Vintage/Current R2R DAC Owners Discussion, Insight, and Review Thread

[glina]

Why ? Don't you believe double slit experiment for quantum mechanics ? : )

I wont argue with quantum mechanics . Quantum effects have been observed, proven and documented. If MojoAudio has measurements or calculations to substantiate his claim, then I guess his products are worth their weight in gold. I'd just like to see proof and not marketing babble.

 

[MojoAudio]

It never fails: I post on a forum and there is always one guy that wants to start an argument.

Since I can't let forum "bullies" get any traction, I'll respond in detail so that all of you reading this thread get a bit more perspective on the topic.

Glina, I've always said: most audiophile specifications are like bikinis: the important things are not what they reveal but what they conceal

For example, how about tube amps in general?

Absolutely terrible specifications compared to solid-state yet some of the most highly regarded systems in the world use tubes.

How about MOSFETs vs BJTs?

When MOSFETs came out there were all sorts of amps advertising specifications like .0000005% THD, yet now the same companies that were manufacturing amps with MOSFETs have returned to using BJTs that don't measure as good.

And for an example considerably closer to home, the specifications on most modern single-bit DAC chips are considerably better than the specs on any vintage R-2R DAC chip, yet most of the people reading this thread prefer the sound of vintage R-2R DAC chips.

Obviously "measurements" don't tell the whole story.

But to get more specific, I'll tell you about some of the experiences we had with anti-resonance devices this year alone.

A few years ago my company tested several of the most acclaimed anti-resonance feet for under $300 and found that our favorite was the Stillpoints Ultra Mini. The improvement in performance over the EAR Sorbothane feet we put on all our products was notable. Something like 10%. Whatever the exact percentage of improvement, we sold Stillpoints Ultra Mini upgrades to customers for the past couple of years, and all of those customers were thrilled with the improvement.

Fast forward to this year to some time between our exhibits at AXPONA and RMAF...

Tony Landry, the owner of Exit Level Audio amplifiers and my co-exhibitor, was messing with his Mystique v3 DAC and switched the Stillpoints for Iso Cups, one of the anti-resonant products he's liked from Herbie's Audio Labs.

Keep in mind that Tony considers our stock Mystique v3 DAC to be the best DAC he's ever heard and that Tony's system is one of the best I've ever heard regardless of price (beats any $250K system I've heard at an audiophile show).

The improvement using Herbie's Iso-Cups over Stillpoints Ultra Minis was notable.

Then I purchased several of the Herbie's products to compare, including different material "balls" for inside the Herbie's Iso-cups.

When we switched from the "black acrylic" balls to the "moss quartz" balls there was another notable improvement in performance.

Frankly Tony and I were shaking our heads, rolling our eyes, and exclaiming things like "What!?!?!?!?" when we heard how much of an improvement it made to chang the materials the "balls" were made out of.

Then we compared our Mystique v3 DAC with and without a Herbie's Sonic Stabilizer on top (audiophile top weight)...once again an notable improvement.

Then we compared one vs two of Herbie's Sonic Stabilizers on top...a notable improvement that had Tony and I once again shaking our heads and exclaiming "What!?!?!?"

Since then we've had several people from the Santa Fe Audio Society over to hear some of the comparisons. All we would do is play the comparison with the Herbie's Sonic Stabilizers: none, one, and two. 100% of the audiophiles that heard our demo were blown away by the improvement in performance. Most of them ended up ordering the same Herbie's products we were using.

So based on that, we had roughly 4X "What!?!?!?" improvements in performance with anti-resonant products this year alone - based on feedback from a few dozen audiophiles I would estimate each of those improvements to be about 5% - so I'm estimating a total of 20% improvement.

Now add to that the improvement the Sorbothane feet we've been using made on our chassis, and the technology we use to lower the resonance in our chassis (2-piece, 1/8" thick, rigid, polymerized aluminum), and that's where I get the other 10% of the 30% claim I originally made. After thinking about it I would say that claim of 30% is on the conservative rather than exaggerated side.

Is it quantifiable a 30% improvement in performance? No.

Did I average and add up the percentage improvement estimated by several dozen audiophiles to get that number? Yes.

So now I'll respond to your last poke where you stated: "makes me question you actually know anything about that device" in reference to the Audio Note DAC.

To begin with, I have the utmost respect for Audio Note, but anyone that has been an audiophile for any length of time knows that they are a "flavor" and that the Audio Note sound leans toward the "hear no evil" and "sweeter" side of the spectrum.

And anyone that knows the manufacturing side of the audiophile industry knows that most of the so-called Audio Note components (oil caps, transformers, wire, connectors, etc) were not made by Audio Note in house, but rather sub-contracted to highly regarded manufacturers that specialize in those products (such as Jensen for their oil caps).

And anyone that is an engineer and/or designer of high-end audiophile gear knows that there have been several advancements in wire, capacitors, connectors, dielectrics, etc. over the past three decades since the PCM-63 was king and that Audio Note DAC was originally manufactured.

I have no doubt that the modern Furutech and Eichmann low-mass RCA connectors perform better than what Audio Note used 30 years ago.

And there is a very wide range of purity and performance from different brands/types of pure silver wire. Stranded vs single conductor...oxygen free vs OCC silver...and don't even get me started on different dielectric materials.

I have no doubt that the cotton dielectric 24AWG single conductor OCC silver wire we use (made by VH Audio) would perform better than whatever silver wire was used 30 years ago to make that Audio Note DAC.

And then there's the Audio Note silver foil in oil coupling capacitors (I believe they were made by Jensen). They were considered among the best-of-the-best 30 years ago but are currently considered by most to be over priced and under performing. Even Audio Note has changed their design over the past three decades.

I have no doubt that modern coupling caps, such as the V-Caps CUTF, Deulund Cast copper or silver, and Mundorf silver/gold foil in oil would all perform better than 30-year old silver in oil Audio Note coupling caps. I've personally tested all of those coupling caps and I've read a few in-depth cap shoot outs online. Every one of those in depth online reviews gave the highest ratings to the three caps I mentioned above and every one of those reviews gave mediocre ratings to the older Jensen and Audio Note silver in oil caps. So apparently my opinion of the vintage Audio Note coupling caps is far from unique.

Of course that's not even mentioning switching the coupling capacitors for amorphous core output transformers made by Lundahl (about $500/pair) or Teramoto (about $1,000/pair). Either one of those would perform better than any of the coupling caps I mentioned. FYI, Exit Level Audio used the Lundhal transformers in the Annapurna headphone amp we used in our award-winning exhibits from 2017. Once again, my opinion of the performance of these component parts is far from a unique one.

So Glina, unless you've had a similar 4+ decades of professionally custom building and upgrading vintage electronics as I've had, and unless you have first hand experience with all of the high priced boutique audiophile component parts I mentioned, you're only going to embarrass yourself making the statements you made

And with that I'll be signing off for a while...even if Glina posts another provocative post directed at me I won't be responding.

Anyone with questions on anything I mentioned feel free to PM me.

FYI, Yildiray and I are already holding PM discussions

 

[glina]

It never fails: I post on a forum and there is always one guy that wants to start an argument.

This is the second time I called you out, both times on a specific topic, and you once again flood us with your self-assessment of superiority without even replying to the questions asked. And then we have quantum physics… I'm really sorry to see you sign off before elaborating on that.

I asked you to justify putting a number to quantify sound improvements. You reply with no evidence whatsoever. No engineer in the world would do such a thing. It's like saying this orange is 30% "oranger" than the other one. Mhmmm. OK. I'm not the one embarrassing myself. I've said nothing false.

You have opinions and subjective experience and I'm all right with that. It's only when you interleave technical facts with marketing talk and dress it up to look "scientific" is where you get my reactions. You're using the forum for advertising. Not everybody will clap along and I guess you'll have to live with that.

 

[MojoAudio]

I know I said I wasn't going to respond, but Glina, you impressed me so much with your ability to put your foot in your mouth while your head was up your @$$, I simply couldn't resist.

It's clear to everyone on this thread that Glina has a chip on his shoulder when it come to me and my company.

In this specific case Glina has taken the thread totally off topic obsessing over a my statement in regards to being able to improve a DAC's performance by "30%" with advanced shielding and anti-resonance technology. How funny that with all the information I recently posted that the main thing he could find to obsess about was that one statement.

As for the "30% improvement in performance from anti-resonant and shielding products" that Glina cherry picked out of one of my posts, I responded quite clearly on where that came from: the average percentage of improvement cited by several dozen audiophiles that heard the comparisons.

Wasn't that stated clearly enough for you Glina?

Fact is that every manufacturer in high-end audio would tell you shielding and anti-resonance are significant. Glina debating my low-ball estimate of 30% just shows how little he knows about advanced shielding and anti-resonance technology.

I could write several pages summarizing the anti-resonance testing we've had done over the years by highly respected laboratories. But that would be off topic, above the technical expertise of most people reading this thread, and could easily be misconstrued as my "bragging" about my company.

But here is one specific example just to put things to rest...

Last year we had anti-resonance testing done at the Sandia National Laboratories. That is the way we found out that our polymerized aluminum composite chassis have 11% lower mechanical resonance than the same exact chassis with an anodized aluminum finish. They shot broad-spectrum high-volume sound at the two chassis and measured each one's parasitic resonance with some test equipment that likely costs more than my home. I received plots and charts of each resonance pattern as well as a detailed summary report by octave. Frankly I understood less than half of the report I got. But the person who did the tests used some type of statistical analysis and came up with the number "11%" based on total average energy.

Of course even with the test results we had to do a series of focus groups with audiophiles to confirm that our new chassis designs sounded better than our former chassis designs.

Is that what you were looking for Glina?

About Glina's last insulting post...

I sincerely think you are one of the few people that reads this thread that questions my expertise. The Mystique v3 DAC I engineered was given quite a few awards from highly respected audiophile publications this past year. Most of these awards came from publications we didn't even do advertising with. Add that to my graduating first in my class with a degree in Computer Electronics Engineering Technology in 2010, and that puts to rest any question of my relative expertise.

What is your education and expertise Glia?

In response to Glina's attacks I cited several well known places that the performance of components is not reflected in their measured specifications, and Glina had ZERO comments to make on all that.

Glina, please, can you explain to all of us how it is possible that single-bit modern DAC chips have better measurable specifications than vintage R-2R DAC chips yet most of the people on this thread prefer the way the vintage R-2R chips sound?

Glina, can you please explain to all of us how it is possible that solid-state amplifiers have significantly better measurable specifications than tube amps yet many of the most highly regarded experts in the audiophile industry prefer the way tube amps sound?

Just like with most forum bullies, Glina cherry picked a few vague or undocumented statements out of context, took the thread off topic, and attacked a person's experience.

Interestingly enough, few if any of these forum bullies will ever respond to questions from the person they attacked, such as the ones I asked above.

Glina, please consider staying on topic and sharing your experiences rather than insulting and attempting to start an argument with people that are trying to contribute.

OK...now I'm off thread for a while.

 

[glina]

I suggest you keep your nerves in check.

Let's leave it to others to judge who is insulting who in this thread. I only asked you "Do you have any measurements to prove this claim? Which parameters exactly have improved by 30%?" and you claimed "the act of measuring them would often corrupt the actual performance.". Can you prove that? If not, I suggest you do not post nonsense like that.

Having pointed out your mistake in suggesting to upgrade opamps on a tube based dac, all I expect from a honorable adversary is to accept they made a mistake. If correcting false information makes me a bully then let it be, I'm a bully.

To put things straight, I never claimed shielding and damping is irrelevant. I'm genuinely interested in your "30% improvement", but looks like I was expecting too much because all you have is opinions, not proof. I appreciate you doing real resonance measurements of different chassis, this is real engineering. Throwing most expensive and boutique parts you can possibly find on top, inside and under a DAC is not engineering.

Why do poor measuring tubes sound better than transistor? Perhaps thanks to, and not despite their imperfections? Still, tubes and especially triodes, are inherently more linear than transistors, which I would argue makes your dismissal of using tubes in DAC output stage questionable.

R2R and DS do measure differently. I'm not talking about basic parameters as THD and SNR, but linearity, settling time and the ability of R2R to hold a perfectly stable output level is where the main differences are. Due to its nature alone, R2R is the purest possible solution for PCM type data, and Delta-Sigma for DSD type data. I actually do think you have the expertise in this field, so I need not elaborate more.

My education and expertise? Physicist and work as Electrical Engineer in manufacturing industry. Good enough for you to talk with?

Have I answered your questions? If not, let me know and I will do so. In a civilized way.

 

[MojoAudio]

And the half-truths, vague answers, and innuendos keep coming...

Since it is on topic, let me clarify some of the half-truths Glina mentioned in regards to tubes and tube DACs.

To begin with most tube DACs have an op amp in them for the I/V (current/voltage) conversion. Nearly every modern single-bit DAC chip I can think of has voltage output only - that means the DAC chip has an op amp in it. This is in contrast to many of the vintage R-2R DAC chips that either have the option for current output as well as voltage output, or are only current output, which requires an external I/V conversion stage.

So all modern tube DACs have op amps in them, if no other place, inside the DAC chips themselves.

With vintage R-2R DAC chips that have a current output option, some tube DACs use tube I/V conversion and tube output stages, and some use an external op amp between the DAC chip and the tube output stage. One of the reasons companies use an op amp for I/V conversion in a tube DAC made with a current output R-2R DAC chip is that the impedance properties of an op amp will yield a more linear I/V conversion than a tube stage. The other reason companies will use an op amp stage between the DAC chip and the tube output stage is that the op amp will drive most tube output stages better than the current output of most DAC chips.

I admit that my generalized comment about op amps in regards to upgrading a tube DAC were generic as opposed to based on my being familiar with the schematics of any specific tube DAC. But that doesn't change the fact that a significant percentage of those of you that have an R-2R tube DAC have an upgradable op amp between the DAC chip and tube output stage.

As for tubes being more linear than solid-state that is a bunch of BS. Yes, if you look at the "curves" on some of these triode tubes they are quite linear, but those curves are optimal measurements taken under the best of laboratory conditions and not what that tube will do in most circuits.

The biggest distortion in a tube stage is not the tube, but the power supply feeding it.

So if you're talking about one of the better Lampizator DACs that cost over $10K and weigh so much you can barely pick them up, then you're talking about a power supply that would come close to the linearity the tubes are capable of providing. If you're talking about a tube DAC that sells for under $2K and weighs under 20 pounds it simply can not have a good enough tube power supply in that chassis to provide a low-noise linear output. Most under $2K tube DACs are a tube buffer stage slapped on the output of a modest DAC stage to "voice" them into sounding more attractive, not to improve their measurable specifications.

That brings me to my main reasons for my not liking tube DACs: NOISE.

The noise from the power supply for the tube in a tube DAC is much higher than the noise in the solid-state power supplies. And the EM fields from the transformers and chokes in a tube DAC are much higher as well.

This leads me back to that whole "shielding" thing and how poor a shield aluminum is. Aluminum is a good electrical shield, but a poor shield for electromagnetic fields, such as the ones created by the power transformers and chokes in a tube DAC. This makes a machined billet aluminum chassis especially poor for a tube DAC, which was the topic of discussion earlier in this thread.

And lastly there is no tube stage with much better than 90dB SNR. That would mean there is no tube DAC capable of resolving even the 96dB resolution of a 16-bit recording let alone HD resolutions. Not to mention the noise from the tubes is multiplied through every amplification stage following.

Just in case any of you don't believe me on this....

Why is there no tube gear used in recording studios if tubes are so low noise and so linear?

And why are all the best-of-the-best audiophile DACs in the world, such as the Reference dCS and Berkeley DACs, all solid-state if tubes are so low noise and so linear?

Apparently I'm not the only person that thinks a tube output stage on a DAC is not the ultimate solution - several of the most renound DAC designers in the world agree with me.

To answer the question I proposed as to why tube amps are considered by many to sound better than solid-state amps, that is actually well know: tubes amps can not pass odd harmonics so they are less fatiguing than solid-state amps. The other reason has to do with the romantic attractive distortion tubes add to the sound. Just like most women look better to most men with a bit of make up, most recorded music sounds better to most audiophiles with a bit of romantic coloration.

To answer the question I proposed as to why single-bit DAC chips have specifications that are much better than R-2R DAC chips yet many people still prefer the sound of the R-2R DAC chips has to do with a few things. The main thing is that many of the specifications they are boasting about are not discernible listening to recorded music in a home. For example, you can't hear the difference between 110dB and 130dB SNR without causing permanent hearing damage. The main reason why some of us prefer the R-2R DAC chips is that there is a distinctive "character" difference between single-bit and R-2R DAC chips. Just like the difference between the "character" of tubes and solid-state, there are simply some people that prefer the "character" of an R-2R over a single-bit DAC. And like most things in the audiophile industry, his has nothing to do with measurable specifications.

Now back to how and why you can't measure the specific performance improvement of anti-resonant products...

In order to test the effectiveness of anti-resonant engineering the component would have to have the vibrations caused by loud music on the chassis. The problem is that there is no test equipment capable of doing any standard tests using music - they all use narrow band calibrated generated signals for testing. So it is impossible to test the actual performance of anti-resonance engineering using standardized testing. If a company wanted to do this type of anti-resonant testing they would have to engineer their own testing methods that would have no correlation with the standard audio testing we are all used to seeing. In other words, the measured tests companies do to confirm the anti-resonance performance of their products are unique and wouldn't make any sense to anyone else in the audiophile industry.

Of course any engineer that understands these standardized audio tests wouldn't embarrass themselves by questioning the fact that there is no standardized method to test the effects of anti-resonant engineering on audio electronics's performance

 

[BucketInABucket]

Could you guys chill out or take this to private messages before this thread gets shut down by a moderator, please? It's getting a bit much.

 

[glina]

MojoAudio

ESS9018/9028/9038 have current outputs.
And what's more important here, PCM1704, PCM63 as well as AD1862 and AD1865 all have current output which can be used for various methods of I/V conversion (resistor, transformer, opamp etc.)
You once again wrote a lengthy chapter of no relevance for the topic. Not in one point of our exchange have we discussed tube-buffered dacs or dacs with voltage outputs. We discussed the Audio Note DAC which has no opamps in neither the early PCM63 versions, nor the newer AD1865 versions. End of topic.

Are tubes more linear than transistor? Yes, in most cases they are. We were not debating power supplies here. Power supplies can improve or reduce performance of both transistor and tube stages. Transistor based designs typically have deep feedback to compensate for other circuit deficiencies. You can't get away with poor supply for a feedback-free tube stage.

And last chapter is a cherry on top of it all. If as you suggest, anti-resonance products are not or cannot be scientifically tested, then how can their manufacturers guarantee their performance when used over a broad range of third party products? Opinions and "subjective listening tests"? This is largely snake oil business.

Your coveted Belleson regs, Sparkos opamps, JL Sound interface or Lundalh transformers on the other hand are highly engineered devices designed by people with real knowledge. Results of extensive and documented work. Measurable and verifiable.

 

[astrostar59]

I agree with GearMe in that there are so many aspects of the application of a DAC chip that the specific "grade" is not particularly important.

And don't even get me started on tube output stages: there is so much noise in the power supplies and tubes in the output stage in most tube DACs that you couldn't hear the difference between the different grades of a DAC chip.

Noise is one thing, timbre is another. IMO tubes can equal / beat the best SS stage if done right. To me they just sound more realistic. They are not tone controls as many tube haters claim. That is why so many top (highly rested by owners and reviewers) amplifiers use tubes. Fets are switching devices. Only saying....

 

[MojoAudio]

BucketinABucket, you are 100% correct.

I apologize for my negative comments off topic directed at Glina, but as you can all see once again, Glina is gunning for me.

I'll be happy to discuss any of this off thread Glina, but I'm not going to let you goat me into some off topic debate or nit pick words with you on thread.

I will comment on anti-resonant products though...

I never said there is no way to test the effectiveness of anti-resonant products. What I stated was there is no way to perform the standard audio tests and measurements on a component in order to evaluate anti-resonant products. This is because you can't simulate the pulse of higher current going through the component while simultaneously simulating the pulse of high-energy sound vibrations effecting the chassis with a wide-band complex musical signal. That's just not how standard audio performance tests work.

These companies that make anti-resonant products do test them and have measurements and data, but the tests they use are all relative to mechanical resonance as opposed to electronic audio tests. It is in no way "snake oil" as Glina stated, but rather a situation where specific anti-resonant products target the specific parasitic resonance frequencies of specific components. This is why different brands and models of anti-resonant products sound better or worse with different components. So it is not a question of "if" anti-resonant treatments can improve the performance of a component, but rather "what" anti-resonant treatment will best dampen the parasitic resonance of that specific component.

And figuring out "what" anti-resonance product is best for a component can also be tested and measured. The problem is that most audiophiles don't have the insanely expensive equipment required to do broad-band resonance testing on each of their components. So most of us are left with subjective trial and error. So anti-resonant products are in no way "snake oil" and in no way "totally subjective" but rather highly component dependent and it is too expensive and inconvenient for audiophiles to do laboratory style tests on them.

So back to my on topic comments about shielding and anti-resonant products making such a significant difference.

I'm not going to go into great depth in this post (you can PM me about it), but there are several anti-resonant and shielding sheeting products made by notable companies like 3M and B-Quiet that are used by military, aerospace, and telecommunications companies that can be added to any DAC to improve performance. The technical skill level is little more than that of elementary school arts and crafts. Oh...you can go through long and intensive comparative testing projects. Most of us neither have the time nor $$$ for that. Or you can simply put anti-resonant sheeting on every open chassis panel (don't cover vents) and shielding sheeting on transformers, around signal cables, and on every IC chip.

Yes, that is a "shotgun" approach, but it is also quite effective.

So for under $200 in materials and roughly an afternoon's work just about any of you can significantly improve the performance of any DAC.

I would very much like to hear from any of you that have done similar anti-resonant and shielding treatments to their DACs...please share your experiences with the rest of us.

When I was commenting about upgrading that Audio Note DAC I made some generic comments. My main point was that it was a DAC well worth upgrading as opposed to replacing. And that not everyone would agree that going from a PCM63 to a PCM1704 DAC chip would be an upgrade. Those should have been the big "take away" from my post.

Yes, in the brief numbered list in that post the word "op amps" did appear among the other upgradable components in the signal path I mentioned. I probably shouldn't be so "generic" in my comments, but the topic was the PCM63 vs. the PCM1704, not the best way to upgrade that specific Audio Note DAC.

And I agree with Astrostar59 about the tonal quality of tubes and that the best of tube amplifiers can rival many of the specifications of a solid-state amplifier.

My point is not that solid-state is good and tubes are bad - quite the contrary. Personally I prefer tube amps.

Nearly every personal system I've owned and exhibit system I've done had tube amps.

My point is that by nature any DAC is a solid-state device - you simply can't get around it. That means it is not a question of "if" you can get rid of all the solid-state devices in a digital system, but "where" the most advantageous place is in the system to introduce tubes.

My point is that putting tubes inside a DAC is asking for all sorts of engineering issues in regards to weight, size, and shielding from noise. Sure there are statement tube DACs, such as the better ones made by Lampizator, but until you get into a tube DAC of the same relative size, weight, and cost as the better Lampizators, you are getting a highly compromised tube DAC, both in terms of power supply and shielding.

Good tube gear costs several times the price and is several times the size and weight of good solid-state gear.

So my main point is that it is easier to get lower noise, higher performance, and similar tonal character with a solid-state DAC and a tube preamp or integrated amp than with a tube DAC.

From my experience, a good solid-state DAC combined with a good tube preamp beats any tube DAC or tube DAC preamp combo I've ever heard. I'm sure there are exceptions, but I've never heard them. It is simply a matter of physics. All things being equal, separate components will perform better. And separate components allow you more flexibility in future upgrades.

In the end it's all a matter of personal preference: a more romantic, softer, sweeter sound vs. a more articulate and resolving sound.

To my ears a system with 100% tubes sounds as unnatural as a system with 100% solid-state.

You can have too much of a good thing

 

[Articnoise]

Hi Yideray,

Please don't be offended, but you've obviously "drank the Kool-Aide."

There are so many factors in all of this.

To begin with, why are you going to spend a small fortune on a milled aluminum chassis? Because you've seen so many modern manufacturers use milled aluminum chassis?

Milled aluminum looks cool and has fairly good at resonance characteristics, but it is a terrible shield.

You would get better performance at a lower cost with a modest off the shelf steel chassis (much better shielding than aluminum) and coating the inside with anti-resonant sheeting.

And that's not to mention using anti-resonant standoffs, such as the EAR Sorbothane or the Stillpoints, and feet under the chassis, and anti-resonance "weights" on top of the chassis. I've experienced over 30% performance improvement with just these simple anti-resonance products.

Then why are you choosing the PCM1704 in the first place?

It is not even a true R-2R DAC chip.

No company can match more accurate than a 20-bit resistor ladder, so the highest performance true R-2R DAC chips are the 20-bit ones, such as the PCM63 and AD1862.

The PCM1704 is a hybrid of two resistor ladders, a MSB and LSB ladder, with an algorithm that combines the sum of the LSB ladder at a fraction of the output voltage to the MSB ladder. To me the sound is a mix of a true R-2R ladder DAC and a modern single-bit DAC. Somewhat slow, smoothed over, and overly refined for my taste.

Personally I prefer the sound of the legendary 20-bit DAC chips, and there are several other people on this thread that have compared and agree with me.

The OPA627 is one of my favorite IC op amps, but it is no comparison to a good discrete op amp, such as the Sparkos Labs:

http://sparkoslabs.com/

Sparkos Labs also makes one of the best ultralow-noise regulator on the market - quite comparable to the Belleson I currently use.

If you're asking the question you are asking about a power supply, then it is obvious you don't know enough to build a proper power supply, and that is likely the most important part of any DAC.

I could write 10X as much on power supply alone, but this is not the tread for it. If you want more info contact me through PM.

And don't get me started on how TOTALLY STUPID it is to put a DAC and a tube preamp in the same chassis. To begin with, the noise in a tube preamp's power supply is extremely high and would require significant shielding and physical distance from the DAC portion. And a tube preamp with a decent power supply would have to weigh over 20 pounds and take up quite a large chassis on its own. Anything less and you'll be killing a significant part of your DAC's resolution.

By nature a DAC is a solid-state device. Most companies that manufacture tube DACs in fact are only adding a tube buffer to the output of a DAC so they can add a romantic attractive distortion. It is a marketing concept. If you have an all solid-state system with high wattage class A/B or class D amplifiers I could see the attraction of this type of DAC, but it still makes more sense to have a dedicated DAC and dedicated tube preamp. Not only would the result be a lower noise floor, this would also give you the opportunity to upgrade each component separately.

And then, what about your DAC's input stage and clocking?

The input stage is every bit as important if not more important than the DAC chip or output stage. A DAC chip can't decode and an output stage can't amplify what it doesn't have.

When doing the R&D for our Mystique v3 DAC we compared several of the most highly regarded and some fairly obscure USB input boards on identical DACs. We even did blind listening tests. There was so much of a difference between so-called 24/192 USB boards it blew our minds. People that took part in the blind listening tests told us the differences were so extreme that they would have believed that we changed speaker systems or headphones with ones that cost several times the same price as opposed to just changing USB input boards.

Oddly enough, the consensus least favorite of all the USB input boards was one of the most highly acclaimed XMOS boards (I can't mention names) that used the same XMOS chip, same brand and model femto clocks, and same brand and model IC regulators as our favorite. That just confirms everything I've been saying about the DAC chip being such a small part of the overall DAC performance.

Our favorite USB input board is this obscure one made by JL Sounds:

http://jlsounds.com/i2soverusb.html

Not only does it have excellent performance, it has a built in L and R channel demultiplexer circuit that will allow you to connect it directly to most of the popular R-2R ladder DAC chips, including the PCM1704. Nice feature.

Just in case some of you are not aware who I am and where my expertise comes from, I'm the owner of Mojo Audio, and my company has specialized in NOS R-2R DACs for nearly a decade. I've personally upgraded the whose who in vintage R-2R DACs and CDPs and personally compared most of the DAC chips, circuits, and component parts people mention in this thread.

And this past year alone, our Mystique v3 NOS R-2R AD1862 20-bit DAC was awarded:

"Best Sound" by The Absolute Sound at AXPONA'17
"Best of Show" by The Absolute Sound at RMAF'17
"Best Innersound at the Show" by Part-Time Audiophile at AXPONA'17
"Star Component" by Audiophilia
"Product of the Year" by Audiophilia

So it would only stand to reason that I must know a thing or two about R-2R DACs

Please feel free to PM me for more info on DAC chips, op amps, power supplies, input boards, etc.

Bottom line is that the best 20-bit and higher DAC chips (such as the PCM1704) used in nearly identical circuits with nearly identical supporting parts will have less than a 20% difference in performance.

Where as the circuits and supporting component parts make up over 80% of any DACs performance.

It’s not “TOTALLY STUPID” to put a DAC and a tube preamp in the same chassis. To use and detached sensitive analogue and digital circuits from the PSUs is beneficial for both SS and tube stages. A DAC is amplifier with a digital board. The analog line stage can be equipped with a VC and voila it’s a pre amp. Why tubes in a DAC? Because of the pros good tubes and tube circuits brings to the table obviously. You don’t like the sound of any tubes? - Ok no problem. But to state that it’s totally stupid to put a DAC and a tube preamp in the same chassis is just plain stupid and ignorant in hobby that is as subjective as high end audio.

A few SOTA DACs that use tubes: Aries Cerat Kassandra Ref2, AQUA La Scala MKII Optologic, Lampizator GG, Aesthetix Pandora, Ayon Stealth.

 

[MojoAudio]

I understand your points Articnoise.

To begin with I LOVE tube amps - especially DHT tube amps. Nearly every amp I've owned for the past 40+ years has been a tube amp.

So if there was a way to make a tube DAC perform as well as a solid-state DAC (all things being equal) I would most certainly be in favor.

And I've owned and/or upgraded several of the most popular non-oversampling R-2R tube DACs and CDPs over the years (literally dozens).

And if you go back about 20 years, I was an avid follower of Lukas from Lampizator - I read most of his blogs and tested many of his circuits.

And from 2009-2013 all the non-oversampling R-2R DACs manufactured by Mojo Audio were tube DACs.

Some even received awards, such as "Best Sound Value" at RMAF'12 by Audiophilia and "Silver Sound" at CAF'12 by AV Showrooms.

So our tube DACs may not have been the best in the world, but they certainly had respectable performance considering the awards they received.

So what was it that changed my mind about tube DACs in 2013?

We were doing the R&D on our Mystique v1 AD1865 DAC and compared the AD1865 DAC chip with a fairly high-end tube output stage to the AD1865 DAC chip in an identical circuit directly driving a direct-coupled output stage (no coupling caps or transformers).

There was no comparison.

Sure the tube output stage DAC had more bloom, a more 3D image, and made bad recordings sound more tolerable (attractive distortions).

But the direct-coupled DAC had significantly better time, tune, tone, timber, attack, bloom, decay, bass and top end extension, harmonic coherency, and spacial cues.

I won't bore all of you with the rest of the R&D we did that year, but the result was that no matter how much we improved the tube output stage the direct-coupled output stage maintained the same advantages.

Since then we've compared our direct-coupled Mystique DACs to several highly regarded tube DACs, including the famous Audio Note AD1865 DAC, and nothing we've heard led us to believe that we should consider engineering another tube DAC.

Notice I used the term "heard."

There is one reason my company has been seriously considering engineering another tube DAC: marketing.

Apparently there is quite a large percentage of audiophiles that want to own a tube DAC.

So we've been planning a version of our Mystique DAC built into a tube DAC/pre/headphone amp for quite some time (ETA 2019).

We've already purchased the rights to the Annapurna tube pre/headphone amp we currently exhibit with to use as the output stage.

But consider the Annapurna tube pre/headphone amp: it's about 23" x 18" x 7" and weighs about 70 pounds.

That would mean to make an integrated tube DAC/pre/headphone amp version of our Mystique DAC that is equal in all ways to our Mystique v3 DAC and the Annapurna pre/headphone amp separates it would have to be 23" x 18" x 10" and weigh about 90 pounds.

Not a particularly practical package.

So our plan is to do what most other companies does when they make a tube DAC: compromise.

We're hoping to create a tube DAC/pre/headphone amp that is close in performance than the original separates, costs about half as much as the separates, and fits in a smaller chassis that weighs less than 50 pounds.

Have any of you ever wondered why nearly all the best components by nearly all the best companies are separates?

Because separate components allow for better isolation of power supplies, and better isolation of noise, while allowing additional space for better performing power supplies, and better shielding from noise.

Simply put a tube DAC/pre amp is a compromise. Period.

BTW, I find it rather hard to believe that more of you are not chiming in on the "separate components perform better" concept I keep bringing up.

 

[yawg]

d.

Why tubes in a DAC? Because of the pros good tubes and tube circuits brings to the table obviously. You don’t like the sound of any tubes? - Ok no problem. But to state that it’s totally stupid to put a DAC and a tube preamp in the same chassis is just plain stupid and ignorant in hobby that is as subjective as high end audio.

A few SOTA DACs that use tubes: Aries Cerat Kassandra Ref2, AQUA La Scala MKII Optologic, Lampizator GG, Aesthetix Pandora, Ayon Stealth.

I've been using an older Cayin DA-2 tubed DAC for a long time now. Every time I listen to a newer "better" DAC I'm underwhelmed. I tried a whole bunch of NOS tubes instead of the two Sovteks which came as stock. Every old set improved the sound somehow but with the German Siemens/Halske CCa everything got perfect.

 

[Articnoise]

I understand your points Articnoise.

To begin with I LOVE tube amps - especially DHT tube amps. Nearly every amp I've owned for the past 40+ years has been a tube amp.

So if there was a way to make a tube DAC perform as well as a solid-state DAC (all things being equal) I would most certainly be in favor.

And I've owned and/or upgraded several of the most popular non-oversampling R-2R tube DACs and CDPs over the years (literally dozens).

And if you go back about 20 years, I was an avid follower of Lukas from Lampizator - I read most of his blogs and tested many of his circuits.

And from 2009-2013 all the non-oversampling R-2R DACs manufactured by Mojo Audio were tube DACs.

Some even received awards, such as "Best Sound Value" at RMAF'12 by Audiophilia and "Silver Sound" at CAF'12 by AV Showrooms.

So our tube DACs may not have been the best in the world, but they certainly had respectable performance considering the awards they received.

So what was it that changed my mind about tube DACs in 2013?

We were doing the R&D on our Mystique v1 AD1865 DAC and compared the AD1865 DAC chip with a fairly high-end tube output stage to the AD1865 DAC chip in an identical circuit directly driving a direct-coupled output stage (no coupling caps or transformers).

There was no comparison.

Sure the tube output stage DAC had more bloom, a more 3D image, and made bad recordings sound more tolerable (attractive distortions).

But the direct-coupled DAC had significantly better time, tune, tone, timber, attack, bloom, decay, bass and top end extension, harmonic coherency, and spacial cues.

I won't bore all of you with the rest of the R&D we did that year, but the result was that no matter how much we improved the tube output stage the direct-coupled output stage maintained the same advantages.

Since then we've compared our direct-coupled Mystique DACs to several highly regarded tube DACs, including the famous Audio Note AD1865 DAC, and nothing we've heard led us to believe that we should consider engineering another tube DAC.

Notice I used the term "heard."

There is one reason my company has been seriously considering engineering another tube DAC: marketing.

Apparently there is quite a large percentage of audiophiles that want to own a tube DAC.

So we've been planning a version of our Mystique DAC built into a tube DAC/pre/headphone amp for quite some time (ETA 2019).

We've already purchased the rights to the Annapurna tube pre/headphone amp we currently exhibit with to use as the output stage.

But consider the Annapurna tube pre/headphone amp: it's about 23" x 18" x 7" and weighs about 70 pounds.

That would mean to make an integrated tube DAC/pre/headphone amp version of our Mystique DAC that is equal in all ways to our Mystique v3 DAC and the Annapurna pre/headphone amp separates it would have to be 23" x 18" x 10" and weigh about 90 pounds.

Not a particularly practical package.

So our plan is to do what most other companies does when they make a tube DAC: compromise.

We're hoping to create a tube DAC/pre/headphone amp that is close in performance than the original separates, costs about half as much as the separates, and fits in a smaller chassis that weighs less than 50 pounds.

Have any of you ever wondered why nearly all the best components by nearly all the best companies are separates?

Because separate components allow for better isolation of power supplies, and better isolation of noise, while allowing additional space for better performing power supplies, and better shielding from noise.

Simply put a tube DAC/pre amp is a compromise. Period.

BTW, I find it rather hard to believe that more of you are not chiming in on the "separate components perform better" concept I keep bringing up.

I’m glad to hear that your Mystique DAC is a successes and have received awards such as "Best Sound Value" – really great!

I have no problem with people liking other gear or tech designs than me. What I can’t understand is things like TOTALLY STUPID of a using tubes in a DAC, even if you think that’s the case. It’s a bit hard to swallow that not long ago you used tubes in DACs yourself and are also planning on doing it again.

I’m absolutely not saying what it’s better to place a pre or headphone amp in the same chassis as the DAC. It depends on total budget, design and implementation. The reason for putting a pre in the DAC is because many people want less boxes and cables and its cheaper (for the consumers) to pump up the analogue line stage a bit and add a VC than make a separate SOTA pre/headphone amp. Less is also more in audio and the line stage will still amplify the signal even if a pre is used, so double amplifying will be the result. A pre amp in a DAC also have the advantage that it’s possible to level match the output signal that goes to the amp, not all amps can handle a 5V input signal for example.

Finally why would an analogue tube stage need to be bigger than an analogue SS stage, if it’s what you are saying?

 

[Articnoise]

d.

I've been using an older Cayin DA-2 tubed DAC for a long time now. Every time I listen to a newer "better" DAC I'm underwhelmed. I tried a whole bunch of NOS tubes instead of the two Sovteks which came as stock. Every old set improved the sound somehow but with the German Siemens/Halske CCa everything got perfect.

I also use German Siemens/Halske CCa in my Pandora. They are great tubes.