Review and Measurements of PS Audio PerfectWave DirectStream DAC

[amirm]

Ted goes on to say:

I fully understand JA's measurements. This is my core area of expertise. Unfortunately it doesn't seem to be Ted's.

Most of JA's review is with regards to frequency response and such which I did not find fault with. I said the frequency response was flat and fine.

Past that, JA measures the response of the DS DAC trying to reproduce a -120 dB tone and fails to do so:

That is perfectly inline with my linearity measurements which show noise taking over at -120 dB:

All the way to the left -120 dB. We see that the response is off the chart showing nothing but noise.

Ted should note that my test uses very sharp filters so that it only allows the test tone at 1 kHz to get through. Despite that, the DAC could not produce a usable output even at that narrow response which got rid of 99% of the noise output of the DAC. JA does the same using long FFT measurement.

This is how an excellent but $99 DAC responds to the same test:

Let me repeat: this is a $99 DAC.

 

[AndrewDavis]

Oh Amir! Don't you know facts and measurements are just objectivism!? And that is bad! Facts and measurements don't capture all the facts...er....Well, anyway....I believe my ears!

What is actually right is how the music makes my emotions feel at that exact moment. You and your decades of experience and carefully gathered data can't convince me otherwise!!!!!

 

[bigjacko]

This is the first time I have seen that amirm has confidently saying there is a difference in audibility. I want to ask what is the limit of our hearing, and can we hear further even lower distortion on top of headphone distortion. By the way, what distortion number can you guys confidently tell that it definitely has distortion?

 

[amirm]

Ted finishes with:

What does he mean they are never full scale? He has not heard of loudness wars? Or the fact that bass signals can have very high amplitudes? One buys a US $6,000 DAC but has to be careful to not run full amplitude music through it? How am I going to play that content? Buy another DAC?

While JA did fail to run full amplitude 1 kHz through the DAC, he did run one at 50 Hz:

0 dBFS is absolutely full level. And produces horrendous distortions, consistent with my testing.

All of this is beside the point that I clearly showed the issues with this DAC being due to use of output transformer. They distort. Everyone knows they distort. And that is what I found. How cout that distortion not be correct to find in measurements and to hear in person?

Very disappointing to hear Ted says he has no intention to investigate. And it is not like he has done anything with JA's test either which was back in 2014. Same outcome today out of a brand new PS Audio PerfectWave DirectStream DAC.

 

[FrantzM]

Yup, the 62 db noise floor is the dead giveaway! In an earlier life coulda been a Goldmund Reference TT.

At least this TT measurements were amongst the best of its kind. This BS Audio seats near the bottom of DAC performance.

 

[HammerSandwich]

The number of super high end DACs devices shown to have simply flawed engineer ing is amazing.

Still true!

 

[Icboschert]

The problem is that there isn't a $5,000 power conditioner attached to the DAC

 

[typericey]

Any reason why PS Audio spec'd the balanced output to be 2.8V and the standard not 4V? Could the lower voltage be contributing to the poor measured performance?

 

[HammerSandwich]

If you can't hear such artifacts, I suggest learning to be more of a critical listener buying affordable, transparent gear, spending the difference on concerts & recordings, and learning not to obsess about hardware. It will do you some even more good.

This message brought to you by Psychological Science Review.

 

[HammerSandwich]

This thread has been quite interesting! My least expected observations:

1. That a $6k DAC's designer would argue, "It's much better than the measurements show, but you can't tell because it's so noisy."
2. That Art Dudley's reviews may be relevant & reliable.

Nope - didn't see either of those coming.

 

[audimus]

Ted’s argument reminds me of the arguments from climate-change deniers trying to use “science” to argue. Using terms that sounds like science but with no understanding of what they mean.

 

[FrantzM]

The problem is that there isn't a $5,000 power conditioner attached to the DAC

True that!

On top of that Amir did use some cheap interconnects as he is wont to do. At least he could have used the BS Audio AC 12 power cable.

 

[amirm]

This is the first time I have seen that amirm has confidently saying there is a difference in audibility. I want to ask what is the limit of our hearing, and can we hear further even lower distortion on top of headphone distortion. By the way, what distortion number can you guys confidently tell that it definitely has distortion?

Let me point out that in this case, the dashboard shows the best case performance of this DAC since it measures at 1 kHz. At other frequencies, distortion rises substantially, making its audibility a simpler task. It still requires careful listening and proper material. And ability to AB against another DAC without such distortions.

 

[Thomas savage]

It's funny how consistent the response is to these reviews, that is when they show up issues like this one has ..

Both the indoctrinated fan base of the products and those who make them seem to come out with quite flawed and illogical defenses .. as if they are suffering the effects of being in some kind of bubble , like those secular communities that believe in all sorts of nonsense and enthusiastically follow their leader with unshakable belief... Cults I think they call them.

It is as if amir is exposing cults , the cult of high end audio. I think they give the Scientology crowd a run for their crazy bucks.

This is the problem with self affirming systems that ignore objective/scientific inquiry, both the customer base and indeed the companies themselves are entirely without objective reasoning . It's created a huge mess that's horrendously over valued/priced and really needs tearing down.

 

[audimus]

Let me point out that in this case, the dashboard shows the best case performance of this DAC since it measures at 1 kHz. At other frequencies, distortion rises substantially, making its audibility a simpler task. It still requires careful listening and proper material. And ability to AB against another DAC without such distortions.

Or being aware of the poor measurement prior created an audibility expectation bias towards the negative that was “confirmed” by listening. Next time, perhaps do the audibility test first before measuring? The very fact that you selected material that you thought would likely expose the issues meant you already had an expectation and confirmation would be a validation of your measurement, a positive reinforcement.

Sorry, you cannot have it both ways. Dismiss any claims of hearing a difference without a controlled, unsighted test but make a similar claim that isn’t free of expectation biases.

 

[HammerSandwich]

It is as if amir is exposing cults , the cult of high end audio. I think they give the Scientology crowd a run for they crazy bucks.

OMG, that's IT!!!

The most common argument against DBT is that it's unlike normal listening - too stressful for valid conclusions, etc. We need to remove the stress. How about if you just listen? No need to decide if X is A or B. Just listen.

But hold these handles while you do, and our E-meter will tell us if you can hear a difference.

 

[amirm]

Or being aware of the poor measurement prior created an audibility expectation bias towards the negative that was “confirmed” by listening. Next time, perhaps do the audibility test first before measuring?

No. My testing was blind. You can have all the knowledge of measurements and still can't game the system.

Knowledge of measurements allows one to find critical content that exposes weaknesses that are case specific. Here, knowing that low frequencies distort allowed me to fine tune the listening tests. Such approach is highly encouraged in standards such as ITU BS1116 and in the industry and research in general. To wit, we always test audio lossy DACs with content that we know cause them grief.

Otherwise, imagine trying to be lucky and find the right tracks out of 50 million available.

Our goal in any audibility testing is to get positive outcome. That way, the conclusions are affirmative and solid. Negative outcomes are not as conclusive.

 

[Thomas savage]

Or being aware of the poor measurement prior created an audibility expectation bias towards the negative that was “confirmed” by listening. Next time, perhaps do the audibility test first before measuring? The very fact that you selected material that you thought would likely expose the issues meant you already had an expectation and confirmation would be a validation of your measurement, a positive reinforcement.

Sorry, you cannot have it both ways. Dismiss any claims of hearing a difference without a controlled, unsighted test but make a similar claim that isn’t free of expectation biases.

In these cases where audible issues are suggested it would be really good to have a blind test performed against a reference product, a DAC in this instance.

 

[Cahudson42]

Thanks Amir! If you have saved only one reader from throwing $6000 down the toilet, it's worth all the flak. Objectivism is never going to cut it with an 'alternative facts' crowd anyway..

May I also suggest a new award? The 'Shattered Panther'. Maybe a single paw left. Or just a bit of a tail. And a pile of rubble. You really do need something better than the 'Decapitated Panther' for this one...

 

[bunkbail]

Let me point out that in this case, the dashboard shows the best case performance of this DAC since it measures at 1 kHz.

Can you please do a quick FFT at 50Hz like JA did, just for peace of mind.