Music in the Round #45 Page 2
As you might expect, the amount of signal processing and number crunching involved is not trivial. Setting up the Realiser A8 includes physical installation and "personalization": measurements of the user's HRTF, headphone/ear calibration, and head-movement tracking. If there's a fly in the Realiser ointment, it's the complex and seemingly obsessive process needed for successful setup and calibration. The 90-page owner's manual (download only, from www.smyth-research.com/dload.html) is methodical and detailed, but rife with obscure acronyms. A glossary and/or index would be greatly appreciated. I could have muddled through, but fortunately, my Realiser was accompanied in person by Smyth Research's Lorr Kramer, who guided me to a successful outcome.
Physical Installation: The package includes the Realiser A8 processor box and power supply, the RC-1 remote control, a TU-1 Head Tracker to be attached to the top of the headphones' headband, a TR-1 Head Tracker Reference for 0° reference, two HTM-1 miniature in-ear microphones, and a set of Stax SRS-2050 II headphones. The A8 has eight channels of analog RCA inputs/outputs, and is inserted between the preamplifier-processor and the power amps. I used the first six ins/outs for 5.1-channel signals. In addition, it has stereo line outputs, analog and digital stereo headphone jacks, a stereo microphone jack, outputs for a tactile transducer, and various other connectors, including those for the head-tracking function. The Realiser cannot accept a digital input.
The Stax 'phones were connected to the stereo line outputs, but other 'phones can be driven from the A8's dedicated headphone jacks. The set-top TR-1 Head Tracking Reference straddled the tweeter housing on the center-channel B&W 802D loudspeaker in my Manhattan system, and perched atop the plasma display of my Connecticut system. The TU-1 Head Tracker was clipped to the headband of the Stax 'phones. When the A8 is powered down or in standby, its inputs and outputs are directly connected by relays to offer a true and transparent bypass.
Personalization: First, each speaker's lateral and vertical angular positions are entered into the A8. Smyth's Lorr Kramer had a nice laser level for this, but a tape measure and a little geometry would work as well, if less quickly. Then, with the little microphone buds nestled precisely in my ears, Kramer began to calibrate the channels one by one, followed by the computation by the A8 of each ear's HRTF and each speaker's Personalized Room Impulse Response (PRIR), and finally, the equalization of the headphones (HPEQ), to ensure that the PRIR was accurately implemented. During the PRIR procedure, to set up the head-tracking system, I was instructed to turn my head to look at each of the three front speakers. Note that the Realiser A8 does not equalize or correct the main system's performance, but some tweaking is possible post hoc. Finally, the PRIR and HPEQ data sets are independently named and stored. The A8 can store up to 64 PRIR and HPEQ files, which can be called up and applied in any combination.
Revelation: All that done, I sat with the 'phones on while Kramer stored the data. All of a sudden, Smyth's test loop of synthesizer music, which is stored in the A8, came bopping out of the left front speaker. "Okay, that's the left speaker," I saidthen realized that I was still wearing the Stax 'phones. Kramer said, "But it's not on bypass." I lowered the headphones and he switched to bypass. It sounded exactly the same. I put the 'phones back on and we repeated the comparison from speaker to speaker. Dammit, it always sounded as if it was coming from "that speaker right there," and not from the 'phones. Back and forth we went, 'phones on and off my head, bypass switched in and out. I was like an infant playing peekaboo: surprised every time. I couldn't believe it.
Realisation: I couldn't wait for Kramer to leave: I wanted to listen to music over headphones. For the first time in my life, headphone listening was not only convincing but enjoyable. The Realiser A8 provided the same balance and soundstaging as my main system. Multichannel recordings were so compelling that I periodically had to remove the 'phones just to convince myself that they were the actual source of this wonderful illusion. Of course, no small part of it was the quality of the Staxes themselves. I blew the dust off my old Grado SR-60s, which indeed were able to re-create the same soundstage, but in direct comparisons the Staxes much more accurately reproduced the harmonic texture of the B&W speakers.
Dmitri Kitayenko and the Gürzenich-Orchester Köln's solid and powerful recording of Tchaikovsky's Manfred Symphony (SACD/CD, Oehms OC 665), and other good but conventionally mixed multichannel recordings, sounded so similar through the Realiser A8 and my main system in full glory that both experiences were completely satisfying. More immersive mixes that place the listener among the performers, such as Matthias Jung and the Saxony Vocal Ensemble performing motets by J.S. Bach (DVD-Audio, Tacet DVD 108), could be positively hair-raising. In fact, the SVS Realiser A8 seemed to offer slightly more stable imaging with such recordings, something I noted particularly with Willie Nelson's Night and Day (DVD-A, Surrounded By SBE-1001-9), where individual instruments are discretely placed around the listener.