Is Audio Science Review going about it all wrong? Or partly wrong? Or all right?

[SIY]

I did not expect that!

My chief weapon is surprise...surprise and fear...fear and surprise.... my two weapons are fear and surprise...and ruthless efficiency.... My three weapons are fear, surprise, and ruthless efficiency...and an almost fanatical devotion to the Pope.... My four...no... amongst my weapons.... are such elements as fear, surprise...

 

[LTig]

I would be cool with it if his equipment does sound different. Tons of people think this and that device sounds musical yet when I listen to them in controlled testing, nothing like that comes out whatsoever. They either distort a lot in which case they sound harsh and lousy which is not musical. Or the distortion is low enough that is not audible.

Harmonic distortion can make subjective sound better and it is used in production of music. BUT: it is used for specific channels and individually so. The mastering engineer finally masters the production so that it sounds best for him (hopefully).

There is no reason to think that adding more distortion of which kind ever to the summing channels could possibly improve the sound compared to what the mastering engineer heard. If adding distortion would improve the sound he would have done it.

Also how could one ever think that each and every piece of music needs exactly the amount of distortion such a "sounded" amplifier or DAC adds to it? We should break out of the circle of confusion (see Floyd Toole) and use only transparent equipment in the play back chain. It is true, bad recordings might reveal their short comings, but it is the only way out.

Yet I can understand that someone with a more narrow taste of music prefers equipment which matches better to the typical short comings of those recordings. It may for example make sense to use the same type of speakers and amplifiers as used by the studios in the majority of these recordings. One just should accept that newer recordings may not sound as good as they are. And one should not tell other people that such a sounded system is SOTA and better than all others - a typical fault in many published subjective reviews.

 

[Shadrach]

I would be cool with it if his equipment does sound different. Tons of people think this and that device sounds musical yet when I listen to them in controlled testing, nothing like that comes out whatsoever. They either distort a lot in which case they sound harsh and lousy which is not musical. Or the distortion is low enough that is not audible.

While I get and agree with your general point , in the specific this is yet another misleading conclusion about audio gear that is just as wrong as any. After all, saying something is musical provides huge marketing value. If controlled testing doesn't show it, then it is a big disservice to customers.

That is in part why I'm here and not there.
I would be fascinated to see if a difference could be heard in an ABX trial. What comes to mind is Bob Carvers challenge. My understanding is that initially Stereophile could detect a difference between the amps in question, but I can't recall whether the first test was ABX or not. It was only after B Carver had carried out null testing and matched the amps that they became indistinguishable.
In the group I used to belong to in the UK we did get a veryfew significant positive results in ABX testing and if my memory serves me it was with valve v solid state. Not having sufficiently accurate equipment to measure I cannot say whether it was because of distortion or not, but they didn't sound unpleasant.

 

[SIY]

More likely, frequency response, at least for the case of power amps. Stick an impedance in series with the output and it will interact with the loudspeaker impedance. And I'll wager that's all Bob Carver did.

 

[Blumlein 88]

I agree with you in many ways. And I also agree with the non-verbal stuff. But I can assure you that the other one doesn't know. I only say that I will do a test with them and that she has to give her findings. I don't give any visual cues, no verbal cues (I know I can't assure that 100%, but you know what I mean). Nothing.
But I get what you are saying. But for the sake of argument say that the other listeners haven't picked up any cues. Would it be meaningful if we both come to the same conclusion (and we both listen one after the other with no interaction between the two of us)? Not in a sort of consensus way but written down without any talking between the two of us.

Depends upon the particulars. Person A goes into the listening room, compares two things with no volume matching, makes notes about perceived differences. Notes one item has more bass and bigger soundstage. Person B goes into listening room does the same thing and leaves notes describing the same effect. Does it mean something?

You don't know. Was one DUT louder than the other so it got turned down by both A and B? Perhaps while it matched by ear it was louder by enough to matter to both person A and B. That could be why they concur. Yet there might be no other difference.

Describing differences under your proposed scenario can work, but only if the real differences are very large. Small subtle differences aren't going to be reliably detected this way. It is an excellent way to convince yourselves you are both hearing the same thing and it has validity. Which can lead you to false conclusions with very high confidence in what you believe to be true.

I know how attractive it all is, but just take the time and match levels. It takes all these other things out of play. Your conclusions are then on more solid ground. I get that even this simple step takes the fun and flow of the auditioning experience out of it. But the shared auditioning experience isn't really doing us any favors. If you like it and enjoy it then all is fine. Until you need to convince someone else.

And volume matching is not some huge chore. Once you done it a time or two there isn't much to it. You only need a very inexpensive multimeter.

 

[garbulky]

My chief weapon is surprise...surprise and fear...fear and surprise.... my two weapons are fear and surprise...and ruthless efficiency.... My three weapons are fear, surprise, and ruthless efficiency...and an almost fanatical devotion to the Pope.... My four...no... amongst my weapons.... are such elements as fear, surprise...

Your Monty Python deflection not withstanding, I hope we see some change in your behavior.

 

[Blumlein 88]

More likely, frequency response, at least for the case of power amps. Stick an impedance in series with the output and it will interact with the loudspeaker impedance. And I'll wager that's all Bob Carver did.

He did put in series resistance. I recall him describing a couple other changes as well. One was to reduce the effectiveness of the power supply at low frequencies, because the reference CJ performed less well than his amp. Even after he added series resistance which would have altered damping of a real speaker. He also was said to have added some coils in the feedback circuitry.

He claimed to have obtained a -70 db null with hands on fine tuning.

 

[SIY]

Your Monty Python deflection not withstanding, I hope we see some change in your behavior.

 

[Bliman]

Depends upon the particulars. Person A goes into the listening room, compares two things with no volume matching, makes notes about perceived differences. Notes one item has more bass and bigger soundstage. Person B goes into listening room does the same thing and leaves notes describing the same effect. Does it mean something?

You don't know. Was one DUT louder than the other so it got turned down by both A and B? Perhaps while it matched by ear it was louder by enough to matter to both person A and B. That could be why they concur. Yet there might be no other difference.

Describing differences under your proposed scenario can work, but only if the real differences are very large. Small subtle differences aren't going to be reliably detected this way. It is an excellent way to convince yourselves you are both hearing the same thing and it has validity. Which can lead you to false conclusions with very high confidence in what you believe to be true.

I know how attractive it all is, but just take the time and match levels. It takes all these other things out of play. Your conclusions are then on more solid ground. I get that even this simple step takes the fun and flow of the auditioning experience out of it. But the shared auditioning experience isn't really doing us any favors. If you like it and enjoy it then all is fine. Until you need to convince someone else.

And volume matching is not some huge chore. Once you done it a time or two there isn't much to it. You only need a very inexpensive multimeter.

Thank you very much for explaining it. And yes I agree.
A multimeter is not a problem, my dad has always had a multimeter (has a fairly new one now) my nephew also got one.
It is the execution that will be the problem.
Am I correct in what I say underneath?
-I have the Stereophile test cd's I think (https://www.stereophile.com/content/stereophile-test-cd-tracks-1-4)
I would need to play the first track and see that it is at the right sound level to test music.
Then I would pull out the speaker cables plugs out of the speaker and then measure how much volt is coming out of the black and red side of one speaker (how much must I set the multimeter voltage maximum so that I won't blow up the meter?), and is it save to touch the plugs ends or will I get a chock if I do that?
Then I connect my other device and repeat the same process but this time when I am measuring the speaker wire plugs I turn on the volume level till it has the same voltage.
After that I plug the cables back in the speaker and write the volume levels needed on a paper and let other's switch between the devices.
Is this correct?

 

[SIY]

Even a bit easier. Yes, play the track and set the level to where you want to do your analytical listening. But then after you change the device under test, no need to disconnect the speaker cables, you can measure the voltage across the speakers to equalize levels.

 

[amirm]

Harmonic distortion can make subjective sound better and it is used in production of music.

It can make what better? I know of severe distortion added to guitar amps and such but no case of music being done and then in mastering dialing in more distortion to make it sound better.

I think this is one of those fish stories that has been said so many times that it is accepted as fact by both subjectivists and objectivists. I don't think it has merit.

 

[Bliman]

Even a bit easier. Yes, play the track and set the level to where you want to do your analytical listening. But then after you change the device under test, no need to disconnect the speaker cables, you can measure the voltage across the speakers to equalize levels.

To be honest. And I am a bit ashamed of saying this. But I don't understand what you mean with this "no need to disconnect the speaker cables, you can measure the voltage across the speakers to equalize levels"

 

[SIY]

To be honest. And I am a bit ashamed of saying this. But I don't understand what you mean with this "no need to disconnect the speaker cables, you can measure the voltage across the speakers to equalize levels"

Nothing to be ashamed of! You're asking the right questions and we're happy to walk you through it. First, save me some reading- what two things are you trying to compare? And what do you want to use as the music source (i.e., CD, FLAC or WAV, streaming...)?

 

[Bliman]

Nothing to be ashamed of! You're asking the right questions and we're happy to walk you through it. First, save me some reading- what two things are you trying to compare? And what do you want to use as the music source (i.e., CD, FLAC or WAV, streaming...)?

I have contacted Okto Research for the Dac Stereo 8 https://www.oktoresearch.com/index.htm .And hopefully I can test it in May-July. The other device I would test it against is the Grace Design m902 https://www.stereophile.com/headphones/406grace/index.html. I would feed both to my cd player Nad c542i optical input. I also would feed both to my Lyngdorf power amplifier http://www.stereomojo.com/LyngdorfSDA2175AmplfierReview.htm . in the case of the Grace design it would by RCA cables I think and the Okto Research by XLR cables. Then the loudspeaker cables would go from the Lyngdorf to my Dynaudio 52Se speakers and my Q Acoustic Subwoofer. And I would use CD's to compare the two.

 

[Blumlein 88]

So set the multimeter to Volts ac. Connect black lead to the black post, and red lead to the red post on the speaker terminals on back of the speaker as shown below. You need not disconnect anything. You can probably put it right into the middle of the terminal going by this picture. You'll read a voltage. Likely less than 10 volts at a comfortable listening level which you will have already established with one of the DACs. If the meter isn't auto ranging set it to the highest voltage setting and reduce it if the voltage is low enough.

So looking at the DACs you can use RCA with the Grace and XLR with the Octo. You could then switch between them at the Lyngdorf amp by switching inputs. Make sure you don't have to turn off the amp before swapping from RCA to XLR input. Doing that with the amp on might result in a pop or other sound you don't want going to the speaker. To be safe switch the amp off swap from one input to another and turn it back on.

One other possible issue is it looks like both DACs control volume to the nearest db. Worst case you will end up with signal different by .5 db, but likely you'll get closer than that. Though whether you'll get within .1 db is not certain. Hopefully you'll at least get within .2 db for your testing. You could do this by burning a CD-R if you have that ability. Trim the file level in software, save and burn the CD. But if levels are close enough you'll not need to do that.

 

[solderdude]

Just make sure you use a voltmeter that has AC-Volt setting ranges that start with 200mV-AC or 2V-AC.
Do NOT use the multimeters that have 200V-AC as the lowest range. These meters won't give correct values at lower levels.

When metering with the speakers connected use a relative low frequency in the range of 200Hz to 400Hz.
Chances are that even when setting the sound pretty loud you will be measuring between 1V and 2V.
When your meter has a 2VAC range this is perfectly usable.
For reaching 0.1dB your metered voltage should not deviate more than 1%
0.2dB = 2.3% voltage deviation, 0.5% = 5.9% deviation.

 

[beefkabob]

Okay here is another one you could try on a black background.

Thanks, but doesn't work on this monitor! I'd need a magnifying glass to show you. The black line between the two white lines is significantly narrower than the white lines. OLEDs are not grids like traditional LCDs, even though they appear to be with an all white background. I will have to try another monitor.

If you had DACs output music straight to an ADC, how meaningfully could you measure differences in the before and after music files between DACs?
Are DACs ever off key? Say it's a 440hz test tone and they play it at 445hz?

 

[RayDunzl]

If you had DACs output music straight to an ADC, how meaningfully could you measure differences in the before and after music files between DACs?

That's been demonstrated here with a single DAC and ADC.

Recording -> DAC -> ADC -> New Recording

Eight times looped before there's a hint of audible difference between the original and 8th generation..

https://www.audiosciencereview.com/...-choose-the-8th-generation-digital-copy.6827/

Are DACs ever off key? Say it's a 440hz test tone and they play it at 445hz?

My CD Player has a speed adjustment knob, which came in handy when playing some CDs made from old cassettes.

---

Accuracy of the playback rate of different DACs would depend on the accuracy of the clock.

I doubt they would noticeably differ, which isn't the same as measurably different.

 

[Bliman]

So set the multimeter to Volts ac. Connect black lead to the black post, and red lead to the red post on the speaker terminals on back of the speaker as shown below. You need not disconnect anything. You can probably put it right into the middle of the terminal going by this picture. You'll read a voltage. Likely less than 10 volts at a comfortable listening level which you will have already established with one of the DACs. If the meter isn't auto ranging set it to the highest voltage setting and reduce it if the voltage is low enough.
View attachment 25158
So looking at the DACs you can use RCA with the Grace and XLR with the Octo. You could then switch between them at the Lyngdorf amp by switching inputs. Make sure you don't have to turn off the amp before swapping from RCA to XLR input. Doing that with the amp on might result in a pop or other sound you don't want going to the speaker. To be safe switch the amp off swap from one input to another and turn it back on.

One other possible issue is it looks like both DACs control volume to the nearest db. Worst case you will end up with signal different by .5 db, but likely you'll get closer than that. Though whether you'll get within .1 db is not certain. Hopefully you'll at least get within .2 db for your testing. You could do this by burning a CD-R if you have that ability. Trim the file level in software, save and burn the CD. But if levels are close enough you'll not need to do that.

Thank you, thank you. I think I get it. Thank you for giving this detailed information.
It is really helpful. I am curious how it will unfold.
Now hopefully I don't have to wait to long to test it out. But again thank you and also the others that helped me.

 

[SIY]

Thank you, thank you. I think I get it. Thank you for giving this detailed information.
It is really helpful. I am curious how it will unfold.
Now hopefully I don't have to wait to long to test it out. But again thank you and also the others that helped me.

Blumlein covered the ground where I was heading quite thoroughly, so the mechanics should be relatively clear. I might note that 1 kHz test tones are common and present no real trouble for most multimeters. Since you're looking for relative rather than absolute levels, even if a multimeter is rolled off a bit, the matching ability will not be diminished.

Here's a general overview on the other side of testing- what variables you need to keep in mind when setting up the test format. And my hat is tipped to you for your curiosity and willingness to dive in on understanding and experiencing this sort of experimentation!