Updated DAC Measurements

I didn't measure the right channel of the Master 7 because of high DC offset, which started around 20mV and kept rising during the tests.  I may have broken the Master 7 in earlier testing by power cycling too rapidly.

Measuring the Master 7 left channel in full balanced connections, I found 0.009% THD.  It required some tricky connections do this measurement with a mono balanced signal.  I ran the left balanced output of the DAC into a balanced input of the Emotiva XSP-1 preamp, then switch the subwoofer outputs to full range, and take the R and L subwoofer outputs.  Testing the Emotiva Stealth DC-1 this way showed 0.0005% distortion, a bit higher than the 0.0003% it measures directly, showing the residual using this setup to be below 0.0005%, plenty good enough for measuring the Master 7 at 0.009%.

But in single ended mode, the same left channel of the Master 7 is measuring a horrible 0.183% distortion.  So I think something may indeed be broken in the left channel, as well as the right.

Master 7 Measurements

Meanwhile the Emotive Stealth DC-1 measured very well, with THD at 0.0003% in direct stereo balanced connections, and 0.0004%-0.0005% when measured through the Emotiva preamp, and 0.0005% in a mono'd measurement through the Emotiva like I did for the Master 7.  In both cases, these are right at my measurement residuals, the Emotiva dac having distortion as low or lower than my Juli@ sound card itself.

Emotiva DC-1 Balanced Measurements

The distortion test spectrum is very clean for the Emotiva DC-1 even going through the Emotive Preamp (so you're seeing the distortion of both devices compounding, but still not much).

Emotiva DC-1 through Emotiva XSP-1 Preamp

So, if low THD is important, and I think it is, why don't I use the Emotiva DC-1, which is as low as I can measure?  I tried that for an hour, but there does seem to be less there there, as compared with the 1704 based DAC 19 (let alone the Master 7).  But there's also a much greater blackness, a softer presentation that can still have shocking, perhaps even more shocking, highlights, flashing at you like on a bright sunny day.  I'm listening to it now and will decide, it's not an easy call subjectively--you think you've nailed it then you haven't.  The loser may take up super tweeter duties, where THD may be less important.

So, my subjective preference still leans toward the Master 7 compared with the DC-1, when I get the Master 7 fixed.  The Master 7 was fabulous sounding, I felt that right away and for weeks, and perhaps it even measured better in the beginning before my re-starting or something damaged it.  I hope it can be fixed to work as good as the specifications, which are only a hair behind the DC-1.

My subjective feeling is that the R2R dac makes you feel the cool breeze of reality all the time, it's right there ahead of you.  With the Master 7 the breeze isn't noticeable anymore unless you concentrate on it, but the thereness is immediate and obvious.

Sigma Delta dacs remove this by somehow equalizing the air pressure, you're in airlock, but slightly removed from reality.  But then at any given moment the doors of the airlock crack open and you've got your blast of reality in one big blast.

I generally feel that 1-bit sigma deltas such as in the Sony SCD-9000ES do this also, but the blasts of ultimate reality are finer, and therefore more musical.

Running through the higher input impedance of the Emotiva Preamp (50k) did not seem to help the Master 7 single ended measurement, in fact it went up from 0.135% to 0.183%...though that is very dependent on conditions such as the peak level, I had seen 0.18% during the earlier set of measurements but it mysteriously improved to 0.135%.*  The balanced distortion (which I was unable to measure before) reduced a huge amount (to 0.009%) because most of the single ended distortion is 2nd and other even harmonics.  I had predicted it would decrease a large amount, and I hoped for better than 0.01% which is I think a minimal requirement for digital converters, and I got the minimum technical performance I wanted even asis.  But I believe both should be much better, even considering the lack of feedback used in the circuitry, because of its sophisticated design, and of course the spec is a lot better.

(*Ahh, perhaps I didn't allow enough warmup to get to the better values the second time around, and perhaps even further warmup would help...save that for a future set of measurements.)

One of the design features, which does make it somewhat more difficult to achieve low distortion, is zero feedback.  In principle, I like zero feedback, but for voltage ouput devices* I wouldn't trade distortion above, say, 0.0005% for it.  And in fact the Master 7 specification was exactly that.

(*The situation is very different for power amplifiers because they provide current to a reactive load.  Zero feedback has special advantages...but also so does super high feedback done correctly, it's only middling feedback that can cause great harm.  Anyway, there's a reason for having more stringent distortion requirements for the line devices than the power devices.  To be fair, only the total distortion ultimately matters, but earlier distortion compounds with and increases later distortion.  So you are much better off keeping distortion low early, and it's usually more economic and practical also, and there may be particular advantages for higher distortion power amplifier designs--though all other things being equal the lower distortion the better always.)

Zero feedback would be more essential for those hoping to get the full 'benefit' of NOS operation.  Only with zero feedback do you get the best time performance to limit of the highest frequency aliases, fwiw.

The highest slewing of transients, etc.

Now high distortion is not unheard of among high end dacs.  It's pretty typical of devices made by certain companies and designers...the same who favor high distortion SET power.  Audio Note (UK) for example.

Stereophile published a review of one of the latest Ear Dacs designed by the famous Tim de Pavarachini who also at least recently follows the "ignore distortion, instead do this..." guidelines, and Atkinson published the distortion spectrum, which featured an impressive sequence of even and odd harmonics diminishing slowly as is typical for non-feedback triode designs.  The 2nd harmonic was around -58dB and the 3rd not much lower, etc.  The second harmonic alone would correspond to a level of 0.15 or something, actually about what I'm seeing as the full THD of the Master 7 (in single ended mode and possibly broken).  So in other words, the harmonic distortion was quite a bit higher, and John Atkinson used the correct impedance and balanced outputs if available with very nice Audio Precision measuring equipment.

Well the subjective review was somewhat unfavorable also, with the reviewer complaining about grit, etc.

But Akinson didn't criticize the distortion levels much.  Instead he focussed on some very arcane aspects of the jitter rejection performance, claiming that was responsible for the negative comments.

Nobody can actually prove this (to critics) one way or another, or even if the Ear dac is audibly better or worse than any other without doing some very hard work, which we can be pretty sure nobody will ever do (dbx and all that).  Then for sure, no one will prove what is the cause of that difference, in the analog amplitude or digital timing aspects.

RMAA

Last weekend I tested DACs using RMAA.  I discussed the results here and gave a few numbers (but was unable to figure out how to post even pictures of the results...something I hope to fix soon).

I'm pretty well aware of many issues doing the measurements (especially, the rather low impedance of my sound card) and I'm planning on doing an improved set of the most important measurements soon including balanced outputs and higher impedance loads.

Meanwhile, here's NwAvGuy's review of RMAA, which seems somewhat more negative than deserved (though all of the substantive points he makes are valid, and he does say many times that this may be your best option under $5000).  It's worth re-reading and re-re-reading.

A bunch of people at Steve Hoffman's forums were testing their DACs last year with RMAA.  Some results interesting, others comical.  The last guy doesn't seem to get that you get virtually perfect results using digital output to digital input, which he must be doing by mistake.

Holt on Polarity

Stereophile has resurrected an oldie from J Gordon Holt on Polarity originally published in 1980.  He is saying that you only hear it when you have asymmetrical non-linearity.

This generates a lot of flack from commenters.

I think Holt is correct in a limited sense.  When people claim to hear differences because of absolute polarity, it's almost certainly one of these (and they are not exclusive):

1) their imagination, because of sighted and uncontrolled testing
2) non-linearity in reproduction

The third possibility--actually reliably hearing polarity differences--occurs only in limited cases:

1) asymmetrical test signals designed to make it more obvious
2) Solo trombone or the like
3) Headphones plus either of the above

That is essentially what the premier study on polarity by Lipschutz in the 1980's determined, using pulse coherent Quad ESL-63 speakers and headphones.

Holt wasn't using test tones or very specially selected music, so his results are quite reasonable, and shows his fundamental honesty, and the relative inaudibility of polarity.

I had my own brush with this recently, first I correctly heard something which turned out to be polarity, and then I thought I heard it again and didn't.

Even if relatively inaudible, a good recording or system should try to get it correct, to the degree it is easy to do so.  If this means reversing speaker cables on each track of a recording...I'd say forget it.

Zero phase shift would be nice too, if not so impossible.  Tradeoffs between lumpy frequency balance or phase shift generally favor prioritizing the frequency balance, and that's why we have things like crossovers and multi-way speaker systems in the first place.

Meanwhile the idea that recording industry inverts polarity as a way of ripping off the public...sounds pretty preposterous.  If ever wrong, one would first suspect error rather than design.  There are far more obvious ways to alter recordings for popularizing and/or degrading the material, such as in frequency balance and dynamic range.  Polarity adjustment might be made simply on the basis of which sounds better, which could simply be random chance because of the difficult of correctly hearing it.  Given that such judgements are possible, and even occasionally used, one would expect error to creep in, even if technical connections and systems were always perfect (and, that's a stretch too).

Flying with the Eagle

After running for a week on the left channel only with no issues, using either speaker, I unplugged the FPB 300 for awhile, connected the right channel, and let it run on the easier left speaker.  It ran for about 20 minutes and shut down.  OK, I've got the RMA, and it's getting packed and going back to Krell for service.  It's been my audio pride and joy for 8 years, far more neutral and dimensional sounding than any other amplifier I'd listened to.

But what to play next?  The Aragon has proven itself to be Acoustat capable, as much as any amp, after all, it's run the Acoustats for about 50% of the time since I got them.  I was almost inclined to do that, but finally I decided to hook up the Eagle 2 I had bought last year, and done some technical testing (it worked, basically, I didn't do much except for quick power testing), and then it sat, since I was a bit afraid of how it would deal with the Acoustats.

But now I have the 3 amp fuses in place, the Eagle was explicitly sold to me as 'working' (the seller said to have been using it for a month), it tested OK, so why not.

WOW!!!

This has changed the presentation entirely.  No longer am I listening to the "Lacking in Transparency(tm)" Acoustats!  My speakers are now ceiling high line source ESL-63's, if there were such a thing!

Allowing a bit of break in, and for the fact it is far more voltage sensitive than the Krell (though having the same 100,000K input impedance--something I now consider a very important factor), before I knew it, I could measure peaks as high as I do with the Krell, yet, it didn't seem to be playing hard at all.

Moxie Power!  As Montcrief describe the Robertson 4010 amplifier (an Iverson design, but made after Iverson bailed from the company Electro Research, selling everything to Robertson to pay for their costs in tooling the EK-1 phono system), I'm sure this would apply even moreso to Iverson's own amp, made under his direction in USA.

(The Robertson was a small 60W amp that Montcrief described as having 10,000W moxie power.  Meanwhile, some big enormous high power amp was described as having a mere 10W moxie power. Montcrief even described a moxie power methodology.  I personally would not get a Robertson because they were into using potted modules to hide their actual parts, potted modules which have been known to crack.  Despite Montcrief's glowing review, the company was not around long.  Iverson's own amp is far more powerful and robustly built.  I consider the Eagle 2 his ultimate masterwork and I believe he did too.  It got the glowing review and personal ownership by J Gordon Holt.  It's simpler than the higher power models.  Iverson and Munro tried to sell me one at an audio meeting in 1988, and I've been very sad I didn't buy it, but now, the magic is in the house.)

This disproves audiophile nonsense about feedback.  This amplifier breathes feedback to the max, that was the idea.  Iverson understand how to do that, stability and linearity and low propagation delay.  Then you can crank up the feedback way beyond what others dare, and get a virtually perfect voltage source.  This amplifier is rated as 800 damping factor and I wouldn't be surprised if it delivers more.  Power supply stability is important too.

Along with that, distortion down to 0.0025%, that's like CD quality, frequency response DC to 100kHz, noise at 90dB (when other amps are quite often down in the 60's).  These specs matter, contrary to audiphile nonsense that 1% distortion is fine as long as it's the 'right distortion.'  No--you want zero distortion.

But still I think the outstanding magic here is the perfect voltage source operation, as illustrated by the super high damping factor.

So here it is, so transparent I could not imagine my speakers sounding like this.  Not only transparent, but smooth and incredibly dynamic.  I don't need those Linkwitz speakers now, my line source electrostats have caught up, thanks to improved amplification.

Objectophiles couldn't believe it, but nothing has proven to me the important of amplifier choice like this.

And just like EKSC says, it's clear, smooth, dynamic.  It's more listenable as well as fun to listen to.  There more there there, and it's popping right out you in more 3D than ever (and intense separation, because there's more there there) but not clobbering you or making you want to cover your ears.

Even the slight warpage of low power college radio is rendered so nicely I can just listen and listen.  The instruments, everything, sounds more real than ever, even in the most sketchy reproduction like this.  You do hear the music, not the source problems somehow.

All that and it's running pretty cool.  One channel got up to 118F at the end of my most intense listening so far, with peaks at 88dBC, which is about my peak on the Krell.  The Krell would have been showing 165F left and 155 right (due to airflow patterns).  And the Eagle 2 cooled down fairly rapidly to  a mere 108F only at a slightly lower level.

It's like it doesn't have to exert itself to handle the highly reactive but non-absorptive load of the Acoustat.  The Eagle just wiggles it's shoulders and it's moxie power does the rest.

Monday: The flying continues.  This is the most transparent sound I've ever heard, there is more there in everything.  I listed to Emerson Lake and Palmer and now Another Green World, and both classical and indie rock radio.

I see now lots of amplifiers have high damping factor, and I'm thinking there's more than one way to achieve that.  I continue to believe the magic of the Iverson designs is pushing feedback to the maximum.  That's how he gets distortion (THD) down to 0.0025% at rated power, which is basically the same as the 90 dB S/N level.  According to Baxandal, feedback is only useful in very small or very large amounts...in the middle it doesn't help (but that's where most amplifier designs are).  Most modern amplifier go after the small and medium amounts of feedback.  Iverson took the feedback thing as far as it could reliably go.

So what I imagine is happening here is that distortion is effectively being 'eliminated' by pushing it back into the noise floor, which is the ultimate limit--can't get THD to go below that.

My Krell FPB 300 is fairly low distortion as modern designs go, but 0.02% at 1kHz is ten times what the Eagle 2 has, and the FPB rises up to 0.15% at 20kHz, despite the fact that power response actually goes to 300 kHz.  So there is rising HF distortion that goes 40 dB above the noise floor at the highest frequencies.

The Parasound JC1 has even higher damping factor than the Eagle 2, but distortion also rises to 0.15%.

Thanks to misunderstandings, audio engineering for high end audio not only downplays the importance of low distortion it seems to bring on more distortion, because it can (according to audiophile mythology, going for too low distortion is bad).

So actually you could say the Eagle 2 isn't magic, it's just one of the few that does things the correct way.  But I still consider that to require a fair bit of magic.  And there may be other magic involved too.

When you go to the upper amount of 'good' feedback (above 35dB) you begin to actually suppress the high order harmonics which are actually reinforced by middling amounds of feedback (5-35dB).  Once you know those high and odd order harmonics down, you get smooth rather than brittle sound, and the Eagle 2 is incredibly smooth as well as being incredibly clear.

Stereophile reviews of the Eagle 2 do indeed rank the sound quality higher than Eagle 7.  I can see why you could imagine needing more power, so that's the reason for the Eagle 7 and other models.  I remember at the time Iverson was showing off the Eagle 2 (and it was indeed a magic show, including arc welding and lots of other tricks) he disagreed that the Eagle 2 could be improved with higher bias.  The higher bias in the Eagle 7 was there because that amplifier needs it...to be almost as good within its rated power as the Eagle 2 is within its power ratings.

But possibly I can crank it up well beyond the 1v input on peaks.  I've been holding back a bit, though sound levels are still reaching 87dB peaks.  This is an old amplifier...far older than the Krell actually, and it probably badly needs a refurb, but still sounding great.  This is not just magic, it's a miracle.

Later I shut the amp down, with Tact gain set to +2dB, playing The Doors album on my table.  Digital output peak of -6dB peak was showing, that would correspond to about 1.25V.  I suspect the amplifier was shut down by current limiter, possibly that monitors regulated voltage as on the Eagle 7, which I have an unofficial schematic for.  There is a LOT of high frequency grundge on this album.  I had played Dylan Blood on the Tracks at +4dB without issue.

Playing with Fire

My thinking that Iverson was going after the high end of feedback, above 35dB, was entirely a guess simply based on the low distortion numbers.

But taking a look at the Eagle 7 unofficial schematic I found, it looks like my guess is correct.  This amplifier has at least 2 more gain stages than usual, and they are triple darlingtons like the outputs.  This has open loop gain out the window and across town.

I also knew a little bit of Iverson's life story.  His father was a TV repairman in the 1940's and 1950's, who was always tinkering with high voltage.  This was too much for the man's wife, who divorced him.  But he lived with his son, John E, who acquired his high voltage passion.  Later, among other things, John E devised the 'corona' speaker, based on high voltage ionization.

This is the story of someone who doesn't just flirt with danger, they live within it.  That's the kind of person who wouldn't feel uncomfortable tinkering with amplifiers with super high open loop gain.  One wrong move, and SNAP!  Not to say that there's necessarily any high voltage here...the Eagle 2 amplifiers operate with less voltage than many amplifiers today because they are lower power than many amps, rated at only 110W RMS into 8 ohms, which corresponds to about 29 volts.

Yin and Yang

But when you have loads and loads of bipolar open loop gain, something special is happening.  As bipolar transistors heat up, their gain increases, so they want to heat up more.  So the electrical force of the amplifier is constantly pushing outward, only to be held back by the feedback which forces it to act like a more perfected amplifier of normal gain.

That feedback however cannot act instantaneously.  In the very short interim, measured in nanoseconds or something like that, the amplifier does perhaps slightly exceed correctness, only to be held back by the propagation delay of the amplifier.  Until just now I had not understood why that was important in an amplifier like this, except for permitting the high feedback to work in the first place.  If the propagation delay is too long, you have to sacrifice bandwidth, and this amplifier retains closed loop bandwidth to 100kHz which is fine.

Anyway, what I've just described, the constantly pushing outwards only to be held back characteristic is exactly the defining quality of maleness, or yang.

The opposite yin quality is exemplified by Class A bipolar amplifiers, a fact which just struck me yesterday.

Class A bipolar amplifier are heating up to the maximum degree when carrying no signal.  When a signal is being amplified, the less-active devices on the negative side of any swing cool down ever so slightly while the signal becomes negative.  So rather than heating up with signal, they are cooling down, and therefore losing gain.  A Class A amplifier is, therefore, always falling slightly behind, and the feedback serves to give it the prod to keep it fully out there.

So now I'm thinking neither rendition is fully accurate, but the two different renditions, yin and yang, are complimentary.

Speakers, however, one might expect to be mostly yin, so a bit of extra yang from the amplifier makes it balance out.

AND that's what I think I hear, the Yang of the Eagle 2 is complementing the infamous Yin of the Acoustat 1+1 transformer coupled electrostatic speaker, together making the most musical sounds I have heard, perhaps not fully accurate, but very punchy, moreso than ever before, with rythm, pace, and transparency I never thought I'd hear without ribbon tweeting or direct coupled amplification.

Emergency Landing

With deep regrets, I decided to take down the Eagle 2 on Tuesday Night (4th of July).  It had shut down the first time on Monday night.  I had first played Bob Dylan's Blood on the Tracks LP at nearly maximum volume (Tact level 98.8) which was skating on thin ice because in principle some pop could come by (though this is a clean nearly new record and none did) and clip the DAC output at 2.5V, which is more than twice the rated power sensitivity of the Eagle (1.0V).  Because the Eagle has slightly more gain than the Krell FPB 300, I was actually playing this album more loudly than I ever had (it's hard to get vinyl cranked up on my system because the phono gain isn't quite high enough) and enjoying it immensely.  But lady luck shone brightly and the Eagle delivered a stirring performance, also with very well defined cymbals in roughly the same plane as Dylan (though, perhaps the drum kit should have been slightly more back, but it was already a bit behind the plane of the speakers.  A friend had asked me to make sure the cymbals didn't sound like white noise, and they didn't.  But on a few vocal and harmonica forte fortissimos, Dylan or his harmonica did come slightly in front of the plane of the speakers.  I am now thinking this was a distortion, possibly current limiting.

I was not so lucky playing The Doors first vinyl album, which I had purchased for the first time ever a few days earlier.  It was mastered louder and so I dialed back the level to 95.5.  But I could see from my currently unused tweeter crossover that there were a lot of spurious high frequencies on this album, moreso than typical SACD's.  But it never seemed like it was going too loud or anything, until it just quit, the Eagle 2 shut down, turning off the power light.  (I turned it off, then was pleased the amplifier came back on again when I tried 45 minutes later.)  After the fact, I noticed that the Peak Peak (as opposed to peak rms) digital level had reached -6dB, which would correspond to 1.25V output into the amp, just above clipping.  But amplifier usually shut down because of overcurrent, not just overvoltage, especially for so small an overvoltage.  The Peak RMS level was less -10dB.

Hoping to avoid that situation, I continued playing the amplifer into the next day.  But then, playing some Tangerine Dream, another shutdown occurred.  This time I had only reached a Peak Peak level of -10dB, which should be well below any kind of clipping.  Though it could be that this amplifier isn't capable of handling the Acoustats, I am thinking it is because it is old and needs service.  The second shutdown having occurred at significantly lower level, it's possible that some part(s) are now failing more each time.  This is a good reason to pull something out of the system and get it repaired before a more drastic failure occurs.

I put the Aragon 8008 BB back online, which has served more time playing the Acoustats than any other amp primarily because it has never shut down at all, let alone shut down in a progressively deteriorating fashion indicating the need for repair.  I've long felt the Aragon may have been somewhat lackluster compared with other amps, even the Parasound amps designed by Curl may be more electronically sophisticated, especially signature products like JC1.  But if indeed the Aragon is lacking in design refinement (and to be fair, the big name designers aren't always perfect either, and some are clearly stupid or worse) which I'm beginning to reconsider seeing its impeccable bench performance of the 8008 when new, it makes up anyway for in incredibly high quality construction.  The relatively simple and proven design probably is itself also a factor in reliability.  When it was sounding positively bad, I had attained a serious underbias (though 'normal' by later specs) by putting on tall feet...it is equisitely sensitive to tweaking like that...which I now think was a design intent to some degree, and I re-adjusted the bias for the taller feet, and everything is great now.  Anyway, it sounds OK after being warmed up, though the deeply caffinated and liquored magic is gone, I'm back on the ground floor.

Speaking of which, I should add that I've also become a bit suspicious of what the Eagle 2 is doing (when it is working correctly).  Somehow, it may be making things sound better than they actually are.  I'm not sure how that could be done, but I'm thinking along the lines of cancelling earlier IM distortion by only creating the reverse polarity of IM distortion, at least what little there is (rated at 0.0035%, which is higher than the THD at 0.0025%, and it's unusual for the IM to be higher than the THD, but both numbers are very small compared with typical ratings nowadays, when people seem to believe that good specs mean bad sound.

Anyway, I really don't know if it uses many stages of darlingtons like the Eagle 7, but that other design is further evidence of Iverson's use of the high feedback amplifier design concept.  I think he must have done that here, and it gives him super low distortion, super high damping factor, and because it is done correctly, super smooth and wonderful sound.  If he is also positively improving the sound somehow, that would be interesting to understand.  But I think it's almost certain to be a product of the basic design of the circuit, if maybe done with some intentionality, rather than a bolted on sonic coloration.  IIRC, Iverson knew lots of tricks and he may have pulled a few ones here.  I don't think it will get tiresome, it continued to draw me into to music I had previously found hard to listen to.

And that's the kind of thing I want to understand more.

But meanwhile I believe I want to refurb my now 3x Eagle 2's and get two of them converted to higher current mono operation for the Acoustats.  I'm going to ask if that would do it.  Whether or not the Eagle 2 is slightly altering the picture, and I suspect it is, it makes everything sound better, more worth listening to.  It might not be the amp to listen to all the time, but it is so much fun, I want to use it mostly, I think.

And get the Krell fixed too, as the next best alternative sonically, when it isn't the best.

My objective DAC tests show my world is upside down

Well, everything I know is wrong.

First Tests (needing correction in various ways)

The best measuring DAC I own is the Emotiva Stealth DC-1, a relatively small but solidly made $499 device which uses sigma delta DACs and (by default) ASRC input.  Changing that to Synchronous (yes, the least expensive DAC I own gives a choice) has virtually no effect (neither S/N or THD, both of which pegged my Juli@ card residuals).  As Julian Hirsh said about the Marantz 10B in 1963, it is so good, I am unable to measure.  (Julian's test gear wasn't that great.  He could have easily measured high-by-todays-standards THD with the Sound Technology ST-1000a, which came out a few years later.  My test gear is aways behind my best equipment)  Anyway the S/N and Dynamic Range peg my limits at 110dB and the THD also at 0.0003%, using RightMark at 1kHz.  The actual performance is better, I just can't measure it.

The spectrum is very clean too (and I'm not sure now which if any peaks come from the input of my analyzer):

Emotiva Stealth DC-1

And that's at the unbalanced outputs.

I am as yet unable to measure balanced anything because I don't have the needed adapters, which I just ordered, for my Juli@ card.

The worst measuring Dac is my esteemed Master 7 Singularity, using default OS (which should have lowest measured distortion and noise).  I measure 0.13% THD, pretty mediocre for digital device.  However the other channel, sadly isn't working at all (it has 1.5V DC offset, hmm) and this channel might have some issues also, so bear in mind that might be a result of some malfunction.  I am hoping for warranty repair at this point, at least for the non-working channel.  (If user error, you pay for all shipping and parts, that may well apply.)

Now in this case the lions share of the distortion, reaching to about -69dB, is second harmonic.  That ought to be cancelled in full balanced mode...which I am unable to measure.  The other harmonics seem below -95dB, which would coincide with numbers like 0.005%.  I'm still hoping for something like that.  The spec was something like 0.0005%.  Go figure.

So, a big disappointment, but if it gets fixed with distortion below 0.01% I'll use it.  (It may already be there in this channel in balanced and high impedance.)

BTW these also measured with coax input...my only choice for these tests, and PLL, which might or might not be the best.  I suspect no measureable effects, and that brings up something else.  The distortion progression, though it does show some non-harmonic distortions, is very clean.  In contrast, the Denon DVD-9000 has pretty good numbers (only about 3x worse than the Emotiva, rather than 50x worse, in distortion, but the noise is 10dB worse...and actually 3dB worse than Master 7), but the upper frequency spectrum (below -110dB) gets hashy looking, a lumpy partly correlated noise (I suspect aging parts).  Meanwhile the Master 7 shows discrete distortion spikes getting better above a thin line of noise--and no hash.

Denon DVD-9000

And, the Emotiva shows pretty much a perfect spectrum above the lowest noise of all, with just a few tiny ticks above the noise level for harmonic distortion, peaking below -130dB, no hashiness, no nothing, just noise at -135dB across the spectrum...which is my residual.

Anyway, as I said the Denon occupies a middle ground, pretty close to the Emotiva and quite acceptible I think of around 0.0012% THD and S/N about 100dB.  I don't understand why the S/N isn't better than that...it might have to do with that hashiness.  But the S/N is still ok.

Still, the lack of hashiness would tilt me toward the Audio GD if it got down to 0.01% in balanced mode.

The DAC 19 measured 0.075% THD, not really all that great, and more 3rd harmonic than second, though just like the Master 7, no hashiness, just a bit too much distortion which looks like analog distortion

While the FPB and Master 7 are being repaired, I'm going to use the DAC 19 becuase it's very low latency matches the Emotiva, and the Denon measured the 300 msec additional latency (variable at different sampling rates) I found I had to compensate for the variable latency--and that's just too much trouble.  The Emotiva is dedicated to the subs which need the balanced connection.  I'm going to get more of those Emotiva's, however, or something just like it.  I would not be above trying something like that on the Krell instead of the subs.

Further Thoughts

I will be adding some pictures later, including the raw data, to this post.

The poor measurements of the Audio GD Dacs probably were made worse by the "punishing" 10k load of my analyzer card: the Julia@ in SE mode.  The 100k load of my Krell and Eagle amplifiers is probably going to improve those numbers by about a factor of 3, so I'd predict 0.02% distortion from the DAC 19.  I am going to try to measure this eventually.  That's as acceptable as it is sounding right now.  And balanced operation should dramatically reduce the mostly even order distortion of the Master 7 to below 0.01%, along with the higher impedance load.  I'm going to measure balanced operation later, and possibly buffered/balanced.  And analog distortion generally falls into the noise at lower levels anyway.

One problem, however, I don't get the benefits of the balanced Master 7 operation with the Eagle 2 amplifier, I might as well just use the Dac 19 for that I guess...or use some kind of intermediate balanced to SE buffer, I could repurpose my bedroom Emotive Preamp for that...except I need it in the bedroom.

I'm thinking Kingwa's discrete FET circuitry could be better.  I'm thinking I'd prefer a solution using the best of all opamps, the OPA 211, using good ole feedback, which is proving itself useful in my amplifier tests.  The OPA 211 uses hybrid germanium technology, and silicon FETs cannot do as well, and combined with top quality design and feedback, distortion should be able to actually meet the 0.0005% spec, I think that's what TI's reference amplifier does.

The graph almost seems to show the spike in 2nd order distortion for one channel in the Master 7...but actually both channels are the same channel, I had to use a Y adapter to connect to the Juli@ because otherwise it gave me 'average' numbers absurdly high.  It could be the peak is entirely due to this.  So I'm not counting the above spectrum and results as the official ones yet.

ASRC - Asynchronous Sample Rate Conversion [updated 7/19/17]

Nowadays I have trouble talking about asynchronous interfaces for SPDIF and AES because everyone else is only thinking in terms of USB interfaces.

USB interfaces are a whole other ball of wax I don't want to think about, along with I2S, as I have detailed in an earlier post, because they inhibit the chaining of digital devices and generally lock you in to proprietary software.  That kills the fun, and right now I can't use those interfaces anyway because my style of digital audio relies upon the chaining of digital audio devices, such as the Behringer 2496 DEQ units I use as crossovers and room correction, and the Tact RCS 2.0 I use for volume, balance, and polarity adjustment (though it is capable of much more).  USB and I2S assume you are connecting a computer to a DAC, with nothing in between.  In principle you could chain USB and I2S devices....but you'd have to write all new operating systems for every step of the chain.


Anyway, SPDIF/AES interfaces can also be implemented in synchronous and asynchronous versions, the asynchronous version generally involving ASRC -- Asynchronous Sample Rate Conversion.

ASRC has always seemed very fishy to me, because it essentially relies upon endless interpolation.  The data presented to the ultimate digital converter is not the data you started with.

No interpolation process can be perfect, and especially when in this case it involves guesswork about incomplete data.  Oversampling is entirely different because it is a kind of feedforward instead of feedback.

In some sense, synchronous conversion is perfect because it does not touch the contents of the data.  It only introduces tiny amounts of timing variation, which could be nearly eliminated by using buffers or other stabilization measures.

Here is a pretty good discussion of ASRC at What's Best Forum.

One digital receiver chip which implements ASRC is the Cirrus Logic CS8420.  While bit perfect transmission introduces no THD or dynamic range limitation whatever, the CS8420 is specified as having 128dB dynamic range and -117dB THD at 1kHz.  That's pretty good, and better than virtually all DAC's, but it is not perfect.  Further it seems to me the imperfection is increased by jittery inputs.

Benchmark DACs may have been among the first to include ASRC--they made a big point about it--but I'm suspicious that earlier ones also had this going back to Mark Levinson and possibly even Wadia (though Wadia seems never to have used the term ASRC).  High end digital pioneers implemented ASRC in their own FPGA's.

When I asked Kingwa about whether the Master 7 Singularity used synchronous or asynchronous receivers for AES and SPDIF and he replied Asynchronous.  However I'm not 100% sure he understood that I was not asking about USB.

In this little discussion, an industry expert says that ASRC is essentially PLL done in the digital domain.  That is exactly the idea that has occurred to me in the past month.  But once again, I must add that 'the digital domain' means that it alters your audio bits, and such alteration can never be done perfectly, and the imperfections relate to stuff like noise and ground loops in the signal that would otherwise cause only tiny alterations in timing.

On the other hand...Emotiva shows impressive improvement using ASRC.  The Emotiva Stealth DC-1 (a inexpensive sigma delta DAC including AES input and balanced outputs which I use for the subs) allows you to turn ASRC on or off.  I have generally turned it off, but maybe I shouldn't?

Emotiva has been using the AD1896 ASRC - same a Benchmark DAC 1.  (Not sure if that's true of Stealth DC-1 though.)

Amirm at WBF isn't impressed by the Jtest results shown by Emotiva.  He says ASRC needs a test to expose its weaknesses the same way Jtest exposes jitter caused by AES/SPDIF.

(I can't believe that Amir and many other posters I've enjoyed reading at WBF are now banned.  I made a rambling post to an ages old thread over there a few days ago, so apparently I'm not banned yet.  You cannot say that Amir is an objectivist flamethrower, like, say, Ethan Winer--though I liked him also.  At one point, I remember a great debate between Amir and Winer but that was at AVS.  So I wonder what's up over there.  It's not just objectophiles who have been tossed apparently.  Other forums were lots of interesting people are now banned include SBAF, over there it seems like over half of all posts are by now banned posters.  But SBAF always sounded tricky.)

Update 7/19/17

Here's one of the best discussions I've found about PLL vs ASRC digital input receivers (aka DIR).


It turns out that my new favorite DAC, the Denon DVD-5000, which was considered State of the Art in 1998 (a mere 19 years ago) uses the very well known CS8414 dir by Crystal Semiconductor, which is a standard digital PLL type of receiver (not ASRC.  The 8414 was a follow on to the 8412 which many still regard as the best of it's type.  The following CS8416 is generally considred not as good.  But all of these chips have a self-jitter specification of 200ps (not wonderful).  That specification tells us very little, however, about how incoming jitter is suppressed.

Within the ASRC type receivers, there is actual considerable jitter suppression done before the ASRC.  The ASRC is said only to encode a tiny portion of the incoming jitter, the "residual" jitter, into the data.  Which, as proponents of ASRC say, is precisely what the DAC chips themselves do.

(So, I ask, why don't they leave the DAC chips to do that, as they previously had, and simply use the "superior" jitter suppression to do what it can do without using any explicit ASRC at all?  I mean this as a serious question, however I would not be surprised if the ASRC chips handle this residual jitter better than the DAC's do.  But where are the numbers?  Where are the measurements?)

Generally the ASRC chips have much better self jitter numbers, like 50pS.  But the newest synchronous receiver chip, the DIR 9001, also has a self jitter spec of 50pS.  This tells us nothing about incoming jitter suppression, and the self jitter specs may have been differently determined, but it is possible that DIR 9001 is better than CS8414.  Or not.  Some people still seek the old Crystal receivers.

Both ASRC and PLL type receivers require some kind of buffer, or the pre-ASRC jitter rejection could not possibly work.  There is some argument at the above link as to how large this buffer needs to be, with some saying it can be quite small.  I think I can see why that might be possible in some cases, but I'm not sure if it can accomodate all cases.

This Hobby is Getting out of Hand

Reading the story of a guy who started loving Magnepan 3.6s and Martin Logans, and then discovered Apogee's, now his favorite of all.  And just a few years into his Apogee kick, he has a whole "forest" of Apogee's, and remarks... "Man, this hobby is getting out of hand."

Back at the AcoustatAnswerMan thread at DIYAudio, poster Kouiky writes incredibly dense, condescending and essentially content free prose relying on marketing buzzwords and inuendo about how Acoustat lovers are fooling themselves into believing Acoustats are the best speakers ever.  Sound Labs are the real deal he claims, but while criticizing others for fantasy he also hardly backs his claims up with any evidence, just rationales.  Blowhard!

This kicks off a certain amount of somewhat less content free arguments (though, from what I've read so far, no one except for me is really explaining why a 'flat' line source speaker like the 1+1 is so wonderful, then I get to my comments on page 106 where I don't do such a good job either).  The bottom line is that flat gives you choices, curved does not, if the speaker is curved what's the point of angling it?

Of course the big thing is price, I got my 1+1's for $600 plus $300 shipping, and a pair of Sound Lab Ultimate 1's go for $45,000, that's 50 times more money and approaching the appraised value of my home a few years ago.  (Since then, to be fair, I've spent another $1000 or so on Acoustat replacement parts, and my home is now appraised at $70,000.)

[Oh, there are cheaper models?  How can I again live with a small speaker shouting up at me from the floor?  I've gotten used to tall towers making waves throughout my little cottage.]

I'm perfectly open to believing the Sound Labs might be better based on reputation and the fact the company is still a going concern continuing to do R&D.  Though I should mention that several of the very experienced audiophiles I know think the Sound Labs aren't very good and the good ole Acoustats are probably better.  But they could be very wrong--by all accounts Sound Labs have gotten better and better and what I consider the best Acoustat product, the 1+1*, hasn't been made in a long time and even then was only created by combining panels that had been designed in the 1960's and 1970's.  I myself haven't heard Sound Labs at least since 1979 when I saw them at a store and I'm not sure I even heard them then, but they looked nice.  It's ultimately true that everything is a compromise, and anyone might like this or that better for unpredictable reasons.  So behind the curtain, some chose the Acoustats, and others would choose the Soundlabs.  Even technically there is some issue with using lots and lots of small electrostats in a big curved frame, as compared to big panels.  Little panels are indeed more subject to things like comb filtering.  Now, I am not saying Sound Labs hasn't solved this problem well enough for most people, and quite possibly me too.  But there might be some who like the other approach.

(*I can't say I've systematically heard all the others.  I only heard 2's and 3's once, and that's it.  But many people who have heard them all, such as Mr. Acoustat and E-Stat, have also settled on the 1+1 as the best model as well as something essentially unique in audio: a 1-way narrow and flat and tall line-source speaker.)

I used to believe I had to have 1+1's instead of anything similar (even Magnepans) for several reasons, one being I have such a small room the speakers need to be narrow (and also to provide viewing area for the TV for my parties).  Now I hope to have projection screen TV system soon, so then I could have significantly wider speakers, but it might still be hard to fit in the biggest Soundlabs, even if I had the money with no other pressing needs.

Poster E-Stat (who can also be seen at other blogs) seconds the notion that Sound Labs are better, but emphasizes the incredible value of the Acoustats...he still has 1+1's in his garage system.

E-Stat (possibly in another website I saw today) tells a microscopically short story of being at HP's house: three incredibly transparent systems.  Will we ever see pictures?  Apparently not that kind of transparent.

One point that is made in the Sound Lab vs Acoustat argument is how the Acoustat has quite an enormous resonance in the bass.  But it turns out that virtually all dipolar speakers do this going back the Peter Walker's Quad ESL 57's.  This is how they overcome the natural bass cancellation that a dipole would otherwise see.  WRT Sound Lab, in the 1980's John Atkinson remarked about the very large bass resonance they had.  Kouiky snorts that since then things have changed, Sound Lab now uses 'distributed resonance' (which he does not define).

The alternative to having a bass resonance is an electronic correction--which is precisely what Linkwitz does with his dipole speakers.  Along with Linkwitz most expensive designs like Orion, you must use a very complex electronic equalizer/crossover and multiple amplifiers.

I've always thought of Linkwitz as the greatest, but when I last heard a friends pair of Linkwitz latest top shelf design...the one that replaced Orion...I thought it was outstanding for a dynamic speaker system, and possibly slightly better than my 1+1's in immediate transparency, dynamics, and good soundingness (though I have equalized my Acoustats to sound equally good by adding in the 'Linkwitz dip')...but not enough so for me to bother with it and concurrently lose the Acoustat advantage in utility in my current nearly near-field 'electrostatic headphones' listening configuration.  I wasn't blown away, though possibly a better configured Linkwitz system would do so (better room, better source--it was a chromecast, and better electronics).  If I were to go to a better electrostatic system--assuming there is one--I'd want to be blown away by it, not just a small step up but a transformation.  And I think a tall line source makes a better 'whole house' system for background music and parties.

One very well heeled audiophile I knew in 1980 who could buy anything he wanted thought nothing had better bass than Acoustats, so he used Model 3's and 4's together as the substitute woofer for his Hill Plasmatronics.  He also had Magnepans but didn't think they're bass was nearly as good.

Anyway, some of my testing in the past couple weeks has been running the Acoustats full range (to see if that would make the amplifier shut down quicker--couldn't tell) and I was shocked, shocked at how good the bass sounded.  I've been equalizing and measuring and re-equalizing over and over since 2009 and still not attained the amazing bass clarity of the Acoustats in my room down to at least 40 Hz with no room equalization at all!  My thinking has run along the lines of snapping up two pairs of the now easily attainable 2+2's to use as subwoofers which would take no more floorspace than my two monsterous SVS PB13 Plus Ultras.  But the SVS are also useful tables in my small room, and I think the sub+electrostat matching thing is a great challenge.

In that other blog somewhere I read that yes, the Acoustats and most other dipoles have large bass resonances.  However, they don't excite the room in more than one direction, thereby avoiding all the room modes associated with the two other directions, which is in most cases FAR more important than the bass resonances in the speaker itself.  I concur and this is the essential part.  This is when the questions about HP's listening room came up because E-Stat said he was there hearing the Sound Labs and said the room was fairly small, but others said the room was supposed to be 25 feet long which eliminates much of the room mode problems.

BTW, here is the page which shows a pretty good guess about Roy Esposito's Air Mod.

From the description of what it does I'm not sure I'm interested, but I'm thinking some similar kind of re-working could deal with the problem I experience (the 100 Hz boom and the 120-200 Hz suckout) and if I could deal with it there it would be better than electronic, because it's a huge waste or impossible to fix a suckout with EQ.

Anyway, back to content free reviewing, the review of the Acoustat 2+2's in 1986 in The Absolute Sound by William T. Semple comes close to that (though, the Manufacturer's reply wasn't any better).  And much of what Semple does say is way out wrong, or just twisted to make it sound worse than it is, such as this, which strangely appears among the list of the speaker's strengths:

"JN wrote that the Acoustat was 'relatively neutral', a sort of backhanded compliment, which is, in fact, equivalent to calling the speaker colored."

Funny I don't recall seeing Mr. Semple reporting in any other issues of The Absolute Sound, and he writes a lot like HP.

But there are a few bits of truth.  The Acoustats are not seemingly as transparent as some speakers somehow, and they don't have the most extended high end.  Andy Szavo says if you find inadequate highs you can adjust the control.  But sadly the HF control doesn't increase the extension of the highs, it merely increases the volume of the highs (update: in the C mod, it may actually increase the extension a tiny bit, but it is already at diminishing returns, and making the adjustment requires messing inside the interface box not just turning a knob...I have never done it myself)  and I can report to you that the volume of highs from the Acoustat 1+1 is not only sufficient, it is slightly excessive, especially on the beam.  Only about 10 degrees off the beam is the quantity of highs not in excess, and that (circa 10-15 degrees off the beam) is precisely where to sit, because most of the alleged problems, such as head-in-a-vice beaminess, go away there.  At 10-15 degrees off axis, you can move around a bit without changing the balance.  And then you also have the opportunity to add a omnidirectional ribbon tweeter, as I have, which is pure magic, and adds back every quality the Acoustats were supposedly missing.

But I think the Acoustats are nevertheless sufficiently transparent to be moreso that 95% of all speakers, including about 95% of high priced (over $10,000) speakers.  Only a select few, such as Quad 988's, and Maggies that have ribbons, are more transparent.  Cones and Horns can forget it with very few exceptions (such as Yg).

And the way to deal with the tiny lack of high extension, is to add a super tweeter, which is not a curse it's an opportunity.  Of course dealers and manufacturers want you to believe your $2400 speakers are all you will ever need.  But if you are a real enthusiast, you will never believe anything like that about anything, and will add some kind of super tweeter, be it ribbon, plasma, or a specially made electrostatic.  As I usually do (right now I'm running the Acoustats w/o supertweeter because everything is a mess because of testing, and I can't actually say I'm missing much).

The slightly imprecise imaging of the 1+1's is actually The Absolute Imaging you will find in concert halls.  The pinpoint imaging of speakers with tiny tweeters is not at all like the real thing.

It is so unusual to have a full range speaker that radiates full range from most of the front.  When you walk up to it, it doesn't get 'closer' like speakers with tiny tweeters.  That makes you think it's less transparent, has less highs, etc.  But again, let me tell you, when you walk forward a few feet in a concert hall nothing much changes either.  So once again the Acoustat presentation is actually surprisingly like the real thing.

I have got to dig up the actual TAS review of the 1+1's, which Semple reports is in the Summer 85 issue.  Thanks him for that bit of info.  Semple makes that sound like a much more positive review, though they had not apparently heard the "C" mod.  Then Semple goes out of his way to actually put down that earlier review, comparing it to something that Julian Hirsch might write.  (That sounds a lot like Harry's writing to me.)

It wouldn't surprise me if Strickland had not kissed HP's butt.  They had a fundamentally different view of where audio was going and should go, with Strickland more sympathetic to the relatively (here we go again) affordable, and HP siding more with the no limits side.  And Strickland more (but not entirely) on the audio objectivist side, along with his friend David Hafler (who was not getting very much attention in the pages of TAS at all).

Problem was, however, electrostatic dipole speakers are not intuitive to most people, and they are much more dependent on proper setup (as in, you get nothing until you get there) than most speakers (which can be fairly good even without the intensive setup).  So Strickland needs high end audio dealers, but the dealers are siding with the more and more easily profitable 'no limits' audio, the kind that lets then get by selling 5 $50,000 speakrs instead of 125 $2000 speakers.  The rest of audio retailing sinks to the Best Buy model which demands easy price-based sales with no support needed.

HP's destruction of Vendetta Research was soon to follow.

Now though I've never seriously considered buying the $50,000 speakers, you can blame me for part of this.  Despite what I'm saying above, I've always enjoyed reading HP and the other high end magazines.  I've been a long time subscriber.  It doesn't bother me so much I'm reading about the unaffordible.  It's like reading a gossip column.  And eventually, things I never imagined having, things such as the Krell FPB 300, became available to me.

But then you could say for me the magazines weren't serving the dealer trade--as I imagine their primary purpose, and I remember when TAS and Stereophile blew past their original concepts and started publishing advertising.  At first and for some time, it was only Dealer advertising.

No, for me the magazines are serving as a gateway to the endless amount of re-cycled audio.  So they really are the 'racing forms.'