MSB 32/384 DAC, Disc Transport, ADC, and Monoblock

MSB Technology made a splash at the Rocky Mountain Audio Fest last weekend with several new digital-audio products that can handle resolutions up to 32 bits and sampling rates as high as 384kHz. Also on hand was the company's newest monoblock power amp.

The DAC IV is MSB's latest digital-to-analog converter, shown here with the optional iLink II iPod dock and sitting atop a Data CD IV disc transport and Power Base 12V power supply (which is included with each MSB DAC IV and Data CD IV). The DAC IV is available in three versions—Platinum, Signature, and Diamond—with increasingly improved performance as you move from one version to the next. The base prices for these DACs are $8000, $17,500, and $26,000, respectively.

Then there are the options available for each DAC IV. Aside from the aforementioned iLink II iPod dock ($2000), perhaps most important is the upsampler ($1000) that converts 44.1, 88.2, and 176.4kHz audio data to 352.8kHz and 48, 96, and 192kHz data to 384kHz, all at 32-bit resolution. Also available is a USB port that can accept a 384kHz input from a computer ($1400).

The Data CD IV reads standard CDs as well as high-resolution audio from CDs and DVDs, including HRx discs from Reference Recordings that currently use 24-bit resolution at 176.4kHz. In fact, it can read and output digital-audio data up to 32-bit/384kHz, though no such commercial discs are currently available as far as I know. Lower resolutions can be upsampled to 32/384 with an optional module ($1000), which is not required if you plan to use a DAC IV with upsampling. The Data CD IV's base price is $4000.

Of course, there is virtually no content recorded at 32/384—at least, not yet. To fill this void, MSB offers two analog-to-digital converters (ADCs)—one model with two channels and another with eight channels—that digitize balanced analog inputs at 32/384. Both start at $16,000, which seems a bit odd until you learn that the 2-channel model uses four converters per channel for improved performance.

At the end of the signal chain, it all comes down to analog, which is where the new M202 monoblock power amps enter the picture. Priced at $17,500/pair, these cylindrical beauties can pump out over 200 watts into 8Ω, and they are capable of driving 1Ω loads. They feature fully balanced circuit paths, and they include balanced and unbalanced inputs that are treated separately.

These 32/384 products lead me to wonder, as I often do, at what resolution does digital audio become indistinguishable from analog? Or is analog always distinguishable from digital no matter how high its resolution? I don't know, and I don't know of any scientific studies that address this question, but it sure is interesting to contemplate. I'd love to read your thoughts on this, so please post a comment here.

Share | |
COMMENTS
Jarod's picture

Hey Scott, I heard you say that there is virtually no content recorded at 32/384, at least not yet. Do you know of any future plans in the works for 32/384 recordings?

uavtheo's picture

Scott and all,

I came across an article somewhere about the distinguishing issue that many characterize as that "digital sound" comes from the anti-aliasing filters. I did a quick Google search as my recollection was that this discussion had come up with a Meridian CD player. I came across this paragraph in Stereophile at: http://www.stereophile.com/cdplayers/meridian_8082808i2_signature_refere... (which may be where I had seen it as I subscribe to Stereophile) "A pair of technical papers in the March 2004 issue of the Journal of the Audio Engineering Society (Vol.52 No.3) indicated the direction Meridian would be taking with the 808.2 (footnote 1) In "Coding for High-Resolution Audio Systems," Bob Stuart examined what was required to get the highest audio quality from a digital system, while in "Antialias Filters and System Transient Response at High Sample Rates," Peter Craven examined the nature of the ubiquitous anti-aliasing filter (in A/D conversion) and reconstruction filter (in D/A conversion) (footnote 2)" If you go to the URL, the footnotes have links to the paper and an audio explanation.

Regardless, this may offer some insight into your question about technical papers looking at distinguishing analog and digital.

Keep up the great work. We all enjoy the podcast and I look forward to it weekly. You've kept it at such a good, high level. Congratulations.

Can I suggest a future show to also look at remote contols? I'm obviously very technical and setup a Niles multi-room IR repeater system and wanted a remote to manage all my different components. I've settled on a Logitech Harmony as I can't stomach having to go back to a dealer for any tweaks when I can do all this myself. But, I'd be curious to see about any WiFi-enabled remotes and remotes that sense whether or not a component is actual on or not. That's the biggest weakness I see in the Logitech Harmony family.

All the best,
Theo

Jarod's picture

Thanks Scott. Thats a good point about what there is to gain at a res as high as 32/384 compared to 24/96. I wonder?

uavShiitaki's picture

When I see people ask when digital will sound indistinguishable from analog, I think it shows how biased they are. The answer is never; because the goal of digital is to faithfully reproduce the original signal, not mimic the transfer function of an analog medium.

I don't know what data density exceeds the capability of human hearing, but I'm confident it's possible with current technology. That technology I do not think is conveniently available however.

At first glance the idea that analog is infinite resolutions sounds good, however audio tape has a bias that caps its frequency range. The magnetic media has limits in dynamics because of magnetic saturation. A recording needle, and the cartridge used for playback, both, have limits to what kind of grooves they can cut and follow. Or another words they are physically limited in both dynamics, and frequency range.

If you love analog, is it because of what you DON'T hear? The point of a recording is to reproduce a work of art, and absolute realism shouldn't be the point. Would the Mona Lisa painting be better if it was a photograph instead?

Scott Wilkinson's picture
Thanks for the references and for the kind words! See my reply to Shiitake below for a bit on anti-aliasing filters.

Doing a show on system control is a great idea; I'll work on it!

Timian's picture

That's a fantastic point, Shiitaki! I have to admit that I've never really thought to question the 'analog = factual reproduction' syllogism, but I'm glad to have it pointed out. To give them the benefit of the doubt, however, I assume that most audiophiles who use this trope understand it to be exactly that. Or so I assume, anyway.

Timian's picture

As I'd never even heard of MSB Technology before this post I can't comment on their performance or sound, but from an industrial design viewpoint the M202 monoblock power amplifiers are really quite stunning.

Scott Wilkinson's picture
Shiitaki, you make many interesting points here. I disagree that the question shows bias, though your answer that they will never be indistinguishable is a valid (though arguable) one.

Regarding the faithful reproduction of the original signal, that is one goal of both analog and digital, and many audiophiles would argue that analog does a better job of that than digital, no matter what the resolution. However, this is not always the goal of digital; I know of many digital-audio products whose goal is to model the sound of analog...in other words, to mimic the transfer function of an analog system.

The human hearing system has a total dynamic range of about 130dB. In PCM digital audio (which is what these MSB products use), the dynamic range is determined by the bit resolution; 16-bit digital audio (as found on CDs) has a theoretical dynamic range of 96dB, while 24-bit audio has a theoretical dynamic range of 144dB, so we should not be able to hear any difference in dynamic range beyond, or even as high as, 24 bits. Granted, the real-world numbers are lower, and most integrated circuits can't reproduce a dynamic range of more than about 120dB (20 bits) anyway. So I think current (and relatively available) technology already roughly equals the capability of human hearing in this regard.

Frequency response is more controversial. The higher the sampling rate, the higher the upper end of a system's frequency response, which is roughly half of the sampling rate. So for CDs, which use a sampling rate of 44.1kHz, the highest frequency that can be reproduced is about 22kHz, which is lowered to 20kHz by a lowpass filter because that is the theoretical top end of the human-hearing range, and most of us can't hear that high anyway. Also, an anti-aliasing filter is used at the input so the analog-to-digital converter doesn't try to convert frequencies higher than half the sampling rate, which results in very audible artifacts called aliasing.

So what's the point of using a sampling rate of 96, 192, or 384kHz? These would allow reproduction of audio at 48, 96, and 192kHz, respectively. Many experts claim we can "hear" these high frequencies, not directly, but as high harmonic components of a complex waveform. Also, high sampling rates allow the use of anti-aliasing and lowpass filters with more gentle slopes, which results in less phase distortion.

I would never say that analog-audio systems have "infinite resolution," at least not in terms of dynamic range or frequency response. As you point out, magnetic tape and vinyl are limited in both respects. But analog has no quantization—the voltage varies continuously, not in discrete jumps as digital always does. so in that sense, analog does, in fact, have "infinite resolution."

If someone loves analog, I don't think it's because of what they don't hear; it's because they like what they hear better than what they hear from digital. I agree that the point of a recording is to reproduce a work of art, and sometimes, absolute realism is not the goal. But in other cases, it is; this depends on the content producer's intent. (Then there's the question of whether or not any audio-reproduction system can be indistinguishable from a live performance; I tend to doubt it.) If Da Vinci had had photography, would he have photographed Lisa del Giocondo instead of painting the Mona Lisa? Both are valid art forms, as are analog and digital audio.

Scott Wilkinson's picture
I don't know of any such plans. I'll ask Mark Waldrep the next time he's on my podcast, since he's a big proponent of high-resolution PCM recording. But even he records at 24-bits/96kHz, which is far less than 32/384. Thus my question, do you gain anything substantial by increasing the digital resolution beyond 24/96?
Scott Wilkinson's picture
I agree that these amps look beautiful! And although I haven't heard these particular products, the MSB equipment I have heard performed exceptionally well.

X
Enter your Sound & Vision username.
Enter the password that accompanies your username.
Loading