dCS Debussy DAC - Review

DarqueKnight

Part 1 of 3

Introduction

I had been searching for nearly a year for a DAC that wasn't a step backward in performance from my Cary Audio CD 306 Professional Version SACD player, that offered DSD playback, that wasn't prohibitively expensive, that worked well with my digital source and other components, and that wasn't an eyesore.

My system integration issues would have been much easier if my computer audio setup actually used a computer as a music file source, rather than using a specialty digital player (Bryston BDP-2). I have discussed my difficulties with other DACs in other threads, therefore I won't rehash them here. Those other threads are listed in the reference section below.

My research indicated that DACs offered by dCS and MSB technology might be worth pursuing. I never received replies to any of my questions that were emailed to MSB. The people at dCS were very patient and thorough in answering my questions, technical and otherwise. One of dCS's personnel even helped me find a used, mint condition dCS U-Clock at one of their dealers at an attractive price.

So, here we are (me and my two channel system) enjoying the best digital playback ever. The Debussy has only been in service for ten days of nearly constant playback but its arrival has been more like getting reacquainted with an old friend rather than forming a new friendship with a stranger. The only time the Debussy has not been playing, since installation, was during cable changes, fuse changes, or setting up listening trials with another component.


Figure 1. Such Good Sound from a very modest, entry level dCS stack, top to bottom: dCS Debussy DAC, dCS Puccini U-Clock, PS Audio PowerBase isolation platform, Black Diamond Racing Mark 4 carbon fiber isolation cones and pucks.

DCS recommends a break in period of 4 to 7 days for a new Debussy. The Debussy sounded harsh in the highs and thin in the bass at first listening, but the harsh highs were gone in 24 hours. The bass didn't fully come into focus until the seventh day. The imaging, sound staging, and overall clarity and detail was better than either of my Cary DACs (CD306PV ($8,000) and DMC-600SE ($8,000)). I was advised by Cary, and others, that I needed to look at five figure solutions if I wanted something substantially better than the CD306PV. The Debussy retails for $11,500. I actually would have been satisfied with a DAC that equaled the CD306PV in sound quality, in addition to offering DSD playback and working well with my Bryston BDP-2 digital player.

Critical listening began on the eighth day after installation.

Listening Evaluation Methodology

My performance evaluation criteria focuses on stereophonic performance. Each component under evaluation plays the same musical selections. I map the location of images in the sound stage both laterally and aerially. I make detailed notes describing the characteristics of all sound images (clarity, detail, sonic weight, tactile sensation, etc.). Then I compare my listening notes. Often, I am able to pick up on performance differences in real time during listening. Sometimes, I do not become aware of differences until I compare my notes. This is especially true of spatial and tactile sensation differences.

Comparison To The Cary DACs

The dCS Puccini U-Clock master clock ($5,500) enhanced the sonic performance of both my Bryston BDP-2 and the Debussy. The BDP-2 does not have a word clock input. However, the BDP-2's USB output can be synched to the U-Clock and then the U-Clock converts the incoming USB digital stream to outgoing SPDIF to the Debussy. The U-Clock was not used when comparing the Debussy to either of the Cary DACs. Also, the CD306PV is not in stock form. Its power line fuses were replaced with HiFi Tuning Supreme fuses and the transport and interior had extensive treatment with Dynamat Xtreme vibration damping material.


Figure 2. David and Goliaths - front to rear: dCS Debussy DAC, Cary Audio DMC-600SE CD player/DAC, Cary Audio CD 306 Professional Version SACD player/DAC.

As would be expected, the CD306PV and DMC-600SE sound very similar, but the DMC has a small advantage in overall clarity and detail. The performance gap between them would have been wider if not for the CD306PV's tweaks.

The Debussy was in a class above the Cary's in overall clarity and detail and in image weight, depth, and stereophonic holography. More detail is not always a blessing however, as the Debussy offered more spotlight on the poor recording quality of some many pop, rock, and R&B recordings. However, on well recorded material, particularly well recorded DSD files, the presentation was outstanding.

With regard to tonal balance and overall "sound", I did not find the Debussy to be a great departure from the Cary's. I just got more of what I like in every aspect of stereophonic performance.

DarqueKnight

Part 2 of 3

Evaluation Summary

The Debussy does not offer user selectable upsampling options. Input is either upsampled to 5 bit 2.8224 Mega samples per second for 44.1k, 88.2k, 176.4k, and DSD64 sources or to 5 bit 3.072 Mega samples per second for 32k, 48k, 96k, and 192k sources. The Debussy accepts DSD and PCM on all inputs (USB, AES, SPDIF).

The enhancements brought by the U-Clock are similar in degree to that of switching between stereo and mono on a preamp.

Further spatial enhancements were realized by changing the stock fuses of the U-Clock and Debussy to HiFi Tuning Supreme fuses. Referring to figure 6, when playing Michael Franks' "Bwana He No Home", with the stock fuse, and the U-Clock out of the chain, the lead guitar was just behind the chair on the right. The lead vocal was just in front of the middle of the equipment cabinet, the drum kit was behind the vocalist, the saxophone was to the left and just behind the vocalist, the piano was co-located with the left speaker, and the conga drums were just in front of the left speaker.

Replacing the Debussy's stock fuse with a HiFi Tuning Supreme fuse resulted in the following spatial differences: the lead guitar moved a foot further back, the vocalist moved a foot forward, the drum kit sounded from a couple of feet beyond the rear wall, the saxophone moved back one foot, the piano moved forward 1 foot and 2 feet to the left of the left speaker, the congas moved forward one food forward. There was more tactile sensation in the bass.

Turning on the U-Clock with its stock fuse, all images remained in the same location, but there was a more holographic presentation with more tactile sensation in the bass and more image weight. Replacing the U-Clock's stock fuse with a HiFi Tuning Supreme resulted in a small improvement in image weight and bass tactile sensation.


Figure 3. The Debussy's remote is machined from a block of black anodized aluminum. It weighs 13 ounces. I copied its codes into my plastic and glass Harmony One remote and put the nice Debussy remote in storage.


Figure 4. Left to right source components: Teres Audio Model 255 turntable with Graham Phantom II tonearm and Ortofon MC Windfeld moving coil phono cartridge, Bryston BDP-2 digital player, dCS Debussy DAC and dCS Puccini U-Clock. My vinyl cravings have decreased considerably since the Dubussy DAC/Puccini U-Clock went active in my two channel system.


Figure 5. The Debussy's sleek form factor and subtle blue LEDs are a perfect aesthetic match for my two channel system.

Nothing's Perfect


Figure 6. The Debussy was a near perfect fit for my computer audio setup.

The Bryston BDP-2 is somewhat picky about who it plays with. With some DACs, there is a click or muted output whenever the next song has a different sample rate. This is due to the DAC needing time to lock on to the new rate. The Debussy has this issue to a small degree. The DMC-600SE had this issue to a large degree. My Cary CD306PV and Cary DAC-100 didn't have the issue at all. This issue would be resolved by being able to insert a 2 second gap between songs. However, this is a minor issue since most of my music lists are all 44.1k PCM or all DSD64.

Since the BDP-2 does not have a word clock input, only its USB output can be properly synched to the U-Clock...but its AES (best) and SPDIF (second best) outputs, via the new Integrated Audio Device upgrade sound card, are the better sounding inputs. However, the USB connection with the U-Clock sounds better than the AES connection without the U-Clock.

If use an AES connection between the BDP-2 and Debussy while the U-Clock is connected to the Debussy, The AES data stream will be synched enough to maintain a lock on the data rate, but I get periodic "dropouts" due to the disparity between the Debussy's clock rate and the imperfectly synched AES input from the BDP-2. The "dropouts" are heard as an intermittent gradual lowering of sound level or as faint static-like noise.

Compared To Vinyl?


Figure 7. So beautiful, yet so barbaric. I wish I knew how to quit you.

I'm ashamed to admit that I occasionally listen to music via dragging a small rock through plastic record grooves. Most of my records, relatively speaking, sound like trash, but the good ones are very good and the great ones are spectacular. The best sounding recording I own is the Classic Records 45rpm 4-disc edition of Dave Brubeck's "Time Out" album. The heavy, realistic, holographic, and exciting, sound field that this record creates is one of the wonders of the audio world. Comparing this record to my DSD rip of the SACD played through the Debussy was like comparing the sound and fury of a thunderstorm to a gentle spring rain. What Confucius said over 2000 years ago is still true today: "The key to enjoying great digital is never to listen to great analog." Bear in mind that I'm someone who viscerally despises the maintenance, nuisance playback procedures, and other rigmarole associate with vinyl records. Also bear in mind that my vinyl gems are few and far between and that the digital versions of most of my records sound as good as, or better, than their vinyl counterparts. There is also the situation of some of my favorite records not being available in digital format. For perspective, the total retail price of my analog setup (turntable, record clamp, tonearm, and cartridge) is $14,594. The total retail price of my digital setup (digital player, DAC, clock) is $20,495.

DarqueKnight

Part 3 of 3

Happy Ending

The Debussy/U-Clock combination was the big payoff to a long and often frustrating search for a satisfying digital solution. In the future, when I am more dedicated to audio than I am now, and when that long-lost rich relative wills me their fortune, I'll probably move deeper into the rabbit hole and further up the dCS hierarchy. I'd love to hear what the "really good stuff" sounds like.

Associated Equipment

Teres Audio Model 255 turntable
Graham Phantom II tonearm
Ortofon MC Windfeld phono cartridge
Sonic Purity Concepts and Design "The Clamp" record clamp
Bryston BDP-2 digital player
dCS Puccini master clock
Cary Audio DMC-600SE DAC
PS Audio PowerBase isolation platforms for turntable, BDP-2, Debussy DAC and Puccini U-Clock
Black Diamond Racing isolation Pits and Mk IV Cones
Pass Labs XP-30 line level preamplifier
Pass Labs XP-25 phono preamplifier
Pass Labs X600.5 monoblock power amplifiers
Revelation Audio Labs Prophecy Cryo-Silver Split Configuration USB cable
Revelation Audio Labs Prophecy Cryo-Silver S/PDIF RCA-RCA coaxial cable
Revelation Audio Labs Prophecy Cryo-Silver S/PDIF BNC-BNC coaxial cable
Revelation Audio Labs Passage Cryo-Silver DB-25 power umbilicals for XP30 and XP-25 preamps
AudioQuest Sky XLR interconnects
AudioQuest Everest speaker cables
AudioQuest LeoPard tonearm cable
PS Audio PerfectWave AC-12 power cords
PS Audio PerfectWave P-10 AC Regenerator
Polk Audio SDA SRS 1.2TL loudspeakers (heavily modified)
Salamander Synergy Triple 30 audio credenza

References

http://forum.polkaudio.com/discussion/168179/the-year-of-dacs/p1

http://forum.polkaudio.com/discussion/166413/bryston-bdp-2-digital-player-review

http://forum.polkaudio.com/discussion/168871/pure-overkill-computer-audio-office-system

http://forum.polkaudio.com/discussion/153600/further-vibration-abatement-for-the-cary-audio-cd-306-pro-version-sacd-player

tonyb

Outstanding and thorough review as always Ray. Have to admit though, I got a chuckle out of that sweet vinyl rig.....from someone who despises vinyl. LOL I don't think I would keep a 14k investment to play a couple albums I thought were great.

What I took away from your review, aside from the stellar performance of the DCS, is that the real estate between that and the Cary isn't all that great in overall sound. Which for me, further cements my thoughts on the SQ of the Cary.

Glad you found your solution, keep playing the lotto and maybe you can move up the chain but at the rate your going even that may not be enough.

DarqueKnight

tonyb wrote: »

I don't think I would keep a 14k investment to play a couple albums I thought were great.

Well, those are the types of things you do when you get caught up in the rabbit hole.

tonyb wrote: »

What I took away from your review, aside from the stellar performance of the DCS, is that the real estate between that and the Cary isn't all that great in overall sound. Which for me, further cements my thoughts on the SQ of the Cary.

Yes, from outside the sweet spot, they are similar in overall sound...on my rig. If sat in the sweet spot and ignored the spatial properties, again they are similar in overall sound. When concentrated on sound stage, image placement, image weight, and fine detail, the Cary units were left at the starting blocks. There have been many "WTH-where did that come from " moments over the past few days when I heard something in very familiar recordings that I had never heard before.

tonyb wrote: »

Glad you found your solution, keep playing the lotto and maybe you can move up the chain but at the rate your going even that may not be enough.

I think I'm set with electronics...for a while. I'd like to turn my attention back to my decades-long quest for a set of conventional speakers whose imaging properties I can live with long term.

tonyb

Curious on something Ray, and maybe you can provide a short answer. Especially since I know before you commit to a piece you investigate it top to bottom and sideways.

What exactly is the DCS u-clock doing that the clock in either of the any other dacs isn't ?

I would assume, and maybe I'm wrong, that the clocks embedded in high priced dacs shouldn't be too shabby. Is it simply a matter of build quality or is there some dedicated process that differs in the DCS u-clock ?

I ask because as per your review, it appears the U-clock is whats making the real solid improvements in SQ.

mrbiron

Great read as always but there is something that i always question when coming across your posts. I know you've answered it before but i'm not going to wade back through it all....

When are you going to try a new SPEAKER? It strikes me as odd that you have tens of thousands of dollars in a system yet you still run 1.2's, heavily modded of course.
I assumethe real issue is you've tailoring the sound to fit the Polks. Bringing in a new speaker would require you to go back through your gear. Audio is a fickle b*tch.

DarqueKnight

tonyb wrote: »

What exactly is the DCS u-clock doing that the clock in either of the any other dacs isn't ?

The U-Clock is providing an enhancement to an already stellar performing DAC. Any thing built to a price point has to make compromises, and therefore has room for improvement.

The U-Clock's more precise clock signal removes another layer of noise and lets more of the music signal shine through. Another very important advantage of an external master clock is synching all the digital components to the same clock signal.

We have reached a point where jitter in modestly priced digital gear is "below measurable levels". Some people take this to mean that the jitter is at a level where it is of no concern. That's like saying that someone with "undetectable" HIV in their blood is perfectly healthy. All "below measurable levels" means is the jitter cannot be measured with the lab instruments commonly used to measure it. More sensitive and specialized tools must be used. Just because the jitter is below measurable levels doesn't mean it doesn't have audible effects. Realizing better sound and spatial performance by adding a higher precision clock to a component that already has "jitter below measurable levels" demonstrates that what can't be easily measured can have significant audible effects.

tonyb wrote: »

I would assume, and maybe I'm wrong, that the clocks embedded in high priced dacs shouldn't be too shabby. Is it simply a matter of build quality or is there some dedicated process that differs in the DCS u-clock ?

The clocks in high quality DACs are excellent, but they are built to a price point. Including a more precise clock in the Debussy would raise the price substantially. A lot of people wouldn't be willing to pay extra for something that is essentially a tweak. Ultra precise clocks are expensive to manufacture. One of the main concerns is building a clock that will maintain its stability and accuracy over time.

The U-Clock uses an OCXO (oven controlled crystal oscillator), which keeps the quartz crystal raised above ambient temperature to improve its frequency stability.

"A crystal oven is a temperature-controlled chamber used to maintain the quartz crystal in electronic crystal oscillators at a constant temperature, in order to prevent changes in the frequency due to variations in ambient temperature. An oscillator of this type is known as an oven-controlled crystal oscillator (OCXO, where "XO" is an old abbreviation for "crystal oscillator".) This type of oscillator achieves the highest frequency stability possible with a crystal.

Achieving better performance requires switching to an atomic frequency standard, such as a rubidium standard, cesium standard, or hydrogen maser."

High quality OCXOs can cost in the thousands of dollars. Furthermore, incorporating one in the design of a DAC presents some challenges due to the heat issues.

https://en.wikipedia.org/wiki/Crystal_oven

tonyb wrote: »

I ask because as per your review, it appears the U-clock is whats making the real solid improvements in SQ.

No, the Debussy is excellent on its own. The U-Clock, like any other high quality enhancement, only enhances the Debussy's performance. Think of the Debussy as a window pane that has a thin film of oxidation. The U-Clock removes that oxidation film. Going to clearer piece of glass (a higher performing DAC) would provide an even clearer view.

mrbiron wrote: »

Great read as always but there is something that i always question when coming across your posts. I know you've answered it before but i'm not going to wade back through it all....

When are you going to try a new SPEAKER? It strikes me as odd that you have tens of thousands of dollars in a system yet you still run 1.2's, heavily modded of course.
I assumethe real issue is you've tailoring the sound to fit the Polks. Bringing in a new speaker would require you to go back through your gear. Audio is a fickle b*tch.

I don't tailor the sound of my system to fit the Polks, I tailor the sound of my system to fit the sound of live music. The sound of a live band in a good acoustic space is the standard by which I measure audio system quality.

Since I have a performance goal in mind, which serves as a measurement standard, I am not going to go in a direction which takes me in a radical new direction. I am going to choose gear that takes me further in the direction I want to go.

I've tried lots of new speakers, some costing in the tens of thousands of dollars. None of them had the sound stage stability of the SDAs. Another issue is that I prefer tall speakers that generate tall sound stages. Larger, high performance loudspeakers are not common and it is not easy (for me) to find places to audition them. For example, from what I have read and from what I have been told, Sonus faber's Aida would be a speaker worth pursuing. They retail for $130K but can be found on the used market for less than half that. There was a pair on Audiogon in March 2015 for $48K, another pair for $53K in September 2015, and there is a pair now for $68K. However, I wouldn't make that kind of investment without extensive listening, and the closest SF dealer to me that has a pair of Aida's set up is over 900 miles away.

tonyb

Thank you Ray, a well thought out answer just like your reviews is much appreciated. I also like a taller speaker for the same reasons btw.

Now, if clocks are built to a price point, which as we all know everything is, whats top dog no holds barred, no price points dictating results ? I'm just having a hard time wrapping my brain around spending 10k roughly on a piece, then having to spend another chunk to overcome a certain portion of that piece.

By all logic, the U-clock too is held back because it too is built to a price point, no ? I would inject the old "diminishing returns" theory into this but I know people will pay for those smaller increases in SQ and I have no problem with that. Life is too short to worry about diminishing returns.

I guess what I'm getting at Ray is, where does it end ? When you spend a million bucks on a clock/digital transmission ? Or when your satisfied with the sound ? There is always going to be better around the corner, different price points....yada yada. I certainly admire your chase for better sound, and god bless you have the ability to play on that level, but will you ever be satisfied ?

Or...is it simply a matter of trying to achieve the best sound one's wallet will allow ? God knows we all do that right ?

Moose68Bash

@DarqueKnight,

Thank you for your extraordinarily clear explanation of the benefits of a U-Clock that is isolated from the DAC.

I also learned from your explanation of the reason you have "stuck" with your SDAs. Given the geometry of my family room, which unfortunately has to also serve as my listening room, my SDAs are as far as I have been able to determine the best speakers for me. Like you, I'm trying to reproduce, as nearly as possible, the sound of a live performance.

You have undoubtedly looked at alternatives to the SDAs much more than I, and I have found only one that really intrigued me. Even my wife was ready "to pull the trigger" on them, but they posed installation problems that made it a "no go" in the end -- e.g., they required two separate amplification systems, one for the full-range towers and another for the bass towers, and we were not happy with any of the possibilities we envisioned for locating all that equipment.

These were a set (four towers) of Nola Grand Reference speakers (Series II, as I recall). We auditioned them in suboptimal conditions (a warehouse), but the realism of the sound they produced even there was absolutely stunning. They are tall, which like you I believe is important, and the speed, frequency range, imaging, etc. were all well beyond anything I had heard.

I noticed a set of these that is an older version ( I think) with some upgrades is no for sale on Audiogon:

https://app.audiogon.com/listings/full-range-nola-speakers-exotica-grand-reference-loudspeaker-system-2015-11-10-speakers-90038-los-angeles-ca

Have you ever considered or listened to one or another of the speakers in this top-end line of Marchisotto's designs?

Nightfall

To me this just proves that there's nothing wrong with the fact that I didn't prefer the Focal 1028's to my vintage Polks. If SDA's can hang on in a system costing as much as Rays - they're better than the Polk brand gets credit for, which people seem to think are only a gateway into high end.

SCompRacer

Excellent job, as always Ray.

tonyb wrote: »

Or...is it simply a matter of trying to achieve the best sound one's wallet will allow ? God knows we all do that right ?

That's what I do. I'm not gonna second mortgage for audio gear.



DIY'ers have been addressing the clock for years, and power, and inputs, outputs...etc.

DarqueKnight

tonyb wrote: »

Now, if clocks are built to a price point, which as we all know everything is, whats top dog no holds barred, no price points dictating results?

The "top dog" would be the bleeding edge state of the art piece, whatever that happens to be at the time. As I discussed in another thread, achieving perfectly shaped digital pulses (square waves) is technically impossible. The slanted sides on the leading and falling edges of the pulses represent error that affects the timing accuracy of the pulses. As you go up in clock quality (and up in price), that error, and the audible noise that comes with it, is diminished. DCS has two, more accurate, clocks the Rossini ($7,500) and the Vivaldi ($13,500). All of dCS's clocks are specified as better than +/- 1 part per million accuracy. Above the dCS clocks are atomic standard clocks such as the Esoteric G-01 Rubidium Clock ($21,000) which is specified as +/- 0.05 parts per billion accuracy.

It is doubtful that the Rossini, Vivaldi, or G-01 clocks would outperform the Puccini U-Clock in my system because only the U-Clock has a USB input that allows my BDP-2 digital player to be synched to it. All of the other clocks are designed to only work with components that have word clock inputs.

tonyb wrote: »

I'm just having a hard time wrapping my brain around spending 10k roughly on a piece, then having to spend another chunk to overcome a certain portion of that piece.

Come on Tony, what you just stated is a basic reality of any high performance product. People spend $100k+ on a Porsche and then drop another $10k on tires, wheels, and shocks. Most people are perfectly happy with the Porsche's stock equipment, but every product that is not bleeding edge state of the art has design compromises, and there are always performance enthusiasts who don't mind spending money to remediate those compromises.

My dCS components came with a full set of power and interconnect cables, and this advisement:

"The cables supplied are commercial-grade parts, as we realise our customers will want to make their own choices."

With that in mind, I did make my own choices, and the retail cost of the choices for replacing the U-Clock's stock cables were as follows:

Revelation Audio Labs RCA-RCA SPDIF 1 meter digital cable $699.
Revelation Audio Labs BNC-BNC SPDIF 1 meter digital cable $699.
Revelation Audio Labs Dual-A Head USB 1.25 meter digital cable $649.
PS Audio Premier SC 1 meter power cord $995.

Sooooooo, $5,500 retail for the clock and another $3,042 retail to replace the clock's $2 "commercial-grade" cables? Yes indeed. Gladly paid with no hesitation at all once I heard the difference.

Other accessory tweaks for the U-Clock:

HiFi Tuning Supreme power line fuse $60.
PS Audio PowerBase isolation platform $995.
Black Diamond Racing Mark 4 carbon fiber isolation cones $135 for 3.
Black Diamond Racing Mark 4 carbon fiber mini pits $135 for 3.

Adding all of the above comes to $4,367, which is nearly 80% of the U-Clock's $5,500 retail price.

I showed the measurable differences in the U-Clock's timing signal with different cables here:

http://forum.polkaudio.com/discussion/165959/digital-interconnect-cables-whats-your-experience/p4

tonyb wrote: »

By all logic, the U-clock too is held back because it too is built to a price point, no ? I would inject the old "diminishing returns" theory into this but I know people will pay for those smaller increases in SQ and I have no problem with that. Life is too short to worry about diminishing returns.

It depends on what you value with regard to stereophonic performance. If I wasn't an imaging and sound stage junkie, the benefits brought by the Debussy, and further enhanced by the U-Clock, wouldn't be worth the extra cost. As I said before, the Debussy and my Cary DMC-600SE and Cary CD306PV are very similar in overall sound. Listening to them outside the stereo sweet spot, I think most people would have difficulty telling them apart. However, inside the stereo sweet spot, the Debussy's superiority in depth, image weight, stereophonic holography, tactile sensation, fine detail, and bass articulation are striking.

tonyb wrote: »

I guess what I'm getting at Ray is, where does it end ? When you spend a million bucks on a clock/digital transmission ? Or when your satisfied with the sound ? There is always going to be better around the corner, different price points....yada yada. I certainly admire your chase for better sound, and god bless you have the ability to play on that level, but will you ever be satisfied ?

It ends for me when I am as close to the standard of live music performance that my funds allow. I am satisfied with my stereo system, but that does not mean I don't enjoy hearing what else is out there that might offer higher performance.

My search for a new DAC was motivated by new features that I needed, not because I wanted to beat the CD306PV's performance. The first thing I did was consult Cary to see if my CD306PV SACD player could be modified to play DSD files. After receiving a "no", I started looking at DSD DACs. The main thing the Debussy offers is that it plays all of my music file formats and it is not a step below the Cary in sound quality and it integrates well with my existing digital equipment. The fact that it surpasses the Cary in sound quality is icing - and a thick layer of icing at that.

tonyb wrote: »

Or...is it simply a matter of trying to achieve the best sound one's wallet will allow ? God knows we all do that right ?

When I am shopping for anything, I have quality, features, and performance goals in mind. The solutions I choose are the ones that meet, exceed, or come as close to, those goals as my funds allow. I know what my favorite musical instruments sound like, therefore it is easy to gauge how close a piece of stereo gear gets me to that standard.

DarqueKnight

SCompRacer wrote: »

DIY'ers have been addressing the clock for years, and power, and inputs, outputs...etc.

Some years ago I replaced the off-the-shelf crystal clocks in my Denon DCD-1650AR and Sony XA7ES CD players and was shocked at the level of improvement. The upgrade clock kits cost around $250, but transformed the clarity, detail, and imaging performance of the players.

While we are on the subject of DIY, it is interesting amusing how anti-audiophile naysayers ridicule audiophiles for things like CD player clock upgrades, power cable, speaker cable, and interconnect cable upgrades, and isolation tweaks when these things all originated in the DIY community, where the driving motivation is increased audio performance for as little monetary outlay as possible.

Moose68Bash wrote: »

You have undoubtedly looked at alternatives to the SDAs much more than I, and I have found only one that really intrigued me.

Finding speakers that exceed my modified SDA's clarity and detail is easy. Finding speakers that match, let alone exceed my modified SDA's stereophonic imaging properties has been impossible. Even when I have auditioned conventional speakers that generated a large sound stage, the illusion is very fragile and quickly collapsed with even small head movements.

Imaging is my top priority, because, you know, that's the main point of stereo.
Clarity and detail are second priorities.

To my knowledge and according to my experience, for most audiophiles, the priorities above are reversed.

Moose68Bash wrote: »

Have you ever considered or listened to one or another of the speakers in this top-end line of Marchisotto's designs?

I have never considered or auditioned Nola speakers. They visually remind me of the Infinity IRS Beta and Infinity RS 1B four tower systems.

SCompRacer wrote: »

I'm not gonna second mortgage for audio gear.

I have read accounts of people doing this and it puzzled me. I couldn't take equity out of an appreciating asset like my home, and spend it on a rapidly depreciating asset like audio gear. It's their money though.

Nightfall wrote: »

To me this just proves that there's nothing wrong with the fact that I didn't prefer the Focal 1028's to my vintage Polks. If SDA's can hang on in a system costing as much as Rays - they're better than the Polk brand gets credit for, which people seem to think are only a gateway into high end.

I always explain it as the SDAs were far ahead of their time. The quality of crossover components, not to mention low noise amplifier designs, available in the 80s and 90s did not allow the SDA designs to be implemented and perform at optimum levels.

I don't have any sentimental attachment to my highly modified SDA SRS 1.2TLs. They are still here because they offer the best stereophonic sound per my requirements. When I began my high end upgrade adventure in 2006, I fully expected the 1.2TLs to be replaced. From a certain point of view, I did "replace" them because the only original parts are the 6.5" drivers, 15" passive radiator, and the MDF cabinets. Everything else, internal wire, binding posts, wood trim, grille cloth, tweeters, crossover, and interconnect cable, were replaced with much higher performing parts. The cabinets, passive radiator, and drivers are not completely stock either. Extensive vibration damping treatments were applied throughout: driver and tweeter retaining rings, Dynamat Xtreme on driver and passive radiator baskets, Armacell gaskets, and Black Hole strips on the cabinet interiors.

When I find a speaker that offers better overall stereophonic performance, and that fits my budget, and that doesn't offend my aesthetic tastes, I'll buy it. Until then, the Mysterious Monoliths will continue to laugh at more expensive stereo loudspeakers that are compromised by the comb filtering effects of interaural crosstalk.













As you are reading this I am "touching" my SDAs.

ken brydson

https://www.youtube.com/watch?v=-FucbvoFFy0

See my sig...

tonyb

Ray,

Thanks again for further explaining the clock details. Please don't take my comments personally, I think we know each other well enough by now.

I'm speaking from the average joes point of view, obviously audio has many levels most of us will never see and that doesn't excuse it's relevance at higher levels either.

Thoroughly enjoy the education though, and the science behind it.

Maybe I'm just wondering something many around here say...."At some point you stop chasing gear and simply enjoy the music".

I'm also wondering if those errors in the digital transmission in the form of square waves measured in the billions + or-, is even audible.

One more thought.....I think if you go searching for new speakers now, you might throw the synergy you have now off kilter triggering another round of gear gettin'.

btw...I agree that although something is measureable, or not measureable, would have total control over the sound. That's probably where the rough edges as you say on the square waves comes in, but when we start talking in the billions + or-, I have to start wondering if something else in the build is responsible for addition audible sound quality improvements.

DarqueKnight

tonyb wrote: »

Please don't take my comments personally, I think we know each other well enough by now.

I'm speaking from the average joes point of view, obviously audio has many levels most of us will never see and that doesn't excuse it's relevance at higher levels either.

Tony, nothing personal taken. Your comment was somewhat puzzling to me because it wasn't clear to me that you were playing "devil's advocate" and speaking from the viewpoint of the average joe.

tonyb wrote: »

Maybe I'm just wondering something many around here say...."At some point you stop chasing gear and simply enjoy the music".

True audiophiles don't chase gear, they chase a sound performance. It is an unfortunate stereotype and misconception that audiophiles are lovers of expensive gear. The truth is that most audiophiles are cheapskates and try to get the best sound possible for as little expense as possible. There are those who routinely switch out gear because they assume that the latest and greatest...and most expensive...must be better. If you ask such people what aspects of stereophonic performance they are trying to improve with a new piece of gear, they couldn't tell you.

tonyb wrote: »

I'm also wondering if those errors in the digital transmission in the form of square waves measured in the billions + or-, is even audible.

It is possible to measure the noise and distortion content in the audio signal delivered to the speaker and see if it falls within the range of human hearing.

tonyb wrote: »

One more thought.....I think if you go searching for new speakers now, you might throw the synergy you have now off kilter triggering another round of gear gettin'.

I prefer speakers, electronics, and cables with a basically neutral and balanced sound, so anything I choose would fit those criteria. Having a sonic performance goal in mind keeps me from going off in radical directions.

tonyb wrote: »

btw...I agree that although something is measureable, or not measureable, would have total control over the sound. That's probably where the rough edges as you say on the square waves comes in, but when we start talking in the billions + or-, I have to start wondering if something else in the build is responsible for addition audible sound quality improvements.

The parts per million and parts per billion clock accuracy refers to how much the clock frequency deviates from the theoretical ideal, not the shape of the square wave clock pulses. The noise induced by clock jitter can "snowball" into other types of noise, similar to the way that mechanical vibration can snowball into electrical noise that has an audible effect.

nbrowser wrote: »

DarqueKnight wrote: »

As you are reading this I am "touching" my SDAs.

...just a tip is all...lol

No pun intended...right?

tonyb

Ha...yeah Ray, I'm not beneath playing Devils advocate on occasion. Your comments do have me wondering about yet another question....

Is jitter able to accumulate in ones system ? For some reason I have it in mind that jitter, if present, is dependent on the receiving end of the down field piece as to how much will get by.

I know you mentioned the snowball effect and that indicates a Yes to my question, but how do you account for the other pieces in the chain and what they do for jitter ?

DarqueKnight

tonyb wrote: »

Is jitter able to accumulate in ones system ?

Yes. Having all digital components synched to the same clock signal addresses the cumulative effect of jitter from differently clocked sources.

tonyb wrote: »

I know you mentioned the snowball effect and that indicates a Yes to my question, but how do you account for the other pieces in the chain and what they do for jitter ?

It depends on how well the other pieces mitigate noise. It is similar to a source component generating a certain amount of electrical noise, then the preamp generates a certain amount of electrical noise, then the power amp generates a certain amount of electrical noise...and along the way, each cable will generate a certain amount of electrical noise. The signal presented to the loudspeaker will exhibit the cumulative effect of noise in everything the signal passed through.

tonyb

By the same token Ray, isn't it at least conceivable that along the way each piece can also reduce jitter further. An opposite of the snowball effect ?

I would think so, of course depending on the gear in question. Also, and forgive me if this is a dumb question because I'm just trying to get a better grasp on this....

When you have a piece that improves the formation of square waves, then it hits another piece in the chain, aren't those square waves formed all over again by the lessor quality piece ? I'm trying to follow the benefit of improving the formation of those square waves to the end, which is the speakers. Would it be logical to keep the best piece in the chain, and it's signal, as close as possible to the last output piece before it hits the amp and speakers ?

Just for the record, I've always had an inclination that jitter was cumulative, I just can't explain it. My warped brain says that if jitter can accumulate down the chain from piece to piece, it can also be reduced in the same manner. Am I wrong to think this ?

DarqueKnight

tonyb wrote: »

By the same token Ray, isn't it at least conceivable that along the way each piece can also reduce jitter further?

Of course. This is what the U-Clock does. It takes the data stream from the BDP-2's USB output, converts it to a re-clocked SPDIF data stream, and sends it to the Debussy.

tonyb wrote: »

When you have a piece that improves the formation of square waves, then it hits another piece in the chain, aren't those square waves formed all over again by the lessor quality piece?

It depends on what the digital component is designed to do. The digital data may or may not be processed in such a way to improve (or diminish) its signal quality.

What you have in a digital playback chain is a digital music file source and a digital to analog converter that converts the digital data in the music file to electrical analog waves that eventually arrive at the speaker to be converted to analog acoustic waves.

In my case, the digital data from the source (BDP-2) is processed by conversion to SPDIF and re-clocked. I could bypass the U-Clock and feed either the USB, SPDIF, or AES data stream directly into my DAC. Before the Debussy converts the incoming digital signal to analog, it is converted to a 5 bit per sample 2.8224 Mega samples per second or a 5 bit per sample 3.072 Mega samples per second signal.

tonyb wrote: »

I'm trying to follow the benefit of improving the formation of those square waves to the end, which is the speakers.

The output of the DAC is not square waves, but sinusoidal waves. The more accurate and clean the digital pulse train that goes into the DAC, the more accurate and clean the analog sinusoidal output will be.

tonyb wrote: »

Would it be logical to keep the best piece in the chain, and it's signal, as close as possible to the last output piece before it hits the amp and speakers?

There shouldn't be a lot of pieces. In the case of a one box solution, the file player, clock, and DAC are all in the same chassis. In the case of a separate digital playback system, what you would have is a file player and a DAC, and maybe a master clock for the player and the DAC.

tonyb wrote: »

Just for the record, I've always had an inclination that jitter was cumulative, I just can't explain it. My warped brain says that if jitter can accumulate down the chain from piece to piece, it can also be reduced in the same manner.

That's true. Again, it depends on how each piece processes the digital data.

[Deleted User]

DK,

Would you mind sharing your selection of tracks? The break down of what you heard different, for better or worse, at what time intervals?

halen

DarqueKnight

halen wrote: »

Would you mind sharing your selection of tracks? The break down of what you heard different, for better or worse, at what time intervals?

I'll list a few. Listing everything, would give me writer's cramp.
All tracks are 44.1k FLAC files ripped from CD unless otherwise noted. DSD64 files are single rate DSD files ripped from SACD.

General Observations

I notice a lot more percussion and background details than before.

With the Cary DACs, I would occasionally hear a song with sound stage depth projected beyond the wall behind the speakers. I hear this fairly often now, on recordings where this amount of depth was previously not perceived.

There is greater distance in sound quality between PCM and DSD, and between PCM of different sample rates. I used to wonder if higher rate PCM was a scam, because I didn't hear a difference between 44.1K, 88.2K, 96k, 176.4k, and 192k PCM. Now, I do. Since the Debussy upsamples all PCM to single rate DSD, one might think that the original sample rate wouldn't matter (much), but it does. The more digital "raw material" you start with, the better the end result.

DSD recordings are such a small part of my music question that I questioned whether I should even invest in a DSD capable DAC. Now, I would totally abandon PCM if it were practical and economical to do so...and if the quality of the recordings took full advantage of DSD's higher resolution.

After a session of good vinyl, I experience "digital depression" where my ears have to readjust to the lower resolution and lower holography of digital. Now, after a session of good DSD, I experience "PCM depression" where my ears have to readjust to the lower resolution and lower holography of PCM. Again, bear in mind that this is with a DAC that upsamples PCM to DSD64.

Isunova Pi by E. S. Posthumus

Thunderous drum beats at the intro have more sustain and decay and more tactile sensation. I can hear more details in the background percussion instruments.

Train A Goin' by Sheila E

The Debussy brings out more of the locomotive's "clackity-clack" sounds as it is moving along the track. The bass beat sound effects at the beginning are heavier, more resonant, and bigger in size.

Till We're Together Again by The Rippingtons

This is one of my favorite recordings for measuring bass transients and bass articulation (bass solo starting at 3:56).

Fireopal (Acoustic Version) by Ottmar Liebert - 88.2K FLAC

The Debussy adds a lot of weight to the flamenco guitar's wood body and string overtones.

You Don't Know What Love Is by Sonny Rollins - DSD64, 44.1k FLAC

This is a mono recording. The DSD has a moderate advantage in bass weight, bass detail, and clarity in the saxophone, particular the mouthpiece and reed noises. The FLAC rip sounds louder.

I wish this album was available in stereo, without all the other instruments bunched up single-file behind Sonny's tenor saxophone.

You Don't Have To Go Home by Boney James

The intro is background noise from a restaurant. I can now make out what one person is in the crowd is saying while I'm seated at the listening position. At 1:44, there is a serrated percussion sound 2 feet in front of the right speaker that used to sound like wood scraping against wood. It now sounds like a grunt combined with the wood sound.

Take Five by The Dave Brubeck Quartet - DSD64, 44.1k FLAC, 176.4k FLAC

I did a listening demo this weekend with the good (44.1k), better (176.4k), and best (DSD64) digital versions I have. As I went up in sampling rate, the images got heavier and more detailed and tactile sensation increased. Then I played the Classic Records 45rpm 4-disc version and watched their teeth drop out of their mouths from the shock.

So What by Miles Davis - DSD64, 44.1k FLAC, 192k FLAC

Same results as with "Take Five" above.

Just the Two of Us by Grover Washington, Jr. - DSD64, 44.1k FLAC

The DSD has heavier images, fuller bass, more clarity and definition, and thicker synthesizer notes, and the percussion instruments, at the sides are clearer, heavier, and more detailed.

Grover's saxophone has a thick, silky smoothness that is missing on the PCM rip.

Give in to Me by Michael Jackson

This is a cold, hard, mechanical sounding recording...that sounds even more cold, hard, and mechanical when put under the Debussy's sonic microscope. Slash's guitar solo beginning at 3:29 is my favorite part of the song. Although I hear more definition and detail in Slash's playing, it's still like watching through a thick pane of ice.

Human Nature by Michael Jackson - DSD64, 44.1k FLAC

Now, the difference between the DSD and FLAC versions is like hearing the same song on mono AM radio and FM stereo. I don't fool with the FLAC version any more.

This Masquerade by George Benson - DSD64, 44.1k FLAC

This is a good PCM recording. However, the DSD's holographic interplay between Benson's singing and guitar notes at the into, not to mention the added weight of the voice and guitar notes, means the PCM version is retired.

[Deleted User]

DK,

Thank you very much!

halen

tonyb

Excellent Ray, thank you for getting in the weeds in your reviews and answers to questions. One can always learn something in audio.

DarqueKnight

Clarification On Digital Clocks

The terms "word clock" and "master clock" are often used interchangeably, but, while a word clock and a master clock each perform the function of synching a group of digital devices to the same sampling time signal, they do so in different ways.

A word clock, such as dCS's U-Clock, overrides and replaces the internally generated clock signal in the individual digital devices connected to it.

A master clock does not replace the clock signal of digital devices connected to it. A master clock synchronizes (slaves) the clocks of individual digital components to match that of the master clock. Therefore, all digital devices run on their own internally generated clock signal, but that clock signal is synchronized to an external reference master clock.

DarqueKnight

More Multiple Versions Listening Fun

DarqueKnight wrote: »

Take Five by The Dave Brubeck Quartet - DSD64, 44.1k FLAC, 176.4k FLAC

I did a listening demo this weekend with the good (44.1k), better (176.4k), and best (DSD64) digital versions I have. As I went up in sampling rate, the images got heavier and more detailed and tactile sensation increased. Then I played the Classic Records 45rpm 4-disc version and watched their teeth drop out of their mouths from the shock.

I revisited the listening exercise quoted above once I realized that I have three DSD versions of "Take Five": A rip of the Analog Productions SACD, a download of the Analog Productions DSD file, and a rip of the Columbia SACD.

When I did a comparison of the eleven versions of this album that I own, I didn't make a rip of the Columbia SACD because I assumed it would be inferior in sound to the Analog Productions SACD rip and DSD download. I was wrong.

DarqueKnight wrote: »

I own the following eleven versions, which are listed below in decreasing order of sound quality on my two channel audio system:

1. Classic Records 4 LP (one sided discs) - 1995 release.
2. Analog Productions 2 LP - 2014 release.
2. Analog Productions SACD - 2012 release.
2. Analog Productions DSD download - 2012 release.
2. My DSD rip of the Analog Productions SACD to a DSF file.
3. Columbia, Legacy LP reissue - 2008 release.
3. Columbia, Legacy SACD - 1997 release.
4. My 44.1 kHz/16 bit rip of the 1997 Columbia Records "20 bit" CD.
5. My CD-R copy of the 1997 Columbia Records "20 bit" CD.
6. Columbia, Legacy CD - 1997 release.
7. Columbia, Legacy FLAC file download, 176.4 kHz, 24 bit - 2013 release.

The list above is revised to:

1. Classic Records 4 LP (one sided discs) - 1995 release.
2. Analog Productions 2 LP - 2014 release.
3. Columbia, Legacy DSD rip of SACD to a DSF file
4. Analog Productions DSD download - 2012 release.
4. My DSD rip of the Analog Productions SACD to a DSF file.
5. Columbia, Legacy FLAC file download, 176.4 kHz, 24 bit - 2013 release.
6. My 44.1 kHz/16 bit rip of the 1997 Columbia Records "20 bit" CD.

I did not do a comparison of the digital files to the disc versions below, since my SACD player is stored in a closet:

Analog Productions SACD - 2012 release.
Columbia, Legacy LP reissue - 2008 release.
Columbia, Legacy SACD - 1997 release.
My CD-R copy of the 1997 Columbia Records "20 bit" CD.
Columbia, Legacy CD - 1997 release.

The 176.4k FLAC file has more detail, weight, and tactile sensation from the 1-2-3 pattern of the acoustic bass. There is more natural resonance from the bass's wood body.

The Analog Productions DSD download and my DSD rip of the Analog productions SACD sound identical, with more ambient cues, more air and detail in the cymbals, and sharper transients in the snare licks compared to the 176.4k version.

The rip of the Columbia SACD sounds louder, has more image and bass weight, and sharper, clearer transients in the drum licks. There is more breath noise from the saxophone mouthpiece and the sax notes have a silkier sheen around them.

Music I Had Forgotten About

Going through my music collection, I found a couple of Mobile Fidelity CDs that I wasn't too impressed with when I first purchased them. On my former Cary CD306PV SACD player, the MoFi CDs had a little more clarity and detail than the regular commercial CDs, but I didn't think they were worth the premium and I wasn't planning on buying any more of them.

The notes below are amended to include the 44.1K FLAC rip of the MoFi CD.

DarqueKnight wrote: »

Just the Two of Us by Grover Washington, Jr. - DSD64, 44.1k FLAC

The DSD has heavier images, fuller bass, more clarity and definition, and thicker synthesizer notes, and the percussion instruments, at the sides are clearer, heavier, and more detailed.

Grover's saxophone has a thick, silky smoothness that is missing on the PCM rip.

The crackle and resonance in Bill Withers' voice (starting at 0:41 and 5:15) that is expressed on the SACD rip is missing from the commercial CD and is smeared a bit on the MoFi CD rip. The last sax solo beginning at 6:12 is surrounded by lush percussion and background vocals. In this section of the song, individual sound images are well separated in space on the SACD rip, separated to a lesser extent on the MoFi CD rip, and background instruments are mashed together behind the saxophone on the 44.1k CD rip.

The MoFi rip is a moderately close second to the SACD rip and the commercial CD rip is retired.

What's Going On by Marvin Gaye - 44.1k commercial CD rip, Mobile Fidelity 44.1k CD rip, DSD64 rip of SACD

Commercial CD rip compared to MoFi CD rip: Bass lines are softer sounding and have less articulation and definition on the commercial CD rip. The crackle and resonance in Marvin's voice is more pronounced on the MoFi rip. The MoFi rip has more clarity and detail in the background percussion instruments from 2:45 to the end. The crowd conversation at the beginning has more clarity on the MoFi rip.

The DSD rip of the SACD has more weight and tactile sensation, particularly in vocals. There is more clarity and detail compared to the MoFi CD rip. The sax notes have more weight and a silkier sheen around them.

ukcolin99

What a great post. Thank you DK & contributors.

DarqueKnight

More Clock Connection Trials

DarqueKnight wrote: »

Since the BDP-2 does not have a word clock input, only its USB output can be properly synched to the U-Clock...but its AES (best) and SPDIF (second best) outputs, via the new Integrated Audio Device upgrade sound card, are the better sounding inputs. However, the USB connection with the U-Clock sounds better than the AES connection without the U-Clock.

If use an AES connection between the BDP-2 and Debussy while the U-Clock is connected to the Debussy, The AES data stream will be synched enough to maintain a lock on the data rate, but I get periodic "dropouts" due to the disparity between the Debussy's clock rate and the imperfectly synched AES input from the BDP-2. The "dropouts" are heard as an intermittent gradual lowering of sound level or as faint static-like noise.

After further experimenting, I found that if I connect the BDP-2 to the U-Clock via USB cable, then connect the BDP-2 to the Debussy via AES cable, without connecting the U-Clock's clock signal output to the Debussy's clock signal input, I get better sound compared to either the USB or SPDIF inputs. The AES output is synched to the U-Clock and I don't get dropouts. Also, the Debussy is able to lock on to the U-Clock's word clock signal through the AES input from the BDP-2, as evidenced by the Debussy's "Word Clock" indicator LED.

It is clear to me now that I can connect the U-Clock either to the BDP-2 or to the Debussy, but not to both of them at the same time. Better sound results from connecting the U-Clock to the BDP-2 source rather than to the Debussy DAC. Also, if the U-Clock is connected to the Debussy rather than the BDP-2, the Debussy will not automatically lock on to a particular music file's sample rate. I have to manually select the sample rate. If the U-Clock is connected to the source, the Debussy will automatically lock on to the music file's sample rate.

I still get a muting of the first half-second or first second of songs that have a sampling rate different from the previous song, as the DAC switches to the new rate. But, as stated earlier, most of my playlists consist of files with the same sampling rate, so this issue doesn't pop up often.

tonyb

I dunno Ray, dropouts no matter how short would bug me especially for such expensive pieces of gear. Kinda like buying a Ferrari and having it sputter occasionally when you step on the gas.

DarqueKnight

It's not a fault of the Debussy, it is a fault of the BDP-2 that does not allow the insertion of gaps between songs. Other BDP-2 owners have complained about the same issue with other DACs. If I play gapless music from a computer running JRiver, I have the same problem. Inserting a 2 second gap between songs makes the problem go away.

I have two older DACs with slower processors (Sony TA-E9000ES and Cary CD306PV) and a newer, mid-level DAC with a slower processor (Cary DAC-100) that do not exhibit this behavior. My two newer, higher performance DACs with faster processors (dCS Debussy, Cary DMC-600SE) are the ones with the hiccups.

A better analogy would be buying a Ferrari and driving it on roads with a few potholes. The occasional bumpy ride would be no fault of the Ferrari.