DACs with internal master clocks: less importance of transport?

Traditionally the CD transport provided the clock signal to a DAC, with in that respect the transport being the master and the DAC the slave. Thus the CD transport was the deciding factor about the amount of jitter (timing irregularities in the digital signal which may have an effect on sound quality). A lower amount of jitter also appears to be the main reason why CD recordings playing from hard drive are perceived to sound better than CD from a conventional transport.

The next few posts will deal with the issue of clocking in the DAC instead of the transport from a few different angles.

First up will be excerpts from an article by Robert Harley from TAS, then I will reproduce a few thoughts and measurements by LessLoss Audio, and subsequently some comments by Michael Ritter from Berkeley Audio Design with respect to the Berkeley DAC containing a master clock that is rather independent from incoming signals, e.g., from transports.

Finally I will present my own observations from transport comparisons on the Berkeley DAC and my own conclusions.

Robert Harley from TAS performed an interesting experiment, as described in this article:

http://www.theabsolutesound.com/articles/do-hard-disk-drives-sound-better-than-cd/

He begins by saying:

"Many contend that hard-disk drives sound better, much as CD-Rs sound better than the source CD from which the CD-Rs are made.

"I tested this contention by comparing the sound of the Sooloos and Qsonix servers to a state-of-the-art CD transport (the Esoteric P-03), alternately feeding the music server under evaluation and the CD transport to an Esoteric D-03 digital-to-analog converter through the same digital interconnect.

"Listening to the same CD from Sooloos and from the transport, I heard a subtle but noticeable improvement in sound quality from the music server. I heard more space, air, bloom, and soundstage depth when the audio data were read from the Sooloos hard drive rather than from the CD. The hard-drive-sourced sound had better resolution of low-level detail, particularly reverb decay, which is why the presentation sounded more airy and spacious. The presentation was gentler, a little more laid-back and relaxed, and had a greater sense of ease. The upper midrange and treble were smoother and more “organic” sounding when the data feeding the DAC was sourced from Sooloos' hard-disk
drive.

"I repeated this comparison with the Qsonix server and heard a similar increase in resolution, improvement in spaciousness and depth, smoother textures, and more relaxed presentation. In fact, the two music servers sounded very much alike. Much of the current speculation as to why this should be the case focuses on the fact that the Exact Audio importing software used in Sooloos and many other hard-drive-based systems can read sections of the disc multiple times if data errors are detected. The implication is that random bit errors affect sonic qualities such as soundstage depth and the reproduction of timbre.

"I disagree. CD's error-correction system can completely correct burst errors of up to 4000 successive bits. My experience performing bit- for-bit comparisons between source data and replicated CDs when I worked as a CD mastering engineer suggests that bit errors are virtually nonexistent. If CDs were rife with errors, computer software couldn't reliably be delivered on CD (CD-ROM does, however, have an extra layer of error detection and correction but is invoked only when discs have been mishandled and become damaged)." (End quote.)

He then continues with a technical argument why you should hear clicks on CDs if there were actual bit errors upon reading -- since we don't hear clicks there are no bit errors.

This, according to Robert Harley, leaves jitter as the sole difference in sound quality.

"The Esoteric P-03/D-03 combination's separate clock link allowed me to test this hypothesis. This separate clock link slaves the transport to the processor?s clock, a feature that greatly reduces jitter in the D/A?s word clock. (The word clock controls precisely when the DAC converts the incoming samples to analog, which is the point where jitter matters.) The arrangement obviates the need to extract a clock from the S/PDIF interface, a primary source of jitter in separate transports and D/A converters. In addition to avoiding the jitter-inducing effects of the interface, another advantage of a separate clock line is that the D/A converter becomes the timing reference for the entire system. That is, the D/A converter generates the timing reference for the crucial digital-to-analog converter stage, and the transport must lock to this reference. In a conventional connection between a transport and D/A converter, the D/A converter must lock to the transport's timing reference. The result is higher jitter in the clock that controls the actual digital-to-analog converter chips. The separate clock line eliminates these two shortcomings.

"I compared the sound from the music server?s digital output to the Esoteric P-03 transport with and without the separate clock line (you can engage and disengage the clock with a front-panel button). Engaging the Esoteric's clock (lowering the jitter) when listening to a CD rendered an improvement in sound quality virtually identical to what I heard when the music was being streamed from the music servers. This suggests that jitter is, indeed, the explanation for why hard-disk drives sound better than optical discs."

From this article by Liudas Motekaitis from LessLoss Audio:

http://www.forum.audiophile.org/ContentDetails.php?CID=14&ID=108

"The entire audio business of CD transports and DACs is built on the totally backward setup of the CD player containing the Master Clock and the DAC being the Slave. This results in the entire palette of innovations to lessen Jitter, starting from air drives to expensive digital cable technologies with complex math to reduce line-induced Jitter, to very carefully filtered power supplies, to all sorts of very necessary things when you want to achieve the least possible Jitter. So we have the worst possible digital scenario bringing in the most possible amount of money, because it is extremely difficult to annihilate Jitter when the CD player is the Master Clock. If you're looking for quality, this is stupid, to say the least!

"The whole setup should be different. As is the standard case in any pro-audio studio, it is always the playing device, the DAC, which is the Master Clock. The clock is located right next to the converter chips. That way, no line induced Jitter can appear. This clock signal is then taken from the DAC device and is used as the clock input of the signal source device, say the computer, the DAT player, or the CD player. Yes, in that setup, the CD player is receiving a more jittered clock than the DAC is, but that doesn't matter, because the DAC is doing the audio playing. When the Jittered audio signal arrives at the DAC, it is quantized into place temporally and is then played, in perfect synch with the clock oscillator, which is right next to it.

"But oh! In that case, you can use a $100 CD-player with a very poor power supply, a digital cable made from your average household extension cord, and still get a better sound than you'd be getting if you spent $30,000 on the best transport and digital cable! Yes, ladies and gentlemen, this is the naked, uncensored, plain truth about CD players and DACs. Their Master/Slave relationship is BACKWARDS and their prices therefore HIGH." (End quote.)

Of course there is a bit of commercial hyperbole here; he leaves out that a cheaper player may also emit much higher RF noise which, next to jitter, may be degrading the sound as well.

However, he then presents real jitter measurements:

"The image below shows the typical scenario using a CD Player and high quality DAC in the traditional way:

CD Player = Digital Master
External DAC = Digital Slave
Digital S/PDIF Cable Carries Digital Signal and Clock.

It is clearly evident that in this configuration the Jitter is worse than 500 ps.

The following image shows the typical scenario using a CD Player with a high quality clock upgrade, and whose external DAC is also slaved to the CD player, which is in principal the exact same scenario as above, only with a better clock. This is also the case when using a $10,000 transport with the best possible internal clock:

High quality CD Player (or High Quality Clock Upgrade)= Digital Master
External DAC = Digital Slave
Digital S/PDIF Cable Carries Digital Signal and Clock.

It is clearly evident that this configuration achieves less Jitter at about 170 ps.

Now we modify the cheapest CD Player we can get our hands on (in this case a JVC model) and use the external LessLoss Audio DAC 2004 in Digital Master Mode. In this way, the Master Clock is in the DAC and the CD player is connected via two cables: the typical S/PDIF Digital cable and a Digital Clock cable. The results would not differ if you used a $10,000 transport with the best possible drive mechanism or other exotic features such as Housing Metal from Mars, etc.:

Lowest Quality CD Player = Modified as Digital Slave
External DAC = LessLoss Audio DAC 2004 in Master Mode
Digital S/PDIF Cable Carries Digital Signal.
Clock Cable (50 Ohm Coaxial) Carries Clock Signal from DAC to CD player.

You can clearly see that this configuration achieves the least possible Jitter at less than 60 ps. Due to noise resulting from the ADC conversion process, it is not possible using this method to make any judgement on anything less than around 50 ps."

More recently, several DACs other than the LessLoss DAC, such as the Berkeley Alpha DAC, have moved the clock to the DAC, making the DAC the master and the transport the slave.

***

Here is an excerpt from a letter from Michael Ritter, Berkeley Audio Design, to soemone who questioned why the Berkeley Alpha DAC does not have a word clock input:

http://www.computeraudiophile.com/f10-music-servers/clocks-and-alpha-digital-analogue-converter-997/

"Regarding the Alpha DAC not having a word clock input, I understand your concern. It's interesting that clocking is perhaps the area of the Alpha DAC's design we spent the most time and effort on.

"We learned from development of the Pacific Microsonics Model Two that the most critical clock in an audio system is the conversion clock - the internal clock in the ADC or DAC. That clock needs to be extremely stable and have as little jitter as possible. Ideally, the clock in an ADC or DAC should not have to lock to an external clock at all, whether it is derived from a digital signal stream or a separate word clock.

[...]

"In designing the Alpha DAC we spent a long time developing a sophisticated proprietary system with extremely transparent audio quality that allows the Alpha DAC's conversion clock to operate virtually as a master clock for the DAC in high isolation from the incoming clock signal." (End quote.)

In the linked post he then goes on to explain why for hi-rez signals such as 24-bit 176.4 kHz or 192 kHz the signal source also needs to have extremely low jitter. He then recommends the Lynx AES Card for hi-rez (his comments were written in 2008, before Berkeley's own USB converter came to market).

Recently I auditioned the Simaudio Moon Neo 260 DT transport (cost 2 grand) in my system since my 20 year old Wadia 8 transport is dying (problems finding tracks) and spare parts are not available anymore since the unit is simply too old. My DAC is the Berkeley Alpha DAC 2.

I listened to the Simaudio transport with stock power chord and with the new Shunyata power chord for digital components (cost 1 grand). You might expect that this power chord which reportedly works very well on DACs could yield sonic benefits if it enhances the performance of the transport and in the process reduces jitter.

Yet my observations were:
Simaudio Moon Neo 260 DT transport with or without Shunyata power chord -- no difference
Simaudio Moon Neo 260 DT transport vs. Wadia 8 transport -- no difference

I established that there were no differences at all on widely varying music material and by listening to a wide variety of parameters, such as timbre, bass, rhythm&timing, micro- and macro-dynamics, spatial depth and ambience retrieval. Once I thought there was a difference I switched back to the other component (transport or power chord) and convinced myself by critical and when necessary, repeated, listening that there was indeed no difference and that my initial perception of a difference had been an illusion.

All this would make perfect sense if the Berkeley DAC handles all the clocking and differences in jitter output by the two transports are thus neutralized. Given these observations and the above comments by Michael Ritter from Berkeley Audio Design about the DAC, I now would also expect that an Esoteric P-03 transport would not have made a difference. This transport costs 14 grand new but had been offered as demo model with full 3-year warranty for 5 grand at Audiogon. It was tempting, but I am now happy not to have gone with the Esoteric. It would still have been much more expensive than the Simaudio transport, it would have been way to heavy (71 pounds), and if I would have needed a replacement of the VRDS-Neo transport mechanism it could have cost me a lot of money after the warranty had been over. I would now also expect on CD playback hardly any difference with such an alleged miracle transport like the Nova Physics Memory Player -- or with a computer server rather than a transport. Again, if the Berkeley DAC does all the clocking, it should not matter.

I am not certain about digital interconnects, however. Ack (Peter) has recently reported a marked difference on the Berkeley DAC with a top-of-the-line MIT interconnect, and the quality of a digital interconnect could decide about the amount of signal reflection, which may introduce quite a few artifacts that potentially could alter the signal, see:

http://en.wikipedia.org/wiki/Signal_reflection

So all in all, the lay-out of the Berkeley DAC designed by people who came from pro-audio, where apparently it is customary to put the master clock in the DAC, turned out to likely be very beneficial. I get uncanny resolution from the DAC while worries about the importance of transport performance are probably a thing of the past. I suppose the same thing should hold with other DACs that have internal clocks that serve as master clocks.

microstrip said:

BTW, how is it possible that the Berkeley DAC does "all the clocking"? Does it send it back to the transport?

Click to expand...

Michael Ritter, quoted above:

"In designing the Alpha DAC we spent a long time developing a sophisticated proprietary system with extremely transparent audio quality that allows the Alpha DAC's conversion clock to operate virtually as a master clock for the DAC in high isolation from the incoming clock signal."

ack said:

Regarding the MA-X digital cable, more of the credit really needs to go to 1rsw here, who turned me on to it quite a while ago; it's just that back then I couldn't hear a difference with the interconnects used at that point. But in general, digital is much more than just clocking and jitter; reflections and noise would also be important factors.

Click to expand...

Yes, I agree.

The Berkeley does quite a few things quite well, from internal reclocking, to careful board layout to isolate noise, high-noise rejection at the input, etc; but it can't do it all. But slaving the transport to the DAC is not something new either in audio, Spectral and Wadia started doing it decades ago.

Click to expand...

You are right. However, I am a bit confused about this; they slave the transport to the DAC by an extra clock link, like illustrated here:

http://www.tnt-audio.com/clinica/diginterf2_e.html

(Go to "clock backwards configuration")

Yet for the Berkeley DAC, if we can believe MIchael Ritter's claims, they have designed a system "that allows the Alpha DAC's conversion clock to operate virtually as a master clock for the DAC in high isolation from the incoming clock signal."

So there would be no 'slaving' of the transport, and I guess this would also touch upon Microstrip's question above.

Regarding jitter, Amir here had also posted a nice article years ago about it, and the main point is that jitter exists everywhere within a circuit and between components.

Click to expand...

Thanks for that info. Googling "amirm jitter wbf" yields quite a few threads.

MadFloyd said:

Did you try the Shunyata power cord on the Berkely and if so, what were your observations?

Click to expand...

Aah, I wish it would have been easy to do so. My situation is a bit complicated. I moved here from Europe in 1997 and my Tice Power Block II still runs on 240 V with Schuko power connections (my DAC and amps are fed by it). So I would need a high-quality Schuko adapter for the Shunyata chord to make it work.

microstrip said:

Al. M,

As said by another poster, the main question is this "virtually as". In electronics, two clocks can not be synchronized by miracle, and if the Berkeley DAC is the master, the transport must be the slave. As this does not seem to happen, there will be two different clock signals with very close frequencies inside the unit. Unfortunately the sentence as you quoted it seems to be an example of ambiguous and misleading marketing, playing with a specif sense of the word isolation and the words virtually as. Perhaps the sentence gets a different and clear meaning in the full text - some times just quoting short sentences from long texts results in an unfair and unwanted change of the original text.

Click to expand...

Microstrip, I did provide the link to the full text of the letter. Here it is again:

http://www.computeraudiophile.com/f10-music-servers/clocks-and-alpha-digital-analogue-converter-997/