Parasound Halo CD 1 CD player
In his August 2008 Follow-Up to Jon Iverson's and Kalman Rubinson's coverage of Sooloos's Source:One (which was the component in the Sooloos system of the time that output an analog audio signal), John Atkinson wrote: "Looking at the [inside of the] Source:One, at the front center you have the network interface card; on the left you have a switching power supply; at the rear center is an industrial PC with its familiar but unused I/O jacks at the front and its vertically mounted RAM board at the back; and at the right is the RME Hammerfall soundcard plugged into the PC's PCI slot. . . . The significance of this photo is that what ostensibly is an audio component is actually a PC dedicated to one specific function. I think this is increasingly what we will see in high-end audio components."
The JA of 2008 was prescient. For the high-performance audio market, it makes a lot of sense to process digital audio data via sophisticated software running on a dedicated personal computer.
Which brings us to Parasound's Halo CD 1 CD player ($4500). Some might find it questionable to release today, as one's first digital-disc player, a machine that plays only "Red Book" CDs, rather than a universal or near-universal (nonBlu-ray) player. It is especially questionable given the market's apparent acceptance of expensive players based on one of Oppo's inexpensive universal players.
That said, I think Parasound's approach will work for music lovers whose large music collections consist mostly of CDs, who want to approach the best CD sound available today, and who aren't prepared to go whole hog into computer audio or a dedicated music server such as Meridian's, formerly Sooloos's, Control:15, both of which require an external DAC. By concentrating on CD playback and ignoring SACD, DVD-Audio, Blu-ray, and even MP3, Parasound can max out the CD 1's CD performance and deliver the goods at the relatively reasonable price of $4500reasonable, that is, for a product that aspires to be "the best."
Like the other Halo models, the CD 1's robust, full-width case is available in black or what I call champagne silver. The review sample came in black. The overall The front panel has a disc slot (no drawer), a vacuum fluorescent display window, an On/Off switch, a button to switch analog output stages between one based on op-amps and one based on discrete components, transport controls consisting of large oval buttons for Play/Pause and Stop/Eject, and smaller round buttons for forward and back Search and Skip. The rear-panel connections are an IEC power cord inlet; an S/PDIF digital output on RCA, BNC, and TosLink (optical) jacks; and single-ended and balanced analog outputs. There are no digital inputs to allow the CD 1 to be used as a DAC. Remote control is standard; the Polarity button toggles polarity between 0° (Normal) and 180° (Inverted). All connectors are robust and of top quality, and the power cord is unusually hefty.
The CD 1 supports neither AES/EBU digital output nor word-clock sync. Parasound's Richard Schram doesn't think much of the AES3 digital-on-balanced-cables standard, which he believes is the result of an engineering compromise in the early days of digital intended to allow European broadcasters to save money by using their installed base of balanced analog microphone cables as digital-audio cables. Schram states:
"A connector carrying high frequencies must have mechanical dimensions that give it the same impedance as the impedance of the cable. The AES/EBU standard's XLR connector is simply wrong for digital audio, in all mechanical respects. The XLR was developed for analog audio, and frequencies lower than digital. When connector and cable impedances are mismatched, the result is reflectionsechoes of the signal travel back and forth, corrupting the real signal. In digital audio this results in jitter, and ringing near the signal edges, which are the transitions between a 0 and a 1. The longer the cable and the higher the sampling rate, the worse the problem."
Fascinating. Barring a deathbed confession by a guilt-ridden EBU apparatchik, I think we'll never know the inside story. But I did think it important to explain that Parasound's decision not to include an AES/EBU balanced digital output jack was not penny-pinching on their part. Anyway, I am aware of no D/A converter that has an AES/EBU XLR digital input but no RCA digital input. Furthermore, I think the inclusion of a BNC jack more than makes up for the absence of an AES/EBU jack.
Inside the CD 1 are a computer optical-disc drive (CD-ROM), an R-core transformer, an analog power supply, an Intel single-board computer running Linux and DSP software from Holm Acoustics of Denmark, and a DAC board, apparently connected to one of the motherboard's USB sockets. A nearby USB thumb drive may contain the computer's own firmware. There is no cooling fan.
Features include two digital power supplies, transformer isolation for the RCA and BNC digital outputs, a linear power supply for the analog section, a choke for AC line filtering, and internal partitions of 3/8"-thick aluminum, all of which are intended to reduce noise. Parasound claims measured jitter of less than 10 picoseconds, which is remarkably low.
The CD 1 spins CDs at four times the usual speed, using the Holm Acoustics software; every bit is read at least twice, and the results must match, before being sent on to the memory buffer. Parasound claims that their proprietary software enables the CD-ROM drive to spin more quietly. I have no idea how that works, but I found the CD 1 to be very quiet.
Reading a CD's Table of Contents takes about seven seconds, so there's a bit of a lag between loading a CD, pressing Play, and hearing music. In the event of an error, the data is re-read several times before the system gives up and begins to interpolate its own approximation of the missing data. Either way, the digital data are played (or sent out via S/PDIF) from solid-state memorywhich means that the Halo CD 1 can be called a memory player. Parasound's product literature seems to indicate that the buffer can hold 30 seconds' worth of music.
I had not before sought out a memory player to review, because I wanted a one-box player that also would be significantly less expensive than what was previously on offer. I believe Parasound can accomplish that owing to having its products manufactured in Taiwan, and the economies of scale enabled by its established dealer base (footnote 4).
Parasound uses Analog Devices' AD1853 DAC chip. Parasound didn't take the approach of balanced dual-differential, complementary digital architecturenot for reasons of cost, they claim, but because they believe balanced digital architecture creates timing errors between channels. After being read, the digital data are upsampled to 352.8kHz. Neither the white paper available from Parasound's website nor the owner's manual mentions the digital filter's being a minimum-phase filter, so I conclude that it is not.
It didn't take too much listening for me to conclude that, for playing regular CDs, the Parasound CD 1 was quite similar in essential sound to Bricasti's justly fêted M1 DAC: Detailed yet smooth, fast yet not fatiguing, and with wonderful frequency extension at the bottom and top. This is as it should bethe Parasound costs about half the Bricasti's price, yet it is limited to CD playback and has no digital inputs or selectable digital filters (including minimum phase), or a volume control.
I think the Bricasti and the Parasound are fairly priced for what they are; the question is whether the Parasound's limited design brief meets your needs (or whether you're prepared to pay for more than one digital solution: eg, an affordable DAC such as Arcam's Sonlink, for lower-resolution, quasi-background music from Sonos and Pandora; and a higher-priced solution such as the CD 1, for when you do nothing but listen.
In addition to the Beethoven string-quartet CDs I mentioned in my last two columns, I listened a lot to the two CDs with which I began this month's columnThe Bespoke Man's Narrative (CD, Mack Avenue MCD 1066), Richard Strauss's Zarathustra paired with Strauss's Don Quixote, with cellist James Kreger (CD, Guild GMCD 7204)as well as to Requiem for a Pink Moon: An Elizabethan Tribute to Nick Drake, by lutenist and singer Joel Frederiksen and Ensemble Phoenix Munich, and many of the favorite CDs I've often written about here. I experimented with the CD 1's selectable output stages, and to my surprise had a preference for the Op-Amp setting that was so slight as to perhaps border on the imaginary.
To directly compare Parasound's Halo CD 1 and Bricasti's M1, I used The Bespoke Man's Narrative and Mary Black's No Frontiers (CD, Gift Horse G2-10002). It was close, but the Bricasti was clearly superiornot so much in tonal quality as in the solidity and dimensionality of the soundstage, that elusive quality of "thereness." The Bricasti filter that sounded closest to the CD 1 was Minimum Phase 0.
If your budget has few or no constraints, it might interest you that the Halo CD 1 performed magnificently as a transport for Bricasti's M1 DAC ($8595). No knock on Musical Fidelity's comparatively affordable M1CDT transport ($999). That said, as a transport, the Halo CD 1 produced a sound unquestionably larger in dynamic scale and in space, as well as having a lower noise floor, and more quiet between the notes. The Musical Fidelity M1CDT was unquestionably more musically engaging than my cheap'n'cheerful Denon Professional DN-961FA, a retired radio-station player, when I used it as the transport. (I've since sold the Denon on eBay.)
I think that Parasound has fulfilled its design goals for the Halo CD 1. If you're prepared to keep playing your CDs and don't care about futureproofing, this is a fantastic player for reasonable money, given the high qualities of its engineering and parts. A quickly rebadged Oppo the CD 1 is not. My vote: Class A. John Marks
Footnote 1: My review sample of the Halo CD 1 was one of the first 25 production units. It suffered some shipping damage on the way here, evidenced by a mechanical 60Hz hum from the power-supply section. I suspect that the force of the impact loosened or relocated a power-supply component enough that it began to buzz as a result of having been knocked into an electrical field. I returned the unit to Parasound. On learning that a hard blow could have this result, Parasound redesigned the power supply around a larger inductor, revised the parts layout, applied these changes to the PCBs of the review sample and all other production units, and returned the sample to me. It performed flawlessly .John Marks