Being an objectivist this feeling is eating my brain... [Headphone amps]

[Ron Texas]

If something is eating your brain, embrace it. When your brain is gone, the worries go and music will sound wonderful. The point is, don't get all wound up about this stuff.

 

[JoostE]

I have the same feelings, but as a biomedical scientist, this happens all the time.

I think something similar is going on here, we simply don't understand the full picture yet. That doesn't make the graphs wrong, nor does it make your brain wrong. The objective measures are just missing a small piece of the puzzle.

 

[cestmoi]

I have the same feelings, but as a biomedical scientist, this happens all the time.

I think something similar is going on here, we simply don't understand the full picture yet. That doesn't make the graphs wrong, nor does it make your brain wrong. The objective measures are just missing a small piece of the puzzle.

As an engineer and also as a scientist, I have to agree...

 

[SimoF]

Often we forget that an AMP with ultra low THD will reproduce accurately the recording but it won't always reproduce what the artists intended, studios might not have the same ultra low THD amp that you use, so the mixing and mastering might be done using a not so ideal amp, and so you'll get the most satisfing result using a not so ideal amp.
Plus, measurements are made in ideal condition, headphones are not a pure resistive load and we don't listen at full volume, amp behaviour may vary.
We should not trash as "Psychology" everything we don't fully understand.

 

[BrEpBrEpBrEpBrEp]

Often we forget that an AMP with ultra low THD will reproduce accurately the recording but it won't always reproduce what the artists intended, studios might not have the same ultra low THD amp that you use, so the mixing and mastering might be done using a not so ideal amp, and so you'll get the most satisfing result using a not so ideal amp.
Plus, measurements are made in ideal condition, headphones are not a pure resistive load and we don't listen at full volume, amp behaviour may vary.
We should not trash as "Psychology" everything we don't fully understand.

I think we can assume that modern mastering engineers are mostly using transparent amps.

 

[SimoF]

I think we can assume that modern mastering engineers are mostly using transparent amps.

In my library there's music from the 70s until now, from all over the world. Music from "Modern master engineers using transparent amps" Is just a small percentage.

 

[solderdude]

Often we forget that an AMP with ultra low THD will reproduce accurately the recording but it won't always reproduce what the artists intended, studios might not have the same ultra low THD amp that you use, so the mixing and mastering might be done using a not so ideal amp, and so you'll get the most satisfing result using a not so ideal amp.

In order to achieve that goal ( 'hear' what the 'artist(s)/sound engineer(s)' intended) would be to exactly copy the interior of all studios that produced the music you own/will own/hear in the future AND use the exact same (active or not) monitors and sit at the exact same position as the mastering engineer would it not ? (assuming your hearing matches that of the folks in the studio).

Isn't the distortion (or lack there-off) of an amp a silly argument in that light, seeing as there are many more aspects as just (generally inaudible) amounts of distortion by amps with different distortion characteristics anyway?

That whole 'argument' has little to nothing to do with the amplifier used but the countless other aspects that impart factors more on reproduced sound.

 

[BrEpBrEpBrEpBrEp]

In my library there's music from the 70s until now, from all over the world. Music from "Modern master engineers using transparent amps" Is just a small percentage.

Me too. Presumably, the CD masters took this into account.

 

[Trouble Maker]

The bleed from one channel to the next could be in phase or out of phase depending on what's in the recording and there is no time delay as with crossfeed

Isn't a phase delay and time delay the same thing, just looking at them from the frequency or time domain?

My intuition wanted to say a time delay is a frequency dependent (or variable based on frequency is a better way to put it?) phase delay. I did a little light digging and found this page. I really liked the graphs about linear phase on it, makes prefect sense to me. https://lpsa.swarthmore.edu/BackGround/TimeDelay/TimeDelay.html
Probably real (complex) systems have variables that make a constant time delay unrealistic, it's going to be variable based (dependent) on the input (speed/frequency) to the system.

Any thoughts?

Note: I'm far enough out from dealing with Laplace transforms and the like that I can't speak to the accuracy of all of the technical details on that page, but some of the graphs were really helpful for me to help ground my intuitions.

 

[Weeb Labs]

You are right, it was done playing a 1 Khz tone and a quick FFT using a microphone, then adjusting the pot manually until getting the value. Even if the microphone or adc are not the best, it should be matched enough in my opinion.

I am rather surprised that nobody has mentioned this since it was posted. Level matching should always be performed to within at least 0.1dB via voltage matching; microphones are inadequate. This is particularly critical when comparing headphones and is the most likely explanation for your A/B test results.

 

[Nango]

In order to achieve that goal ( 'hear' what the 'artist(s)/sound engineer(s)' intended) would be to exactly copy the interior of all studios that produced the music you own/will own/hear in the future AND use the exact same (active or not) monitors and sit at the exact same position as the mastering engineer would it not ? (assuming your hearing matches that of the folks in the studio).

Isn't the distortion (or lack there-off) of an amp a silly argument in that light, seeing as there are many more aspects as just (generally inaudible) amounts of distortion by amps with different distortion characteristics anyway?

That whole 'argument' has little to nothing to do with the amplifier used but the countless other aspects that impart factors more on reproduced sound.

I think lots of individuals want to be even better than the artist and the sound engineer and therefore we like to play around with virtual effects and DSP.

While in sociology meantime the scientific community understands that the individuals today "don't want to just live their lifes but to play and "perform" their lives" most of music consumers also want to "make or live an experience" while listening to the music and not just to listen.

 

[pozz]

Isn't a phase delay and time delay the same thing, just looking at them from the frequency or time domain?

My intuition wanted to say a time delay is a frequency dependent (or variable based on frequency is a better way to put it?) phase delay. I did a little light digging and found this page. I really liked the graphs about linear phase on it, makes prefect sense to me. https://lpsa.swarthmore.edu/BackGround/TimeDelay/TimeDelay.html
Probably real (complex) systems have variables that make a constant time delay unrealistic, it's going to be variable based (dependent) on the input (speed/frequency) to the system.

Any thoughts?

Note: I'm far enough out from dealing with Laplace transforms and the like that I can't speak to the accuracy of all of the technical details on that page, but some of the graphs were really helpful for me to help ground my intuitions.

Sorry, I was less clear than I thought.

There are two time delays at play here: the time delay introduced by the crossfeed, and the time delays between the L/R stereo channels which create the stereo soundfield.

• Crossfeed time delay is based on the time it takes for the L signal to hit the L ear and then wrap around the head to hit the R ear (and vice versa). There will be difference in time per frequency, but I don't know if designers take that into account. I would imagine some do, some don't.
• Stereo is based on time/amplitude differences in signal between L and R channels. These of course vary over frequency. Thing is, if you overlay a copy of the L signal over R (unintentionally through circuit level crosstalk), since L and R carry similar information, there will be some cancellation and boosting. Depending on where the cancelations and boosts happen it might affect the mid or side portions of the stereo soundfield. But the crosstalk has to be at very high levels for you to start noticing. Definitely within -20dB, and easily within -10dB.

Often we forget that an AMP with ultra low THD will reproduce accurately the recording but it won't always reproduce what the artists intended, studios might not have the same ultra low THD amp that you use, so the mixing and mastering might be done using a not so ideal amp, and so you'll get the most satisfing result using a not so ideal amp.
Plus, measurements are made in ideal condition, headphones are not a pure resistive load and we don't listen at full volume, amp behaviour may vary.
We should not trash as "Psychology" everything we don't fully understand.

The more you know about psychoacoustics, a branch of psychology, and understand the levels discussed in this thread, the less you will hold onto that position.

 

[SimoF]

In order to achieve that goal ( 'hear' what the 'artist(s)/sound engineer(s)' intended) would be to exactly copy the interior of all studios that produced the music you own/will own/hear in the future AND use the exact same (active or not) monitors and sit at the exact same position as the mastering engineer would it not ? (assuming your hearing matches that of the folks in the studio).

Isn't the distortion (or lack there-off) of an amp a silly argument in that light, seeing as there are many more aspects as just (generally inaudible) amounts of distortion by amps with different distortion characteristics anyway?

That whole 'argument' has little to nothing to do with the amplifier used but the countless other aspects that impart factors more on reproduced sound.

Exactly. There are too many factors to consider before telling someone "it's just in your head". I was not surprised when I read the OP post, and I was not surprised when, after a level matched blind test bestween the SMSL SP400 and the Magnius, that should be two SINAD champs, they were very different and the SMSL was way better.

 

[solderdude]

The question remains... when both are low output resistance and have low distortion and a wide enough bandwidth what could cause 2 amps to sound different (subjectively).
In this case define 'way better'.

What headphones ? used differential or SE ?

 

[Pennyless Audiophile]

As a matter of fact I tried to measure it with my oscilloscope (not good) over the last months... nothing relevant. Which is not surprising because we are talking about a 10-12 bit ADC.

Really curious if anyone else has experienced the feeling of wanting to like a superiorly-engineered product but faced this same dichotomy.

All the time. I have shelved my Klipsch RP-600M for Wharfedale Diamond 220 and Tannoy Gold 7, both well measured components/systems, but the emotion that the 600M give me with some music, is unmatched. I tried to replicate the Klipsch sound with an EQ, I didn't succeed. Better, the emotional impact of the 600M wasn't replicated even if the sound was similar.
I didn't do any double blind so there may be all sorts of biases (even if all three look gorgeus!), but I definitely wanted to like the Diamond 220 better and I didn't. Today I rotate the two in and out of the systems every few weeks.

 

[SimoF]

The question remains... when both are low output resistance and have low distortion and a wide enough bandwidth what could cause 2 amps to sound different (subjectively).
In this case define 'way better'.

What headphones ? used differential or SE ?

"way better" : I don't need be concentrated to perceive differences.
differential, Focal Utopia, Hifiman HE560.

 

[Pennyless Audiophile]

I am rather surprised that nobody has mentioned this since it was posted. Level matching should always be performed to within at least 0.1dB via voltage matching; microphones are inadequate. This is particularly critical when comparing headphones and is the most likely explanation for your A/B test results.

I am asking just to understand, where exactly should I measure the voltage? Why is it better than the sound pressure?

 

[Pennyless Audiophile]

In my library there's music from the 70s until now, from all over the world. Music from "Modern master engineers using transparent amps" Is just a small percentage.

Having had quite a few acquaintances in the industry, I am sure that many "modern master engineers " may purposely use noisy electronics to give an "vintage" or "analog" vibe to the music.

 

[levimax]

I have the same feelings, but as a biomedical scientist, this happens all the time.

I think something similar is going on here, we simply don't understand the full picture yet. That doesn't make the graphs wrong, nor does it make your brain wrong. The objective measures are just missing a small piece of the puzzle.

The perspective of "There is something new and undiscovered to be measured to improve the sound quality of audio amps" keeps the searching / upgrade cycle going. I have spent some time and effort setting up quick switching level matched blind amp tests (which takes some effort to do right) and I have come to the conclusion that transparent audio amps are a "solved problem" and have been for a very long time (at least 70 years for the consumer). While at first I was disappointed (I love to buy and build amps of all sorts) it quickly became a huge relief as I know I am not missing anything in regards to power amps and helps me focus on things that do matter like rooms and speakers.

 

[solderdude]

Having had quite a few acquaintances in the industry, I am sure that many "modern master engineers " may purposely use noisy electronics to give an "vintage" or "analog" vibe to the music.

But that would be embedded in the recording process most likely by using plugins.
They don't use noisy amps between the recording console and monitors they use.