The TiDEQ can be configured as a single-channel crossover, but not stereo. In this configuration, we tried our crossover summation test. This simulates the two-way acoustical summation of an ideal speaker system. The 12-dB/Octave crossovers had a typical dip at the crossover point, typical of an analog crossover. But this could be corrected to 1.5 dB by inserting the right amount of delay. The 24 dB/Octave crossovers summed to an excellent 0.5 dB without using delay. This is a nice advantage over analog crossovers.
Each of the 30 filters plus the three parametric filters is capable of +6, -18 dB boost/cut. This non-symmetrical parameter helps avoid clipping problems and allows the input level to remain high. Using a lot of boost would force very low input levels, which would reduce the digital resolution.
GeneralCrossovers-To use the system as a digital, two-way crossover, you need two of them. Phoenix Gold doesn't claim to incorporate crossovers, just adjustable high-pass or low-pass filters. While the 24 dB/Octave slope of the high pass filter can be used for a subsonic filter, a steeper slope would be desirable. This unfortunately is a standard slope amongst most subsonics filters.
Input level adjustment-There is no adjustment of input level. If the input is set too high, the unit clips. While 4 volts is the maximum level, signal levels at 3 volts could cause clipping if too much boost is used in the equalizer settings.
Output Attenuators-When we forced the unit to clip with the output attenuated, the clipping LEDs failed to come on. We recommend leaving the output attenuators at 0 dB to prevent this situation.
Source impedance-Our measured noise levels increased by 5 dB, going from a source impedance of 20 ohms to 600. Using a low output impedance deck is desirable.
GUI-The graphical user interface works well. The only concern is the 600 x 800 pixel resolution may not be usable with some on-board displays.
Q-This is an excellent idea and makes for a very versatile equalizing process. Unfortunately, Phoenix made a mistake with their Q definitions. Their reported Q of 7 should really be 4.2. That means that all of the other Q measurements are off. This doesn't affect anything sonically, but they need to revise their software and owner's manual.
DSP Scaling-Normally the volume control is ahead of the TiDEQ. With normal listening levels, the input may be in the range of 15mV. This forces the equalizer to work with much less resolution, such as 15-16 bits. For the purist, I'd recommend controlling the volume after the TiDEQ to improve SNR and distortion, while keeping the digital signal near maximum.
Size-The unit measures in at 14" L x 10.25" W x 2.25" H. A little large, but it does house 68 filters, plus time delay!
Listening TestOur listening test revealed little problems. Sound quality was good otherwise; no anomalies were detected.
A very slight audible hiss was heard with the amplifier set to 1-volt sensitivity even though our SNR measurements were above 102 dB.
The majority of the noise issues were detected while making adjustments. With the computer connected to the RS-232 interface, we also heard motor noise from the CD player. Disconnecting the computer eliminated this source of noise. Also, isolating the computer from the 12 volt ground solved this noise problem as well. Not to worry.
Another noise problem was also noticed with the computer connected. Changing the filter settings via the Windows program caused a quiet chirping noise, but disappeared with ground isolation.