Jump to content

Why do some audiophiles spend thousands of dollars on a DAC? Are they searching for a "sound signature" they like, or just greater "accuracy"?


Recommended Posts

2 hours ago, Grant Slack said:

Hello AV,

 

First, we have to establish which of the 'small differences' are in the sound waves at all, and not just a cognitive consequence of sighted listening.

 

This same comment applies to a lot of posts that have been made in this thread so far.

 

cheers,

Grant

You are stating that if it doesn't appear in the wave form that it is cognitive consequence of sighted listening.

 

So you have a conclusion established based on a bias from the get go.

  • Like 2
Link to post
Share on other sites
  • Replies 375
  • Created
  • Last Reply

Top Posters In This Topic

Top Posters In This Topic

Popular Posts

A consumer DAC is much more than just a chip.   If you take the PoV that anything in a hifi system degrades the sound ... people search for components that deliver the minimum degradation in

Yeah, honestly @MLXXX the whole premise of allowing YouTube viewers to judge sound quality of different DACs from a recorded video, played back on their device is massively flawed.  Others have alread

There's a 'Dorothy Dixer' if ever I heard one!     Andy  

4 hours ago, aussievintage said:

Now let's see if someone embarrasses me by hearing big differences :)  

 

 

I volunteered my wife to take the test. She listened to the D10 through the speakers but not the headphones.    She can hear no difference at all.   

 

Oh well, even if nothing else, if someone else hears the difference, it will be a point of comparison on the quality of DAC required to discern the difference.  or ...  there may truly be no audible difference, of course.

Link to post
Share on other sites
1 hour ago, muon* said:

You are stating that if it doesn't appear in the wave form that it is cognitive consequence of sighted listening.

 

So you have a conclusion established based on a bias from the get go.

 

Measurement-focussed people always have a bias, Ian!  xD  ("If it can't be measured ... it can't be heard. ")

 

Andy

 

  • Like 3
Link to post
Share on other sites
13 minutes ago, aussievintage said:

 

 

I volunteered my wife to take the test. She listened to the D10 through the speakers but not the headphones.    She can hear no difference at all.   

 

Oh well, even if nothing else, if someone else hears the difference, it will be a point of comparison on the quality of DAC required to discern the difference.  or ...  there may truly be no audible difference, of course.

 

Am obliged to you aussievintage, not only that you took the trouble to listen, and to encourage your wife to listen, but that you had the courage to report that there was  no (or extremely little) audible difference. I say "courage" because in an audiophile thread to admit to not hearing a difference can invite sneers!

 

It was probably a good move to choose file 4 for the comparison, the file most affected by the multiple re-recordings.  As it seemed the same (or almost the same) as the original, early on in your listening session while your ears were still fresh, it would probably have been a waste of your time listening to files 2 and 3 in the same listening session.  (Perhaps on another occasion when your hearing was more receptive (e.g. after a particularly good night's sleep), file 3 or even file 2 might sound slightly different to you than file 1.)

 

In my own listening tests I used software players that could be switched on the fly between two files when looping a short segment of the music (Audacity, or foobar 2000 with ABX plug-in). That made it easier for me to zero in on any apparent difference.   It was an immediate A B test of just a few seconds' worth of music.

Link to post
Share on other sites


7 minutes ago, MLXXX said:

In my own listening tests I used software players that could be switched on the fly between two files when looping a short segment of the music (Audacity, or foobar 2000 with ABX plug-in). That made it easier for me to zero in on any apparent difference.   It was an immediate A B test of just a few seconds' worth of music.

 

I might mix my own file containing repeated excerpts from each file to do a quick AB between them.

 

8 minutes ago, MLXXX said:

Perhaps on another occasion when your hearing was more receptive (e.g. after a particularly good night's sleep)

 

Certainly will experiment again later

Link to post
Share on other sites
15 hours ago, aussievintage said:

I was only talking about the small measured differences.  So consider them established.

But not established as audible purely due to the sound waves.

 

14 hours ago, andyr said:

If you are suggesting that the listening group that I referred to in one of my posts all knew that amp #2 would cause the singer to jump forwards of the plane of the spkrs - which thus caused us to hear this effect when we plugged it in ... I am afraid you are dead wrong.  None of us - except the amp designer - had ever heard this amp before; we had absolutely no idea how it would sound.  (And the amp designer didn't forewarn us.)

The cognitive consequences of sighted listening would have dominated if the listeners knew which was which.

 

12 hours ago, muon* said:

You are stating that if it doesn't appear in the wave form that it is cognitive consequence of sighted listening.

 

I neither said nor meant that. I'm only saying we need to be sure that the listening test is not sighted, to avoid the cognitive consequences of sighted listening.

 

12 hours ago, muon* said:

So you have a conclusion established based on a bias from the get go.

Certainly not. Since your first statement is wrong, your second is too.

 

I am surprised and disappointed that two members immediately jumped to label me a 'measurement guy', disparagingly and wrongly. 

 

I don't see you fellows doing that to AV, who, in his quote above, says he took measurements to establish the existence of differences.

 

All I am saying is that, when you do a listening test, you have to do it well, or it will be invalid. The usual methods are quite terrible and lead to illusory perceptions and wrong conclusions. Thinking that the usual methods are perfectly fine is the real way to "have a conclusion established based on a bias from the get go."

 

My point was never about making measurements.

 

Regards

Grant

  • Like 1
Link to post
Share on other sites
9 minutes ago, Grant Slack said:

But not established as audible purely due to the sound waves.

 

I never said it was.  You apparently only read selected parts of the conversation

 

10 minutes ago, Grant Slack said:

I don't see you fellows doing that to AV, who, in his quote above, says he took measurements to establish the existence of differences.

 

 

I did NOT.  You have  not understood what I wrote.

 

 

Please, re-read the whole exchange.  I am sure it will become clearer.

 

 

 

 

 

Link to post
Share on other sites
23 minutes ago, Grant Slack said:

 

The cognitive consequences of sighted listening would have dominated if the listeners knew which was which.

 

 

I find that a bizarre statement, GS.  (Perhaps you need to take your head out of you know where and look around.  xD )

 

Yes, we knew which amp was being used - but, after listening to amp #1 for a couple of hours, we had no idea of how amp #2 was going to sound.

 

Andy

 

  • Like 1
Link to post
Share on other sites


12 hours ago, aussievintage said:

 

I might mix my own file containing repeated excerpts from each file to do a quick AB between them.

Yes, when the audible differences between two audio files are very subtle, quick A B tests of short excerpts may be an efficient and effective way to hear them.

 

__________________________

 

I find foobar 2000 (with an ABX plug-in which can be downloaded separately) very useful in that you can hunt around in the file for a "revealing" part of it. And then just repeatedly play that segment and swap between the 2 versions of the file.

 

Also, if you want to prove to yourself that you are not just imagining the small differences you hear, you can activate the full ABX functionality and test yourself on a blind basis for that revealing part of the recording.

 

Edited by MLXXX
Link to post
Share on other sites
5 minutes ago, andyr said:

I find that a bizarre statement, GS.  (Perhaps you need to take your head out of you know where and look around.  xD )

 

Yes, we knew which amp was being used - but, after listening to amp #1 for a couple of hours, we had no idea of how amp #2 was going to sound.

 

 

To be fair Andy, with sighted testing, once one or two people, particularly if they are regarded as the gurus of the group, expresses an opinion that the soundstage is much better, or whatever, the expectation is set, and it is set on the amp they can see playing at the time.  In groups of humans, these ideas feed on each other to the point that incorrect findings can result (be heard).

  • Like 2
Link to post
Share on other sites
9 minutes ago, andyr said:

Yes, we knew which amp was being used - but, after listening to amp #1 for a couple of hours, we had no idea of how amp #2 was going to sound.

This line of discussion is going well off-topic, but as it is being discussed I'd note  said that when a group of people exchange comments about how something sounds subjectively, that does tend to detract from the scientific rigour of the exercise.  As human beings we like to agree with each other if we possibly can.  There are some individuals who seem to habitually go against the flow, but perhaps they wouldn't be invited to a group listening session in the first place!

 

Faced with the situation in the future of demonstrating a new amp in front of a group of people it could be more persuasive from an evidence viewpoint if the participants were asked to write down their impressions of the new amp on pieces of paper, without consulting amongst themselves first.   If all participants independently reported the same distinct and unusual change [not just a generic change such as "lower noise floor", or "more dynamic"] in writing, that would carry a fair degree of weight I would think.

  • Like 1
Link to post
Share on other sites

To be honest I struggle to see why a DAC costing a few k would be 'that much better' than one that's a few $100. Would love to be able to demo a few in person. Then could decide for myself. 

 

That said, dedicated DAC (Topping D50s) is MUCH better than my on-board PC DAC, and better than Chromecast audio in-built DAC. Next purchase will probably be SMSL SU8 V2 which is around $350AUD. 

 

Anyone know of a place in Melbourne where you can go listen to a range of DACs (perhaps even bring your own speakers) before purchasing?

Link to post
Share on other sites


22 minutes ago, aussievintage said:

To be fair Andy, with sighted testing, once one or two people, particularly if they are regarded as the gurus of the group, expresses an opinion that the soundstage is much better, or whatever, the expectation is set, and it is set on the amp they can see playing at the time.  In groups of humans, these ideas feed on each other to the point that incorrect findings can result (be heard).

Well said.  I hadn't seen this before making my post just above.

  • Like 1
Link to post
Share on other sites
14 minutes ago, MLXXX said:

Yes, when the audible differences between two audio files are very subtle, quick A B tests of short excerpts may be an efficient and effective way to hear them.

 

__________________________

 

I find foobar 2000 (with an ABX plug-in which can be downloaded separately) very useful in that you can hunt around in the file for a "revealing" part of it. And then just repeatedly play that segment and swap between the 2 versions of the file.

 

Also, if you want to prove to yourself that you are not just imagining the small differences you hear, you can activate the full ABX functionality and test yourself on a blind basis for that revealing part of the recording.

 

 

 

I   just loaded file 1 and 4 into audacity.  I set it looping on the tinkly bit (is that a glockenspiel?) and just solo'd between each track.  There IS a difference in detail.  It is subtle but pretty much what I thought I heard yesterday just playing the files.

 

I notice that you supplied file 1 Inverted.  When I play both together, they null out, and I guess what I am then hearing are the residual differences.  If so, there is quite a lot of info that is different.    Not so easy to hear amongst the rest of the correct sound though.  I guess there's a little lesson there.

 

 

Link to post
Share on other sites
28 minutes ago, aussievintage said:

 

To be fair Andy, with sighted testing, once one or two people, particularly if they are regarded as the gurus of the group, expresses an opinion that the soundstage is much better, or whatever, the expectation is set, and it is set on the amp they can see playing at the time.  In groups of humans, these ideas feed on each other to the point that incorrect findings can result (be heard).

 

 

6 minutes ago, MLXXX said:

 

This line of discussion is going well off-topic, but as it is being discussed I'd note  said that when a group of people exchange comments about how something sounds subjectively, that does tend to detract from the scientific rigour of the exercise.  As human beings we like to agree with each other if we possibly can.  There are some individuals who seem to habitually go against the flow, but perhaps they wouldn't be invited to a group listening session in the first place!

 

 

Thank you guys - but, being a lot older than teenage, I am well aware of group dynamics.

 

And when everyone in the group simultaneously lets out a surprised 'Wow!" (when Amp #2 is plugged in ) ... no-one is doing any "influencing" of anyone else!

 

In fact, as Amp #2 was known to be Class A, SE and only 25w, all of us (except perhaps Hugh Dean, the designer) expected it would not fare well, driving the large Dalis - compared to the Bryston, with its several hundred watts.  That is why we were amazed with the forward projection of the singer's voice, relative to the Bryston.  Of course, turning the volume up, Amp #2 (Hugh's "Glass Harmony") was heard to suffer, with the Dalis - but that was expected; the unexpected thing was the forward sound stage that it delivered.

 

Andy

 

  • Like 1
Link to post
Share on other sites


20 minutes ago, aussievintage said:

I notice that you supplied file 1 Inverted.  When I play both together, they null out, and I guess what I am then hearing are the residual differences.  If so, there is quite a lot of info that is different. 

A large part of that residual signal would be attributable to the very slight differences in frequency response.  Some could be due to harmonic distortion.

 

I did notice that if I used a 1kHz test tone at full scale (0dB), its 3rd harmonic was easily seen on a spectrum display. However the actual music was all well below 0dB in level.

  • Like 1
Link to post
Share on other sites
11 minutes ago, andyr said:

And when everyone in the group simultaneously lets out a surprised 'Wow!" (when Amp #2 is plugged in ) ... no-one is doing any "influencing" of anyone else!

My comment was more a generic one than aimed at your particular set of facts.  Generally it would be better (more persuasive from an evidence viewpoint)  to ask people not to make any utterances (including gasps, or shouts of "wow"!) until after committing their impressions to paper, independently.

 

Your listening group situation seems to have been at the extreme end of the spectrum - a very dramatic difference.  There was an apparent gross change in the stereo (or "sound stage") imaging, a change much greater than one would have expected.

Link to post
Share on other sites
59 minutes ago, MLXXX said:

 

Your listening group situation seems to have been at the extreme end of the spectrum - a very dramatic difference.  There was an apparent gross change in the stereo (or "sound stage") imaging, a change much greater than one would have expected.

 

 

Correct - an amazing (and completely unexpected!) difference.

 

 

  • Like 1
Link to post
Share on other sites
On 14/11/2020 at 2:25 PM, Grant Slack said:

Toole is a strong advocate for (well designed) tone controls, separate from EQ, because his not-inconsiderable experience with production ‘standards’ has taught him that we frequently get variations between recordings that make tonal adjustments worthwhile.

 

One could argue that he is saying ‘adjust to taste’, but I think he is saying that, once you get used to good recordings on a well-balanced system,  that don’t need adjustments, you will appreciate being able to adjust unbalanced recordings, not ‘to taste’, but ‘towards reference’.

 

Hi Grant.

 

Technically this makes sense to me but in practical terms my first thought is that constant EQ'ing to taste is probably too hard, too time consuming and too error prone especially if done by hand. Thus my "dog chasing it's own tail" comment.

 

But, I think that a computer algorithm employing some fuzzy logic could be used to EQ each track of a music collection to some sort of reference and preferably one the user has some sort of control over.

A similar example is Volume Leveling and specifically I know of and use this feature of JRiver which levels volume to the industry standard R128.

https://wiki.jriver.com/index.php/Volume_Leveling

Although the implementation is not always perfect, for me it is a game changing feature because I don't have to change volume between albums\tracks and it is great for comparing different masters of the same music because everything is played back at virtually the same average volume.

Edited by Satanica
  • Like 1
Link to post
Share on other sites
2 hours ago, Satanica said:

But, I think that a computer algorithm employing some fuzzy logic could be used to EQ each track of a music collection to some sort of reference and preferably one the user has some sort of control over.

 

The problem is that the computer algo cannot be aware of the content of the recording.   No particular frequency distribution can be assumed.   A recording of flute, or bagpipes, or a recording of a double bass, for example, will all be very different.  

Link to post
Share on other sites
25 minutes ago, aussievintage said:

 

The problem is that the computer algo cannot be aware of the content of the recording.   No particular frequency distribution can be assumed.

 

If it reads the data (bits) then why can't it?

It would be doing a similar but different analysis as the Volume Leveling I outlined earlier.

Perhaps the logic would need to be too fuzzy to be logical. :wacko:

I don't know, just an idea.

Edited by Satanica
Link to post
Share on other sites
1 hour ago, Satanica said:

 

If it reads the data (bits) then why can't it?

It would be doing a similar but different analysis as the Volume Leveling I outlined earlier.

Perhaps the logic would need to be too fuzzy to be logical. :wacko:

I don't know, just an idea.

There are plenty (100's) of mastering plugins that aim to match the characteristics of a reference or bunch of reference recordings including eq, Would probably work in Jriver Here is one

https://www.masteringthemix.com/products/reference?currency=AUD

 

heres a list https://www.musicradar.com/news/10-intelligent-mixing-plugins-that-will-make-your-life-easier-and-save-you-time

 

  • Like 1
Link to post
Share on other sites
1 hour ago, Satanica said:

It would be doing a similar but different analysis as the Volume Leveling I outlined earlier.

Volume levelling as described in the link you provided would be very straightforward.  A fully defined, "mechanical" exercise.

 

In the realm of photography, software may be able to correct wrong exposure and/or wrong colour temperature automatically.  It can do a remarkably good job.

 

However with music recordings, microphones for instruments are usually flat; for pop music vocalists they may accentuate the mid-range.   What would an algorithm be expected to identify exactly and "correct"?  I strongly suspect you'd need human intervention at this stage of technology.   Fully automated  processing would tend to be hit and miss, I'd say.

Link to post
Share on other sites
17 minutes ago, frednork said:

There are plenty (100's) of mastering plugins that aim to match the characteristics of a reference or bunch of reference recordings including eq, Would probably work in Jriver Here is one

https://www.masteringthemix.com/products/reference?currency=AUD

 

heres a list https://www.musicradar.com/news/10-intelligent-mixing-plugins-that-will-make-your-life-easier-and-save-you-time

 

 

I think I understand how this would work but if I understand it correctly it would require a lot of work.

Link to post
Share on other sites
11 minutes ago, MLXXX said:

However with music recordings, microphones for instruments are usually flat; for pop music vocalists they may accentuate the mid-range.   What would an algorithm be expected to identify exactly and "correct"?  I strongly suspect you'd need human intervention at this stage of technology.   Fully automated  processing would tend to be hit and miss, I'd say.

 

Yes I mostly agree. It was simply an idea against not constantly EQ'ing to taste using tone controls as pointed out by @Grant Slack as recommended by Toole. I know for me I have no interest in constant and continual tweaking, do you?

 

Not "correct" so much but "change" to a target defined by the user. Most of the music I listen to is full-range band metal\rock but as @aussievintage pointed out if the content is of a single instrument like a flute for example then EQ'ing to a target would not be so viable. In any case for any kind of content there would need to be strict EQ limits in place.

Edited by Satanica
Link to post
Share on other sites
1 hour ago, Satanica said:

If it reads the data (bits) then why can't it?

 

Because it does not know the correct profile for the unknown recording it is looking at.

 

13 minutes ago, MLXXX said:

However with music recordings, microphones for instruments are usually flat; for pop music vocalists they may accentuate the mid-range.   What would an algorithm be expected to identify exactly and "correct"?  I strongly suspect you'd need human intervention at this stage of technology. 

 

This is a good example of what I mean.

 

 

3 minutes ago, Satanica said:

Not "correct" so much but "change" to a target defined by the user.

 

My experience with my digital phono preamp, is that you are changing the target every second recording or so.  But yes, you could have preset profiles for "symphony orchestra" "heavy metal band" "acoustic folk" etc etc

Link to post
Share on other sites
3 minutes ago, Satanica said:

I know for me I have no interest in constant and continual tweaking, do you?

No. I limit myself to adjusting the gain.

  • Like 1
Link to post
Share on other sites
On 16/11/2020 at 10:14 AM, aussievintage said:
On 16/11/2020 at 10:01 AM, Grant Slack said:

I don't see you fellows doing that to AV, who, in his quote above, says he took measurements to establish the existence of differences.

 

 

I did NOT.  You have  not understood what I wrote.

 

Apologies. Quite right. When you said they were “measured differences”, I mistakenly thought you were being literally true and had measured them. ;) 

 

My beef wasn’t with you anyway: I just felt it wasn’t fair that others (not you) were labelling me a ‘measurement guy’ when I hadn’t been talking about measured things at all, and you had been. It just seemed a bit ironic, as well as unfair.

 

Cheers

Grant

Link to post
Share on other sites
12 minutes ago, Grant Slack said:

just felt it wasn’t fair that others (not you) were labelling me a ‘measurement guy’

 

The biggest mistake is the need to label people one way or the other.   

  • Like 1
Link to post
Share on other sites
21 hours ago, Satanica said:

 

Yes I mostly agree. It was simply an idea against not constantly EQ'ing to taste using tone controls as pointed out by @Grant Slack as recommended by Toole. I know for me I have no interest in constant and continual tweaking, do you?

Hi Satanica,

 

perhaps you have made a simple, occasionally-needed solution look much more tiresome and complicated than it would really be.

 

Toole is talking about a scenario where you have a good, well balanced system in place, with room modes handled effectively, perhaps with fixed EQ, and a smooth accurate treble which does not behave like a laser beam. Most music is going to sound fantastic on this system: music featuring strong bass is not going to turn everything to mud as your room modes take over, and music featuring exciting treble is not going to cut your ears in two. You won’t be rushing to the tone control for “continual tweaking”.

 

In this scenario, you will occasionally put on an album where you can tell that it was produced with a tonal imbalance of some sort, and you will tweak the bass/treble up/down a bit, once, and still enjoy the music, rather than bin it.

 

And it is not meant to be “EQ’ing” in the sense we usually mean, with multiple bands of variable widths etc that would be a nightmare to fiddle with while the music plays: he is talking about ‘adjust bass’ and ‘adjust treble’ controls, simple as that, but they are well-designed filters.

 

With today’s programmable remotes and smartphone apps, it wouldn’t even involve getting up.

 

cheers

Grant

Edited by Grant Slack
Link to post
Share on other sites
23 minutes ago, Grant Slack said:

Hi Satanica,

 

perhaps you have made a simple, occasionally-needed solution look much more tiresome and complicated than it would really be.

 

Toole is talking about a scenario where you have a good, well balanced system in place, with room modes handled effectively, perhaps with fixed EQ, and a smooth accurate treble which does not behave like a laser beam. Most music is going to sound fantastic on this system: music featuring strong bass is not going to turn everything to mud as your room modes take over, and music featuring exciting treble is not going to cut your ears in two. You won’t be rushing to the tone control for “continual tweaking”.

 

In this scenario, you will occasionally put on an album where you can tell that it was produced with a tonal imbalance of some sort, and you will tweak the bass/treble up/down a bit, once, and still enjoy the music, rather than bin it.

 

And it is not meant to be “EQ’ing” in the sense we usually mean, with multiple bands of variable widths etc that would be a nightmare to fiddle with while the music plays: he is talking about ‘adjust bass’ and ‘adjust treble’ controls, simple as that, but they are well-designed filters.

 

With today’s programmable remotes and smartphone apps, it wouldn’t even involve getting up.

 

cheers

Grant

 

Yes, I understood this is was what you meant. I'm yet to come across recorded music that makes me want to "bin it" so to speak based on audio quality and\or mixing but I get your point (and Tooles). I think the kind of your recordings that irk me the most are from the 60's and prior and I rarely listen to music before 1970. Also, I'm kinda happy just to accept what is there knowing that sure there might be enjoyment improvements here and there with some EQ but I think some enjoyment too comes from just pressing play and letting whatever unfold.

  • Like 1
Link to post
Share on other sites
11 minutes ago, Yamaha_man said:

They are just trying to find the digital equivalent to vinyl or R2R. 

 

Well that explains why I gave up on expensive DAC's a long time ago. :yes:

Link to post
Share on other sites
3 minutes ago, Satanica said:

 

Well that explains why I gave up on expensive DAC's a long time ago. :yes:

Sometimes after drinking 12 of these you don’t want to go near your turntable because when you break the cart you could buy a cheap brand new Kia for the same price as the replacement cart.

Expensive/quality DACs have their place and value. 

image.jpg

  • Like 1
Link to post
Share on other sites
20 minutes ago, Satanica said:

 

Well that explains why I gave up on expensive DAC's a long time ago. :yes:

But we get awfully close! 😂

Link to post
Share on other sites
1 minute ago, evil c said:

But we get awfully close! 😂

 

Joking aside, I have been listening to a lot of 24 bit 96kHz vinyl lately.     Really can't complain.  Sounds just like analogue vinyl to me.   But the DAC is an order of magnitude cheaper than the title suggests/questions might be needed.

Link to post
Share on other sites
  • Recently Browsing   0 members

    No registered users viewing this page.




×
×
  • Create New...