Index to this Issue Home Page

 

Product Review - A Comparison of Three Outboard Digital-to-Analog Converters: Bel Canto DAC-1, MSB Link DAC, RAudio DAC-0 - January, 2001



Paul Knutson


Divider

 

Bel Canto DAC-1

Digital Inputs: RCA type and Toslink

Output Level 2Vrms

Dynamic Range 120 dB

Input Sample Rate 32 kHz - 96 kHz

Size: 3 1/2" H x 9" W x 3 1/2" D

Weight 3 Pounds

MSRP $1,295 USA

E-mail: info@belcantodesign.com

Website: www.belcantodesign.com

 

MSB Link DAC, Modification by Dan Wright

Price: $399 for MSB DAC plus ~$200-250 for upgrades

MSB Website http://www.msbtech.com 

Dan Wright E-Mail modwright@yahoo.com 

Dan Wright MSB Mod website http://themorgue.org/audio/msb-link/ 

 

RAudio DAC-0 by Whitley Ltd.

Digital Input: RCA Type

Output Level 0.5Vrms

Dynamic Range 120 dB

Input Sample Rate 32 kHz - 96 kHz

Size 3" H x 6" W x 8 1/2" D

Weight 3 Pounds

MSRP: $2,700 USA

E-Mail raymond@raudio.com 

Website: http://www.raudio.com 

 

Introduction

One particular shelf of my equipment rack is reserved exclusively for the digital-to-analog converter (DAC).  The constant ďcomings and goingsĒ of DACís on this shelf resemble the action at In-and-Out Burger on a Saturday afternoon.  These frequent changes are a direct result of the plethora of noteworthy, affordable DACs on the market today, and my corresponding desire to hear them and tell you about them.  Having so many superb, relatively affordable DACs to choose from is not a bad problem.  It does, however, increase the difficulty of deciding which one will allow you to enjoy your CDs the most over the next few years as next generation format wars rage on between the behemoth electronics companies.

Since one of my goals is giving credible reviews, I try to maintain a complete reference system so that anything coming in for review can be evaluated within a consistent context of high-performance associated equipment.  My reference components must be able to stand the test of time.  Since the rest of my system has been so stable, the lack of a ďreferenceĒ DAC results in some mild anxiety.  (As if we audiophiles donít have enough reasons to be anxious already.)  If I had to choose one DAC for my reference right now, it would be the Perpetual Technologies P-1A/3A combination, even though it hasnít been in my system too long.  I really admired that combo when I first reviewed it a number of months ago, and its performance continues to impress me.  To refresh your memory of what Stacey Spears, John Johnson, and I each had to say about the Perpetual pair, go back and check the review.  Iím lucky enough to still have the P-1A/3A on hand and it served me well as a basis for comparison at different points during this review.

The problem is that choosing a reference DAC from amongst todayís offerings is in many ways like touring Napa Valley in search of the ďperfectĒ Cabernet.  Typically, the process involves sampling as many as possible, noting that nearly all are wonderful in some way, and ultimately choosing the one that appeals to you most based on your own personal tastes.  But how do you know that another winery wonít release an even better bottling a month from now?  The answer is Ė you donít.  Ugh! 

The lack of a stable reference DAC is a dilemma that probably faces many of you reading this.  Believe me, I sympathize with the audiophile angst that lingers out there in todayís rapidly evolving digital audio market.  I read about it all the time on various internet newsgroups.  A fair amount of e-mail comes my way from readers who are clearly riding on the DAC merry-go-round.  Frankly, they have no idea when to jump off.  Despite these challenges, weíre in the midst of exciting, evolving times in digital audio, and here at Secrets weíre glad to be covering it for our readers. 

Iíve recently had the great privilege of listening extensively to three notable DACs Ė the Bel Canto DAC-1, RAudio DAC-0, and a MSB Link DAC modified by Dan Wright.  As I received these DACís for review, there was something about each that thrilled me Ė sonically, technically, or both.  My former ďreferenceĒ DAC of a couple years ago, a Stan Warren modified Monarchy M-33, was a fine performer, or at least I thought.  Not surprisingly, however, each of the DACís Iíve just finished evaluating bettered my old (and long since departed) reference DAC in many important ways.  The three DACís mentioned above are the featured attraction of this comparison review.  

Introducing the Contestants Ė Technical Overview

Iíll provide some background on each DAC in this survey, but in lieu of covering ground that is well-trodden, especially with respect to the Bel Canto and MSB, Iíll keep my technical overview of these DACs as relatively short and sweet as possible.

Bel Canto DAC-1

In designing the $1,295 DAC-1, Bel Canto eschewed making a strong visual statement in favor of making a strong audible statement through the innovative application of the latest digital conversion technology.  Itís evident that much design forethought went into this 24/96 upsampling DAC.  Incidentally, itís the first DAC from Bel Canto, and it makes a rather auspicious debut.

The Bel Canto DAC-1 features separate digital and analog power supplies, 24/96 upsampling via the popular Crystal CS8420 sample-rate converter chip, and digital conversion via the Burr Brown PCM1704 DAC.  From there, one of its more important design elements comes into play as the digital filter begins a slow-rolloff at 48Khz, which is said to provide a more natural sound than the dreaded ďbrick-wallĒ 22 kHz sharp-rolloff filters found in many DACs.  While I canít explain exactly why this is so, the more I learn about what really impacts the sound of various DACs, the more Iím aware of the extreme importance of design choices made in the digital filter realm.

Any additional attempt at providing a technical description of the DAC-1 would simply compel me to plagiarize from the Bel Canto website, so if you want to dig deeper I suggest that you check out their well-designed site on your own.

MSB Link DAC Ė with modifications by Dan Wright

The version of the MSB Link DAC used in this review is the Mk II, which I own personally.  Itís about 18 months old.  Presently, the MSB Link DAC is being offered to consumers only as the Mk III version.  From what Iíve read, the Mk III version allows for the installation of a variety of factory upgrades, but otherwise itís the same machine in stock format.  My non-upsampling MSB Link DAC Mk II uses the Crystal CS8414 input receiver, and the Burr Brown PCM1716 DAC that is 24/96 compatible.  Unlike the Bel Canto, the MSB doesnít upsample to 24/96 automatically, but it will receive and process 24/96 digital output from a Pioneer DVD player, for example.

If your interest in the MSB Link DAC is keen, you have no doubt read at some point about the various upgrades performed on the MSB by outside vendors such as Dusty Vawter at Channel Islands Audio.  Realizing that a cottage industry had sprung up for modifying the Link DAC, MSB decided to move strongly into the Link upgrade market itself.  For more information on this, see our Editor's earlier review in Secrets of a fully decked-out MSB Link DAC III.  The modifier of my Link DAC, Dan Wright, is likely unfamiliar to many of you, but that may change as more people find out about what he does.

I got to know Dan through e-mail discussions after reading messages he posted to the popular internet audio discussion site called Audio Asylum (http://www.audioasylum.com).  Amongst other topics, Dan wrote about various upgrades that he performed to his own Link DAC, and he indicated a willingness to perform these mods for others.  Danís posts were intriguing, and it was evident that he knew his way around a circuit.  I liked his inquisitive nature in trying new things to enhance the performance of his Link DAC.  Since I wasnít spending a lot of time listening to my stock Link DAC, I figured there was no harm, and potentially lots of sonic upside, to letting Dan work his magic on the Link DAC that was spending most of its days in my storage room. 

In our discussions, I found Dan always enjoyable to correspond with and very responsive by e-mail.  I also appreciated that his prices are more than fair for the extensive nature of his upgrade and the quality of parts used.  Basically, for around $200-250, depending on the work you want done and the parts you choose, Dan will upgrade the stock MSB Link by focusing on two main areas:

            Power Distribution

                        Filter capacitors

                        Diodes

                        Regulators

                        Other capacitors

                        Cutting out parts that donít need to be there

             Analog Output

                        Op-amps (ask him about the Linear Technologies Ė I did, and they are great)

                        Resistors

                        Capacitors

                        Cutting out parts that donít need to be there

                        Controlling RFI (radio frequency interference)

On the surface, that list may look like a major overhaul, but itís really not.  The Linkís basic digital conversion architecture remains intact.  The upgrades simply allow it demonstrate more of what itís inherently capable of doing.

Aftermarket upgrades are not uncommon, and thereís a simple explanation to why the MSB Link DAC in particular has been a popular component to tweak.  The reason is that parts selection is necessarily limited when the manufacturer is releasing a product at a budget price point.  At a retail price of $399, the Link DAC III certainly falls into the ďbudgetĒ category by any reasonable standard in the world of high-end audio. {Note by Editor: One of the main differences between DACs is not the DAC chip itself, but all the connecting parts, including capacitors, resistors, and power supplies.}

MSB made good parts choices to begin with when designing the Link DAC.  However, if they started dropping in upgraded, boutique parts, the retail price would have to increase exponentially to the cost of those parts.  If that happened, the Link DAC wouldnít be a budget component for long.  Like it or not, thatís how the economics of audio manufacture work.  Dan Wright, on the other hand, can selectively target the most essential components in the Link, then replace stock parts with the highest quality replacement parts possible from companies such as Black Gate, Caddock, Linear Technologies, and others.  This process works because Dan doesnít generally make his money on the parts themselves, but on the labor he charges to install them.

RAudio DAC-0

The RAudio DAC-0, which is designed and manufactured in Hong Kong, is the ďlone wolfĒ of this group in many respects.  It offers a few unique technical design elements that Iíll explain shortly.  The RAudio DAC-0 is also the one DAC in this survey that you are least likely to be familiar with as the company is still working to develop its international distribution system.  Therefore, Iíll spend a bit more time discussing the design of the RAudio than the others.

Foremost of the unique design features is the use of a battery-based 12-volt DC power supply.  The DAC-0 is supplied with a small, external AC-to-DC power supply that you use to charge the batteries housed in a somewhat larger, gray plastic box.  An indicator light on the power supply tells you when the batteries are fully charged, at which point you unplug the charging unit, then connect a small umbilical power cord from the battery-box to the DAC-0.  Voila! The result is a power supply that is fully independent from your house wiring.  The battery charge lasts long enough for any marathon listening session Ė I got about ten hours of run time before feeling that a juice-up was needed.  Personally, I didnít care for the extra effort required in the battery charging process (I find audio ďfiddlyĒ enough already), but I appreciate that a fundamentally different design such as the DAC-0ís battery-based power supply may require a bit more exertion.

Ideally, the battery-based power supply should offer two distinct advantages: 1) your DAC is not directly impacted by the quality of power flowing from the wall socket, which is often poor, and 2) your DAC is not feeding digital hash and noise back into the power lines, which DACs are prone to do.  Think of it as the ultimate form of digital isolation.

The RAudio DAC-0 shares with the MSB the use of the Crystal non-upsampling CS8414 input receiver and Burr Brown PCM1716 DAC chips.  Therefore, the DAC-0 is also 24/96 compatible, but it doesnít get to 24/96 on its own (no upsampling) like the Bel Canto does.

The analog output section of the DAC-0 is fully encased in epoxy, keeping it a mystery to the user.  Raymond Chan, the designer, regards the analog section as the true ďessenceĒ of the DAC-0ís performance.  I donít entirely agree with the secrecy thing, but I respect it.  There are only so many ways to ďskin the catĒ, as they say Ė if Mr. Chan has conceived of a new and better one, then by all means itís his right to maintain the propriety of it. 

One key difference of the DAC-0ís analog section is that it outputs only 0.5V RMS.  By comparison, other DACs, including those in this survey, typically output around 2.0V RMS or more.  Other than the fact that I needed to use more of the volume control on my preamp, I found no significant advantages or disadvantages to this lower output during my listening tests.

Finally, any true minimalist will love the DAC-0.  On the back of the DAC-0 youíll find an RCA coaxial digital-input jack, left and right RCA analog outputs and the receptacle for the power cord from the battery-based supply Ė nothing else whatsoever in the way of controls, lights, or connectors.  Then again, what more do you really need?  Itís quite similar, actually, to the paucity of unnecessary controls on the Bel Canto.  If the goal was keeping things simple, which I support, the DAC-0 is a success. 

I need to point out that Iíve had three samples of the DAC-0 during my review period, which was long, indeed.  The first DAC-0 didnít work because it was damaged during shipping (hey, itís a long trip from Hong Kong!).  The second sample was in the early stages of audition when Mr. Chan contacted me to advise me of any important upgrade.  He asked that I listen to the newest version, which I agreed was the best way to proceed.  Without question, it sounded better than the second sample.  I experienced no functional problems with the third sample, and itís the one that I describe in this review.

Partly because of the lack of any extraneous controls, the DAC-0 looks smashing.  Itís a modestly sized wood enclosure finished on all visible sides to a piano-gloss sheen.  The DAC-0 appearance is a good example of the potential beauty of simplicity.  The pictures Iíve included donít quite do it justice, but youíll get the idea. 

At $2,700, the RAudio DAC-0 is by far the most expensive DAC in this survey.  Iím not going to question any companyís pricing strategy, but in the long run the RAudio will be pressed on price by the competition.  The distinct trend in digital audio today is better performance for less money, and all DAC manufacturers will need to be aware, at a minimum, of the changing landscape.

Iíve touched on the technical design, but Iím not going to spend any time describing for you the looks or basic features of the Bel Canto DAC-1, or the MSB Link DAC.  Each of these DACs have been covered at length in both print and electronic media, and Iíve got nothing particularly original to add on either topic.  Suffice to say that both look fine to me, and both have the necessary connectivity and functionality to make themselves right at home in my system.  Besides, if I could read your mind right now, Iím sure that itís gently encouraging me to get on with it.  

A few Notes on Comparison Reviews

This comparison review has been a lot of fun, but at the same time itís been a major challenge.  Anyone who tells you itís easy to draw firm conclusion when doing a comparison review is naive.  There are lots of variables.  Counting pregnant chads in Florida is infinitely simpler.

Any component, whether itís a DAC, speaker, amplifier, or something else, has a huge number of setup variables that will directly impact listening impressions and subjective conclusions.  On the surface, it may seem that evaluating a DAC is fairly straightforward, but thatís not the case at all.  Many questions face anyone that attempts to review DACs in a comparison format, including the following:

        Does the DAC have the ability to drive your reference preamp adequately? (hint: this played a big role for one DAC in this particular comparison)

        What type (AES/EBU, RCA, BNC, Toslink, etc.) and brand of digital cable are you using?

        Are you using the stock powercord, or an aftermarket cord?

        What type of isolation feet have you tried, if any?

        Did you use an external jitter-reducer in conjunction with the DAC, or not?

More questions abound, but you get the picture.  My point is that you need to take my opinions for what they are: just opinions, knowing that your mileage may vary.  If that sounds like a cop-out, Iím sorry, but Iím not writing this review to tell you what DAC you should purchase.  My purpose is to pass on what I heard from each DAC during many months of listening and comparing.  Yes, I believe I can discern the sonic differences between these DACs as well as anybody, and Iíll do my best to convey those to you, but itís still just one reviewerís opinion.  Bottom line is that Iíd encourage each of you to do your own listening. The point is that DACs do sound different from one another.

Also, please note that Iíll end this review with a brief list of setup information for each DAC covering some of the questions posed above.

Listening Tests

Bel Canto DAC-1

The Bel Canto DAC-1ís sound was immediately striking Ė in the best sense of the term.  It presented a roundness and wholeness to vocalists and midrange-based instruments that the other two DACs didnít quite match.  There was a sense of coherency, of delivering a musical message with the lines of aural communication wide open.  Itís a listening experience that I found immediately accessible.  The impression was further reinforced over the course of many months.

The DAC-1 presented a subtle, nearly imperceptible sense of a warm, luminescent aura around notes, particularly on solo piano recordings.  The DAC-1 proved particularly inviting for this type of music.  The entire harmonic structure, from the primary tones to the over-and-undertones, was presented by the Bel Canto intact.  Top-to-bottom balance was good, with no part of the spectrum drawing undue attention to itself.  The lowest bass was a touch reticent, which may be attributable to the necessary restrictions of the power supply in a chassis as compact as the DAC-1.  Despite this modest shortcoming in the lowest bass region, the Bel Canto did not sound light or thin.  Timbre was true, and natural.  There was a fine sense of air and space around notes. 

One of the best ways to convey the character of the Bel Canto is to tell you what it didnít do Ė at no point in my listening did the DAC-1 shut me out of a listening session by exhibiting an impersonal, cold sterility.  In fact, the DAC-1 had quite the opposite effect.  Its inviting nature made me look forward to hearing it.  It was always ready to play music, and many styles of music were rendered well by the DAC-1.  The overall character of the Bel Canto made it very easy to listen to, and not fatiguing even during long listening sessions. 

If you think the adjectives Iím using to describe the DAC-1 point to a colored, soft presentation, let me assure you thatís not the case.  The Bel Canto is not overly rich, syrupy or slow Ė it simply did not go overboard in that regard.  Just as you would expect (and hope), the bright, harsh CDs in my collection still sounded that way.  The Bel Canto, thankfully, didnít go so far as to fundamentally alter the inherent sound of a recording.  Truly bad recordings still sounded bad, but good recordings were oftentimes downright enchanting through the DAC-1.

The presentation of the DAC-1 was further back in the soundstage than the other DACs in this survey, or the Perpetual Technologies P-3A reviewed in September.  The soundstage was modestly more narrow, as well.  More complex, dense, musical passages lacked perhaps a bit of color and contrast, and overall performance energy was conveyed with less enthusiasm when compared to the best DACs Iíve heard.  These are really just minor quibbles and only the type of drawbacks that a listener would notice when they are listening in the overly attentive way that reviewers often do.

Many audiophiles, including lots of reviewers, loosely toss around the phrase ďitís all about the musicĒ, when, in fact, their listening habits and playback gear suggest quite the opposite.  Iím not damning that general orientation.  Audio is, after all, a hobby for most of us, and hobbies should be pursued from whatever angle brings joy to the hobbyist.

With that as a backdrop, however, Iíd suggest that the Bel Canto is a DAC for those that truly are more concerned about the music, as opposed to the more ďhi-fiĒ aspects of high performance audio.  As Iíve already said, the DAC-1 doesnít obscure detail, or gloss over important musical nuance Ė it certainly gives you all the music.  At the same time, it does have a way of making most of your discs sound somehow more enjoyable, which no doubt results from specific design choices that Bel Canto made when developing the DAC-1. How much of this was attributable to the 24/96 upsampling, I canít say, but overall I spent time with more of my CD collection when auditioning the DAC-1. (Remember, upsampling actually creates new samples of music in the waveform, based on mathematical assumptions called algorithms, whereas, oversampling, which usually just creates more samples that are strings of zeros, is for the purpose of making it easier for the filters to remove noise.) 

MSB Link DAC Ė with modifications by Dan Wright

The Dan Wright modified MSB Link is the most immediately exciting DAC in this survey to listen to.  Micro and macro dynamic contrasts pop out at you, dazzling your aural sense.  The soundstage is wide, deep, and detailed.  Notes move, bounce, and flow in a seemingly unimpeded fashion.  I suppose the MSB exhibits an excellent sense of what our audio reviewer friends in England prefer to call PRAT (Pace, Rythym and Timing).  PRAT is a term that Iíll leave in the Mother Country for now, but at least I now know what theyíre talking about  . . . I think.

Compared to the Bel Canto and RAudio, all this energy that the MSB exhibits comes across as a bit raucous.  On the other hand, the result is more like what happens when you experience live music.  Thereís no doubt in my mind that many of my hard-bop jazz and rock albums were better served by the moddedí MSB than the more docile contenders in this report.

The MSB favors the primary tone of each note, but man, does it ever nail it!  The Bel Canto does a better job of reproducing the full harmonic envelope, while the MSB hits the primary notes with more conviction.

After listening to all types of music, I concluded that the tonality of the MSB favors the top and bottom ever so slightly to the mids. It doesnít sound as pretty as the Bel Canto and RAudio in the vast middle portion of the scale.  If anything, the mids of the MSB come across as somewhat more direct, and tonally pure, but not as organic and inviting as the Bel Canto.  From the highest-highs to the lowest-lows, the MSB offers a mightily engaging listening experience, however, and one that after many months I am no less thrilled by.

I spent a lot of time with my stock MSB Link prior to having Dan Wright perform the mods.  In its stock form, I guess you could say that I appreciated the MSB, but I was not taken aback by any aspect of its performance.  In fact, I failed to write my intended review last year because nothing about the stock MSB moved me enough to put pen to paper and tell you about it.

My ďho-humĒ attitude about the MSB quickly changed when I received the modified Link DAC back from Dan Wright.  Forgive me if I donít run through the same old litany about the ways that the sound improved.  Suffice to say that all aspects of the MSBís performance took two giant steps forward, propelling the modified Link into my preferred zone and prompting me to include it in this review.   At $399 for a new Link DAC III that Dan Wright can modify (or a used Link for a cheaper price) plus approximately $200 for modifications, this is an outstanding value that is unmatched in my experience.

What Dan Wright (and others like Dusty Vawter at Monolithic Sound) do to extract such outstanding performance from the MSB is a testament to the quality of the original design.  Without a solid starting design most modification jobs would sound like junk  . . . or at least modified junk.  Dan Wright started with a good basic design, then made it a lot better.  The result speaks for itself, and Iím thoroughly impressed.

RAudio DAC-0

The RAudio DAC-0 is a natural, neutral sounding D-to-A converter.  It shares with the Bel Canto an extreme ease of listenability over long periods.  You donít fully appreciate this type of sound until you are forced to live without it, when you revert back to an old-technology, poorly-designed DAC that reminds you of everything it was about digital playback that you didnít like in the first place.

With its battery-based power supply, the DAC-0 was impressively grain-free, just as the designer intended, Iím sure.  There was precious little of the whitish edge or rough texture hanging around upper-register notes that is often assumed to be inherent in digital playback.  By presenting such a pristine top end, the DAC-0 constructs a convincing argument that digital nasties may be as much the fault of the power you feed your digital playback system as the playback technology itself.

My earliest listening sessions with the DAC-0 left a clear impression of its even tonal balance.  It didnít unnaturally editorialize any part of the musical spectrum.  In fact, ďbalanceĒ is a term that popped up frequently in my listening notes during my time with the RAudio.  Through the RAudio, I had the distinct impression that instruments and vocals, no matter which octave they participated in, occupied their own place in the performance.

Compared to the other DACs in this survey, the RAudio afforded a somewhat more full, and inviting, mid-bass region.  In my system, I detected a consistent, very slight rise in the region somewhere around 120 Hz, give or take a handful of Hertz.  Because of the thinness of some recordings, the effect was not entirely unwelcome.

Because this is a comparison test, you are likely interested in knowing other ways that the DAC-0 compared to the Bel Canto and MSB: the overall timbre of the DAC-0 was good, if just a bit darker than the MSB Link and not as inviting as the Bel Canto.  Soundstage width was not up to the standard set by the MSB, but probably a bit more extended than the Bel Canto.  Soundstage depth, however, was a different presentation entirely in that the RAudio brought things a couple steps forward, laying the music on a more level plane, close to even with the speakers.

I ran into some difficulty with the RAudio DAC-0 because of its relative lack of dynamic contrast in my system.  It seemed to play within a tighter range of soft-to-loud, regardless of what the recording called for.  To place this in a clearer perspective, if the most dynamic DAC in this survey (MSB Link) covered a hypothetical loud-to-soft range of 0 -100, the RAudio was something like 0 - 70.

The other drawback I heard in my system with the DAC-0 was a relative absence of the lowest bass octave.  None of the three DACs in this survey do low bass like the Perpetual Tech P-3A with the Monolithic P3 power supply, but the DAC-0 didnít seem to give me any meaningful action from about 45 Hz down.  This low-bass obstacle, combined with the lack of dynamics mentioned in the preceding paragraph, prompted an e-mail exchange with the designer of the RAudio DAC-0, Raymond Chan.  The result of our brief exchange was the realization of something that I should have expected: a possible impedance mismatch.

According to Raymond Chan, the DAC-0 prefers to see a preamp input impedance of approximately 100 kOhms.  My current reference preamp, the Monolithic Sound PA-1, shows the source component an input impedance of 20 kOhms.  See the problem?  We have a classic case of a potential impedance mismatch (the bane of every audio reviewerís existence), and as was demonstrated here, it can be nearly fatal to the component being reviewed.  The DAC-0 was obviously running out of steam as it attempted to drive my reference preamp, which as you will recall from my review of the Monolithic is a completely passive design until the volume control exceeds the 12 oíclock position, which I never do anyway. {Editor's Note: Just as with a power amplifier having difficulty driving a low impedance speaker - say 3 Ohms - a DAC can have difficulty driving a low impedance preamplifier input, in this case 20 kOhms. The battery power supply in the RAudio is not strong enough to drive a low impedance preamp input. You can have the same problem with a preamp trying to drive a low impedance input on a power amplifier. One of the symptoms is not hearing much bass output.}

Armed with this troubling revelation, I am faced with a couple shaky paths to conclude my listening test comments on the RAudio DAC-0:

1.      Do I declare my extensive listening tests invalid and simply conclude that I have no idea what the DAC-0 sounds like?  Hardly.  Apart from my impedance-impaired judgement of the dynamics and the low bass, I believe that the DAC-0 sounds like what I heard and reported;

2.      Do I change my reference preamp mid-review, thereby changing my system, and try to draw accurate sonic conclusions about the DAC-0 in the context of this new system?  Nope, I wasnít gonnaí do it.  It simply wasnít practical to start over with all three of these DACs using a new reference system.  Also, in the greater scheme of things, my position is that you should not have to change your reference system to accommodate a component with non-traditional specifications or compatibility.

My purpose isnít to be overly critical of the DAC-0, because it did a number of things well, but in my system, in this particular instance, it was not an ideal electrical match with my reference gear.

Conclusion

Iím wowed in many ways by both the Bel Canto DAC-1 and the Dan Wright-modified MSB Link DAC.  With its supreme musicality, the Bel Canto epitomizes much of whatís right about digital audio playback today.  The DAC-1 doesnít try to hit you over the head with its hi-fi sound.  It takes an entirely different route.  The Bel Canto merely goes about its business, and that business is making wonderful music.  The modified MSB Link DAC, on the other hand, gets my blood pumping and feet jumping, making it an absolute blast to listen to.  The amount of improvement wrought by Dan Wrightís modifications left me shaking my head in amazement.  Itís a very exciting DAC, and itís positioned at the pinnacle of the value scale.

Between the Bel Canto and MSB, there is no ďwinnerĒ, per se, because the strengths of each are so distinct.  Simply put, they sound different.  Which DAC you prefer would ultimately come down to personal preference.  Personally, Iíd have no problem using either of these DACs as my primary playback source for the foreseeable future Ė they are both that good.

Because of the impedance mismatch problem, Iím unable to draw any final conclusions about the sound of the RAudio DAC-0.  The designer of the DAC-0, Raymond Chan, advised me that a revised version of the DAC-0 (called the DAC-i) is in the final stages of development and will debut at CES 2001 in Las Vegas (which Iíll be covering, by the way).  I donít know the nature of the changes, but perhaps the impedance issue will be addressed in terms of a larger power supply.

Even though it wasnít included in this comparison review, the Perpetual Technologies P-3A/1A combination, which I reviewed separately earlier, is right at home with the best of this group in terms of overall performance.  In fact, taken as a package, the Perpetual P-3A/1A remains a price-to-performance sonic standard to which any DAC in this survey (or elsewhere) would do well to try to emulate.  Iíll continue to use the P-3A/1A as a benchmark.  Standing alone, the stock P-3A barely reaches the performance level of the Bel Canto or modified MSB, if at all, but the addition of the P-1A changes the game significantly and swings it in Perpetualís favor.

Iím hoping to have a couple more of the hot, new DACs to tell you about in early 2001.  If I still have the DACs in this survey to compare them to, Iíll be sure to report on how they all stack up against each other.  We could only be so lucky to have something even better come along and dislodge these DACs from their respective perches Ė we shall see.  Truly, these are good times in affordable digital audio.  

Set-up Information and Suggestions

1)     At the suggestion of both Bel Canto and MSB, I tried utilizing the TOSlink input of each DAC that receives the output of a Pioneer DVD player.  No problems with this setup, but it didnít hook me.  The music sounded somewhat less substantial, but itís difficult to describe what I mean when I say that.  Maybe it was the type of TOSlink cable I tried (Monster Cable), but somehow I doubt it.  In both instances, I ended up preferring the sound of the coaxial input received from the output of my very fine G&D Transforms CD transport using a DH Labs digital cable.  Plus, the RAudio accepted only coaxial, so using TOSlink would have further skewed the validity of comparisons;

2)     Jitter reduction boxes are a very good idea with all these DACs.  I first tried the Monarchy DIP.  The MSB seemed to respond to this the most favorably of all.  The RAudio showed some true improvement in top-end extension, while the Bel Canto sounded only slightly better with the Monarchy DIP than without it (which is likely a testament to the jitter management capabilities of the Bel Canto after the signal has arrived).  Next, I tried the Perpetual Tech P-1A  . . . and Zowie!  Without the upsampling functions, using only the jitter reduction of the P-1A, it bested the Monarchy DIP in all respects.  The sonic improvements were similar to those rendered by the DIP, only more dramatic.  The Bel Canto showed a more marked improvement in clarity and openness with the P-1A as the jitter reducer than with the Monarchy DIP;

3)     A real test came when I used the P-1A to send a 24/96 signal to the various DACs.  Doing so, the MSB Link really came alive.  It retained the same punchy, fun character it exhibited with the Monarchy DIP, but the overall musicality became closer to that of the Bel Canto.  The P-1A and Bel Canto were a very good match, too, but I preferred to let the Bel Canto perform the 96 kHz processing, sending it just a 24 bit/44.1 kHz signal out of the P-1A.  Interestingly, this is the same experience I had with the P-3A.  When I sent full 24/96 from the P-1A to the Bel Canto, my perception was that things became a bit too heavy and dark.  The RAudio responded well to receiving the 24/96 output from the P-1A.  The DAC-0 took on a more open, expressive character, and the drawbacks mentioned in the listening notes above were mostly minimized but not eliminated.  Once again, Iíll point out that the base sonic character of these DACs didnít change with the insertion of either the Monarchy DIP or Perpetual P-1A. 

4)     The only DAC in this survey that allowed me to utilize an aftermarket powercord was the Bel Canto.  I ended up really liking the Bel Canto with the Harmonic Technology Pro-AC11, but other power cords I auditioned (TG Audio HSRi, Nordost El Dorado) also easily outperformed the stock powercord in my tests. 

5)     Always experiment with different types of isolation feet and bases in your own system.  In my tests, I found the RAudio to sound best resting on a trio of Black Diamond Racing #4 cones, but avoid Vibrapods at all costs Ė they bog the DAC-0 down substantially.  The Bel Canto sounded best just sitting there on its own four feet, believe it or not.  I did like the effect of a small sandwich bag containing a lead shot and sand cocktail placed on top the DAC-1 for damping purposes.  The MSB did well resting on Black Diamond Racing #3 cones, but it was with the Nordost Pulsar Points that I obtained the best overall tonal balance.  I tried various damping items on the cover of the MSB, but none of them really improved the sound Ė they only changed it.  Thus, I ended up with nothing on the cover of the MSB.

 

UPDATE - January 21, 2001

I wrapped up my DAC comparison review, at long last, just after Christmas 2000.  It was posted to the site right around New Yearís Day.  Shortly thereafter, in early January, a group of us from Secrets made our way to CES 2001 to check out all thatís new and exciting in our world.

While at CES, I spent some time with Bel Canto and RAudio, each a manufacturer of one of the DACs in the comparison review.   Both of these companies gave me news about their DAC that potentially impacts the conclusions I reached in review.  Therefore, I want to share that news with you right away. 

Bel Canto DAC-1 Ė Power Supply Update Coming 

The Bel Canto DAC-1 is having its power supply re-worked.  My understanding is that existing units can be upgraded, and new units will ship with the new power supply as standard.  Expect a modest retail price increase. 

Whatís most interesting, and has me awaiting the upgrade anxiously, is that my criticisms of the DAC-1 were mild, but I have reason to believe that my concerns will be addressed directly by what Bel Canto is doing to improve the power supply. 

As an ego pump, Iíd like to believe that the changes are being made because of what I said in the review.  Thatís a pipe dream, however, because Bel Canto announced the availability of the upgrade pretty much simultaneous with the publishing of my review.  Maybe it was my sublimiliminal (oops Ö in honor of the inauguration this weekend) communication with the Bel Canto designers during my listening tests.  Yeah, thatís it. 

Youíll recall from the review that I found the Bel Canto to fall just short of the MSB in the areas of dynamic contrast and PraT (thereís that English phrase again).  Also, the Bel Canto showed a slight lack of color, contrast, and energy when the music became very dense and complex.  Finally, the low bass was a touch reticent compared to my friendís Theta GenV. 

Can you guess the good news?  Yep, you guessed right.  The criticisms leveled at the Bel Canto are just the type of thing that a beefier power supply will tend to address.  Here comes my sermon again: you canít underestimate the importance of the power supply in any audio component.  Itís absolutely, positively vital. 

Iíll evaluate and report on the upgraded Bel Canto DAC-1 as quickly as possible.

RAudio DAC-0 Ė a new name, DAC-i, and a new Analog Output Module

As announced at CES 2001, the RAudio DAC-0 is now officially re-named the DAC-i.  The primary change between models is that the DAC-i features a new, and reportedly much improved, analog output stage. 

The new analog output stage is retro-fitable to the DAC-0 by removing a few screws and disconnecting and re-connecting a few wires that have male/female fittings on the ends.  It sounds simple, and it is.  The factory will ship all DAC-is with the new module already installed, of course, but Iím going to install it myself.  The upgraded component set was given to me by the very kind designer of RAudio, Raymond Chan, with whom I had a nice visit in the RAudio demo room at CES. 

Referring once again to my review, you recall that I wasnít able to complete my evaluation of the RAudio DAC because I encountered a serious impedance mismatch between it and my preamp.  The inability of the DAC-0 to drive my preamp resulted in truncated dynamics and low bass.  Despite the impedance issue, I could safely evaluate the essential tonality of the DAC-0, and I reported it as I heard it, but I simply couldnít take a true measure of the DAC-0ís overall performance.  The necessary remedy was to change to a preamp with higher input impedance, but changing oneís reference system mid-stream isnít the best way to conduct a comparison review.  So, my grade on the DAC-0 at the end of the review was a big, fat ďIĒ for incomplete. 

To complete my assessment of the RAudio DAC-0, Iím planning to install the upgrade (thereby rendering it a DAC-i), then listen to it again in the same reference system as the original comparison review.  Thatíll be interesting.  Finally, my plan is to listen to the DAC-i in the context of a friendís system who has a tube preamp with adequately high input impedance.  Changing reference system review contexts will nullify any conclusions about the DAC-i in relation to the Bel Canto DAC-1 and modified MSB Link, but Iíll be able determine with finality whether the DAC-i is truly better than itís predecessor. 

Stay tuned, and I will report on each of these upgraded DACs as soon as possible.  For now, good listening.

 

- Paul Knutson -

Paulís Reference System for this Review

G&D Transforms Reference One CD transport (Nordost El Dorado power cord)

Monarchy DIP outboard jitter attenuator (Harmonic Technology Pro-AC11 power cord)

            Or

Perpetual Technologies P-1A digital correction engine (Monolithic Sound P3 power supply)

One of many DACs (see review)

Monolithic Sound PA-1 Linestage (HC-2 power supply with TG Audio HSR power cord)

Wright Sound Labs WPA 3.5 monoblock power amps

Silverline Audio SR-15 speakers

Sound Application CF-2 power conditioner (analog equipment)

PS Audio P300 w/multiwave (digital equipment)

DH Labs D-75 BNC-to-RCA coaxial digital cable

Nordost Silver Shadow RCA-to-RCA coaxial digital cable

Analysis Plus Silver Oval-In interconnect

Analysis Plus Silver Oval speaker cable

Divider
© Copyright 2001 Secrets of Home Theater & High Fidelity
Return to Table of Contents for this Issue.