AudioSceptic
Major Contributor
Here we go...
-
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
- Thread starter amirm
- Start date
Beat on what area specifically? With what input? Most DAC chips have a pathetic 8x digital filter and then resort to sample-and-hold (zero order hold), linear interpolation or other utter crap to take the rate further. Meaning that the reconstruction accuracy is poor at best with 44.1k input... And then they use some simplified modulator that has spurious tones and such. Or like ESS, wandering performance and noise bumps (thanks to DPLL/ASRC).
I wonder if you wouldn't like it if you didn't know it was playing...
I actually wonder how you believe you can divine these differences, but I'm sure it doesn't involve any controls...so nevermind...
It also wasn't the latest version with ladder DAC.It's totally understandable as at that time the super-performer AC power cord Resolution Audio is currently offering was not available.
'The very special performance characteristics of this hand built power cord are achieved by pure copper strands that are drawn in a way that forms the wire into a cross sectional molecular shape. This allows noise coming from the component’s power supply to quickly escape the component on the skin of each wire strand. In doing so, the BlackJack™ helps to keep the component’s own power supply noise from contaminating the music. This processed wire also has the great advantage of time smearing any noise traveling in the wire in the opposite direction thus reducing the amplitude of noise getting into the audio component from the wall outlet.'
YesYou didn't mess about, did you?
Did the lights go dim when you switched those on?
AudioSceptic
Major Contributor
Sound good? You do understand that that is going to be a matter of opinion, i.e. subjective?Yes, you need to both measure well and sound good, both at the same time!
Many times I found out later. I don't have a reason to prefer something, this is just what I've established after building DACs with many of those chips since early 90's and having tens of DACs made by others. It is just my personal preference anyway, I don't mind someone else liking what ever they do. And I wish I could pick and select features from different brands and combine them to something custom, because each has their own strengths and weaknesses. So I've kind of concluded that I rather stay away from any of the chips and deal with "discrete" instead.
Yes, absolutely. I emphasized that this is my personal subjective preference. Other people will very likely have different opinion.
To me, this THD/SINAD discussion is "history repeats itself" type. In the 70's first Tapio Köykkä was building circlotron-type(*) tube amps and complained that new transistor amps with extremely low THD sounded bad and distorted transients. And everybody was like, it cannot sound bad, the THD is absolutely minimal, here are the measurements! Until Technical Research Centre of Finland did some research on this, and Leinonen, Otala & Curl published a paper how to measure this type of transient distortion. Later also came so called Otala & Lohstroh amplifier that fixed these issues (Harman/Kardon and Electrocompaniet productized this).
So overall, IMO, there's no single measurement that exhaustively describes performance, but instead it is very complex matter with multiple aspects.
I've started using TIM measurement for DACs too. Here's just one:
The four high peaks are part of the test tone. Rest of the lower level peaks shouldn't be there.
*) At the moment Yamaha A-S2000 -series amps are using MOS-FET based circlotron circuit, but I don't know of other commercial implementations at the moment.
So overall, IMO, there's no single measurement that exhaustively describes performance, but instead it is very complex matter with multiple aspects.
I've started using TIM measurement for DACs too. Here's just one:
The four high peaks are part of the test tone. Rest of the lower level peaks shouldn't be there.
*) At the moment Yamaha A-S2000 -series amps are using MOS-FET based circlotron circuit, but I don't know of other commercial implementations at the moment.
How about doing some actual controlled testing? It isn't even about preference, it's more about a detectable difference. Building them doesn't make your ears and brain work differently. Uncontrolled tests are completely worthless to anyone, including yourself, in this context.
AudioSceptic
Major Contributor
Good point. I'm sure many of us remember the TIM claims/findings in the 70s. The Pro-Ject chart doesn't really tell us much though, unless we can compare it with some others, especially those with poor SINAD but high perceived SQ.
Building a DAC that ends up in disappointment is not so great experience, lot of wasted time. My office desk headphone setup has three DACs I can easily switch between and my loudspeaker setup has eight DACs I can easily switch between (preamp has only eight inputs). And for me, the first seconds of hearing something are the ones that matter most, even when not comparing to anything. Worst DACs are just in the storage gathering dust, not connected anywhere.
My listening tests are worthless to anybody else, and the way I listen to my hardware is anyway my own thing and I don't really care much what someone thinks about my listening methods because it is none of their business anyway.
Then why share these impressions at all? If you don't care, then I sure don't...
Maybe designing it to meet specs, and learning how to properly do comparison testing rather than trying to create a certain sound would have cut down on frustration?
Because if something sounds bad to me, it sounds bad to me and for me it is unacceptable.
Just like if people vote one way or the other in elections.
Technically by measurements, they meet the specs (as defined here on ASR). If I take two DACs that measure practically the same regarding THD/noise, they don't sound the same. Yeah, I think I have a hunch which measurement results matter in that respect, but it's not the THD/noise, given the figures aren't bad.
And I think I certainly know how to do proper comparison testing. Even standard compliant, although I have not built a standard compliant listening room for that purpose (but I've been in such many times in the past).
After several decades I've settled to approaches that work for me.
Good if you always manage to buy, or design and build something that both sounds and measures perfect for you.
I have some bunch of such measurements now. And over time will collect more.
And no, SINAD + TIM doesn't yet define the sound either. It is still just scratching the surface, just twice as much compared to SINAD alone.
Miska, which measurements do correlate with the sound quality differencies that you hear? Obviously TIM is one of them, what others? Do you use AP analyzer or do you know how to perform them with the AP that Amir has? It would be nice if you could help him to develop/find the most important tests!
Personally, I am mostly interested in performance of redbook and 24/96. Does MP3 or AAC set specific demands/problems?
Personally, I am mostly interested in performance of redbook and 24/96. Does MP3 or AAC set specific demands/problems?
Last edited:
Yes, I get that it sounds bad to you...whatever that means. Since it is a non controlled/supported impression, I'm back to wondering why anyone should care.
Ummm...i don't like lima beans and cilantro. Lima beans and cilantro are both crap. Everyone should know...
If I said that about a DAC...suddenly I'd get responses like... What didn't you like? How was the bitterness? Too much plankton taste in the quiet parts? I've heard that from others, I definitely should wait to see if you like the brussell sprouts...
Science isn't determined by vote...we aren't voting on the best sounding DAC. Lots of places for that...
No need to. If enough people think the same way it begins to matter.
You may be a minority that doesn't like it, that's ok. It could be also that you are majority that doesn't like it and that's ok too. From statistical point of view it would be interesting to know how many people have similar dislike. You're not gonna sell Pepsi or Coca-Cola based on measurements.
Engineering sciences yes, political and economical sciences are a little bit less straightforward in that respect. And in nature, many things happen almost randomly, or are more like making a bet on something.
My question is more like for what purpose does a DAC exist? For measurements or for listening enjoyment? IMO, neither is excuse to do badly on the other. I've many times emphasized that I want both excellent measurements and good sound. But it is not given that good measurement results lead to good sound. It is much harder to get both.
Last edited:
OP
- Thread Starter
- #940
It doesn't really. Or we would conduct audio research that way instead of controlled testing.
Similar threads
