i. Example: The late, lamented NHT xD DSP-controlled speaker system was often compared (favorably) to the far more costly Wilson Watt/Puppy speaker system.
ii. Amplification products from TacT and Lyngdorf, while strong performers in their own right, become even stronger when coupled with their manufacturers’ DSP-driven room/speaker EQ systems.
iii. Example: speaker system from manufacturers such as Emerald Physics (various models), Legend (the Tikandi speaker), and Lotus (the Granada speaker) use DSP-driven crossovers to achieve results that have won acclaim from hard-core purists.
iv. Observation: DSP will never be a replacement for well-designed and well-engineered products and good listening room acoustic (it can’t “turn a sow’s ear into a silk purse”), but it can solve problems that traditional acoustical/electrical/ mechanical engineering methods alone cannot solve (for example, providing a method for having ultra-steep crossover slopes while preserving phase coherency).
a. Who pioneered it?
i. Tomlinson Holman (the TH in “THX”), professor of film sound at USC.
ii. Dr. Chris Kyriakakis, professor of electrical engineering at USC.
b. What’s special about it?
i. Works for two-channel or multi-channel applications.
ii. Rather than striving for “perfect” results in one listening seat, strives for “very good-to-excellent” results in multiple listening seats.
iii. Uses multiple room/speaker measurements, and does not average them, but rather applies fuzzy logic techniques to model the room/speaker interface as an identifiable set of problems that ear affect sound quality across multiple listening positions.
iv. Adjusts both frequency response and time response. creating custom correction curves for each channel (and the subwoofer, if any).
v. There are over 1000 EQ points applied per channel, with EQ points weighted by octave (more in the low end, fewer in the high end).
vi. Can accommodate multiple “target curves,” if desired, but typically at the OEM level (i.e., listeners cannot create custom curves of their own).
c. Pro’s and Con’s
i. Really works: noticeably smoothes frequency response and improves sound quality for non-sweet-spot listening positions.
ii. Can solve problems you might not think DSP could fix (e.g., corrects loose bass, tightens up reproduction of textures, can—under the right circumstances—increase apparent transparency, etc.
iii. Offers sufficient transparency to allow the “personality” of loudspeaker systems and electronics to show through.
iv. Some Audyssey add-ons seem pretty gimmicky.
v. For better or worse, buys into various elements of “THX dogma.”
vi. The standard Audyssey target curve will not please everyone.
a. Developed in Denmark
b. What’s Special About It?
i. HolmImpulse freeware—measures frequency and impulse response
ii. Holm Mic 1 calibration microphone
iii. DSPreLab software—similar to HolmImpulse, but fully integrated with DSPre 1 hardware product
iv. Works with full-range (so-called “1-way”) speakers, or can serve as combination correction/electronic crossover system (for so called “2-way” or “3-way” active speaker systems).
v. Correction schema—measure at 50-100 cm from speaker (user control measurement/correction range). Goals are to correct:
1. Impulse response
2. Frequency response
3. Phase response
4. Big picture: build a “Flat Filter” for your speaker.
vi. Graphic EQ system allows elaborate “target curve” reshaping options
vii. Room Adaption (sic) system measures at sweet spot and four other locations (sweet spot, left, right, above sweet spot, and in front of sweet spot)
1. Corrects channels individually
2. Can use averaging, but user controls percentage of sweet spot priority
3. Adaption typically stops 150Hz (can be expanded up to 5kHz)
4. Can vary adaption from full to nil, so listeners can add correction gradually.
5. Can set max. gain/cut limits.
6. Can control “aggressiveness” of correction (making correction curves more or less fine-grained)