• 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!

Do you need linear power supply for DACs?

svart-hvitt

Major Contributor
Joined
Aug 31, 2017
Messages
2,375
Likes
1,253
I don't know why I am being forced to believe I can hear something I can't hear? If (you think) you have good hearing, totally fine. I can't even hear some lossy codecs in high bitrates, and resampler resampled to and from prime numbers with multiple passes. You can sell this stuff to believers, but sorry, I am not one of them. I am talking about upsampling, not 3D processings and such.

On a general note on what is audible: Training (preferably under guidance) can help a lot. In audio, we’ve known for ages that training is needed to hear certain things. So you should ask: Am I a trained listener?
 
Last edited:

March Audio

Master Contributor
Audio Company
Joined
Mar 1, 2016
Messages
6,378
Likes
9,317
Location
Albany Western Australia
A bit far-fetched conclusions to draw correlations between the two, given more differences than similarities.

That's for example based on my background on finding the needle in acoustic haystack.



Both are by the way coming to home networks too. Just check out Merging's NADAC family.

But yeah, my NAA is designed to work over WiFi and such.
I'm not even sure what that means, but yes you do seem somewhat pre occupied with tiny and inaudible ;)

Does nadac benefit from silver ether net cable?
 

Miska

Addicted to Fun and Learning
Audio Company
Joined
Feb 20, 2019
Messages
615
Likes
448
Adding computer processing does not reduce cost.

Yes it does, because if you use one for listening music, you already have made the investment.

I do realize that many DAC manufacturers are scared about my things. :D

The ones that I was confident made a difference were the ones that had measurable in band effects (roll off)

Not long ago you claimed that different filter responses don't make audible difference, none of the ESS filters roll off before 18 kHz and only one below 20 kHz.

Of course if you can't hear the difference it doesn't mean that nobody else can. Also depends on what properties you are sensitive to, and how much critical listening practice you have. Maybe someone else would have a different preference than you on the filter?
 

March Audio

Master Contributor
Audio Company
Joined
Mar 1, 2016
Messages
6,378
Likes
9,317
Location
Albany Western Australia
Yes it does, because if you use one for listening music, you already have made the investment.



Not long ago you claimed that different filter responses don't make audible difference, none of the ESS filters roll off before 18 kHz and only one below 20 kHz.

Of course if you can't hear the difference it doesn't mean that nobody else can. Also depends on what properties you are sensitive to, and how much critical listening practice you have. Maybe someone else would have a different preference than you on the filter?
If. Pc is not the only source. Plus you are ignoring cost of software to perform the processing and of a pc more powerful than may otherwise be necessary.

No I did not claim that at all.

You are making assumptions as to what dac I was listening referring to.

Aahhhh so now I can't hear...... Predictable....

So how about a straight answer from you. Do you think the differences in distortion between the two plots you posted earlier are audible.

Yes or no?

PreBoxS2-44k-THD-graph.png


PreBoxS2-DSD512-THD-graph.png
 
Last edited:

bennetng

Major Contributor
Joined
Nov 15, 2017
Messages
1,634
Likes
1,692
On a general note on what is audible: Training (preferably under guidance) can help a lot. In audio, we’ve known for ages that training is needed to hear certain things. So you should ask: Am I a trained listener?
I indeed trained myself. For lossy encodings I trained myself by using lower bitrates first so that I know how the artifacts sound like, then gradually using the higher quality settings. A very small amount of lossy encoding in the highest possible bitrate are indeed not transparent to me, but the percentage is very low:
https://hydrogenaud.io/index.php/topic,39970.0.html

I also used the same tool that Amir introduced to train myself, and I did that before the born of ASR. IIRC using that tool I can hear up to 11 different EQ band IDs. That's my limit.
https://www.audiosciencereview.com/...an-blind-speaker-testing-system.33/post-21816

About resamplers, I trained myself at lower sample rate or lower quality settings, then gradually increased the settings.

So yes, training can improve hearing, but the improvement won't increase infinitely.
 

Miska

Addicted to Fun and Learning
Audio Company
Joined
Feb 20, 2019
Messages
615
Likes
448
If. Pc is not the only source. Plus you are ignoring cost of software to perform the processing and of a pc more powerful than may otherwise be necessary.

That is up to everybody to decide. For me, I only have Google's Chromecast Audio dongles in addition to computers, and that goes through a PC too using Toslink, for performing digital room correction and upsampling. On the go I listen from a laptop.

No I did not claim that at all.

You are making assumptions as to what dac I was listening referring to.

Aahhhh so now I can't hear...... Predictable....

I'm assuming yours? Or you didn't even try to listen the ESS chip filters? How come?

You have very selective hearing and memory, because you were constantly repeating that digital filters make no audible difference.

But how predictable, that only filters running on a computer are in your opinion inaudible, and if the on-chip filters have any faults or short-comings they are inaudible. But otherwise they are audible, but only if they are on-chip. :D:D
 

Miska

Addicted to Fun and Learning
Audio Company
Joined
Feb 20, 2019
Messages
615
Likes
448
About resamplers, I trained myself at lower sample rate or lower quality settings, then gradually increased the settings.

For lossy codecs, resamplers and delta-sigma modulators, it is also important to know what kind of material best triggers certain types of deficiencies and then listen. Much easier to spot.
 

March Audio

Master Contributor
Audio Company
Joined
Mar 1, 2016
Messages
6,378
Likes
9,317
Location
Albany Western Australia
That is up to everybody to decide. For me, I only have Google's Chromecast Audio dongles in addition to computers, and that goes through a PC too using Toslink, for performing digital room correction and upsampling. On the go I listen from a laptop.



I'm assuming yours? Or you didn't even try to listen the ESS chip filters? How come?

You have very selective hearing and memory, because you were constantly repeating that digital filters make no audible difference.

But how predictable, that only filters running on a computer are in your opinion inaudible, and if the on-chip filters have any faults or short-comings they are inaudible. But otherwise they are audible, but only if they are on-chip. :D:D

It is for the individual to decide but your assertion was its cheaper which is not the case.

How come? Because I know the differences are a dragon chase.

You really need to stop that. I have made no such claim. Making things up as a form of attack will get you nowhere.

I have only talked about the audibility of transition band signals at very low levels, as they are with real music, and the effect of low level out of band spuria.


But going back to the question of audibility, do you think the difference in distortion of the 2 plots you posted is audible?

Yes or no?
 
Last edited:

Miska

Addicted to Fun and Learning
Audio Company
Joined
Feb 20, 2019
Messages
615
Likes
448
Ahh damn. it is?

It took them just about 10 years to bring it in, maybe in another 10 years it'll be as good as Apple's one in macOS or the one in Linux kernel.

Although looking at the rate which Windows 10 is deteriorating after each update, I'm not so sure about that.
 

Miska

Addicted to Fun and Learning
Audio Company
Joined
Feb 20, 2019
Messages
615
Likes
448
Then some funny stuff for you. They are CD flac rips.
index.php

Yeah, I've found bunch of tracks from my library where 3 dB headroom is not enough... One is even "hires". For most material it is though. Overall the statistics in this respect for RedBook material doesn't look very rosy indeed. And number of DACs just clip hard, some leave headroom. I think Wolfson was first DAC chip that offered configuration option for -2 dB headroom, but it was disabled by default...

1552575624935.png
 
  • Like
Reactions: MWC

Miska

Addicted to Fun and Learning
Audio Company
Joined
Feb 20, 2019
Messages
615
Likes
448
I have only talked about the audibility of transition band signals at very low levels, as they are with real music, and the effect of low level out of band spuria.

And you hearing differences in the filters is different in which way? Can you clarify?
 
OP
amirm

amirm

Founder/Admin
Staff Member
CFO (Chief Fun Officer)
Joined
Feb 13, 2016
Messages
44,368
Likes
234,384
Location
Seattle Area
Just a note that I measured up to 1 Mhz to address arguments that these power supplies do good way up there, not because it is something that this site is about. It is definitely not a sign of new strategy to care about ultrasonics in these ranges. For the sake of high-res music, I might care up to 90 kHz but nothing beyond.....
 

bravomail

Addicted to Fun and Learning
Joined
Oct 19, 2018
Messages
817
Likes
461
Thx for review, Amir! I think clean strong power supply will have more effect in amp section (or in pure amps). DAC is already very minimalistic case of power usage. Will also like to see amperage and voltage produced from PS (or maybe they already in your graphs, not sure).
 

redjr

Member
Joined
Oct 20, 2018
Messages
72
Likes
63
...... I have no interest in this DSD upsampling stuff at all.....

Actually, even an inexpensive DAC that incorporates a good SRC chip(to convert to DSD) in its design and implementation, will improve the sound rather significantly. It helps to reduce jitter.
 
OP
amirm

amirm

Founder/Admin
Staff Member
CFO (Chief Fun Officer)
Joined
Feb 13, 2016
Messages
44,368
Likes
234,384
Location
Seattle Area
Will also like to see amperage and voltage produced from PS (or maybe they already in your graphs, not sure).
The supplies don't have current meters so I don't know what they put out. I could try to tap in the middle but have to wire up a fixture and I am too lazy to do so. :)
 
Top Bottom