Arcam Alpha 9.

Distributor: Audiophile Systems, Ltd., 8709 Castle Park Drive, Indianapolis, IN 46256; 888/272-2658

Price: $1,599

Source: Distributor loan

Reviewer: Karl W. Nehring

In 1996 Arcam introduced a range of upgradeable CD players, starting with the Alpha 7, which was upgradeable to the next model in the lineup, the Alpha 8. Both of these models have been reviewed previously in these pages. The Arcam Alpha 9, subject of this review, is the culmination of this progression, and is presented by Arcam as a "24-bit" CD player. The main feature of this unit is its digital-to-analog converter, which is based on a new DAC design, the dCS Ring DAC. Cosmetically and functionally, the Alpha 9 is identical to the Alpha 8 and 9, and it shares the same transport mechanism (Sony CDM-14) and mother board.

[ILLUSTRATION OMITTED]

The dCS Ring DAC is said to be based on a high speed autocorrelation DAC architecture originally developed in the UK in the 1980s for military aircraft radar. The device is said to be capable of handling 24-bit digital words at up to 192 kHz sampling frequencies. In addition to this new DAC, the Alpha 9 also features the Pacific Microsonics chip set for HDCD capability.

I had mixed feeling about the aesthetics and overall "feel" of this unit as I unpacked it. For nearly $1,600, I expected to be deeply impressed by the look and feel of this unit, but when I took it out of the box, I was a bit underwhelmed. The unit looks clean and functional, but does not give the impression of being particularly heavy duty. The drawer, for example, seemed little more confidence-inspiring than that in the $300 Weltronics unit that I have been using now for several years. However, the look and feel of the unit grew on me over time. It is certainly clean and functional, and not at all flashy.

I inserted the Alpha 9 into a system that comprised the AVA Fet Valve EC preamplifier, AVA Fet Valve 550hc power amplifier, and Legacy Focus loudspeakers. The first thing I did, after putting in a music CD just to make sure everything was working properly, was sit down and listen for a while to see whether any particular aspect of the sound would jump out at me as an improvement over the inexpensive Weltronics CD8. Well, the sound was just fine, thank you, but there was nothing that leaped out at me as any kind of big improvement. I know that the writers in some of the audio magazines are "immediately blown away" by the sonic splendors they claim to hear when they plug an expensive piece of gear into their systems for the first time, but I suspect I am a wee bit more circumspect than some of these fellows.

At any rate, I figured that any sonic improvements would have to reveal themselves under careful comparative listening sessions later on down the line. For now, I just wanted to let the Alpha 9 play for a while so I could get used to its operation.

Although I heard no immediate sound improvement when I put the Alpha 9 in my system, I soon discovered an area in which its performance was head and shoulders above that of the humble Weltronics unit. Using the CD-Check disc recommended by Robert Thompson, I discovered that the Alpha 9 was able to sail through tracks 2-4 without a glitch, making strange sounds only on Track 5. (By contrast, the Weltronics, although it would actually play all five tracks, started making funny noises on Track 2. This is not very good performance, although I must say that I have never noticed any tracking problems on any of my music CDs, even the old ones).

Some readers at this point might be tempted to think that of course a brand-new $1,599 player would perform better than a three-year-old $300 player on this test. I invite those folks to skip down to the review of the similarly priced ($1,495) Parasound CDP-2000, which did no better on this test than the Weltronics. In fact, this level of performance is on a par with the CD player in the dash of my Honda Odyssey and the under-S100 portable boombox on which I play CDs in my office.

Once again we have confirming evidence for the old hypothesis that money cannot buy you happiness -- in fact, in the case of the Parasound CDP-2000 it does not even necessarily buy you respectable error concealment/ correction performance. But in the case of the Alpha 9, it does.

Those who read Robert Thompson's series of articles about CD player performance in previous issues will already have some familiarity with error concealment/ correction issues, but for those who have not read those articles, the following overview by RT may prove useful. Allow me to put RT's remarks in context by explaining that I had asked him why players that did so lousy on the CD-Check Disc (such as my boombox, car CD player, and the Parasound belt-drive unit) seemed to play any music CD that I would stick in them with no problem. This was his reply to that query:

RT: Some players will "play" through Track 5 or other tracks (on CD-Check), even though they are producing audible errors on those particular tracks. Other players when confronted with so many errors will "lock up" or go into a continuous "loop-mode," it is very player model/ brand dependent. I am surprised that you don't have a least a few CDs in what I assume to be a large CD collection that cause your players some problems, (I'm pretty sure that it's large, since you also do music reviews as well!). Perhaps this is a testament to the way you store and handle your discs vs. the way others do, i.e., a person who leaves discs lying around in a car without their protective jewel cases, etc.

The vast majority of discs in anyone's collection will be perfectly reproduced by the vast majority of CD/DVD players out there. Even players that can't make it cleanly through Track 2 of CD-Check can reproduce without audible effects a large majority of music CDs. However, CDs that aren't in good condition will allow a player that has better tracking capabilities to shine above another with inferior tracking capabilities.

CD information is grouped and processed in "blocks" or "frames" containing 192 bits of audio information each. Some 7,350 blocks of data are scanned by a CD player every second it is playing a disc. If only one bit out of a block is erroneous, the whole block is considered to be in error, called a "block error." Dividing the total number of block errors for a disc by the number of seconds in the disc's playing time yields a figure of merit for pressing quality called the "block error rate" (BER), measured in blocks per second -- the lower the figure, the better. A typical CD in good condition has a BER of 30. The best in a published survey of 30 discs had a BER of 6.7. Even the longest available CD from that group, Beethoven's Ninth Symphony on Denon, had a BER of only 11.2.

A typical new or well-cared-for CD generates about 81,000 block errors in 45 minutes. In David Ranada's experience, all of these errors are exactly corrected, (see the April 1984 issue of Stereo Review starting on page 48 for further information - - much of what I've quoted here comes from that article). There are no concealments -- the audio information is all perfectly reconstructed. The BER has to be VERY high (over 100) to increase the probability of concealments, and even then interpolations are not likely unless there is some damage on or in the disc that is visible to the naked eye. Thus, CD error rates (for new and well-cared-for-discs) are low enough that there is no reason for any concern about their audible effects.

You may wonder, then, why I'm concerned and wrote a series of articles about this subject. Well, the reason is that not all people keep their discs in "good condition." Eventually, every owner of a player will want to play defective or damaged discs at one time or another. The ability of a player to play such discs without audible errors is one of the most important factors that distinguishes one player from another.

Well, enough of that digression. Suffice it to say that the Alpha 9 did well on the CD-Check disc, much better than the Parasound or than the Weltronics. But how did it do in comparisons to these other units when it came to playing music?

To answer that question, I first used several CDs of which I have duplicate copies -- both regular and HDCD-encoded -- to compare the sonics of the Alpha 9 with the Parasound. Although there were times during casual listening sessions when I thought that the Alpha 9 might have just a bit more smoothness in the treble than the Parasound, under more controlled comparisons I could detect no sonic difference between the two players. The Alpha 9 was, however, appreciably faster to start up, faster to cue tracks, and at least for my tastes, more accommodating to use with its front-loading drawer mechanism than the top-loading Parasound and its rather silly disc stabilizer. Add to those advantages the Alpha 9's marked superiority in error concealment/correction, and the choice between it and the Parasound seems fairly clear, at least if I were spending my money.

I also compared the Alpha 9 with the Weltronics, using non-HDCD discs. As much as I would love to be able to tell you that my golden ears could hear the difference wrought by the more elaborate and expensive circuitry in the Alpha 9, I can't. Using identical discs, and matching levels as carefully as I could, I could not pick one unit over the other on non-HDCD discs.

For the extra money over the budget Weltronics, the Alpha 9 offers HDCD capability, considerably better error concealment/correction, a better display, and superior fit and finish. Whether these advantages are sufficient to justify the considerably higher price tag of the Alpha 9 is a subjective judgment that the prospective purchaser will have to make for him- or herself. -- KWN