• 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!

Those of you who believe measurements aren't the whole story, do you have a hypothesis why that is?

Again, “anyone” or “all” cannot be true. Think about forum discussions on beat rate?

show me a single discussion where a mechanical watch fan (like me) claims they are measurably better.

after this claim, i am sorry to say the rest of your argument is based on that logical fallacy.
 
R.jpg
 
I prefer the Matrix theory. We are foiled by thinking our world can be rational, but someone running the simulation can just change rules as they please. So measurements are the whole story except when they change something in the Matrix. Only we don't have black cats to let us know.
 
show me a single discussion where a mechanical watch fan (like me) claims they are measurably better.

after this claim, i am sorry to say the rest of your argument is based on that logical fallacy.

To be precise, you wrote:
Anyone that is into mechanical watches will openly admit it is EXCLUSIVELY about sighted bias.

I replied:
Think about forum discussions on beat rate?

That's one of the most common topics you see, often in the context of measurements!

 
To be precise, you wrote:


I replied:


That's one of the most common topics you see, often in the context of measurements!

From the 10hz piece:

How does all that relate to accuracy? To be honest I am not sure. Theoretically it should improve accuracy. By how much it is hard to say. Breguet doesn’t publish those details, but I have a feeling that the 7727 should be at least appreciably more accurate than lower frequency movements. In the end, you aren’t gonna get a watch like this for perfect timing, but rather for the technology and visuals.

I think @pablolie 's point was just that people are open that they prefer mechanical watches over quartz watches for pretty much any reason you can think of OTHER than accuracy. And they're pretty open about it - as you can see above. This is in contrast to audio, where people assert better accuracy / fidelity based entirely on subjective evaluations.
 
From the 10hz piece:
Which is why I included it. No different than this review.

I think @pablolie 's point was just that people are open that they prefer mechanical watches over quartz watches for pretty much any reason you can think of OTHER than accuracy.
I am not sure when the comparison over quartz occurred. I was saying that mechanical watch fans care about accuracy. That’s no different than tube fans wanting to measure performance. @pablolie instantly rejected my claim that anyone focuses on accuracy of different beat rate.

This is in contrast to audio, where people assert better accuracy / fidelity based entirely on subjective evaluations.

Yes, again, which “people” ? In what context to the original poster’s question?

Here at ASR, I have answered the question posed by OP by stating my hypothesis is that the benefits of sighted bias are under appreciated and we only think of the weaknesses in controlled testing. Here, at ASR. The thread got bumped and @atmasphere replied and I decided to respond to the OP.

So I have answered quite clearly that sighted bias is a good thing. This is the arbitrary “more musical” experience.

In your case, all I seem to read are “people” and “they” who have annoyed you in the past, and it really doesn’t get to the OP’s question or post, does it?
 
Again, “anyone” or “all” cannot be true. Think about forum discussions on beat rate?


The title is


I am in that first category of recognizing that measurements aren’t the whole story. That’s because the author says “story” as opposed to what is audible.

Measurements don’t capture the value or added experiential factor of sighted bias. It is real.




Who are “they” ?

Is that relevant to the question that the OP asked?

The snake oil audiophile is a personality I don’t like either. What I am saying is that some of us on ASR are technical, measurement focused, but still make hard earned cold cash decisions with sighted bias as a factor.

If I look at what I own versus sold off, it’s not the highest SINAD product or highest preference score speaker. It’s not even ergonomics or reliability. There is definitely sighted bias at play and I am happy and comfortable with that perspective.

Last, while the artist doesn’t care what you buy and you don’t care what you buy, if you can IMAGINE the possibility of a person who understands that stream and media are the same but that playing the disc that they bought from the hands of the artist brings more joy, you are understanding what I am saying.

I don't think what you say is problematic or anti-science at all. You're essentially acknowledging that differences you hear might be imagined, but also taking advantage of the fact that placebo is real, and it can sometimes consistently make you prefer one product over another.

It's a question I've asked myself many times. If sighted bias consistently increases the sound quality of a certain component/speaker, is it really so problematic to try and take advantage of that? For example, I'm in love with the aesthetics of the B&W 802/801 speakers, and there is probably some small %(5%?) of me that believes their unique voicing is a product of internal blind testing. Might that be enough to consistently push them over the edge for me under sighted listening conditions?

The (weird) way I've come to view what I hear is:

1. Soundwaves bounce off my pinna and hit my eardrum
2. Eardrum sends a signal through a DSP with many filters based on both my conscious and subconscious mind.
3. DSP module then sends the processed signal to the sound interpreter, and this is what I "hear"

Blind listening eliminates the second step, but if that DSP is going to be in the signal path for 95% of the time I'm listening, perhaps it's worth considering? The argument against this logic is that the brain's dsp is fleeting and fickle, whereas the eardrum is more consistent. I tend to agree with this argument, but there's a part of me that wonders.
 
Mechanical watch collectors are similar to audiophiles who still use vinyl. Both are legacy technologies. No matter how much you improve a mechanical watch or a turntable, neither will ever come close to matching the performance and accuracy of a $100 quartz watch or a $100 DAC dongle. Both can be appreciated as marvels of mechanical engineering. To me (another mechanical watch collector) there is something incredibly satisfying about looking at miniscule parts I can't even see properly with my aging eyes working in perfect synchrony. Just like I appreciate turntables because they are so satisfying to look at.

The difference is that (as Pablo says) mechanical watch enthusiasts don't claim these devices are more accurate than digital counterparts. Or even more functional. It is 100% pure aesthetics. This kind of honesty is why we are tolerated, even if some ASR members might think we are clinging to outdated technology and wonder what is the point. Turntable enthusiasts on the other hand, DO claim better performance than digital. That's why we have a problem with them, I guess.
 
It is if the assertion is what is audible.

In other words, exactly what I wrote.

It is not if the assertion is what is enjoyable.

Sighted bias affects preference? Stop the presses.


I bet you can do fMRI of someone who enjoys mechanical watches and see the difference in enjoyment/pleasure centers with the sighted bias.



You are buying a physical CD or Blu-Ray from the artist him/herself who hands it to you, versus an audible stream.

It’s the same content.

When you stream the content, you may recall the positive experience with your visit. When you pull out the CD that was touched by your favorite artist, it may more strongly evoke that positive feeling.

MAYBE for you, it doesn’t matter. But the question isn’t if you prefer the sighted bias. It’s if you can imagine someone ELSE finding more pleasure listening to the music from the disc that they got from the artist versus stream, despite the bits being the same, it helps you understand how *here* at ASR where we DON’T have snake oil or magic hand waving, might still opt to buy something that isn’t audibly different for reasons that can only be identified with sight.

This all seems quite banal. You're noting that factors other than the sound can influence 'enjoyment'. Well, yes. It's the very reason we control for those factors if we care to find out if the SOUND determines enjoyment.

All it means is, be careful what you attribute your 'enjoyment' to.
 
The difference is that (as Pablo says) mechanical watch enthusiasts don't claim these devices are more accurate than digital counterparts. Or even more functional. It is 100% pure aesthetics. This kind of honesty is why we are tolerated, even if some ASR members might think we are clinging to outdated technology and wonder what is the point.

But we see turntable fans talking about more precise isolation or better wow and flutter, the same way we talk about mechanical watches with better movements, beyond aesthetics.

Turntable enthusiasts on the other hand, DO claim better performance than digital. That's why we have a problem with them, I guess.

Who are you referring to when saying “them”?


The problem with these threads is that there may be a genuine question. A genuine answer may be offered but then it is met with “but others have answered ____” which leads to problems.
 
This all seems quite banal. You're noting that factors other than the sound can influence 'enjoyment'. Well, yes. It's the very reason we control for those factors if we care to find out if the SOUND determines enjoyment.

All it means is, be careful what you attribute your 'enjoyment' to.

Yes. Now read the OP’s question.

“Those of you who believe measurements aren't the whole story, do you have a hypothesis why that is?”

This is because factors other than sound, and therefore factors other than measurements affect how you perceive the sound.

The most prominent factor is sighted bias, which again, genuinely impacts what you hear.
 
You are right, DvsF isn't the only issue, but I think it is the most important. There's a little program called Distort in which you can add distortion and see how audible it is. But it does not have DvsF as a parameter- its just flat across the band. The designer (on ASR BTW) thought that DvsF was too arcane. But most solid state amps with feedback have had this problem in the last 70 years so I don't think that assessment is accurate. But it does show you how hard is is to hear distortion if DvsF is flat.
I have plenty of different distortion tools myself being a musician with rather harsh taste. It's not easy at all to hear distrortion until extremely high levels. Also, having a bit more distortion above, let's say 10kHz, does pretty much nothing as there is very little musical information and energy there.
 
show me a single discussion where a mechanical watch fan (like me) claims they are measurably better.

after this claim, i am sorry to say the rest of your argument is based on that logical fallacy.
High beat rates are simply easier to adjust for positions, because the interference with balance motion caused by dynamics affect a smaller percentage of the oscillations. There are disadvantages in terms of durability, but my 1968 Zodiac SST and my ~1970 Ebel Chronometer both beat at 5 Hz (36000 bph) and are still going strong. I have a Zenith Captain from 2010 and an Ebel Chronograph from 1986 that both use high-beat Zenith movements, and they also run great. There is a measurable explanation, and exacting the best possible operation from an obsolete technology is interesting and valuable for those relatively few who care.

I completely agree that mechanical watches are the archetype for eschewing measurements in favor of aesthetics (and I'm thinking of aesthetics broadly here in terms of artistic taste, not just outward visual appearance, but de gustibus non est disputandum). That said, just as with playing vinyl records, there's value for those few who care to making them work as well as possible within the context of their obsolescence. And there are some advantages: Lack of dependence on batteries and electronics and gradual rather than sudden failure as a result of neglect (although I don't know anyone, ever, who prophylactically replaces batteries to prevent sudden failure) and no dependence on software. My most accurate mechanical watch is a Ulysse Nardin Marine Chronometer, which uses a hairspring carved from a silicon wafer using a laser--certainly an advanced technology in its own right. It's as accurate as a typical quartz watch, but that makes the point--it takes that level of execution to merely match what a $5 quartz movement can do routinely. Yet I value the thought and design execution that led to that result, and aesthetically it's a beautiful watch with a beautifully crafted movement.

If we think of it in terms of requirements and user needs, most people don't need a watch at all and the use case for wearing one is to be able to see the time without having to dig for a separate device. Very few people have any use case at all for real accuracy in a watch. The main use case for "nice" watches is aesthetic, in the same way that people might wear other jewelry or drive expensive cars as a matter of aesthetic or artistic expression. Sure, it may also be a display of wealth for its own sake, but who among us is completely above that? These are the same influences that may drive someone to own and use a turntable for vinyl records even if they don't already have a library of vinyl records.

It could be nostalgic, or it could be a high-touch reaction to high tech. It could be rebelling against yet another battery-powered piece of electronica that has to be recharged every day or two, or suddenly stop working because some pimply faced adolescent in Cupertino (or wherever) decided that they know better my needs than I do, or because some other pimply faced adolescent IT guy thinks he's saving the world from cyber attack by wiping my personal phone just because I didn't update the operating system far enough quickly enough for using the walled-off-anyway government email app. (If it seems like I'm grousing about that one, you earn no bonus points.)

I've never seen a mechanical watch fan claim that a mechanical watch is a more accurate timekeeper than any common quartz watch. But I do not agree that this is the only measurement that matters, or that it is even the most important measurement to most watch owners--and I include owners of watches antithetical to many mechanical watch enthusiasts like the Apple Watch. I certainly do note that most mechanical watch owners do indeed claim that they are better than common quartz watches for a range of reasons unrelated to timekeeping accuracy.

The issue with measurements is when we give them moral authority beyond their technical meaning, and both the wristwatch and audio hobbies highlight that. Many who claim that mechanical watches are bad do so because they think they represent conspicuous consumption--they are making a social rather than a technical argument, or stated differently a ethical argument that the things they value are the only things that matter. Many who eschew measurements in audio use that as the basis for justifying vastly expensive equipment and tweaks, and when we laugh at their self-delusion we are making a moral rather than a technical argument. We believe that the delusion they seem to fall under is morally bad rather than merely silly--and it may be in some cases where the purveyors of these things argue in bad faith against their own technical understanding and beliefs just to promote sales. We should be careful about separating our moral judgements from our technical judgements.

I agree that audio enthusiasts at the high end often believe the schtick, while watch enthusiasts at the high end more clearly distinguish their aesthetic and social intentions from mere timekeeping accuracy. (Neither groups clearly distinguish between aesthetic and social intentions, but that makes them members of the human race.)

Rick "much ado about nothing" Denney
 
Last edited:
I have plenty of different distortion tools myself being a musician with rather harsh taste. It's not easy at all to hear distrortion until extremely high levels. Also, having a bit more distortion above, let's say 10kHz, does pretty much nothing as there is very little musical information and energy there.
The problem with a lot of 'hifi' amplifiers is distortion (if feedback is used) tends to rise at a much lower frequency. If at 10KHz I agree that's probably not a problem for most people, but at 1KHz it is. As I mentioned earlier, this would put a 7th harmonic at the upper end of our ear's most sensitive region (Fletcher Munson curve). Most solid state amps made have distortion rise at frequencies this low, so no surprise that solid state gets accused of 'harsh and bright' by tube aficionados. Tube amps do this too if they have feedback, but owing to a much more profound 2nd and 3rd harmonic, are better at masking this problem.

FWIW, everything I'm talking about here is very measurable. I don't think there are things we hear that can't also be measured. But the significance of what these measurements can tell us is lost on many, often including those making the measurements. So the idea that we hear things we can't measure persists.
 
The problem with a lot of 'hifi' amplifiers is distortion (if feedback is used) tends to rise at a much lower frequency. If at 10KHz I agree that's probably not a problem for most people, but at 1KHz it is. As I mentioned earlier, this would put a 7th harmonic at the upper end of our ear's most sensitive region (Fletcher Munson curve). Most solid state amps made have distortion rise at frequencies this low, so no surprise that solid state gets accused of 'harsh and bright' by tube aficionados. Tube amps do this too if they have feedback, but owing to a much more profound 2nd and 3rd harmonic, are better at masking this problem.

FWIW, everything I'm talking about here is very measurable. I don't think there are things we hear that can't also be measured. But the significance of what these measurements can tell us is lost on many, often including those making the measurements. So the idea that we hear things we can't measure persists.
Having trouble buying that. First Amp review I came across.
At 1khz the 7th is 30db below the 2nd and 120db below the fundamental, inaudable. At 2khz the 7th is at 14khz and getting lost in fletcher-munson, inaudable.
 
Having trouble buying that. First Amp review I came across.
At 1khz the 7th is 30db below the 2nd and 120db below the fundamental, inaudable. At 2khz the 7th is at 14khz and getting lost in fletcher-munson, inaudable.
Sounds like the 2nd might be able to mask the 7th. But Amirm does not show distortion vs frequency in his reviews so that bit is inconclusive.

Uh, FWIW Dept.: The Fletcher Munson curve where this is important is roughly between 3 and 7KHz. Try listening to a 7KHz tone sometime; I think most people will agree that its not something you want to play loudly. This is the reason that many alarms are in this range- so they get heard.

Your comment about 'inaudible' needs a bit of unpacking. You seem to be treating the harmonic as if its a separate signal, which it is not. Its a modifier of the existing signal and part of it when the signal has exited the amp. Because of that they are not masked by the music so much as they are masked by lower ordered harmonics (the 2nd and 3rd). IME its a common mistake to look at harmonic distortion components as separate signals- they are so small, how could anyone hear them?? But what is being ignored here is that harmonics define tone colors of musical instruments. Take a look at the waveform of any musical instrument and then ask yourself what will happen if you add more complexity to it in the form of higher ordered harmonics.

The answer is simply that since the ear interprets all forms of distortion as tonality, the addition of higher ordered harmonics will make the instrument sound brighter. If odd orders are added, probably harsher too.

Another thing to keep in mind (and this is easily proven with very simple test equipment) is that the ear uses higher ordered harmonics to figure out how loud sounds are. Given a +120dB range, that means it has to be keenly sensitive to their presence.

Most of this is basic to music theory so music majors will have no trouble following along. The tricky bit is understanding how we sense volume and that's easily proven as I just mentioned.
 
Sounds like the 2nd might be able to mask the 7th. But Amirm does not show distortion vs frequency in his reviews so that bit is inconclusive.

Uh, FWIW Dept.: The Fletcher Munson curve where this is important is roughly between 3 and 7KHz. Try listening to a 7KHz tone sometime; I think most people will agree that its not something you want to play loudly. This is the reason that many alarms are in this range- so they get heard.

Your comment about 'inaudible' needs a bit of unpacking. You seem to be treating the harmonic as if its a separate signal, which it is not. Its a modifier of the existing signal and part of it when the signal has exited the amp. Because of that they are not masked by the music so much as they are masked by lower ordered harmonics (the 2nd and 3rd). IME its a common mistake to look at harmonic distortion components as separate signals- they are so small, how could anyone hear them?? But what is being ignored here is that harmonics define tone colors of musical instruments. Take a look at the waveform of any musical instrument and then ask yourself what will happen if you add more complexity to it in the form of higher ordered harmonics.

The answer is simply that since the ear interprets all forms of distortion as tonality, the addition of higher ordered harmonics will make the instrument sound brighter. If odd orders are added, probably harsher too.

Another thing to keep in mind (and this is easily proven with very simple test equipment) is that the ear uses higher ordered harmonics to figure out how loud sounds are. Given a +120dB range, that means it has to be keenly sensitive to their presence.

Most of this is basic to music theory so music majors will have no trouble following along. The tricky bit is understanding how we sense volume and that's easily proven as I just mentioned.
One thing that just occurred to me, the phase of harmonics matters a lot for timbre too. I have never seen measurements of distortion harmonics' phase, is there a reason for that?

But I am not totally convinced about the amplitude of harmonics not really mattering. From a certain point of view, it doesn't matter if it's a separate signal or part of the same signal... a harmonic at -120dB is going to be inaudible regardless of where it came from or what its pitch is.
 
So I have answered quite clearly that sighted bias is a good thing. This is the arbitrary “more musical” experience.
Hmm, arguable. If you want to call it that, placebo makes things sound better with sighted bias. But there is also nocebo, making things sound worse.

You don't have to go far to find people who think digital sounds worse, or certain types of DAC sound worse, etc. There have been threads like that on here in the past few days. Those people will be harmed by sighted bias if they happen to see gear that they're supicious of.

If you are able to use cognitive bias to your advantage, great, do it. But I don't know that it always or even usually makes things sound better.
 
But I am not totally convinced about the amplitude of harmonics not really mattering. From a certain point of view, it doesn't matter if it's a separate signal or part of the same signal... a harmonic at -120dB is going to be inaudible regardless of where it came from or what its pitch is.
I'm probably misreading this as the first sentence seems to contradict the second.

However I agree you're not going to hear a harmonic that low on a sine wave by itself. But when we're talking about music, you will hear it as a change in tonality. If you think about it as a signal to hear, well, that isn't how it works. Our ears interpret harmonics as tonality. So a higher ordered harmonic will cause a musical instrument to sound brighter. But if you look at a musical instrument waveform, is the amp making harmonics of only fundamental tones or is it also making harmonics of other harmonics inherent to the instrument? This is the bit that stymies and is why its important to keep distortion as low as possible.
 
Harmonic Distortion added in the signal chain could add or subract from the harmonic distorion produced by the speaker, with unpredictable results, depending on the phase of the distortion product generated by the speaker on its own in the room vs what is added in the signal (which might be more stable).


Experiment:

Using REW, send a 450Hz tone at -20dB to the system (JBL LSR 308 speakers), take an RTA of the speaker sound with UMIK-1.

Note high level of 3rd harmonic distortion in the speaker output.

It is clearly audible.

Oooh that third sounds soooo good! Let's try this other amp!

index.php



Let's say the other amp adds 0.018% third harmonic distortion at -40 degrees.

New measurement of the speaker in-room:

index.php


Holy mackeral Saffire! Dat turd is gone!
 
Last edited:
Back
Top Bottom