Main Menu, the REAL-Q2 walks you through each step of the process. In the beginning you are prompted to load your desired curve. The REAL-Q2 automatically calibrates itself to your reference microphone(s), equalizes the room to your specified response curve, then analyzes and memorizes the total response of the system and room. The whole process takes approximately 8 minutes, and at the end your system is equalized, you get a report on the inherent high and low end roll-offs of your system, and the REAL-Q2 is ready to maintain your specified response curve indefinitely.

2. Real and Adaptive EQ Screen

The Real and Adaptive EQ screen displays the REAL EQ, the Adaptive EQ, and your choice of frequency response curves. After the REAL-Q2 memorizes the REAL equalization as described above, the Adaptive EQ maintains that equalization. The screen at left shows a sample REAL-Q setup, with a flat REAL EQ (large empty squares), the modifications made by the Adaptive EQ (small filled squares), and the internal curve of the REAL-Q2 (shown as REQ in the soft key menu). You can also choose Room curve, or the Total curve of the entire system. The small arrowhead at the bottom of the screen points to one of the EQ sliders, and the line of data at the top of the screen shows the filter number, the frequency, and the Depth of the corresponding REAL and Adaptive EQ sliders. The soft key menus at right are switchable to allow you to move quickly between screens without going back to the Main Menu.

3. RTA

The Real-Time Analyzer displays the 31-band frequency spectrum for each channel. A powerful feature of this RTA is the simultaneous display of the Graphic EQ filters, shown at left (small horizontal lines at each frequency. You can view the RTA and adjust the EQ filters in the same window. In addition, you can adjust the vertical scale from a full 60 dB to 30 or 15 dB, and choose between Absolute/Normalized display, Peak/Hold, and Fast/Slow response. You can also enable either pink or white noise to each channel.

Simulated LCD screen shots. Screens not shown: Change Password, Global Parameters, Preset Curves, Hardware Setup, and About REAL-Q.

REAL-Q2 Features:

• 24-Bit Digital Resolution
• 31-Band, 1/3-Octave Digital REAL and Adaptive Graphic Equalizers
• 31-Band, 1/3-Octave Full-Featured Real-Time Analyzer
• Automatic Room Equalizing
• Multiple Memory Presets -- load curves on-the-fly
• Displays REAL EQ and RTA simultaneously for easy adjustment
• Pink & White Noise Generators
• Ultra-Bright LCD Display
• Dedicated Context-Sensitive Help Key
• Fast, Easy "Grab-a-Filter" Operation with Cursor Keys & Data Wheel
• Password Protection
• Automatic Clip Level Adjust
• True Input Level Metering: -30 to +30 dBV
• Inputs for up to Two reference microphones

The REAL-Q2 is perfect for:

Installations:

• Concert Halls
• Stadiums & Sports Arenas
• Churches
• Conference Rooms
• Schools
• Courtrooms

Live Sound

Indoor and Outdoor Performances

...Or any situation where consistent frequency response is important

The REAL-Q2 works within the sound field, independent of all other signal processing.

Engineering Specifications

Graphic Equalizers

• 31 digital filters on ISO center frequencies, width selectable from .5 to 1 octave in .01 octave increments, + or - 15 dB boost & cut, 20 Hz to 20 kHz
• Independent or simultaneous display and control of Master & Adaptive EQs
• Simultaneous display of RTA and Master EQ

Real-Time Analyzer

• 31-band FFT-based 1/3 octave digital analyzer, 20 Hz to 20 kHz
• Absolute/Normalized, Peak-Hold, and Fast/Slow display
• A, C, or Flat weighting
• 60, 30 or 15 dB full scale displays with adjustable positionFilters
• High Pass with cutoff frequency selectable in 1/3 octave intervals between 20 Hz and 1 kHz, 12 dB/octave roll-off
• Low Pass with cutoff frequency selectable in 1/6 octave intervals between 3.15 kHz and 20kHz, 12 dB/octave roll-off

Noise Gate

• Threshold: -1 to -95 dB (relative to peak amplitude)
• Release Time: 0 to 999 mSec.

Memory Presets

• 19 user defined curves
• 1 factory default
• 1 most recent configuration

Password Protection

Line Input/Output

• Input impedance: Balanced 10K Ohms nominal, Pin 2 high
• Output impedance: Balanced 10 Ohms nominal, pin 2 high
• Input/Output max. signal level: Balanced +30 dBV peak
• Bypass: True power-off bypass
• Output drive: +29 dBV peak into 600 ohm load
• Headroom: +25 dBV peak at 4 dBV nominal input
• I/O Connectors: XLR-3

Microphone Preamp

Balanced only with selectable phantom power

Performance*

• 24-Bit digital resolution
• Frequency response: <0.25 dB, 20 Hz to 20 kHz
• SNR: > 100 dB typical (with Noise Gate)**
• THD: <0.02% @ 23 dBV at 1 kHz
• Dynamic Range: >100 dB

Power Supply

100-240 VAC, 50/60 Hz, 75 Watts max

Dimensions

2-U rack-mount, 19 x 3.5 x 13 in. (48.3 x 9 x 33 cm); 18 lb. (8.2 Kg) nominal

Reference Microphonenot included

* Tests performed using Audio Precision System One model 322 or equivalent

** Signal-to-noise ratio is the ratio of the maximum undistorted signal by specification (26dBV RMS sinewave) to the noise floor