Review and Measurements of PS Audio PerfectWave DirectStream DAC

[scott wurcer]

He does it so well that many of his viewers will not realise just how disingenuous he is,the utter nonsense he can speak and that he is selling them snake oil.

He asks questions toward the end. I suppose one could ask him, "What if it sounds 'good' due to the distortion of the Gilbert cell or the level dependent frequency response of an output transformer?" Both easily measurable and possibly above established limits of audibility.

I would differentiate between effects boxes and true snake oil like Synergistic Research or the $2000 ground boxes full of dirt.

 

[ahofer]

To me he seems to be struggling with his answer. The appeal to “common sense” doesn’t shed any light at all on the question.

I was waiting for “does it sound good due to a high noise floor?” But he went the 2nd order distortion route instead.

 

[pozz]

@amirm
From Facebook MQA Audio Group:
“PS Audio is having a private sale on their Directstream DAC to existing customers. These DACs are awesome and for THIS price it's hard to beat. This DAC does full MQA unfold and sounds awesome! Almost $3k off!”
Now $3.999

Will you be the culprit?

In my local secondhand listings at least 5 Directstream DACs were on sale last time I checked.

 

[Cortes]

The master of audiophile spin strikes again! I really do admire Paul McGowan's style and delivery, he is a master communicator with a genuinely engaging and avuncular style that just draws people in. He does it so well that many of his viewers will not realise just how disingenuous he is,the utter nonsense he can speak and that he is selling them snake oil.

yes, he would be a very high profile politician.

 

[Anechoic Reflections]

this is the first time Paul mentioned ASR in his video

I'm actually surprised he mentioned ASR. More people are going to check out ASR and they might like what they see.

I watched Paul's videos* because I also have an interest for stereo hifi. One day I decided to get into headfi and I was so, so very close to buying one of Schiit's products because Paul recommended it. But when researching, I found ASR. Thank God. I will never look back.

*Please note that Paul's videos wasn't my only source of hifi information, they were just entertaining

 

[Anechoic Reflections]

The master of audiophile spin strikes again! I really do admire Paul McGowan's style and delivery, he is a master communicator with a genuinely engaging and avuncular style that just draws people in. He does it so well that many of his viewers will not realise just how disingenuous he is,the utter nonsense he can speak and that he is selling them snake oil.

Yeah, and it is important to keep in mind that he can delete any comment he does not like. So it alway seems like everyone agrees with him.

 

[graz_lag]

yes, he would be a very high profile politician.

Indeed, particularly within the casa blanca's current administration ...

 

[amnesia0287]

I wont claim to have listened to nearly as many dacs as many people here and I’m sure I’m not as discerning, but I have loved how my DS sounds since I got it and haven’t regretted buying it once. I cant argue with the measurements, though measuring a single build of any device is hard to use as reliable information. The review points to the power supply being the flaw, but never considers that the power supply in the testing unit could even potentially be at fault rather than the design as a whole. It also doesn’t consider that because of the fundamental design differences to other dacs (it works quite differently from any other dac that I’m aware of) and that such differences could require changes to how the output was measured as unless I’m missing something the DS is directly outputting pure dsd (different rails for 1/0) rather than converting to a more linear analog signal using ics or any traditional dacs.

is this better? Hell if I know, but in my system I love how it sounds better than any of the dacs I recall trying previously.

what I am really excited about is I ordered the BHK Signature pre (trading in old receivers is great value cause they give original msrp, though that’s true of everything even on the geek out v2+ se I wasted money on receivers just lose value faster than most anything else). which has 5 inputs, so combined with roon I could conceivably do pretty fast a/b testing of up to 5 dacs. I’m still planning to add a few more dacs over time just to see if I like anything better and to see if i perhaps I prefer different things on different dacs. Right now I’m thinking I’ll probably get a qutest and a holo spring or maybe may and then 1-2 more traditional dacs, maybe the RDI and/or something like one of the mytek dacs (at least 1 thing with native mqa support) as well as some more low end stuff to see how much difference it makes to me.

I will probably sell some stuff after just cause I won’t use it all long term, but no measurements are gonna make me suddenly stop liking how my DS sounds, given what I paid (not full price) and how it matches a lot of my gear plus how much I like it, I highly doubt the DS will be among them. But I also get not everyone is in the same situation to waste money like this. No matter how much I like it, I could NEVER call the DS a value, and I’m sure you can get 98% of the way there for 1/10th the price.

regardless though, I feel like the biggest issue with threads like this is there is still 0 data to actually correlate “bad measurements” with bad sound. Especially these specific measurements. Yes, you also can’t correlate those same measurements to good sound either, but ultimately you would find people with preference on both sides because perception varies. “Tube sound” is generally associated with certain types of harmonic distortion, people still like it. Lots of distortion people can’t even hear.

oh also, I’m curious if hdmi i2s might measure better. I’ve seen a lot of people saying that was how they got the best quality out of then ds. Someday I’ll burn the money to get an i2s card for my pc.

 

[Human Bass]

A Dac cant output dsd or any other digital format. It outputs an analog signal.

 

[JohnYang1997]

Ah shit here we go again.

 

[amirm]

I cant argue with the measurements, though measuring a single build of any device is hard to use as reliable information.

Not when my measurements show the tested unit to have half as much distortion as the manufacture states:

I measured 0.016% distortion+noise, they spec 0.030%.

I know measurements are not easy to understand but I do my best using annotations to make them so. If you are going to critique viability of all the work and expense they take to produce, least you can do is read what is there.

 

[amnesia0287]

I didn’t argue anything you said, but nothing you said means “sounds bad”

 

[amirm]

The review points to the power supply being the flaw, but never considers that the power supply in the testing unit could even potentially be at fault rather than the design as a whole.

There is no flaw in the power supply. I am assuming you are talking about this:

This is a highly sensitive measurement and is showing that the power supply tones are at -105 dB. While they could and should do better, it not a sign of anything being broken. PS Audio doesn't publish any measurements of this type so my assumption is that they don't measure their equipment this accurately so this is part of the design.

With distortion products being 40 dB higher and in the most sensitive region of our hearing, what the power supply is doing here is not material.

What you say is the most common talking point by owners when something doesn't measure well. "Oh, it must be a broken sample." I have repaired hundreds of pieces of electronics over the years and can tell something being broken from a mile away.

But if you think that is still the case, why not contact PS audio and ask them to produce the above noise measurement? That would be something constructive than sewing doubt in the measurements as you are doing.

 

[amirm]

It also doesn’t consider that because of the fundamental design differences to other dacs (it works quite differently from any other dac that I’m aware of) and that such differences could require changes to how the output was measured as unless I’m missing something the DS is directly outputting pure dsd (different rails for 1/0) rather than converting to a more linear analog signal using ics or any traditional dacs.

The purpose of a DAC is to take a digital sample, and produce an analog voltage on the output. Instrumentation does exactly that: generates digital samples and measures analog output. How a DAC goes about its job, is its business, not ours. Or yours.

Remember, the manufacture nor the designer brought up this point. They know better than trying to sell this point. Sadly some others have with ill effect for their reputation.

Finally, as I showed, my measurements are actually showing similar numbers they produced. The difference here is that I put them in context of nearly 200 other DACs and showed how bad they were. If we are not supposed to measure such things, how the manufacturer did?

 

[amirm]

is this better? Hell if I know, but in my system I love how it sounds better than any of the dacs I recall trying previously.

What you "recall" has no value. Your short-term memory lasts just a few seconds for large differences. For small differences which is what I am talking about, you need to have switching time of well under a second to tell differences.

But sure, audiophiles as a group do not have good ability to hear artifacts. Instead, they are influenced by many other things including stories they read from designers, look and feel of the product, different mindset when listening, etc. So it is not surprising that like the sound. What is surprising is that you think a) my listening tests must not be good even though I am trained for such things and b) measurements must be wrong.

You can't fight science with non-science. Perform your listening tests using accepted methodology and then we can put value on them. Otherwise it is like someone saying rubbing garlic on your toes cures your cold. Maybe it does but without controlled testing, we can't rely on it and prescribe it to the world.

And that is what we do here. We follow decades of established process in audio engineering and science in how we evaluate audio. You want to dismiss all that, then consider how willing you are to do the same with your health. Don't try to be your own brain surgeon as the saying goes.

 

[amirm]

regardless though, I feel like the biggest issue with threads like this is there is still 0 data to actually correlate “bad measurements” with bad sound.

Not at all. I presented results of my listenings and they agree with measurements. That you like the sound simply means you have well recorded music and you are attributing that to the DAC. Have someone blind test you against a $200 DAC with levels matched and you will be so lost, you won't be able find your own home.

A christian and an atheist will never agree on whether God exists. I don't know why those of you who have never studied audio science, nor believe in it, want us to accept your data points (i.e. "I like the sound") by us. We believe your audio evaluations are completely faulty. And can bring expert witness after expert witness from professional research and engineering to testify the same. Who can you bring? A software person who has designed a DAC for the first time? And didn't understand that sticking a transformer on the output of the DAC causes severe non-linearities leading to distortion? And destruction of small detail in music?

We can completely explain why you can't hear the distortions (audiophiles are poor in this regard). You can't explain what we do so I suggest not trying.

 

[amirm]

“Tube sound” is generally associated with certain types of harmonic distortion, people still like it. Lots of distortion people can’t even hear.

More folklore. I have tested countless tube/high-distortion product. With critical listening ability, they all sound worse than solid state products. Point me to a single peer reviewed research article that shows such preference in controlled testing. None exists.

Distortion in this product is not tube like anyway. It is way too low to have an effect that typical audiophile can hear. To wit, you are not.

 

[amirm]

oh also, I’m curious if hdmi i2s might measure better. I’ve seen a lot of people saying that was how they got the best quality out of then ds. Someday I’ll burn the money to get an i2s card for my pc.

Don't bother. I have tested it on other products and it makes no difference at all. https://www.audiosciencereview.com/...face-better-for-dacs-than-s-pdif-or-usb.7105/

If I^2S does measure better, how come PS Audio doesn't provide any?

 

[amnesia0287]

What you "recall" has no value. Your short-term memory lasts just a few seconds for large differences. For small differences which is what I am talking about, you need to have switching time of well under a second to tell differences.

But sure, audiophiles as a group do not have good ability to hear artifacts. Instead, they are influenced by many other things including stories they read from designers, look and feel of the product, different mindset when listening, etc. So it is not surprising that like the sound. What is surprising is that you think a) my listening tests must not be good even though I am trained for such things and b) measurements must be wrong.

You can't fight science with non-science. Perform your listening tests using accepted methodology and then we can put value on them. Otherwise it is like someone saying rubbing garlic on your toes cures your cold. Maybe it does but without controlled testing, we can't rely on it and prescribe it to the world.

And that is what we do here. We follow decades of established process in audio engineering and science in how we evaluate audio. You want to dismiss all that, then consider how willing you are to do the same with your health. Don't try to be your own brain surgeon as the saying goes.

it has nothing to do with training. Factually what sounds good to you may not sound good to me. I plan to try more dacs and see what is out there and see if I find things that do sound better and more resolving. I’m not arguing against scientific method. I’m not saying it would sound better to anyone else. I made sure to say recall specifically BECAUSE I have seen you make the same argument in the past and wanted to make clear I had NOT done extensive a-b testing (thus my excitement about the preamp, and you can complain all you want about the dac, but I trust BHK to make a good preamp so now I CAN do a/b tests).

I’m not attacking you or your methodology. A sample size of 1, especially for something hand assembled, doesn’t guarantee all units have the same issue.

all I know is I love how the ds sounds and I love that it can give me goosebumps with songs that have never done so when listened via any other means. It’s not objective at all, I never tried to argue that it was, but if measurements were all that mattered there would be a best set of headphones and a best dac and a best preamp and so on, but no one can agree on any of those because ultimately our perceptions vary.

for mastering music or like in a studio reference setting, I could absolutely see where there would be problems inheran to this dac design, but that’s not my use case. My use case is enjoyment and i do enjoy it. Lately i also enjoy kpop even when i know it’s commercial garbage. Can’t explain preference. I just like it.

 

[amirm]

it has nothing to do with training. Factually what sounds good to you may not sound good to me.

Training allows you to find distortion in music. I suggest studying the science on that before saying such things.

Your second statement is definitely false as well. Even in speakers which all sound difference, controlled testing shows that trained listeners agree with non-trained as well other groups of listeners: http://seanolive.blogspot.com/2008/12/loudspeaker-preferences-of-trained.html

HAR is Harman's trained listeners. They are far more critical when it comes to speaker flaws. But their ranking of which speaker is best and which is worse, is the same as all the other groups from university studios to audio dealers and audio marketing people.

There is a lot to audio science that you need to learn to make a proper argument in this forum. Repeating stuff you have read online, intuited, etc. has little value. Go a bit slower, be less critical and see what we have to offer. I guarantee that if you do that, in a few month you will know more about than anyone around you.