How can DAC's have a SOUND SIGNATURE if they measure as transparent?

[sonitus mirus]

It's not simply that the assertions are unproven, but also that such claims are extraordinary and without necessary proof.

 

[Harmonia]

If you haven't had any particular reason to look into the psychoacoustics side of things, which most of us haven't, you are likely to feel like a lot of blanks get filled in when you do. The brain is a tricky beast, and plays games that good marketing and salesmen will capitalize on. We are all about subjective testing and comparisons, but only when they are controlled with enough rigor to be meaningful.

I agree with what you said. It means to consider the psychoacoustics side of things. I will search this topic and read it later. The human brain is a slave to hormones.

I approached high-end audio easily at first because the prices have gone up so absurdly these days, so I thought, what is a normal price and what do you think?

But now I am learning through this conversation. I learned about many topics that I didn't care about, and I will study about them. I will dig deeper into the "psychology" part that you guys keep talking about.

A filter on the ADC side is to reduce aliasing. Which isn't distortion exactly. Properly designed the filters at both ends let you record, transmit and reproduce the original signal that was sampled within the bandwidth you want. Filters really aren't a personal preference. Since some odd filters cause some mildly audible effects some treat them this way. Personally I think multiple filters on the DAC output is a bad idea. If you want to alter the EQ doing so with precision using EQ makes a lot more sense than messing around with various output filters.

Thanks for the detailed explanation. I understand that the filter is created to produce proper ADC results within the human audible frequency range. But what if there is a subtle difference in sound? It's not a huge difference, but that's what I meant when I said that there is a personal preference part.

If there is a correct answer, there is no need for various filters. Depending on the recorded sound source condition, the filter may need to be changed, and it is made so that the filter can be changed according to the variable. That's why I said "personal preference" in that sense.

Honestly, most people will hardly hear the sound difference due to the filter difference, so it doesn't mean much.

As you mentioned, setting the EQ properly is more important for sound reproduction. That's why I also use ROON to adjust the EQ after setting the hardware. That way, I can change the sound reproduction results in physically impossible situations.

That's why I recommend people to use ROON or a product that can modify the EQ. As I mentioned in other posts, people in my area who use stereo hi-fi speaker systems have a strong stereotype that "you should never use EQ."

I actively insist that it should be used.

Another example is that if you set the woofer properly in a 2.1 channel setup, you can fill in the lack of bass due to the speaker system and space. But the funny thing is, dealers absolutely do not allow you to install woofers. Do you know why?

"If you lack bass, you should use a larger tallboy speaker. You can't hear proper sound with a woofer." They say this. Because woofers are much cheaper. Compared to large tallboy speakers.

They unconditionally recommend a system that makes them more money. That's why I don't use woofers in my area. They look at you like a weirdo if you use a woofer.

No, if the sound is awkward, just use two woofers. You can fill in the lack of bass if you set the crossover well. Even Wilson Audio training says to actively use it depending on the situation. If you set it well, you can make up for the lack.

The easiest thing to do among the options we can choose is software EQ. ROON EQ is good.

Anyway, I'm not too old-fashioned. I'll learn what I have to learn?

 

[Harmonia]

Hallelujah! One of us!
However, you do realize, don't you, that we will hold you to a higher standard because of this? You're one of our family now. Supper is at 5:30. Be sure to wash your hands.

I don't believe you and I don't disbelieve you. (No, my name isn't Schrödinger ) But until you undergo a controlled double-blind test, your assertions are unproven. And that's not to say that they won't be proven. Perhaps they will. That's the point where we investigate the "why" ... not right now.

Thank you for welcoming me. I like this kind of conversation. I also know that I can't give you the results that meet your standards. I will continue to learn, so let's have a fun conversation. I know that I have a lot to learn. I will learn humbly and diligently!

It's not simply that the assertions are unproven, but also that such claims are extraordinary and without necessary proof.

I know. Later, when someone measures DL400, if they can check DSD, it would be good if they could check it. I wrote it with that kind of meaning, so don't worry too much about it. As I wrote in the text, most people won't know the difference between DL100 and DL400 when using FLAC.

 

[JeremyFife]

Okay. Are you saying that you have to use a filter? Thanks for the feedback.

I prefer a filter that is close to linear, but I also know that this is not necessarily the right answer. And I know that filter preferences can change depending on sound taste. I first try using a filter that is as linear as possible, and then try other filters.

If I can turn off the filter, I turn it off. Then I choose a filter and listen. I test various cases.



I know the basic role and content of filters. Digital filters are a type of low-pass filter, and they gradually attenuate signals in the high-frequency range. They are used to reduce distortion (aliasing) according to the sampling rate in the digital signal processing process and adjust the sound quality.

I know. I know, but if I can turn off the filter, I turn it off, or if not, I use the most linear filter possible and listen first. Then I apply the filters. That's just how I usually test new products.

Filters are a matter of personal preference.



Oh, so that's why it can't be deleted. Got it. I'll be careful when I use it!



I know you guys like the measurement, but I just said the result according to my taste. Don't think it means anything. I listen to it because I have a lot of DSD sources and I like listening to DSD format more than FLAC. Whether DSD is theoretically meaningless or not, I listen to it because I like DSD better with my ears.



It's been moved. I thought I wrote it wrong. I didn't see the video. I'll watch it later.



Right. I think that your emphasis on objective results through "measurement" is a good tool for reaching a reasonable conclusion. As I mentioned in other articles, I unconditionally look at the measurement values of products that others have measured and tested.

I graduated from an engineering college, not a liberal arts major. I like measuring. I also like scientific theories. I also enjoy explaining things theoretically.

That's why I'm posting and reading your articles.

The "analog" content that I've experienced so far is the most important point that you guys talk about, such as human illusions and placebo effects, which are the brains that make illusions or can't distinguish.

However, just because I didn't physically measure it, my experience isn't necessarily meaningless. I've had various experiences, not just listening to music.

I've had more diverse experiences than you think, but these experiences aren't meaningless.

I'm talking to you guys right now to approach my experiences theoretically.

And I'm learning along the way.

I'm not just spending my time building my audio system right now, but I'm also writing diligently to learn more things that will help me objectively in my future audio experiences.

Anyway, I admit that you guys are right. That's why I'm going to buy an SMSL DAC and a Class D amplifier.

Hi, perhaps this will help;

Before you make any measurements you first calibrate the equipment. If anyone here made measurements with unknown or DIY equipment then the results would not be accepted. First, and correctly, we'd need assurance that the equipment was working reliably - only then would we move on to discussing the results
That's standard engineering practice (and common sense).

Our ears/brain is a highly unreliable, uncalibrated measurement system with huge variation between systems (people).
So, before we can use our ears/brain and certainly before we can share results we have to add controls and calibrate:
that's level matching (0.1dB), double blind, fast switching and multiple measurements

Without these controls, any listening test is unreliable and cannot be shared.

 

[krabapple]

However, just because I didn't physically measure it, my experience isn't necessarily meaningless. I've had various experiences, not just listening to music.

I've had more diverse experiences than you think, but these experiences aren't meaningless.

But did you get the 'meaning' right?

If you listened to a DSD release and liked it better, that doesn't necessarily mean that DSD has a sound, much less a better one.

That's where blind comparison and level matching come in, to test for common logic errors like that. (You also need to be comparing music that's the same *except for one being in DSD and the other not*)

If you can't apply those controls....and most of us can't ....you can either find some measurements that support your claim, or, temper your claims.

 

[Mnyb]

when doing format comparisons one usually has to do the conversions yourself and not trust different commercial sources for the “same” track even then it can be some snags DSD conversions may involve level changes of the resulting file itself .

Do DAC’s typically play DSD and pcm at the same level ?

Why the rigorous precautions?

In normal listening situations it’s really not possible or extraordinary unlikely for any human to hear differences between Redbook or any better format.
So experimental errors or any far fetched coincidences will dominate over true detections by just the laws of probability.

In the spirit of this tread it can be so that your DAC or streaming system is not transparent it favours some format IE don’t fullfill the criteria in op of being reasonably well constructed or you or it use unorthodox filters choices . Or one of thousands other issues more probable than a true detection

You can also game the listening situation listening to reverb tails at -90dB with the volume fully cranked to insane levels and give yourself a serious case of FOMO

Hence why in the end of the day it’s just easier to measure DAC’s when reviewing them than try to conduct listening test , again on the original topic .

 

[terryforsythe]

In this video the host presents results from an experiment he ran on whether people can hear an audible difference between an origial music file and that same music processed with a DAC. Specifically, he spliced up an audio clip and put it back together with some portions being from the original file and some portions having been processed through a DAC and other gear, which were moderately priced and "not high end really at all". He asked people to listen to the clip and vote on how many transitions between DAC and no DAC were in the clip. In other words, indicate how many times they could hear even the most subtle differences due to the DAC. The results were as would be predicted with a statistical p-value of less than 10^-9, which is overwhelmingly conclusive. The lower the p-value, the greater the statistical significance of the observed result.

The video is well worth viewing by anyone who has an interest in this subject. Given the sheer number of posts in this thread, that seems to be many, many people.

 

[terryforsythe]

The picture below probably helps to explain why people hear, or rather think they hear, inaudible differences between different audio equipment. Our brains are great at playing tricks on our perception.

 

[krabapple]

THERE ARE FOUR LIGHTS!

 

[andyclev]

All DACs sound different when listening tests are done sighted and not level matched.
There is nothing strange about that as the human brain is involved as 'test equipment'.
99% or more of DAC owners test sighted and not accurately level matched and consider sighted tests valid or have a warped idea of what 'blind testing is.
It's these reports and the ones BS audio magazines (paper or web) post, that enhances the notion of DACs, cables, players, decent amps sounding different making it believable and 'factual'.
Perhaps a handful of folks really puts effort in doing a valid (blind) test with statistical relevance. Those that do become 'enlightened' but will be scorned by those that don't. They will be told blind tests are flawed because (insert many reasons) or you are deaf.

It becomes more difficult (if not impossible) to discern between well designed DACs (proper filtering, freq. response and distortion level) when DAC's are level matched and one doesn't know which DAC is playing.

There may be cases it may well be that some DACs sound audibly different (to certain young and trained folks) if they measure substantially different.
With that I mean considered become audible thresholds are crossed in one or more areas.

How I deal with 'sighted' listeners ?
I encourage them (shortly) to do a proper blind test and when interested will give some pointers to test a bit more rigid (but not court style proof delivering).
Sometimes I tell them I can't tell the difference or give an explanation (depends on the person across you).
NEVER tell them they can't or don't hear a difference because the heard so is true, You can't fool their ears after all.
The vast majority thinks I am nuts, an ignorant EE, hardcore objective idiot, deaf or have no idea how music should sound and continues to ignore the message.
They CLEARLY and UNMISTAKABLY heard it and thus it is real.... how could it not be.
Then I go on ignore on the web and when meeting face to face will 'nod' kindly and listen to their babbles (like they do to mine) and friendly part ways.

The only way to get another camp to 'think' a bit more is show them in a test. Have done that in the past (with cables) and the fun part is that they admit 'grudgingly' that they can't hear a difference but the vast majority thinks it was only in a 'rigged' test and don't change their minds one little bit.

These folks exist.. let them live their lives and you yours.
You can't easily convert someone's religion, best to just let them go on thinking their hearing is 'superior' and there are things 'we' cannot measure (yet) or don't know how to measure.

Late to this party, but just wanted to tell you I appreciate your post. I bought a Peachtree Nova300, and then stumbled upon its review on this site, which was given a poor recommendation due to the DAC's measurements. Spooked, I bought a used Topping DAC, in perfect shape, and per this site it measured excellent (highly recommended, resting in the upper side of the blue scale). Results for me? Zero, zip, nada, inaudible, and that really surprised me, as I was expecting at least "something". The money was well spent. The Topping sounds fantastic, just like the Nova300, so I'll give the Topping to my daughter to use as she rolls through her audio journey.

 

[solderdude]

Nova 300 is fine... decent DAC and good power. Even the headphone amp in it is fine.
No worries about the 2nd harmonic being at -85dB at near 0dBFS. This is fine and 'excellent tube-amp-alike'.
The higher order distortions are at an inaudible < -100dB so also very good.
Sure ... one can find devices that perform a bit better in specs but this one is fine and hopefully will give you many years of trouble free service.

 

[SkyTrax79]

Here we are talking about relatively recent DACs, but I wonder how many people would succeed in a blind test between different generations of CD players, like those with no oversampling vs. 8X and even between the old 14 bits vs. 16 bits one. If no one is able to hear the difference that would mean that all the manufacturers would have invented these refinements only to silence audiophiles ? They are refinements that bring nothing more than a supposed improvement in sound.

When I was younger we only had access to audio magazines and I will always remember a test in a magazine that was not even dedicated to audio but still did serious tests and they had tested 20 models of 1987 CD players, 10 brands with each one a low end and a high end model, for example a Yamaha CDX-500 and a Yamaha CDX-1100 and all the participants, who ranged from simple music lovers to sound engineers and audiophiles, had identified the high end model of each brand. Later when I grew up and I had several CD players over the years, from an old Sony with no oversampling one to the most recent and dozens of PC sound cards, I wondered if the tests that I read when I was younger had been rigged because I never heard a difference between any of them.

Finally, when we read reviews of computer websites that test different sound cards, like by reviewers that are not audiophiles but that can say that this or that sound card have a particular sound, as I have already seen that an Asus Xonar DG had more mids than a Xonar DX, the 2 tested flat without effect in the same setup, would that mean that these testers are lying deliberately? I wonder for what purpose.

 

[Somafunk]

Here we are talking about relatively recent DACs, but I wonder how many people would succeed in a blind test between different generations of CD players, like those with no oversampling vs. 8X and even between the old 14 bits vs. 16 bits one. If no one is able to hear the difference that would mean that all the manufacturers would have invented these refinements only to silence audiophiles ? They are refinements that bring nothing more than a supposed improvement in sound.

Last CD player I listened to (not in my car) was back in very early 2000’s before I went all Mac based/active speakers etc.

It was a Technics of some model bought in 1991/92 ish so perhaps a SL-PG 400 something with M.A.S.H, prob cost £180ish and if you were to hook it up to my system below I doubt I could tell the difference.

A mate still uses my technics separate system/tannoy 607 speakers bought back in 1991/92.

 

[Blumlein 88]

Here we are talking about relatively recent DACs, but I wonder how many people would succeed in a blind test between different generations of CD players, like those with no oversampling vs. 8X and even between the old 14 bits vs. 16 bits one. If no one is able to hear the difference that would mean that all the manufacturers would have invented these refinements only to silence audiophiles ? They are refinements that bring nothing more than a supposed improvement in sound.

When I was younger we only had access to audio magazines and I will always remember a test in a magazine that was not even dedicated to audio but still did serious tests and they had tested 20 models of 1987 CD players, 10 brands with each one a low end and a high end model, for example a Yamaha CDX-500 and a Yamaha CDX-1100 and all the participants, who ranged from simple music lovers to sound engineers and audiophiles, had identified the high end model of each brand. Later when I grew up and I had several CD players over the years, from an old Sony with no oversampling one to the most recent and dozens of PC sound cards, I wondered if the tests that I read when I was younger had been rigged because I never heard a difference between any of them.

Finally, when we read reviews of computer websites that test different sound cards, like by reviewers that are not audiophiles but that can say that this or that sound card have a particular sound, as I have already seen that an Asus Xonar DG had more mids than a Xonar DX, the 2 tested flat without effect in the same setup, would that mean that these testers are lying deliberately? I wonder for what purpose.

More likely than lying, though that happens too, is using a very poor methodology. If levels are not carefully matched, then the test is useless. Notice how almost none worry about this. Precisely matched is what is needed too, not matched by ear. If you really want results worth paying attention to then you need blinding. Without those any comments by people are not worth the time to read or watch.

Also there were never very many CD players without some over-sampling. Philips never made one. Sony only did so briefly. Most people have likely never heard a non-oversampling CD player even going back into the late 1980s.

 

[ahofer]

would that mean that these testers are lying deliberately? I wonder for what purpose

A combination of bad methodology (as suggested by @Blumlein 88 ) and the fact that the business model of equipment reviewers depends on a) there being differences in equipment to review, b) there being a steady supply of equipment to review, and c) there being a steady supply of advertising from equipment makers.

PS, in this sense the methodology becomes a “feature, not a bug”, which explains its durability in audio vs other fields of endeavor.

 

[DSJR]

Okay. Are you saying that you have to use a filter? Thanks for the feedback.

I prefer a filter that is close to linear, but I also know that this is not necessarily the right answer. And I know that filter preferences can change depending on sound taste. I first try using a filter that is as linear as possible, and then try other filters.

If I can turn off the filter, I turn it off. Then I choose a filter and listen. I test various cases.



I know the basic role and content of filters. Digital filters are a type of low-pass filter, and they gradually attenuate signals in the high-frequency range. They are used to reduce distortion (aliasing) according to the sampling rate in the digital signal processing process and adjust the sound quality.

I know. I know, but if I can turn off the filter, I turn it off, or if not, I use the most linear filter possible and listen first. Then I apply the filters. That's just how I usually test new products.

Filters are a matter of personal preference.



Oh, so that's why it can't be deleted. Got it. I'll be careful when I use it!



I know you guys like the measurement, but I just said the result according to my taste. Don't think it means anything. I listen to it because I have a lot of DSD sources and I like listening to DSD format more than FLAC. Whether DSD is theoretically meaningless or not, I listen to it because I like DSD better with my ears.



It's been moved. I thought I wrote it wrong. I didn't see the video. I'll watch it later.



Right. I think that your emphasis on objective results through "measurement" is a good tool for reaching a reasonable conclusion. As I mentioned in other articles, I unconditionally look at the measurement values of products that others have measured and tested.

I graduated from an engineering college, not a liberal arts major. I like measuring. I also like scientific theories. I also enjoy explaining things theoretically.

That's why I'm posting and reading your articles.

The "analog" content that I've experienced so far is the most important point that you guys talk about, such as human illusions and placebo effects, which are the brains that make illusions or can't distinguish.

However, just because I didn't physically measure it, my experience isn't necessarily meaningless. I've had various experiences, not just listening to music.

I've had more diverse experiences than you think, but these experiences aren't meaningless.

I'm talking to you guys right now to approach my experiences theoretically.

And I'm learning along the way.

I'm not just spending my time building my audio system right now, but I'm also writing diligently to learn more things that will help me objectively in my future audio experiences.

Anyway, I admit that you guys are right. That's why I'm going to buy an SMSL DAC and a Class D amplifier.

Please chaps, don't forget the MUSIC in all this analysis of 'sound.'

 

[krabapple]

Here we are talking about relatively recent DACs, but I wonder how many people would succeed in a blind test between different generations of CD players,

Vanishingly few if any.

Here we are talking about relatively recent DACs, but I wonder how many people would succeed in a blind test between different generations of CD players, like those with no oversampling vs. 8X and even between the old 14 bits vs. 16 bits one. If no one is able to hear the difference that would mean that all the manufacturers would have invented these refinements only to silence audiophiles ? They are refinements that bring nothing more than a supposed improvement in sound.

When I was younger we only had access to audio magazines and I will always remember a test in a magazine that was not even dedicated to audio but still did serious tests and they had tested 20 models of 1987 CD players, 10 brands with each one a low end and a high end model, for example a Yamaha CDX-500 and a Yamaha CDX-1100 and all the participants, who ranged from simple music lovers to sound engineers and audiophiles, had identified the high end model of each brand.

Yeah, more details on that anedcdata from 'a magazine not even dedicted to audio'. Was it even blind and level-matched?

Later when I grew up and I had several CD players over the years, from an old Sony with no oversampling one to the most recent and dozens of PC sound cards, I wondered if the tests that I read when I was younger had been rigged because I never heard a difference between any of them.

Hmm d'ya think?

Finally, when we read reviews of computer websites that test different sound cards, like by reviewers that are not audiophiles but that can say that this or that sound card have a particular sound, as I have already seen that an Asus Xonar DG had more mids than a Xonar DX, the 2 tested flat without effect in the same setup, would that mean that these testers are lying deliberately? I wonder for what purpose.

Yes and no. Some are lying deliberately, some are just not testing properly.

 

[krabapple]

Please chaps, don't forget the MUSIC in all this analysis of 'sound.'

Who scared you?

 

[DSJR]

Who scared you?

Nobody in particular

 

[ICIETDIYEUR]

Hello.

I like math and logic and I asked myself the following question:

we take 1000 people totally at random and make them take an 'ABX test' between two DACs, then we repeat this same experiment a dozen times.

We add up the results to find out which of the two devices was unanimously preferred (by listening, of course).

It can be seen from the results that the device chosen is not the one that obtains the best results in the measurements:

- What can we conclude from this ?

- What is important ?

Be objective but above all honest and don't tell me that it's not possible because the measurements say the opposite, DACs have fairly small measurement differences between them nowadays and there is still a very large number of people who are convinced that they hear differences.

My final question is: Are we 100% confident that we are currently measuring everything that should be measured ?

NB: I don't know (what else to measure) but I'm not convinced although I don't question the objective evidence of what we measure nowadays.

PS: I liked the 'intervention' of a member of the forum who said first to listen to a device before proceeding with the different measurements so as not to influence his subjectivity (his impressions) because I have the impression that it is often the opposite that happens quite often...

Regards.