Amir at ASR claims Uptone won't sell the ISO regen to him...

[jabbr]

8 hours ago, Superdad said:

 

 

Differences between USB cables are also revealed using with eye-pattern sweeps.  Here is the before and after with the ISO REGEN:

BEFORE:

 

AFTER:

 

 

 

 

 

There is a clearcut improvement in the eye pattern after. 

 

I'd say that the device is working as advertised. Now @JohnSwenson has speculated that this would result in a decrease in jitter at the DAC chip. If that were demonstrated, it would be clear evidence that the device was actually doing something to the DAC. I have said many times that the DAC manufacturers should make more effort to ensure that DACs are not subject to differences in jitter of inputs that are within spec. i.e. in an ideal situation the two above inputs would yield the same SQ. But the world is not ideal and Uptone's business is making improvements. I will say that @PeterSt is one who does spend a great amount of effort on this issue, has his own isolation solution, and whos NOS1a DAC sounds fantastic. Is this due to USB isolation or general attention to detail and excellent engineering? Who knows probably more the latter but probably somewhat due to the former  

 

How should the DAC chip "jitter" be measured? I will give my standard answer that I would expect to see a high resolution phase noise by offset frequency plot as is standard for low "jitter" clock oscillators. These plots are industry standard. So that's what I would like to see. Of course the equipment to do this has been expensive. For my own experiments and development work I have obtained the parts to resurrect an HP 3048A which interested parties can Google. I think that doing something like this is certainly within the skill set of @JohnSwenson  but don't know anything about "Amir" except that he was a Windows software guy but I haven't seen any of the type of measurements I would expect to see to answer the issue (a simple spectral analysis doesn't cut it). In any case he's not here and there are many measurements which can be made for many reasons.

 

Now furthermore I would look for differences in the close-in phase noise at the DAC chip ... but just try and make those measurements  (I have some ideas but not something that I'm prepared to go into details at the moment)

 

So what would we see at the DAC output? Basically a very pure tone fed into the DAC would show a widening of the very high resolution spectrum coming out aka "linewidth". Thats one thing, there are others. 

 

In any case lets stop getting personal here guys... there's a lack of technical discussion about why the measurements that are being proposed elsewhere either are or aren't appropriate.

[jabbr]

6 minutes ago, plissken said:

Here is an eye pattern w/o any added gizmo. Image from Tektronix.

 

 

So my question is: What device is being shown in the eye patterns supplied by Alex?

 

What would the eye pattern test look like with a good USB hub?

 

Where are the analog output measurements showing a DAC incapable of keeping it together?

 

As long as Mask Compliance and Mask Margin are not in the keep-out area you are good to go with either eye pattern. 

 

The regen essentially is a good USB hub

 

I wouldn't expect it to help in every situation and for every DAC. As I've said (repeatedly), I view this issue primarily as the responsibility of the DAC. I know that most DACs use a module such as he Amanero (or others) that might go for $50-100 retail and in many cases might use the clocks that come on these modules as their DAC master. So in many cases this "USB isolation" etc may be suboptimal for audio.

 

I entirely agree that there should be proper measurements showing an improvement at the analog output of the DAC. I've proposed some measurements (not BTW what JA or Amir have published).

 

In a perfect world you are correct. We should be good to go but we aren't and, for example, nowadays with a $20 Crystek having the specs it has there is no reason to have a less than cheap DAC than meets less than these specs... but we aren't in a perfect world and there is opportunity for improvement in the electronics ... and note that e.g the DSC1 we have a great idea of what each of the parts that go into a DAC is.

 

Regarding above post, I am sure that JA has an agenda. I don't know Amir enough to say. Do you think I have an agenda? (if you do you'd be surprised). In any case lots of folks have agendas ... let's try to be objective and rational.

 

[jabbr]

11 minutes ago, Sal1950 said:

"Using this long cable and my MacBook Pro running on battery power, I tested the Regen with a PS Audio PerfectWave DirectStream DAC and a Meridian Prime D/A headphone amplifier, using my Audio Precision SYS2722 analyzer and the same test signals I'd used for the JitterBug review. Sad to report, I found that the Regen made absolutely no difference in the D/A processors' analog output signals. With one exception: the levels of the power-supply–related spuriae in the Prime's output when powered by its wall wart were slightly higher in level with the Regen than without it."

John Atkinson

 

 

Ha ha, ok! Do you have any idea about why the PS Audio DAC might not show a difference?

[jabbr]

3 minutes ago, Superdad said:

John is designing, and after his move will start building, an advanced variation of the John Miles TimePod--the long ago affordable unit that Microsemi now sells at an outrageous price.)

 

Very nice approach ... in fact ... basically using a digital phase comparator assumed to be on an FPGA in order to create a phase error measurement system. There are several ways to do this, either with an external known clock and also using a known delay. In any case a fruitful approach ... my own interest in the HP 3048a is that the accuracy of any such system needs to be calibrated against a known system ...

 

It will be great when we can replace the current battery if measurements with better measurements more capable of making subtle distinctions.

[jabbr]

24 minutes ago, PeterSt said:

 

I would be careful with this.

The FPGA shows how much jitter ?

 

We can measure that

 

So there is this whole field of digital phase comparison needed for phase array radars, GPS systems, SDR (software defined radar) and these are one of the main drivers of high end FPGA. 

 

In any case John Miles is a real guy who made a real system that's been tested against everything from HP née Agilent nee Keysight and got picked up by Microsemi so at least some people think he knows what he's doing

 

 

[jabbr]

3 hours ago, lmitche said:

That's right, I forgot that for some they need to see one of these: 

Before they trust their ears.  Have fun with that!

If you aren't going to trust your ears you need one of these:

and 

 

[jabbr]

56 minutes ago, wgscott said:

What would be of great benefit would be to depersonalize this issue.

 

If folks could agree to do that, then the next step would be to determine what measurement or set of measurements both "sides" of the debate could agree in advance would be decisive either way.

 

The third step would be to find some competent, neutral party that has access to the equipment, to cary out the experiments.

First step -- absolutely necessary

 

Second step -- Entirely agree -- at the very least we might discuss & agree upon a reasonable set of tests

 

Third step is harder because the person or organization doing the testing needs to be funded and therin lie the issues with impartiality 

[jabbr]

1 hour ago, Daudio said:

 

Uh, oh, could it be another "My meter is Bigger then Yours" battle brewing ? 

Haha just giving a hint at what degree of difficulty these tests might entail  

[jabbr]

2 minutes ago, Jud said:

We've already had independent blind testing by two people (I was one) in which the test design most closely resembling the production version of the ISO Regen was decisively chosen in a few minutes of listening.

 

So, assuming this is valid with a larger number of evaluators, a proper measurement would be able to distinguish something that is audible. eg if there are phase error differences that would lend support to the proposition that altering phase error is audible.

[jabbr]

35 minutes ago, Jud said:

 

That's what the ideal state would be - if measurements such as you and John have proposed (or that @PeterStmay have already done or be contemplating) would turn out to correspond to the blind test preference.

If the measurements don't correspond to a valid audible difference then they are irrelevant for our purposes here. 

[jabbr]

2 hours ago, JohnSwenson said:

The test @jabbr mentioned, playing a pure tone through the DAC and doing a spectrum and looking at changes in the "skirt" is probably a very good way to test this, but with existing test gear the skirt from the test equipment is going to be wider than what you are trying to see, again not a particularly useful test.

 

It seems to me that the most promising measurement system is to use a cross correlation phase noise system that has been designed to work at audio frequencies. This at least theoretically has the resolution to distinguish differences in the audio out due to jitter changes in a good DAC, but nobody has built one. That will have to be something else I look into building.

 

 

I have a few questions.

 

The suggestion I made of a measurement on the analog output of the DAC assumes the jitter would measurably widen the frequency band. This would be with a high resolution spectrum analyzer. What information leads you to state that existing spectrum analyzers do not have sufficient resolution? What resolution are to estimating would be needed?

 

In the second case you are measuring the phase error of the DAC clock. which isn't itself in the audio frequency range. Certainly frequency offsets in the audio frequency range are easily measurable.  The internal mixer of my system has a minimum frequency of 5 Mhz and so should be fine with most DACs. One can use an external mixer when lower frequencies are needed. (e.g. Rutgers' appoach)  In any case these measurements can be made with current equipment, and have because, well, DAC clocks have been measured. Are you looking at measuring something else?

 

Not that a digital approach wouldn't be great, just will postpone the measurements until some date after which the measurement system is developed.

[jabbr]

2 hours ago, unbalanced output said:

 

@jabbr, the idea is good however it may be trickier than it seems. When calculating an FFT, you're implicitly  smoothing out the effects of jitter

 

 

4 hours ago, Don Hills said:

thought it extended all the way down to 0 Hz.

What is the other name for a constant phase error?

 

Ok so there are two different measurements that we've been discussing and I need to clarify what is what:

 

1) Measure "jitter" or more properly phase error in the DAC clock: This is done with phase error measurement equipment. There are several ways to measure phase error using both analog and digital techniques, so-called "vector network analyzers", the venerable HP 3048A and newer equipment such as the John Miles TimePod as well as offerings from Keysight nee Agilent nee HP.

 

Phase error is provided as a plot : http://www.crystek.com/crystal/spec-sheets/clock/CCHD-575.pdf -- in this case the Agilent E5052A was used to take the measurements. See how the phase error rises as the offset frequency goes down? Thats called "close in" phase noise. Yes this phase error goes down to an offset frequency of zero. But phase error is always measured with respect to the clock frequency e.g. 11 Mhz, 22 Mhz. "Jitter" is 11 Mhz +/- offset where offset varies from 0 to ...

 

2) Measure the effects of jitter at the analogue output of the DAC: I've been discussing "line width" and widening of the peak in a number of recent posts. If you take a pure tone -- and for the sake of our discussion here, a 1KHz sine wave tone generated at 24 bits 352 kHz, or DSD256 or DSD512 (for the sake of our discussion) and take this digital stream and send it to the DAC, in an ideal situation the DAC will emit a pure 1kHz sine wave.

 

Phase noise, specifically close-in phase noise will cause a distortion in the output of the pure sine wave. What does this look like? It causes the "peak" in the spectrum to widen. That's called "linewidth". It goes up with increasing phase error. If we take the DAC output and run it through a spectrum analyzer such as the Audio Precision that "Amir" uses (quotes because i've never met him and he's not here  then what would be a thin line at the 1kHz frequency will spread out. This is caused by applying the phase error curve as above, to the pure tone.

 

Hopefully this makes more sense, and (2) is proposed as a way for folks who have a spectrum analyzer to measure the effects of clock jitter on the analogue output

 

3) I'm not being coy here: if the linewidth of a pure tone does not measurably widen with a high quality measurement e.g. not single Hz increments but let's say 0.1 Hz or better increments, then jitter at the DAC clock is not having an appreciable effect on the analogue output. I don't know that 0.1 Hz increments are the specific resolution needed but you know, basically if the 0.1 Hz phase error is not significant your clock is doing a great job

 

[jabbr]

11 hours ago, pkane2001 said:

What baffles me to no end is how so many folks get convinced that measurement couldn't possibly be useful in predicting audio performance. I'm sure that some snake oil peddlers would like you to believe that. But objective, repeatable measurement is key to testing the performance of any electronic device. This is overwhelmingly the practice of all competent electronics engineers and designers.... except, it seems, in hi-end audio.

 

If the device 'cannot be measured' or 'the measurements don't demonstrate what it does' then how do you know that it does anything positive to the signal? To me, this is a warning sign that the manufacturer or the designer don't know what they are doing or trying to hide something. Especially if they tell me that I wouldn't understand even if they tried to explain it. Try me: I'm not that dense.

 

There are two sides to this coin:

 

1) Measurements are only good if the correct measurements are done. Any scientist worth their NaCl knows that someone doing a measurement with an agenda will produce skewed results. Even the choice about which measurements to perform or even which equipment to use can affect an agenda and skew results.

 

2) I agree that it is incumbent upon vendors to produce measurements, especially if the vendor is espousing "white papers" and theory of operation. We see all too much voodoo enshrined in pseudo-science and complicated engineering talk. Explanations that the measurements won't demonstrate are fine unless the vendor is espousing a particular mechanism of effect. 

 

3) I know that I discuss a lot of theory but I'm not selling anything, so my advice is free  and you get what you pay for  ... that said, I've considered commercializing certain things, but if I ever put my name behind a product (as opposed to consulting), it would be accompanied by extensive measurements (and also listening experience -- of course!)

[jabbr]

1 hour ago, mansr said:

Nice straw man. I've never suggested that everybody must understand and personally validate the design of everything they use. It is quite sufficient that somebody does this. The problem with many audiophile widgets is that nobody has done anything to validate their operation. In some cases, the manufacturer even actively prevents others from testing it. Would you buy a car from a manufacturer who refused to let it undergo standard testing? What of a car manufacturer who was found to be rigging tests in its favour? Do they get a free pass, or are they hauled in front of court?

 

Ah... a car manufacturer might not let you or I perform tests on a car that we intended to publish hence the concept of the independent testing entity as imperfect as that is. We don't have standards (yet) for what tests are appropriate. Do you have any suggestions?

[jabbr]

7 minutes ago, Don Hills said:

 

For 1): ... So to my layman's mind, the other way of looking at this is that the clock is constantly speeding up and slowing down by random amounts. These speed variations are more pronounced over longer time periods. An analogy might be driving a car, and randomly stepping on the accelerator or brake. The longer you step, the greater the speed deviation. This neatly leads to your 2), where the analogue signal output of the DAC also varies in "speed" (frequency) as the clock speed varies. As you say, this is measurable.

 

Yeah ok that distills a lot of math and physics down -- the clock speeding up and slowing down isn't "white noise" random, rather "pink" or 1/f random, that is to say the speeding up and slowing down is greater for lower frequencies (at the same base clock frequency).

 

7 minutes ago, Don Hills said:

 

Things get interesting when you try to determine what amount of variation is likely to be audible. There have been tests performed with different types of phase noise, and the worst audible case is still orders of magnitude beyond the performance of any competent DAC.

 

People have focussed on the baseline phase noise which is indeed low for "competant DACs". If you look at the close in phase noise it is drastically higher -- how many DACs report their 0.1Hz phase error? What dB would you expect the limit of audibility to be? hint: I bet vast majority of DACs would exceed a reasonable number you might propose.

 

7 minutes ago, Don Hills said:

 

For example, most people seem to be quite comfortable with the "jitter" (wow and flutter) performance of turntables, although they are audibly bad. This is easily shown. Play a 1 KHz tone from a test LP. It sounds ok at first, then you start to notice small anomalies. It doesn't quite sound as solid and substantial  as it should. Then change to the same tone from a test CD or other digital source. The difference is not subtle. But the effect disappears when you compare music on the turntable and digital source, even though the same effects are present in the musc as were audible in the test tone.

 

This is slightly different: If you were to plot phase error against offset frequency, a turntable and clock oscillator would have different curves -- and what I think we are hearing when there is "wow and flutter" is a particular resonance due to the physics of the turntable. Where they are the same is that both will generally have increasing phase error with close-in offsets! Some clocks can have resonances etc, I am describing the general appearance of general phase error plots of clock oscillators (as demonstrated by Crystek and others)

 

In any case yes, the degree to which this is audible, or limits of audibility are not yet known (at least by me)

[jabbr]

I have discussed the method to measure the effects of "jitter" specifically phase noise on the analog output of a DAC. My proposal is to feed a pure tone of, for example, 1kHz, digitally encoded, into a DAC, and perform a high resolution spectral analysis. It has been questioned as to whether existing equipment would have sufficient resolution in order to do this, and that equipment designed for the audio range would need to be developed.

 

Here is a technical discussion from 1984, the HP 3561a: http://www.hpl.hp.com/hpjournal/pdfs/IssuePDFs/1984-12.pdf , in which these exact issues are discussed. Notably:

 

frequency range: 125 microHz (that's right!) to 100 kHz

frequency resolution within this range: 640 microHz

 

So the equipment needed to do these sorts of measurements has been available well since the 1970s but this for 30 years. Additional specs and details are in the above link.

 

This simple and relatively specific measurement that I am proposing has not been widely discussed, or perhaps not yet discussed, so I will write this up as an article.

[jabbr]

17 minutes ago, PeterSt said:

Hey, you know what ? I just made a new USB cable. It is even better than its predecessor. No, typo, it sounds better, it *is* not better.

Who says it sounds better ? me. Maybe you too after trial with money back guarantee.

If your DAC had better isolation, the cable wouldn't make a difference

[jabbr]

30 minutes ago, PeterSt said:

 

Zero.

 

How many DACs report their phase error at any other distance ?

Also zero.

 

And otherwise ... link ?

That's because if I gave you the ability to measure your phase error in real time as you made software and other tweaks ... then you'd never get anything else done ... ever  

 

[jabbr]

30 minutes ago, PeterSt said:

Hey, I use 3 cascaded isolators and the sound is crazy. But it is too difficult to make error-free (for commercializing it). With saying this, all I challenge for is that you guys tell me that the isolation is only so-so.

AND THIS IS TRUE.

I think I have an idea why. It is, I think, a limitation from using most if not all of the USB to I2S interfaces -- you don't need I2S anyways. But this complicates everything. My idea is to start at the DAC and work backwards -- anyways an idea that I'm devoting a whole bunch of time on -- and yes I won't claim it's "better" there would be measurements. You are, at least and rare among manufacturers, interested in and working on these issues IIRC you were the first to point out the Intona device to everyone. Your NOS1a does sound fantastic so the very hard work shows but everything , perhaps, can be improved on -- well until we have proper measurements who knows  

[jabbr]

3 minutes ago, Sal1950 said:

How many rounds of tests were made?  What were the specific "blind" conditions. What were the final voting numbers.

It was @Jud with a bottle of bourbon -- how many rounds did he drink? We hope he didn't go blind  

 

From a technical POV, we call this an out of focus group

[jabbr]

33 minutes ago, pkane2001 said:

 

Please explain how an optical isolator will not stop ground loops or leakage currents.. By the way, current audio standards require well under 100Mbps bandwidth.

This isn't an easy question to answer because of the peculiarities of USB audio transmission. Corning makes a fiberoptic USB extender but the 5v signal needs to travel alongside. Why @PeterSt's three levels of isolation don't "work" ... entirely ... is a bit perplexing and the answer may get complicated, and I think might involve re-engineering the USB interface itself (the boards at the DAC end). At this point we could readily test out USB vs fiberoptic Ethernet (on the same interface) and this would help determine how much is "galvanic" isolation vs "other factors". For me, being able to measure changes in such factors at the level of the DAC clock, as well as analogue output, is essential to ending what will undoubtedly be a permathread until then. In the meantime, for example fiberoptic Ethernet, is "full galvanic" and we will undoubtedly see proliferation in this area -- at least if Merging/Ravenna et al have their way

 

But in terms of isolating I2S or DSD lines, for example, the GMR isolators have a greater bandwidth than optical in terms of what are commonly available.

[jabbr]

http://www.adnaco.com/products/s3a/

 

[jabbr]

9 minutes ago, Speed Racer said:

It is not a simple inline USB optical isolator.

Nothing here is simple  

[jabbr]

23 minutes ago, esldude said:

I think this is very misguided without some reason to believe such close in differences are problem. 

 

Lots of ways to think about this.  A one degree C difference in temperature changes the speed of sound enough that it alters the relationship between direct and reflected sound to a level many times greater than a small perturbation of clock jitter at .1 hz.  Or even 1 hz.  This is just barking up the wrong tree. 

 

You don't know what you are trying to achieve.  Is it 100 dbc/hz at 10 hz offset, is it 100 dbc/tenth hz at a .1 hz offset?  And why other than better must be better?

 

It is really very simple: you need to measure a quantity before determine the extent that it is audible. What I am saying follows directly from the mathematical curve which is measured. Feel free to load the curve (e.g. the Crystek) into a simulator and show yourself what happens.

 

But that is beside the point which I am making -- the point here does not have to do primarily with audibility. It is primarily to determine whether the device has a measurable effect. If it has an effect which is measurable then we can discuss whether the effect is audible.

 

Since you feel this is misguided, do you have a specific proposition in which you would otherwise make measurements to determine whether a specific device has an effect on the "jitter" on the DAC's clock ... oh and don't try to use a sound card with an ADC because you will need to first ensure that the clock of the ADC is better than the clock of the DAC -- those are basics.

[jabbr]

20 minutes ago, esldude said:

Here is the type plot you are discussing of a couple pedestrian pieces of gear.  64k FFT of a 12 khz tone with 48 khz sampling.  Firstly because the FFT is not 1 hz you adjust by reducing the number by 1.3 db for 1 hz equivalence.  This is also a voltage measurement so for power relative to the carrier you divide these numbers in half.  The dotted line at 10.3 hz offset would give just short of 45 dbc/hz.   The inserted result at 1.5 hz is just short of 40 dbc/hz.  Now the measuring ADC is not an AP unit.  Specs are it has about half the jitter of the DAC.  Suggestions are your measuring device be 10 db better which this one isn't.  So some of that is the ADC. 

 

Does the Regen make a significant audible difference that somehow results in the same measurement on such a device or slightly worse?  So the improvements are all at something less than 1.5 hz and very audible?  Seems like a real reach to me.

 

 

Are you kidding? You aren't even doing the test I proposed aside from ? 40 dBc -- that's terrible. I certainly hope its the ADC otherwise the DAC is really bad.

 

Do you see the specs of the Crystek clock? That's a $20 part. You have no way to get close to measuring it.

 

 

If this is a phase noise plot, then the increase as offset goes down is what is seen. In this case you can't determine whether the plot is primarily of the ADC clock or the DAC clock and so there's no way to determine whether some device like the Regen is having any effect.

 

In any case if you apply this phase error curve to a single tone you will see the significant distortion I am discussing assuming your simulation or actual measurements were of sufficient resolution.