Audioquest Pearl USB Cable Review

[Curvature]

This is also diametrically opposed to the notion that your other stuff needs to be sufficiently "resolving" to hear improvements. So either the gear needs to be broken or exemplary.

Exactly.

 

[GXAlan]

But isn't the right question how to identify poor devices, not how to fix the problems they introduce? A correctly designed device is very tolerant. A poor one requires special upstream and downstream care.

Agree 100%. I have spent a lot of money on restoring vintage gear and one thing I don’t do is do silly things like add IEC AC power cable option or upgrade binding posts. I’ll leave the hardwired lamp cord.

I think I ask that question more out of interest/curiosity than in the context of purchasing decisions. Behind every legend is a nugget of truth while myths are completely fictional. I have to wonder, in the 80’s when fancy speaker cable became a thing, were the devices so poorly designed that there in fact were audible differences as amplifiers went into oscillation or were heavily load dependent? When transistors replaced tubes, how did the transistors of the era perform? Did any of them have “solid state glare” that we can measure?

People have long talked about the warmth of the Marantz sound signature and that Denon sounds different. Here at ASR, we can see that the digital filter and HDAM modules result in worse performance which potentially means that people were right when they said they heard differences. It just so happened that it was in the direction of less transparency. But at least you can say that we can measure and generate an explanation beyond mass delusion.

 

[KMO]

It is not a big difference. USB interference is situation dependent and comes and goes. I will update the review thread when I get home. Note the well designed usb implementations don't have this issue. They always work right.

So do you have any sort of methodology to look for interference "coming and going"? Could the Mojo 2 have instead had a good day like this one and passed USB with flying colours in its test?

Or was the interference visibly coming and going during both tests, and you just chose to publish the bad state in the DAC test, and the good state in the cable test?

 

[amirm]

Or was the interference visibly coming and going during both tests, and you just chose to publish the bad state in the DAC test, and the good state in the cable test?

And motivation for that would have been what exactly????

But no. Whenever a dac shows this kind of problem, I try many methods to reduce or eliminate it. What you saw in the review was my best effort. It just happened that during cable test the problem was not there. Which was a good thing as it made the comparison so much more reliable.

 

[TabCam]

It is not a big difference. USB interference is situation dependent and comes and goes. I will update the review thread when I get home. Note the well designed usb implementations don't have this issue. They always work right.

I would say that 9 dB in SINAD for one channel and 5 in the other channel is quite a difference. Also, the intended use for the Chord Hugo is mobile with occasional desktop usage. The Chord Qutest Is the desktop version in an even cheaper price range.

Would the measurement error also have an effect on the linearity? Thanks for updating the review in advance and most of all for the time and effort in this site!

 

[amirm]

So do you have any sort of methodology to look for interference "coming and going"? Could the Mojo 2 have instead had a good day like this one and passed USB with flying colours in its test?

Once again, large number of DACs i test don't have this variability. To the extent they show it at any frequency, it is a problem. Since you all don't have the instrumentation I have, best to stay with products that are solid as a rock here.

 

[Thomas_A]

Since grounding of the unit made things worse, the internittent noise is probably a ground problem. If there is no problem with the connection/connector of the cable, it is a unit issue. I had intermittent noise issues using a Voltex surge protector in the system. Using a standard connection the noise disappeared.

 

[amirm]

Would the measurement error also have an effect on the linearity?

No. All the measurements were done using TOSLINK which didn't have this issue. Here is linearity:

look at top right.

 

[KMO]

And motivation for that would have been what exactly????

Well, it would make sense. Presumably the "clean" state was more stable and hence be better at showing cable difference or lack thereof.

Whereas when reviewing the DAC, you don't want to be giving the false impression that it can maintain the clean state. You did mention that it wasn't particularly stable.

 

[KMO]

Since you all don't have the instrumentation I have, best to stay with products that are solid as a rock here.

I'm not sure you grasped the point of the question. You spotted that the Mojo 2 wasn't as solid as a rock during its test, but was there a chance you could have missed it, due to it having a good interference day? Could it have maintained this test's performance for the entire DAC suite without you realising it wasn't solid?

Can we actually tell which products are as solid as a rock, rather than just how they happen to perform on that test day?

 

[HarmonicTHD]

I'm not sure you grasped the point of the question. You spotted that the Mojo 2 wasn't as solid as a rock during its test, but was there a chance you could have missed it, due to it having a good interference day? Could it have maintained this test's performance for the entire DAC suite without you realising it wasn't solid?

Can we actually tell which products are as solid as a rock, rather than just how they happen to perform on that test day?

Personally I would weed out any „unstable“ product which didn’t perform on test day. Because as a consumer how can I be sure at what day it might perform and and what day it might not.

More interesting are the „false positives“, that is the products which perform on testing day but are actually a dud. Oh well that risk is always there as one might al always get a unit as a consumer which is different from the tested one. I guess I am willing to take the chance and if the product is clearly at fault then just return it.

 

[AudioSceptic]

I looked at that tester. It is a ton of money for little benefit in our situation.

What would be the point anyway? What matters is the *audio* output of devices connected using it, and that is what it's being sold for in the first place, not as a data connection. Audio data is such low bit-rate that it could, for example, fail to reach USB 2 spec but still be fine for audio.

(I see that Sam Adams has said much the same.)

 

[AudioSceptic]

This is also diametrically opposed to the notion that your other stuff needs to be sufficiently "resolving" to hear improvements. So either the gear needs to be broken or exemplary.

So? This is audiophool "logic" we're talking about here, IOW magic.

 

[AudioSceptic]

Once again, large number of DACs i test don't have this variability. To the extent they show it at any frequency, it is a problem. Since you all don't have the instrumentation I have, best to stay with products that are solid as a rock here.

I'd be concerned that, if variability exists at all, it might get a lot worse at any time.

 

[AudioSceptic]

I'm not sure you grasped the point of the question. You spotted that the Mojo 2 wasn't as solid as a rock during its test, but was there a chance you could have missed it, due to it having a good interference day? Could it have maintained this test's performance for the entire DAC suite without you realising it wasn't solid?

Can we actually tell which products are as solid as a rock, rather than just how they happen to perform on that test day?

What do you suggest? That Amir runs multiple test suites at randomly chosen intervals over several days? That he'd be able to test only a fraction of the products he currently does?

 

[Atanasi]

The ground noise issue with Hugo 2 seems similar to the review of E1DA 9038D.
IVX recommended to always use the balanced input of AP, which provides better ground isolation and can accept unbalanced signals.

Ok, guys it seems I've found the reason for always worse Amir's tests for unbalanced DACs and better for balanced. The problem is Amir uses unbalanced APx555 inputs for unbalanced DACs. Only balanced inputs of APx555 have good enough CMRR to measure DACs or amps whatever, and actually, no reason at all to use unbalanced inputs ever(just forget about them) - only balanced XLRs.
Dear amirm, Unbalanced BNC inputs tied to the APx555 chassis and could not be used for non-isolated sources like AC-line powered PC, if you like to use that BNCs for any unknown reason(because any even 30 years old AP has excellent balanced XLRs), please use it with battery-powered sources. Balanced XLRs inputs are referenced to each other but not to the chassis intentionally to let you measure precisely sources with strong common-mode noise(XLR+/XLR-/GND-chassis-AP to GND of your PC, for instance, to USB-C shield). I can't and I don't push you to remeasure my DAC but for the future of the audioSCIENCEreview.com better if you'll correct your setup.

 

[KMO]

What do you suggest? That Amir runs multiple test suites at randomly chosen intervals over several days? That he'd be able to test only a fraction of the products he currently does?

No idea - depends what exactly the nature of the problem is.

He says "USB interference is situation dependent and comes and goes. [...] Note the well designed usb implementations don't have this issue."

So this is a generic problem? Does he have a way to detect those well-designed USB implementations? If not, doesn't he need one?

Or should there be a general disclaimer that we shouldn't put too much store into comparing very-close USB results because its performance isn't generally consistent? (Unlike TOSlink?) We should accept that there's a +/-6dB error in USB SINAD measurement - depending on whether a poorly-designed USB implementation was having a good or bad day during the test?

 

[Thomas_A]

I always ran on battery powered mac for my simple measurements but when I recently did it for my NR1710 AVR I forgot the ususal issues with USB.

A few measurements of the Marantz NR1710 slim line AV amp (update jan 2024, power amp section)

An update of measurements in post 16. I just made some simple measurements via pre out of my slim line Marantz NR1710. Not having much advanced equipment but my 18 year old M-Audio Transit DAC and ADC was used to test it in various ways. It only supports 44,1/16 bit when used both ways...

www.audiosciencereview.com

 

[Geert]

He says "USB interference is situation dependent and comes and goes. [...] Note the well designed usb implementations don't have this issue." So this is a generic problem?

It is a generic problem, for DAC USB implementations that are sensitive to ground loops. Other (better) designs are less sensitive for this problem.

Does he have a way to detect those well-designed USB implementations?

Yes, they don't show noise issues when connected via USB

Or should there be a general disclaimer that we shouldn't put too much store into comparing very-close USB results because its performance isn't generally consistent?

Note that it's not just a problem with the measurements. With a device that's sensitive to ground issues an end user might experience problems during daily use.

We should accept that there's a +/-6dB error in USB SINAD measurement - depending on whether a poorly-designed USB implementation was having a good or bad day during the test?

Of course it would be better if the root cause was identified but that kind of troubleshooting takes time. But it would be unfortunate if bad designs slip through the measurement protocol if they're lucky enough.

 

[amirm]

The ground noise issue with Hugo 2 seems similar to the review of E1DA 9038D.
IVX recommended to always use the balanced input of AP, which provides better ground isolation and can accept unbalanced signals.

But that setup would be contrived and not represent how the product gets used by you all.