The Analog DAC

docvale · Jun 9, 2012

MSB released a new DAC, the Analog DAC. Obviously nobody can have already listened to it, but does anyone know whether it has to be considered the new top of the line or the entry level?

Price seems to be cheaper than the Platinum Signature IV, I say "seems" as if you want the matching power supply, rather than a computer standard supply, you need to add almost $3K

The most interesting thing is that it seems to include the femptosecond control clock. The optional Galaxy clock in the Platinum IV is $9K by itself

so, if included (maybe in an easier realization) this could make this DAC a piece of gear to crave for...
They also contemplate an analog volume control for preamp bypass.
http://www.msbtech.com/products/analogDac.php?Page=platinumHome

catastrofe · Jun 9, 2012

Based on MSB's own writeup, this is their new entry level unit.

docvale · Jun 9, 2012

well, they advertise it as "The Most Technologically Advanced DAC in the World!"

They stress the fempto-jitter stuff and, if you consider to use the matched power supply and to add all the 3 input sets, it will retail for more than the Platinum IV.

Anyway, I'm a little confused.
The Platinum (which is considered by many, audiophiles and professionals, the best DAC in the world) is a fairly young product.
The introduction of the Galaxy clock is also super recent. The new DAC is not reported to have new ladder chips, I think that most of the difference should be in the treatment of the signal in the digital domain (re-clocking, 80bit processing), plus the chance to tailor the flexibility according to the buyer needs: from a USB-only DAC (a la Ayre), to a multiple input + no need of preamp unit.

hiroshige · Jun 9, 2012

My understanding is that this is the replacement for the MSB PowerDac, which is the entry level before you get into the Platinum, Signature, and Diamond. But I see where the confusion comes from based on on the incorporation of some of the new tech into the entry level dac.

opus111 · Jun 9, 2012

docvale said:
Anyway, I'm a little confused.
The Platinum (which is considered by many, audiophiles and professionals, the best DAC in the world) is a fairly young product.

I must confess I'm skeptical of this claim that its widely considered the best. Do you have links to listening tests and/or reviews to back it up? From reading their own blurb it looks like the DAC is designed towards measurements (beating an existing chip) rather than by listening.

hiroshige · Jun 9, 2012

I don't know if it's widely considered the best since this is rather subjective but if going solely by reviews as you ask, it is considered one of the better dac's. HiFiCritic which doesn't accept advertising in their magazine (I suppose to provide more objectivity) gave it their highest rating they have ever given for a dac. Review can be downloaded at http://www.msbtech.com/reviews/HFC21_complete_med-rez-Review.pdf

Other reviews of the dacs and links to the review on their own website: http://www.msbtech.com/reviews/reviewsHome.php?Return=msb&Product=dac

docvale · Jun 9, 2012

opus111 said:
I must confess I'm skeptical of this claim that its widely considered the best. Do you have links to listening tests and/or reviews to back it up? From reading their own blurb it looks like the DAC is designed towards measurements (beating an existing chip) rather than by listening.

Even if I provided 50 reviews saying this, it couldn't be a statement anyway: our hobby is filled with relativity, regardless the fact that this word make an oxymoron when matched with "high fidelity".

Anyway, I can tell you that in the main Italian audio magazine, the least measurement guy of their board, the vinyl-dude Marco Benedetti, defined the Platinum IV as the best DAC he'd ever listened to.
He stated that in his personal system it outperformed the dCS Debussy and the Boulder 1021, which he had tested in more or less the same period. For those who speak Italian: http://www.mondoaudio.it/img/uploaded/immagini/MSB Tech/MSB_DAC_IV_Prova_Audioreview_Marzo2011.pdf

Also, the audio engineer Marco Lincetto, head of the Italian audiophile label Velut Luna, considers this DAC as his personal reference for his listening sessions.

I know a number of forumers, ranging from technology nerds to unbiased music lovers, that put this DAC at the very top.

ToneAudio raved for it, and the reviewer was another vinyl enthusiast... Alon Wolf has it in his test room (together with the mighty Pacific Microsonics)...

Notwithstanding, I'm writing all these things just because you debated what I reported. Personally, I've listened to all the various dCS Debussy, Playback Des MPS-5, Platinum IV, Meitner MA-1, Emm Labs DAC2SE, Bricasti and Weiss Medea, but all in uncontrolled settings (different systems, most of the components listened for the very first time), so I cannot tell which is the best performer.
I'd probably pick the cheapest one on the second-hand market

opus111 · Jun 9, 2012

Thanks

The Martin Colloms write-up is interesting in that the MSB beats the Metrum by 20 points on CD. But the price difference is around 26dB.

The only mention of an MSB DAC on MC's list (on his site) is from a 2008 rating (provisionally) at 55, which is a lowish score considering the prices of their units. Its beaten handsomely by both Rega and Naim. His previous highest score is from the Metrum Octave (he's not updated his list recently enough to include it, but the review of the Metrum gives his score, I recall its 180 on CD). So what stupdendous innovation have MSB developed in the past four years to almost quadruple their score? This enquiring mind wishes to know.

Another interesting thing from that review - no FFT plots, but a linearity plot. The linearity plot was the measurement I was hinting at in my earlier post. Why does MC not present any other plots, rather than the very measurement the DAC is designed to ace? Finally MC lacks credibility in his measurement technique for showing a THD @20kHz. Such a measure is fairly meaningless for a DAC without qualification - such as the sample rate being employed.

I shall look at the reviews you cited with interest.

Phelonious Ponk · Jun 9, 2012

"The analog DAC." I wonder. If i created "The digital Turntable," would objectivists....no, never mind. They'd just laugh.

Tim

Bruce B · Jun 9, 2012

Well the best PCM A-D I've used is the MSB Platinum Studio ADC with Galaxy clock and the best PCM DAC I've heard is the MSB Diamond DAC IV. I think I've heard just about every DAC >$10k in a controlled setting. MSB is on to something!!

Soundproof · Jun 10, 2012

They could be on to something.

A femtosecond is the SI unit of time equal to 10-15 of a second. That is one quadrillionth, or one millionth of one billionth of a second. For context, a femtosecond is to a second what a second is to about 31.7 million years.

Audiophiles could also stop worrying, and enjoy the music.
Yeah, right, in 31.7 million years.

(While sensitive, our nervous system isn't designed for this level of granularity in stimuli. But who cares?)

jkeny · Jun 10, 2012

Soundproof said:
(While sensitive, our nervous system isn't designed for this level of granularity in stimuli. But who cares?)

I've seen you mis-interpreting jitter before - as if the timing error directly translates into a timing error in the analogue output! This is wrong as the clock is used at the ADC to determine the point in time when a sample of the analogue waveform is taken & at the DC end when that sample value is translated back into the correct analogue value in the CORRECT POSITION in the analogue waveform. So you should be able to see that slight variations in this timing cause the analogue waveform to be wrongly recreated in complex ways. A simple summary the correct sample at the wrong time is just as wrong as the wrong sample value at the correct time.

Phelonious Ponk · Jun 10, 2012

jkeny said:
I've seen you mis-interpreting jitter before - as if the timing error directly translates into a timing error in the analogue output! This is wrong as the clock is used at the ADC to determine the point in time when a sample of the analogue waveform is taken & at the DC end when that sample value is translated back into the correct analogue value in the CORRECT POSITION in the analogue waveform. So you should be able to see that slight variations in this timing cause the analogue waveform to be wrongly recreated in complex ways. A simple summary the correct sample at the wrong time is just as wrong as the wrong sample value at the correct time.

Then through a comparison of the analog waveforms prior to conversion to digital and after conversion back to analog, it should be very easy to demonstrate the impact that these infinitesimally small variations in digital timing have on the resulting sound. A little more analysis should be able to determine the audibility of these variations.

Tim

Soundproof · Jun 10, 2012

jkeny said:
I've seen you mis-interpreting jitter before - as if the timing error directly translates into a timing error in the analogue output! This is wrong as the clock is used at the ADC to determine the point in time when a sample of the analogue waveform is taken & at the DC end when that sample value is translated back into the correct analogue value in the CORRECT POSITION in the analogue waveform. So you should be able to see that slight variations in this timing cause the analogue waveform to be wrongly recreated in complex ways. A simple summary the correct sample at the wrong time is just as wrong as the wrong sample value at the correct time.

I haven't misinterpreted anything, and don't have a dog in this hunt, compared to you.

I am, however, supremely grateful that I can enjoy my music without worrying about stochastic interference at levels where the mind's ability to discern variations has long ceased being relevant.

We would have to believe that correct decoding of the digital information could be affected by variations at the femtosecond level, which is such an absurd notion that it is laughable, but probably marketable.

opus111 · Jun 10, 2012

Bruce B said:
MSB is on to something!!

Would you like to offer any speculation as to what that something might be? In what aspects did you find the MSB to be superior to other offerings you've heard?

jkeny · Jun 10, 2012

Soundproof said:
I am, however, supremely grateful that I can enjoy my music without worrying about stochastic interference at levels where the mind's ability to discern variations has long ceased being relevant.

We would have to believe that correct decoding of the digital information could be affected by variations at the femtosecond level, which is such an absurd notion that it is laughable, but probably marketable.

Well, if you understand what I said then your statement is illogical - of course any variation in the timing of when the sample should be converted back to the resultant analogue values WILL result in a different analogue waveform - QED. What I think you might be saying is how audible is this incorrect translation - a different question to both of your statements so far. Let's try to be clear about what is being said.

jkeny · Jun 10, 2012

tomelex said:
Very nice simple explanation for timing. Can you elaborate, in a simple manner as you did above, how you think the waveform can be recreated in complex ways, say even if the timing errors were much larger, if it helps explain it simply?

Thanks,

Tom

Sorry, no I can't but what I'm trying to explain is that we don't hear jitter, we hear it's effects! These effects are complex. Lets take a very fast transient, such as the initial attack of a drum rim shot. If the timing of the samples are off then this steep analogue waveform will be effected much more than a low frequency sound which has a shallower waveform, in theory. So what are the audible effects of such variations in distortion throughout the waveform? I believe we will get a better handle on this in due course but at the moment we are too focused on single tone frequency analysis for our validation of fidelity.

jkeny · Jun 10, 2012

Phelonious Ponk said:
Then through a comparison of the analog waveforms prior to conversion to digital and after conversion back to analog, it should be very easy to demonstrate the impact that these infinitesimally small variations in digital timing have on the resulting sound. A little more analysis should be able to determine the audibility of these variations.

Tim

Yes, in theory, but in reality show me a valid analysis using music as the source! For instance how are you going to compare the two analogue waveforms, convert them to digital?

jkeny · Jun 10, 2012

tomelex said:
OK, but really, I am having a hard time understanding all your talk about fast transients, such as initial attack of a drum rim shot...........just how fast do you think it is.........you are limited in redbook digital to about 20Khz waveform response! Thats like 17 microseconds rise time.

Tom

What I was trying to underline in my simple explanation is not the absolute differences but the variation in differences between Hf & LF with respect to how jitter can effect them. It then goes to a correct model of hearing & what is discernible & what isn't. We are much more sensitive to variations in pitch than to absolute differences in pitch, for instance, so I guess the variations in the context of the musical stream is why I said it was a complex issue & I couldn't explain it in the absolute terms & values that you seem to want.

EDIT: Also, I don't know the answers to a lot of this but I trust my ears & those of others. So I await the measurement techniques & those with expertise in this area to catch up with our hearing!

EDIT, EDIT: Maybe looking at new ways of measurement might be the clue to unlocking all this, I don't know - see Wavelet analysis here http://www.regonaudio.com/Audio Measurement via Wavelets.html
Of course that's if you think there is anything to be unlocked

jkeny · Jun 10, 2012

From here: http://www.regonaudio.com/Jitter.html

Again, this is not the issue of "timing" in the sense of precise bass rhythms. Rather, I am talking about the resolution of individual instruments in an orchestral section, the separation of lines in complex musical textures, and so on. Musical people often comment that digital sounds homogenized, incapable of separating the woodwinds sufficiently or of reproducing string sections. And no wonder: the limited bandwidth of CD digital has jittered the signal to the ears, no matter how perfectly the CD's internal clock performance and how low its jitter overall.

The fact that the ear is not sensitive to very high frequencies does not mean it is insensitive to timing errors. The one would follow from the other only for a linear system. But human hearing is highly nonlinear.

So, Tom, instead of thinking in the frequency domain, the answer to your understanding jitter might instead lie in the time domain?

The Analog DAC

Well-Known Member

Member

Well-Known Member

Member

Banned

Member

Well-Known Member

Banned

New Member

WBF Founding Member, Pro Audio Production Member

New Member

Industry Expert, Member Sponsor

New Member

New Member

Banned

Industry Expert, Member Sponsor

Industry Expert, Member Sponsor

Industry Expert, Member Sponsor

Industry Expert, Member Sponsor

Industry Expert, Member Sponsor

Similar threads