• WANTED: Happy members who like to discuss audio and other topics related to our interest. Desire to learn and share knowledge of science required. There are many reviews of audio hardware and expert members to help answer your questions. Click here to have your audio equipment measured for free!

Review and Measurements of PS Audio PerfectWave DirectStream DAC

No problem. But this is high school science - i.e. the importance of the experiment procedure. The full details of setup and procedure can sometimes explain results and conclusions.

So without the full details of setup and procedure, there is always this question mark. As mentioned, the devil is usually in the details.



Amir already said he cannot write a bible for every review and that is 100% understandable.

It is POSSIBLE that if Amir included details and photos of his setup of the DirectStream DAC measurements that it's designer could pick holes in his testing setup and method... in the same way holes have been shown by Amir's measurements in the DAC's performance...

And now imagine this was done for every DAC Amir has measured?
In this case Amir's results equal or exceed claims of the company making it. What is there to pick holes in?
 
It is POSSIBLE that if Amir included details and photos of his setup of the DirectStream DAC measurements that it's designer could pick holes in his testing setup and method... in the same way holes have been shown by Amir's measurements in the DAC's performance...

Well, the designer could obviously post his own test, with equivalent hardware and methodology, and explain (for example) that the tested sample is defective, or give specific instructions for Amir to replicate the results... Apparently, he has chosen not to do that, saying he did not want to get in a war of ideas, which strongly indicates that he doesn't have his own set of measurements to oppose to the current set. If I had designed and measured a device that performed poorly on Amir's tests and well on mine, I would certainly discuss the issue in private with Amir, providing my measurements and addressing potential issues.

Strangely, it appears that it almost never happens with those high-end boutique manufacturers...
 
give specific instructions for Amir to replicate the results... Apparently, he has chosen not to do that, saying he did not want to get in a war of ideas

Yes but don't forget (as I already mentioned) Amir said he doesn't want to write a bible for every review...

And without full details and photos of Amir's setup and procedure, it's similar in both ways...

I am not taking anyone's side. But it's important to see the overall picture...

Amir could share a lot more. Some manufacturers could share a lot more.

They both have their reasons...
 
They both have their reasons...

Yes, Amir shares test results for free and he is open to corrections (several already by very qualified members of this board) and yes, high-end manufacturers want to keep charging n times k$ for the best sound in the world. :) Different set of reasons indeed :)
 
This leads to a situation where I don't know what to believe anymore. Because I don't know all these technical stuff.

<snip>

Because then we have to choose whose measurements we believe. This is frustrating for someone with little knowledge.

This is a common problem because it is impossible to know everything about everything. So what can one do?

1. Consider the source of the data. Do they appear to have access to the equipment and measurement gear? Do they appear to know what they are talking about? Do they have academic or professional qualifications or related job experience to talk about the issues they do? Do they have a financial stake in the outcome or a review? Are they employed by the gear manufacturer or competitors? What do their motives appear to be?

2. Test the data they provide. Do the specs they quote appear to conveniently match data provided by the manufacturer? Do they use phrases that match ad copy or press releases? Is the data wildly optimistic/pessimistic or does it lie somewhere in the middle of the normal range of $hit and SOTA? If it is beyond SOTA, has a specific breakthrough been achieved? What are peer reviewed journal articles (AES) saying? Is there "hard data" or just impressions/subjective opinions being expressed?

3. Consider the format. Is the data being published on a site that features ads from the manufacturer? Are all reviews on the site positive or are there some critical pieces as well? How does this align with advertisers?

4. Consider the quality of dissenting views... and how they are handled. Are the challenges to the reviews serious (ie. matching in technical quality) or frivolous? Are responses to challenges dealt with head-on or does the reviewer attempt to shift the subject, assign blame, ban the critic, etc?

There are many more dimensions to this, but these will develop as you become more comfortable with questioning the qualities of the source, the data, the publisher, and the responses to reviews.
 
Last edited:
Amir shares test results for free and he is open to corrections

The number of corrections possible will always be limited, without full details of setup & procedure. So we go in a loop :)

Cannot dismiss the importance of full details of setup & procedure, even if Amir does this all for free... as I mentioned, as per high school science experiments, the proecdure can explain a lot about results and conclusions...

Same applies to manufacturers of course.
 
Last edited:
I owned one of these for over a year and I hated the sound. It was very bright sounding. I spoke with the company and they kept telling me it needed more "burn in" time. Load of crapola. In the end, I had to get ride of it.
Why did you buy it in the first place? Reviews? You can't have chosen it in a demo, surely?
 
The number of corrections possible will always be limited, without full details of setup & procedure. So we go in a loop :)

Cannot dismiss the importance of full details of setup & procedure, even if Amir does this all for free... as I mentioned, as per high school science experiments, the proecdure can explain a lot about results...

I have had zero problem replicating his setup and results. Everything needed is in the review, and the project files he freely shares have all the setup details.

Replicability is a total non-issue.
 
Amir could share a lot more. Some manufacturers could share a lot more.

They both have their reasons...
Where are the settings for Stereophile tests? JA shares far less than I do. You have no channel to even talk to him about what he measured let alone have him measure anything else.

As for manufacturers, they could do a lot more? You don't say. They are doing absolutely NOTHING. Most of the gear I measure has no specs whatsoever. Or has specs for chip, or some other vague number here and there. If you genuinely believe they should do more, then you need to direct your ire at them, rather me when I am the one publishing 100 times more data than they are.

Why don't you come out and say what is bugging you? Clearly you are trying to criticise my work beyond the merits. Major companies like Benchmark who do their own measurements have no issues whatsoever with measurements I perform and level of documentation provided. But somehow you feel that is enough. Show some transparency here as it is sorely needed after reading your continued critical commentary.
 
Where are the settings for Stereophile tests? JA shares far less than I do. You have no channel to even talk to him about what he measured let alone have him measure anything else.

100% agreed....


As for manufacturers, they could do a lot more? You don't say. They are doing absolutely NOTHING.

I did say... I agree (obviously)

Why don't you come out and say what is bugging you? Clearly you are trying to criticise my work beyond the merits. Major companies like Benchmark who do their own measurements have no issues whatsoever with measurements I perform and level of documentation provided. But somehow you feel that is enough. Show some transparency here as it is sorely needed after reading your continued critical commentary.

You seem to be taking my "the devil is in the details" (regarding full transparency of details of setup and procedure) personally?

I would think (hope?) you agree with the statement?

I am NOT saying your setup and procedure IS wrong. If you think I am saying this, you have imagined it...

You missed my post yesterday where I said "I like this site..."

Not sure why so moody today :)
 
Not only does JA share too little on tests, I've caught him *numerous* times over the years just flat out *lying*, seemingly at the behest of a treasured, very profitable advertiser. This is an old one, but happened to come up in emails with a client recently, from JA's review of Levinson 31.5 transport:

"And even more important with respect to sound quality, the transport has changed conceptually. That master oscillator controls not the transport mechanism but the output stage, a flip-flop chip that could be considered a 1-bit FIFO. The output data thus have the maximum timing precision, which is where it is needed. The mechanical drive is slaved to the output stage and therefore needs to be able to respond to its demands, which is why a 2x drive mechanism, with its inherently fast response, comes in useful. (This topology, referred to by Madrigal as Closed-Loop Jitter Reduction, or CLJR, was developed for the less expensive No.37 and Proceed CDD transports.)"

No way is JA this ignorant of the techno garbage there, and it's awfully close to the babble spouted by Harman's marketing people, trying desperately to rationalise their inexcusable choice to switch from CDM4 to the criminally horrible chinese-made CDM12 series garbage laser/mech, and also to avoid having to admit that they had very deliberately ripped off Ed Meitner's C-Lock-T reclocking circuitry to avoid infringement suits. I've seen original schematics for both systems, and Harman/ML should seriously have paid a license fee to Ed.
 
Major companies like Benchmark who do their own measurements have no issues whatsoever with measurements I perform and level of documentation provided.

Just out of interest, Benchmark seem to like your AHB2 measurements (stellar performance).

But does @John_Siau agree with all of your DAC3 measurement results?
 
It says "EE/CS"...

4 years, if you start at start of 76 and finish end of 79...




What makes you think this? Something in the LinkedIn that you screenshot?

Amirm is right. Ted Smith is not listed in the official MIT alumni directory.

Actually, it's even questionable as to whether Ted Smith ever attended MIT since the directory lists students who did not finish their degrees.
 
Last edited:
His preamps are used to handle a single microphone for a single instrument or voice. They act as sound processors, as most dedicated microphone preamps for studio use. If the mixing engineer thinks that something sounds better when using such a preamp he uses it - and I'm fine with that. Hey, even the mastering engineer does process the sound.

But when the finished product is delivered it should not be necessary to process it more, other than for room correction. And if it is necessary I prefer a niceness knob in the preamp to add whatever I like, since it depends on the recording.

Yes, I agree, and I've had the opportunity to use some Neve equipment but, like you, I want my audio interfaces to be transparent. Also, the same studios that use his preamps and EQ modules have them racked alongside converters renowned for transparency from the likes of Apogee, Prism Sound, RME, etc.

But Neve's commentary in that video struck me as similar to the language that boutique equipment designers use. I think the big difference is that he knows exactly what he is doing and why he wants that sound. He obviously uses measurements to back up what he is hearing and he is quite open about the reasons for his design choices. Nevertheless, Neve is a cult and many a follower will have a strong psychological investment in his designs.

I think that designers who depart from accepted engineering goals (which we can easily measure with something like the AP) need to justify why they have made those decisions and not just pass them off as "near-perfect analog" or "perfection based conversion that uncovers all the missing information hiding in your digital audio media" which are claims made for the DirectStream DAC. Perfection seems to be a slippery concept . . .
 
That worries me.
I have come here because I wanted another tool in search of best value/performance gear.
Me personally doesn't know half what is written here. But I used his measurements to know how good a product is. Just like magazines use words to describe how good a product is.
But now I read on this other site that maybe the measurements are not done in the right way. This leads to a situation where I don't know what to believe anymore. Because I don't know all these technical stuff.
So it would be great if Amir would address these points.
Now it seems like there can be more different measurements taken and that the analyzer was not calibrated right. And we are again back at the start. Because then we have to choose whose measurements we believe. This is frustrating for someone with little knowledge.

Keep in mind, that Sbaf is heavily biased toward the outdated measure by hearing method and old audiofool claims, probably hard to accept the sad truth, that most of their precious gear is just junk. Of course they flame amirm, it is easier to face the facts and move on. Most of them thinking that they have better hearing than a dedicated measuring tool, that is the problem.
I like that site, there are many good conversation, good data about headphones, but i cant really trust their dac/amp nonsense.. Also really strange that they are actually really serious about the measurements in the case of headphones, while in the case of DAC they saying the opposite.

I think the best option is to using SBAF for getting info about headphones and using ASR for getting info about source components.
 
Last edited:
But the criterion for High Fidelity should, IMO, be accurate reproduction.
This is entirely measurable to a level way beyond the capability of anybody's ears.

This is a reasonable engineering goal but not a satisfying definition/goal in the real world. Eventually what matters is how a piece of equipment sounds to somebody, not how it is measured. And just because something is easily measured does not mean it is the right measure. If something is measured beyond human capabilities ti hear then that should not be part of the ranking in some composite number.

Take an analogy. Say there is a site that tested the traffic resonance of all bridges in a country and ranked them. One might say that a bridge that minimized the resonances (from traffic alone, don’t want to complicate with resistance to winds, earthquakes, etc) was a better engineered design. That is fine. You may have measurements that detect the tiniest of resonances that cannot even be felt. You can even proclaim that it is the right criterion.

But, if you cannot establish any correlation with that measurement with safety of that bridge (except at the extremes), then one cannot make any claims about the quality of the bridge from a safety perspective.

Right now, these equipment rankings with SINAD and a thumbs up and thumbs down create incorrect inferences as it relates to audibility of equipment, none of which have perfect transparency. It also creates openings for companies like PS Audio or NAD to dismiss measurements and/or obfuscate the issue.

For example, the limitations of this methodology to be understood correctly is precisely at the bottom of questions like the below that was asked in another thread.

Does anybody here already actually own an Okto Research DAC8 Pro? Would be interested in knowing if it sounds as good as it measures
Many thanks

This sort of a thing needs to be avoided on its own with better explanation and understanding of what the reviews mean.
 
Thanks for quoting that and I appreciate his response.

I am puzzled what he wants me to change my opinion about. Most of the review is measurements which are not opinions. PS Audio's own website shows very high distortion of 0.03% for this product. My measurements actually showed better outcome than that!

So to the extent the PS Audio forum members (and here) want an answer from him, we have that. The device generates a lot of distortion. It also has high level of noise which again, he agrees with.

The opinion part then is what "sounds good." To me, and the owner of this DAC, the PS Audio DS DAC doesn't sound good. Its distortions due to non-linearities created by the output transformer is unacceptable.

He says he likes the sound better with the transformer. My only ask of him is to have someone test him blind with levels matched. If he can't be bothered to do that, then according to accepted audio science and engineering, his opinion doesn't hold water. So it is not a counter to conclusions reached by us here. That's all.

It is the panther. He wants you to change your opinion to another panther. ;) Maybe a transformers panther.

1569368387967.png
 
@amirm

I am not sure how to read this:
Conclusions
Many of you would have guessed that PerfectWave DirectStream DAC would not do well on the bench.

This reads like; because this is marketed as High End / Audiophile we already know that it is going to perform badly
 
@amirm

I am not sure how to read this:
Conclusions
Many of you would have guessed that PerfectWave DirectStream DAC would not do well on the bench.

This reads like; because this is marketed as High End / Audiophile we already know that it is going to perform badly

This type of thinking is without scientific merit.

Anybody with a reasonable grounding in science will understand that "expected results" shall not influence the outcome. Indeed, a good scientist will derive as much pleasure from proving the null as proving the hypothesis.

Why? Because science is based on the premise that knowledge is superior to ignorance.
 
@amirm

I am not sure how to read this:
Conclusions
Many of you would have guessed that PerfectWave DirectStream DAC would not do well on the bench.

This reads like; because this is marketed as High End / Audiophile we already know that it is going to perform badly



That is not true, there are many good performing high end stuff here. Bbut most high end brands building boutiqe DACs based on nonsense, without any real knowledge, so thats why the statement. Ususally they try to sell their products using many dumb audiophile terminology, for example smooth, analog sound, etc.
 
Last edited:
Back
Top Bottom