Objective proof the UpTone Regen ISO can improve a DAC's output(*)

[Superdad]

28 minutes ago, plissken said:

 

And I believe that Amir did point out that he is very zoomed in and that what your PS is doing, while not optimal is simply inconsequential to SQ. It doesn't technically improve things with the Behringer so the advice is the $$ needn't be spent. 

 

That is an assumption on his part based on his simplistic tests. He focuses strictly on residual noise.

He rejects the premise that improved USB signal integrity and impedance match have an impact on DAC function.

Sorry for him that lots of people can clearly hear the effect but that he can't measure it. 

 

And you know, UpTone is no longer the only firm in this space--and there are now plenty of streamer and DAC manufacturers paying attention to the sonic effects of USB SI.

Spoke to the engineers at Aurender two years ago at RMAF (Same show where we chatted with Amir for 15 min. by the elevator--which he later characterized as an in-depth meeting an assesed John as an "amateur engineer.") The Aurender guys said they quite specifically focused on high SI--for its impact on SQ from the DAC.

 

I think Amir needs to broaden his thinking and methods. 

[scan80269]

1 hour ago, pkane2001 said:

 

Thanks, Scan! That makes sense.

 

What is still not apparent to me is what other effects ISO-R has on noise and leakage currents beyond what might be solved by LPS-1 (or a battery). While it might be that LPS-1 will fix the AC leakage currents that show up in Amir's measurements with SMPS, and the 5v line is also cleaned up, I don't see any other effects of ISO-R on noise or jitter.  Also, the AC leakage currents don't seem to be a problem for a few of the other DACs that he had on the bench, so there may not be a need in LPS-1 for these?

 

Sorry for the long post here, but there are many factors at play that I'd like to comment on.

 

A USB bus powered DAC has an advantage over a self-powered (by DC or AC) DAC in that the former can help eliminate at least one potential AC leakage loop from the system.  By design a USB bus powered DAC draws its power from the USB bus, and not from a separate DC power source possibly originating from AC.  Powering a self-powered DAC via isolated DC power source (e.g. UpTone LPS-1 or Vinnie Rossi MINI PURE DC-4-EVR) or a battery can achieve the same specific advantage.

 

ISO Regen by its design offers USB signal regeneration as well as USB bus power regeneration.  Whether these features are beneficial to a particular DAC will heavily depend on how susceptible the DAC design is to upstream USB crap.  Not all DACs in the market today are equally immune to dirty USB from upstream such as from a notebook.  How much the ISO Regen can "de-crap" the upstream USB source will also depend on the upstream USB source itself.  The ISO Regen represents a specific (but high) level of USB signal integrity (via the UpTone-selected USB hub chip) and a specific (low) level of embedded jitter in the USB signals (via the Crystek CCHD-575 low phase noise oscillator reference used for USB port clocking) at the downstream (DAC-facing) USB port.  These specific signal quality levels are not guaranteed to help each and every USB DAC.  Within my limited collection of DACs with USB interfaces I have examples of one DAC benefiting tremendously SQ wise with ISO REGEN + LPS-1 while another DAC yields no perceptible SQ improvement given the same treatment.

 

There are way too many fallacies on both proponents and opponents of USB decrapifiers.  I will list just a few that came into my mind today:

 

Over-generalizations / Over-simplifications:

* Heard no improvement with device added to my DAC ==> device must be worthless
* Heard big improvement with device added to my DAC ==> device must be universally beneficial with any DAC
* A measurement shows no improvement with device added to my DAC ==> device must be worthless
* A measurement shows visible improvement with device added to my DAC ==> device must be universally beneficial with any DAC

 

Fallacies:
* No difference observed in a measurement ==> cannot sound different
* Difference observed in a measurement ==> must sound different
There are plenty of real-world counterexamples to either of these
* Relying solely on measurements to judge the sonic merit of a product
* Relying solely on listening to judge the technical merit of a product
* Underestimating the complexity of a modern digital audio system and the ways SQ can be impacted

 

Quote from Chris Connaker: "Not everything that matters can be measured and not everything that can be measured matters".

 

Other thoughts:

* Single instrument measurement (with a single graph as output) is far from being an all-encompassing representation of DAC output

* USB de-crapifiers are NOT a panacea

* Sonic benefits of USB de-crapifiers can vary by both DAC and USB source used in the system, with results ranging from big to none

* USB DAC/DDCs are not all equally good at immunizing themselves from upstream USB crap

* Design know-how on effective immunization is relatively recent and not yet widespread (though should become more so with time)

* More USB DACs existing today show vulnerability to upstream USB crap than those that don't (to different extents)

* Galvanic isolation (or lack thereof) is not the only factor impacting SQ of a USB DAC

* Implementation of galvanic isolation can represent a tradeoff, for example jitter aggravation, especially if not carefully designed

* Designing the ability to block AC leakage loops into any device is NOT trivial

 

 

My concerns with Amir's measurements are not so much that they are invalid or improperly done, but rather in how he interprets the measurements and reaches over-generalized conclusions with them.

 

 

[Sal1950]

6 minutes ago, Superdad said:

 

That is an assumption on his part based on his simplistic tests. He focuses strictly on residual noise.

He rejects the premise that improved USB signal integrity and impedance match have an impact on DAC function.

Sorry for him that lots of people can clearly hear the effect but that he can't measure it. 

 

And you know, UpTone is no longer the only firm in this space--and there are now plenty of streamer and DAC manufacturers paying attention to the sonic effects of USB SI.

Spoke to the engineers at Aurender two years ago at RMAF (Same show where we chatted with Amir for 15 min. by the elevator--which he later characterized as an in-depth meeting an assessment of John as an "amateur engineer.") The Aurender guys said they quite specifically focused on high SI--for its impact on SQ from the DAC.

 

I think Amir needs to broaden his thinking an methods. 

Say there slick, we're still waiting on the posting of your measurements confirming your claims Amir's are in error.

You can blow smoke telling the folks here that he doesn't know what he's doing or how to measure, but how about throwing up some results backing up your mouth?

Maybe cause last I heard neither you nor John even owned the required equipment?

Your posting nonsense.

 

[Superdad]

Brilliant Sam!! 

 

[EDIT: Looks like Sal inserted himself between this and @scan80269'a post I was complimenting.]

[Superdad]

7 minutes ago, Sal1950 said:

Say there slick, we're still waiting on the posting of your measurements confirming your claims Amir's are in error.

 

No need. Amir's own graph and set up makes the error plain as I originally explained. Two SMPS. Big leakage loop spurie. 

[plissken]

9 minutes ago, Superdad said:

That is an assumption on his part based on his simplistic tests. He focuses strictly on residual noise.

 

He rejects the premise that improved USB signal integrity and impedance match have an impact on DAC function.

 

I think Amir needs to broaden his thinking and methods. 

 

He's focusing on the claims made on your own site. He's done two measurements: Jitter and 60 Hz Mains suppression.

 

When do you think you are going to have some detailed measurements of a DAC's output that isn't what Amir measured with the Schiit Modi 2? 

 

As long as an inspec USB cable (90ohms?) is used there shouldn't be any impedance mismatch. John Swenson has even said a compliant cable at 3 feet or less is going to give you the best S.I. 

 

I don't think it's on Amir to proof your own product however. 

 

Is there anything hes measured so far that you could, in a material manner, disprove?

[plissken]

1 minute ago, Superdad said:

 

No need. Amir's own graph and set up makes the error plain as I originally explained. Two SMPS. Big leakage loop spurie. 

 

No, the only leakage loop spurie is with YOUR SMPS. The USB bus power and his lab power supply don't exhibit this.

 

Is this one of those hidden camera shows?

[scan80269]

2 minutes ago, Superdad said:

 

No need. Amir's own graph and set up makes the error plain as I originally explained. Two SMPS. Big leakage. 

And all Amir needs to do is a simple control experiment: replace the Mean Well SMPS powering his ISO-Regen with a battery and re-measure.  The humps at 60Hz and its harmonics should drop noticeably.

 

[scan80269]

1 minute ago, plissken said:

 

No, the only leakage loop spurie is with YOUR SMPS. The USB bus power and his lab power supply don't exhibit this.

 

Is this one of those hidden camera shows?

Alex can clarify, but I don't think the ISO-Regen was designed to be effective in block AC leakage loops.  To design a firewall against AC leakage loops apparently requires complex designs involving supercapacitors.  For an ISO-Regen to be effective against AC leakage loops it will need to include all the electronics of the LPS-1 supply.  ISO-Regen does implement galvanic (DC) isolation between its upstream and downstream ports, but AC leakage blocking is not covered by this feature.

 

I also don't think Alex will dispute that the Mean Well SMPS included with ISO-Regen can contribute to the formation of an AC leakage loop.  In my example above, the Mean Well SMPS is within the path of that AC leakage loop.  With multiple components involved in forming a specific AC leakage loop, we need to be careful when pointing fingers at just one of the components.

 

[pkane2001]

29 minutes ago, plissken said:

Nothing I've seen or heard so far would convince me it's a problem for most DAC's. The power supply in my Emotiva DC-1 is pretty robust and most likely takes care of the issue. 

 

I'll add my Gustard X20Pro to this list: I see no AC-related frequencies in the DAC output. I do see a slight bump of a few dB over the noise floor at 120Hz, but this is related to my ADC (which is using an SMPS). The reason I know this is that only my ADC is running at 60Hz, the rest of the equipment is powered by a 70Hz AC power.

[plissken]

15 minutes ago, scan80269 said:

Sorry for the long post here, but there are many factors at play that I'd like to comment on.

 

A USB bus powered DAC has an advantage over a self-powered (by DC or AC) DAC in that the former can help eliminate at least one potential AC leakage loop from the system

 

Agreed and in the past I've recommended Intel Atom and Z8*** based systems that could be either driven with a battery or with a linear regulated supply. I also have recommended CMNR supplies. 

 

15 minutes ago, scan80269 said:

ISO Regen by its design offers USB signal regeneration as well as USB bus power regeneration.  Whether these features are beneficial to a particular DAC will heavily depend on how susceptible the DAC design is to upstream USB crap.  Not all DACs in the market today are equally immune to dirty USB from upstream such as from a notebook.

 

Whom besides Amir has given us a glimpse into how DAC's behave on a Laptop? If we want to call that 6 DAC's now that he's measured using the J-Dunn test (John Atkinson uses this test heavily) while not a trend does show a strong indicator that many DAC's deal just fine with the condition of the SI of the USB out on his Laptop. 

15 minutes ago, scan80269 said:

 Within my limited collection of DACs with USB interfaces I have examples of one DAC benefiting tremendously SQ wise with ISO REGEN + LPS-1 while another DAC yields no perceptible SQ improvement given the same treatment.

 

What DAC would that be the shows the improved SQ? It's one of the things that I have asked for multitudinous times of Alex and never received an answer until Amir did the benching he did. 

15 minutes ago, scan80269 said:

 

There are way too many fallacies on both proponents and opponents of USB decrapifiers.  I will list just a few that came into my mind today:

 

Over-generalizations / Over-simplifications:

* Heard no improvement with device added to my DAC ==> device must be worthless
* Heard big improvement with device added to my DAC ==> device must be universally beneficial with any DAC
* A measurement shows no improvement with device added to my DAC ==> device must be worthless
* A measurement shows visible improvement with device added to my DAC ==> device must be universally beneficial with any DAC

 

I know MANSR, BE17, ELSDUDE, myself haven't personally went with those generalizations. It's been understood that there may be some DAC's so poorly implemented as to benefit. It's not cut and dry. 

 

15 minutes ago, scan80269 said:

Fallacies:
* No difference observed in a measurement ==> cannot sound different
* Difference observed in a measurement ==> must sound different
There are plenty of real-world counterexamples to either of these
* Relying solely on measurements to judge the sonic merit of a product
* Relying solely on listening to judge the technical merit of a product
* Underestimating the complexity of a modern digital audio system and the ways SQ can be impacted

 

Here's another: *It can't make a device sound worse. 

 

15 minutes ago, scan80269 said:

Other thoughts:

* Single instrument measurement (with a single graph as output) is far from being an all-encompassing representation of DAC output

* USB de-crapifiers are NOT a panacea

* Sonic benefits of USB de-crapifiers can vary by both DAC and USB source used in the system, with results ranging from big to none

* USB DAC/DDCs are not all equally good at immunizing themselves from upstream USB crap

* Design know-how on effective immunization is relatively recent and not yet widespread (though should become more so with time)

* More USB DACs existing today show vulnerability to upstream USB crap than those that don't (to different extents)

* Galvanic isolation (or lack thereof) is not the only factor impacting SQ of a USB DAC

* Implementation of galvanic isolation can represent a tradeoff, for example jitter aggravation, especially if not carefully designed

* Designing the ability to block AC leakage loops into any device is NOT trivial

 

 

My concerns with Amir's measurements are not so much that they are invalid or improperly done, but rather in how he interprets the measurements and reaches over-generalized conclusions with them.

 

 

 

Amir seems to be open to feedback. 

[scan80269]

18 minutes ago, plissken said:

 

He's focusing on the claims made on your own site. He's done two measurements: Jitter and 60 Hz Mains suppression.

 

When do you think you are going to have some detailed measurements of a DAC's output that isn't what Amir measured with the Schiit Modi 2? 

 

As long as an inspec USB cable (90ohms?) is used there shouldn't be any impedance mismatch. John Swenson has even said a compliant cable at 3 feet or less is going to give you the best S.I. 

 

I don't think it's on Amir to proof your own product however. 

 

Is there anything hes measured so far that you could, in a material manner, disprove?

Regarding jitter, Amir (and/or his followers) have adopted an assumption regarding the total correlation between incoming USB signal jitter and DAC analog output jitter.  Has it occurred to you that some DACs may be capable of complete de-correlation between the two two jitters, i.e. high incoming USB jitter doesn't necessarily translate to high analog output jitter?  Better DAC designs should be able to take a boat load of bad incoming USB jitter and signal integrity (waveform, etc.) and still output excellent analog with no hints of the USB crap coming through.  With such DACs I would not expect the ISO-Regen to yield much tangible SQ benefit, if at all.

 

 

[kumakuma]

27 minutes ago, Sal1950 said:

Say there slick, we're still waiting on the posting of your measurements confirming your claims Amir's are in error.

You can blow smoke telling the folks here that he doesn't know what he's doing or how to measure, but how about throwing up some results backing up your mouth?

Maybe cause last I heard neither you nor John even owned the required equipment?

Your posting nonsense.

 

 

Sal, welcome back from the ASR ghost town! Out of curiosity I checked that place out and it looks like there's about ten active users and one new post every three hours in the forums. I don't think Alex needs to worry much about anything posted over there.  

[plissken]

4 minutes ago, scan80269 said:

Regarding jitter, Amir (and/or his followers) have adopted an assumption regarding the total correlation between incoming USB signal jitter and DAC analog output jitter.  Has it occurred to you that some DACs may be capable of complete de-correlation between the two two jitters, i.e. high incoming USB jitter doesn't necessarily translate to high analog output jitter?  Better DAC designs should be able to take a boat load of bad incoming USB jitter and signal integrity (waveform, etc.) and still output excellent analog with no hints of the USB crap coming through.

 

 

It has occurred to be because Benchmark Audio connected a 100' cable with very high BER and their DAC was entirely stable and they did this years ago. 

 

My suggestion: Get an Ethernet based DAC with an ungodly amount of RAM for buffering. 

[plissken]

Just now, kumakuma said:

 

Sal, welcome back from the ASR ghost town! Out of curiosity I checked that place out and it looks like there's about ten active users and  about one new post every two hours in the forums. I don't think Alex needs to worry much about anything Amir posts. No one's reading it.  

 

They're reading it here....

[scan80269]

2 minutes ago, plissken said:

 

It has occurred to be because Benchmark Audio connected a 100' cable with very high BER and their DAC was entirely stable and they did this years ago. 

 

My suggestion: Get an Ethernet based DAC with an ungodly amount of RAM for buffering. 

In that specific case with Benchmark Audio, do you know if the digital audio data were transmitted over Ethernet using TCP or UDP protocol?  Just curious.

 

I agree on the benefits of massive RAM buffering.  When my Auralic Aries implemented RAM buffering with a major firmware update some time ago there was a step-function jump up in perceived SQ.

 

USB implementations can sound great, but perhaps they are harder or trickier to implement well vs. Ethernet, so we end up with a lot of mediocre or even below-par implementations in USB DACs.

 

[kumakuma]

5 minutes ago, plissken said:

 

They're reading it here....

 

I'm having a hard time keeping everyone straight. Are you one of those guys who has a secret identity over there? 

[Superdad]

1 minute ago, kumakuma said:

 

I'm having a hard time keeping everyone straight. Are you one of those guys who has a secret identity over there? 

 

Yes, plissken is "Jinjuku" at ASR (he is more polite here than there).  Dennis (esldude here) is Blumlein88 there.

[plissken]

4 minutes ago, scan80269 said:

In that specific case with Benchmark Audio, do you know if the digital audio data were transmitted over Ethernet using TCP or UDP protocol?  Just curious.

 

 

Benchmark doesn't have any Ethernet enabled DAC's. 

[Sal1950]

14 minutes ago, kumakuma said:

 

Sal, welcome back from the ASR ghost town! Out of curiosity I checked that place out and it looks like there's about ten active users and one new post every three hours in the forums. I don't think Alex needs to worry much about anything posted over there.  

Alex and friends seem to be.  LOL

Put up some simple measurements and Alex trys to discredit them screaming foul, you don't know what your doing or how to measure correctly, etc , etc.  All this coming from someone who doesn't own the gear, let alone how to even begin to use it.

As I said, nonsense.  LOL

[pkane2001]

5 minutes ago, Superdad said:

 

Yes, plissken is "Jinjuku" at ASR (he is more polite here than there).  Dennis (esldude here) is Blumlein88 there.

 

It's nice to match monikers to their owners on other forums  Provides more insight into how people think and interact with others. 

[Superdad]

16 minutes ago, plissken said:

What DAC would that be the shows the improved SQ? It's one of the things that I have asked for multitudinous times of Alex and never received an answer until 

 

A great many DACs benefit--and not just the cheap ones.  All you need to do is look at the scores of reports in the USB REGEN and ISO REGEN Listening Impressions thread.  People report great success with DACs in the $5,000-$50,000 range.

[kumakuma]

6 minutes ago, Superdad said:

 

Yes, plissken is "Jinjuku" at ASR (he is more polite here than there).  Dennis (esldude here) is Blumlein88 there.

 

Got it. I guess "plissken" wasn't available as a handle when he joined Amir's site.  

[scan80269]

I have two main setups:

 

1. Auralic Aries Femto <> SBooster VBus2 isolator <> USB cable <> Auralic Vega DAC <> XLR cables <> Jensen PI-2XX isolation transformer <> XLR cables <> Auralic Taurus II headphone amp <> Sennheiser HD 800 headphones with Stefan AudioArt Endorphin balanced cable

 

2. NUC6i7KYK fanless PC (running HQPlayer) <> USB cable <> SU-1 DDC (modded) <> HDMI cable <> Holo Audio Spring Level 3 DAC <> XLR cables <> Auralic Taurus II headphone amp <> Sennheiser HD 800 headphones with Stefan AudioArt Endorphin balanced cable

 

#2 has delivered quite audible SQ improvements with ISO-Regen (LPS-1 powered) inserted between PC and SU-1 DDC (most notably in sibilant reduction and upper/mid treble smoothness)

#1 has consistently delivered no perceptible SQ improvements with ISO-Regen (LPS-1 powered) inserted between Aries streamer and Vega DAC

 

Many times when I tried to improve or tweak the Auralic Aries/Vega combo it has not yielded any tangible SQ benefit.  The Sbooster VBus2 isolator delivered slightly less treble harshness, and so did inserting an EMO systems EN-70HD Ethernet isolator at the Aries Ethernet input, but just about every LPS I tried for Aries replacing Auralic's own LPS have been backward steps.

 

[pkane2001]

50 minutes ago, scan80269 said:

My concerns with Amir's measurements are not so much that they are invalid or improperly done, but rather in how he interprets the measurements and reaches over-generalized conclusions with them.

 

Scan, I respect your knowledge and expertise. I don't necessarily agree with Amir's conclusions, either, but by doing the initial measurements and sharing them with others he has began to advance our knowledge and understanding, well beyond what was previously available -- mostly marketing generalizations and noise. Other measurements and tests will come with time, but at least for me, his measurements and the discussion they engendered have already helped my understanding quite a bit.