Music in the Round

The other option is to connect the USB-XIII and the Merlots directly to the Baetis server via USB and use Playback's ASIO driver (footnote 3). This worked beautifully, and quickly became my preferred configuration for ease of use and musical enjoyment. It also let me make A/B comparisons with the exaSound e38 D/A processor by linking the outputs to the two DAC setups and using input switching on the preamp. The differences weren't huge, and less than I'd expected, but they were in favor of the Playback setup. Soundstage detail was equal, but the exaSound's was a bit more forward; the Merlots' seemed more naturally arrayed. The exaSound's bass was a little emphatic compared to the Merlots' more rounded depiction. The tonal balance remained unchanged.

Overall, the Playback Designs combination of Syrah server, USB-XIII interface, and three Merlot DACs was completely as advertised: beautiful sound, uncomplicated setup and use. In addition, it's network accessible, and if you already own a compatible server, you can use just the USB-XIII and the Merlots. If you already own a Merlot DAC, your upgrade path to multichannel is clear.

AudioQuest Niagara 5000 Low-Z Power/Noise-Dissipation System
I've long been skeptical of the audible advantages claimed for power-conditioning devices, power cords, speaker cables, and interconnects. This is not to say that better cables make no difference at all, but that, for me, their most important functions are durability and security of connection. I've found any sonic differences among them to be elusive, dependent on room and system, and so much less significant than the differences in sound produced by changes of electronics, speakers, and room acoustics that paying attention to them isn't worth my while. Power-conditioning devices offer clearly useful features: power distribution, protection from line noise and power surges, and battery support for power sags or outages. But differences in sound? I haven't been convinced.

So when Stephen Mejias, then AudioQuest's VP in charge of Communications (footnote 4), showed me the company's Niagara 5000 Low-Z Power/Noise-Dissipation System at last fall's CEDIA Expo, I was fascinated by the glossy box, but only as a snazzy power distributor. It looks just like the impressive Niagara 7000, with four of AQ's High Current/Low-Z Power outlets (for demanding loads such as power amps), and eight of the company's Ultra-Linear Noise-Dissipation outlets (for sources). At half the price and less than half the weight of the Niagara 7000, the Niagara 5000 differs from it primarily in lacking the Dielectric-Biased AC isolation transformers on the eight noise-dissipation outlets. Michael Fremer positively reviewed the Niagara 7000, and Herb Reichert very positively reviewed the Niagara 1000. I refer you to their full reviews for technical information. However, despite those published encomia, it wasn't until I bumped into Mikey and Herb at an audio event that my curiosity got the better of me. They didn't want to discuss MQA (the ostensible topic of the event), vinyl, or even music—all they wanted to talk about were their experiences with the Niagaras.

My main component rack contains all of my source, processing, and communications components, and one stereo power amp for the rear channels. For this, the 12 outlets of a single Niagara 5000 ($3995) are just right. But at the other end of my room are, usually, four mono power amps and two subwoofers, so even a second Niagara 5000 wouldn't be just right: Of its 12 outlets, only four are High Current/Low-Z and thus suitable for power amplifiers, monoblock amplifiers . . . or powered subwoofers. Stephen and I concluded that, to optimally use the Niagara system, I needed a Niagara 5000 at the equipment rack and, at the speaker end of the room, another 5000 and the single High Current/Low-Z outlet of a Niagara 1000 ($995). Moreover, AQ's Garth Powell maintains that, to get the best from the Niagaras, they should be the only power devices in the system, and should be connected to AQ's NRG Edison AC outlets ($149 each) via AQ power cords.

You see where this is going. I hired a local electrician to install NRG Edison AC 20A receptacles in the dedicated-line wall boxes at each end of the room. AQ's instructions are clear enough for anyone who's installed ordinary receptacles, but the small cost of installation was worth not having to do it myself. AQ also supplied an NRG 10 power cord ($1099/3m) for the Niagara 1000, and Hurricane power cords ($1599.95/2m, $1149.99/1m) for the two Niagara 5000s. Installation was a matter of unplugging everything, then replugging it all into the Niagaras, which was simple but not easy: All connections into and out of the Niagaras and Edisons were secure to the point of requiring young Stephen Mejias's brute strength.

The overall cost of the Niagaras, cords, and outlets—somewhere in the neighborhood of $13,000—led me to expect great things. But the simultaneous installation of such a multiplicity of new power devices means that I can comment only on the effects of the total AudioQuest system, not on effects of any individual AQ model. That said, compared to the decent wiring but negligible power conditioning I had before, the AQ Niagara system really did improve my system's sound.

With the volume control at a normal listening level and no music playing, I heard nothing different at my listening position. However, when I placed an ear close to each speaker's drive-units, the noise from the tweeters was reduced, and the noise from the woofers was now completely inaudible. This may seem inconsequential, but it isn't. My normal volume setting is the result of determining a comfortable, realistic sound level that also permits hearing sufficient low-level detail. System noise competes with the resolution of low-level detail, and often drives us to turn up the volume just to appreciate the music's rich complexity. One consequence of the lower system noise with the Niagaras is that my new normal listening level is 3–4dB lower than before. This is a satisfyingly real change but not a huge one—it takes an increase in sound-pressure level (SPL) of about 10dB to double a sound's perceived loudness.

However, lowering the reference level by 3dB is the equivalent of adding 3dB of power headroom to the system, which effectively doubles the usable power. Each of my Classé Sigma Mono amps is specified to output 350W RMS into 8 ohms or 700W RMS into 4 ohms, so I already had plenty of power on tap—the extra headroom didn't affect my listening at normal levels. Sometimes, however, I like to revel in a massive sound, and play stuff at high levels just because I can. Now when I do that, my system seems to have no limits at all—the only things that constrain me are fear of damaging my hearing and bothering my neighbors. These phenomena may be psychophysiological, but they're also real.

I love what the Niagara 5000s and 1000 and the rest of the AudioQuest kit have done for the sound of my system, though it comes at a considerable price. However, most audio systems, even multichannel ones, aren't as complex as mine—in those, a single Niagara 5000 with an NRG receptacle and power cord should produce similarly gratifying results. The security of nondestructive surge protection and the convenience of solid power distribution were expected comforts; the real pleasure of the Niagara 5000, et al, has been in the sound—and that I didn't expect.

---

Footnote 3: I also tried this with Roon, but apparently, Playback's ASIO driver, which worked with JRiver, is not yet compatible with Roon.

Footnote 4: AudioQuest, 2621 White Road, Irvine, CA 92614. Tel: (949) 585-0111. Web: www.audioquest.com