Group Test: USB gadgets (AQ Jitterbug, Uptone Regen, iFi iUSB3.0, iPurifier, iPurifier2, UPDATE on p.15…)

[mmerrill99]

Dm, I sent you a PM

[mmerrill99]

They don't reflect mine.

 

Two problems with this.

 

Firstly the data is supplied (I have no doubt in good faith) as a direct indication of the products worth and effectiveness. The implied statement is that a quiet power supply is beneficial. To have any credibility the data needs to be good. It looks dodgy at best.

I agree - if you are presenting data as a means of comparison between devices then that data needs to be as accurate as possible.

 

Secondly, I'm sorry to say that your ears are a poor judge of technical quality. Without objective data to back it up you are just choosing what you subjectively prefer, and / or are biased to be pre-disposed to.

 

Nothing wrong with this of course, but don't confuse it with actual technical efficacy or actual improved sound quality.

Don't agree. What you hear & evaluate is the judge. Yes we can be swayed by bias & pre-disposition, just as easily as we can be swayed by measurement bias & pre-disposition. We are not talking about 'technical' quality as this becomes a meaningless term when we don't know what measurements equate to good sound.

 

So we have to live with the fact that we are using our auditory processing mechanism for judgement & evaluation of what we here in the knowledge that it is imprecise & flawed but we really don't have any choice!

[mmerrill99]

Then you dont understand what I said.

 

Measurements are what they are, they arent subjective. They are just factual information.

Sorry but "factual" is a misleading term. The measurements chosen to be performed on a device are chosen from a limited set of ossible measurements & are therefore very much subject to editorialising by the tester. Furthermore, the measurement results are 'interpreted' by said tester (often others have different 'interpretations) & therefore also subjective in nature. So, I would state that the term "factual" is a very overused one & needs to be qualified as I've begun to do above

Feel free to have personal subjective opinions, its fine, it really is. There are not however necessarily representative of technical or audio quality. They are meaningless beyond the individual. They really are only what you like.

I find that when I find what I hear in agreement with a sufficiently large enough group of people, I consider that what I'm hearing is not "meaningless beyond the individual"

 

Your couching of measurements as "factual" & hearing as "meaningless beyond the individual" is the usual distortion seen on audio forums.

 

Both approaches contain the same balance of wrong & right as each other - neither is superior & both act as useful guides but the strengths & weknesses of both have to be always borne in mind.

[mmerrill99]

The problem with the listening reports & comparisons of these devices is disentangling the 5V power supply improvement from the improvement wrought by isolation & regeneration/reclocking.

I'm not exactly sure what DAC these devices are being tested with but it seems to require 5V Vbus input. It would be good to also use a DAC which is self powered to evaluate the isolation/reclocking aspect of these devices

[mmerrill99]

I don't think this is correct? On both my Chord Hugo and not Ifi Nano iDSD LE dacs, neither require a 5Vbus throughput for handshake with PC. Ground, definitely.

 

Yes, just ground is needed.

5V Vbus if not used for powering the DAC (bad idea) is often used for sensing USB has been connected (not such a good idea, either but it is in the USB standards so any USB device not adhering to this can be said to be non-compliant with USB standards)

[mmerrill99]

Hi Bogi,

I don't disagree that many DAC's need the 5Vbus, but none (that I know of) need it for handshake. Thus the 5Vbus does not need to be continuous throughput.

I'm afraid that's not correct - the XMOS USB device senses the Vbus 5V on one of it's pins (as the connected signal) & doesn't operate without this signal. This can be provided by the 5V from a self-powered device but this is not the USB specified standard & means that if a computer goes to sleep, on wake-up it's highly likely that the USB audio device operating in this way will not actually enumerate with the PC & the USB audio device and/or the PC have to be rebooted in order to establish connections again.

 

You are correct that it's not a handshake in the strict sense of a messaging protocol handshake but it is an electrical protocol, as such.

[mmerrill99]

Yes, some DACs function with USB cables, where VBUS is broken.

Although it is nonstandard USB implementation, in the case of DACs it is seen rather an advantage.

 

Yep, I agree - it can be beneficial to the sound. Why? Probably got to do with the nature of the signals that are being carried in the USB cable.

 

BTW, I find it hilarious that what Plissken would define as "competently designed" USB DACs (because he thinks having no 5V Vbus makes them "beyond USB improvement") includes these nonstandard USB implementations. Just shows how lost he is - chasing his tail around in ever decreasing circles - where his definition of "competent" now means breaking the standards, lol.

[mmerrill99]

My Auralic Vega is used for listening impressions. Does not use 5V.

Yea, thanks but not sure this is the case with DM's setup - his seems to need 5v & he's the one doing the tests & comparisons.

 

All I'm saying is that the influence of the 5v on sound quality needs to be disentangled from the influence of the isolation & rocking aspects in order to be able to judge comparative performance.

[mmerrill99]

I have both. The DP-777 doesn't use USB 5V, but most of my portable USB DACs did use the USB 5V.

Ok, thanks

For all the USB DACs out there, may be >90% will use USB 5V, so "USB Power" is one of the main factors and can't be ignored in real life.

Really, I'm not sure about that >90%?

Here's something to think about - is a USB DAC that uses Vbus as it's power source a good design choice for optimal sound? Great for portability but not so much for great sound. If the objective of these USB gadgets is best sound then maybe using USB DACs which are powered by Vbus is not a wise choice?

[mmerrill99]

I would have agreed, before hand, about the preference for not powering a DAC via the Vbus, but I have to admit that I am quite impressed, if not surprised, with the quality of sound from the ifi nano idsd LE by powering it by the Vbus. It's definitely better than the battery, although I use a LPS-1 via Regen to power the ifi.

I'm not sure I follow - Vbus powering iFi nano is better than what battery power?

In fact, one could make the argument that it may be better to have a DAC that is powered via the USB Vbus, for it would be easy then to power with the galvanic isolated supply Uptone LPS-1.

Again, I don't get this - feeding a cleaner supply into where Vbus once ran is not the best design choice - poor wiring, poor connections, non-optimal routing, etc.

What could be lost in SQ because of the proximity of the clean 5Vbus to the data? I still break the 5Vbus, by eliminating it in the USB cable between the Intona and Regen.

Yes, running a VBus wire alongside a twisted differential signal pair IMO, is not a good design choice from a the signal integrity viewpoint - ripple noise on vbus will not couple into both members of the differential pair equally hence common mode noise rejection will not be as effective.

It does trump the battery supply only power.

What battery supply?

[mmerrill99]

The battery supply built into the ifi. Can run on battery or 5Vbus only. Not saying the battery supply design is optimum on the ifi??? But the 5Vbus power via LPS-1 provides better SQ than the battery supply only on the ifi.

Right, battery in the iFi is not as good as supercapacitor power supplied via LPS-1. I find it confusing to call this 5Vbus as it has no relation to VBus, only uses the same wiring?

I agree, the design choice of power supply via 5Vbus, we would think would not be of optimum choice, but I gotta say, on the ifi at least, it works very well.

Sure!

[mmerrill99]

5Vbus and VBus are one and the same. I just like to distinguish that you have the spec USB 2.0 5V as power for the VBus.

Hold on - the way I thought you were using the term 5Vbus was to describe how power was being delivered from the LPS-1 along the Vbus wire or Vbus pin in the USB plug. This is very different to the VBus power that comes from a normal USB socket. This is all I meant & why I found it confusing a term.

 

But maybe you are using it as a term to describe the cleaned up VBus that is done inside the iFi - I'm not sure - I haven'tlooked into the iFi nano?

I suspect that the effect of the Vbus 5V power has a negative effect on SQ in the ifi when operating under battery power, presumable it is still trickle charging the battery. Really to test properly the battery of the ifi as opposed to the LPS-1/Regen/Vbus power, one would have to cut off the Vbus so as to not influence the sound when being powered by the battery. Then the difference wouldn't probably be so much. But what a pain in the ass to have to do every time you want to use the DAC in desktop application.

[mmerrill99]

@DAW1d: Your trouble doesn't seem to be about USB noise. If it would be then you would hear it through headphones too. Look for a solution to avoid ground loop - probably between your speaker amplifier on some other audio component. That's other topic than this thread is discussing.

I would disagree, Daw1d's issue would appear to me to be ground related - his headphones have no connection to mains ground whereas his amplifier/speakers would have. I suspect that it is leakage current noise that he is perceiving.

[mmerrill99]

Evaluating sound performance through listening requires effective methodology and adequate music.

Otherwise you're just "tasting" which is what a lot of audiophiles and reviewers do.

 

R

 

Funny but I judge food & meals in exactly this way - by tasting & am not interested in the chemical makeup of the food - it tells me nothing about how it will taste to me. I'm not interested in an "effective methodology" as it takes the life & fun out of what should be a pleasurable & enjoyable pastime.

 

Does my mood & the company & chat during a meal influence my judgement? Sure it does. If I go to the same restaurant many times with different people & in different moods does this lead me to a general idea of whether I like the food or not.

 

I find this works with many choices that I make. Can I get it wrong? Sure, but it's life & I try to enjoy the rollercoaster ride.