• 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!

Master Thread: Are measurements Everything or Nothing?

You are not hearing any difference between DACs unless it is trivially obvious with measurements. That fact is not debatable without a significant amount of proof. No amount of anecdotal evidence will ever suffice.
So, I post that I heard the difference between a solid state output DAC A and a tube buffered output on DAC A and that I found that DAC B sounded identical to the tube buffered output of DAC A and this is your response? Would it not be expected that the tube buffer would change the way the output measured?

I've seen people say that our ability to measure exceed our ability to hear. What I've not seen is someone that can correlate measurements to how a component will sound.

Are you familiar with Iconoclast cables that are designed by Galen Gareis? I love his scientific approach to design and find it very interesting that he offers multiple conductor materials even though he makes it very clear that he does not know how to design for them. The point is that he can hear something that he doesn't know how to measure or design for. He seems like someone that should know as much from a scientific design perspective as anyone in the industry. If he doesn't think that measurements tell the whole story then it seems to be at least plausable.

How does someone "know" if a particular set of measurements for a given thing are sufficient to account for everything without testing things and seeing if the measurements can explain everything that has been experienced?
 
1.) Yes. You may well mislead yourself, believing your results to be worthwhile, factual or correct when they are not. You may even misidentify the nature of your experiment, causing you to embark on a totally fruitless quest.

2.) You may then mislead others. I don't mean mislead out of malice, I mean mislead others due to a confidence that is unfounded.

Witness the "cold fusion" debacle of a few years back. Only when controls were examined in greatest detail was there a realization that the "breakthrough" did not exist. I know that the enjoyment of audio doesn't rank up there with the importance of cold fusion, but the principles of disciplined examination and cross-examination are the same.

Jim
How does someone use measurements if they never do anything to develop a knowledge of what they mean in a practical sense? I agree that if the conclusions drawn are overly specific they will almost certainly be misleading. For example, I've seen people make statements about how a larger conductor gauge in speaker cables made a specific difference in their system when the cables being compared had nothing in common beyond being long and having two ends.

I had a Krell S-300i amplifier that I was comparing to my Pathos Classic One Mk III and they very sounded different to me, but I really struggled to choose which one was better. Then I rolled the tubes in the Pathos based on a recommendation and the transformation was significant enough that I sold the Krell soon after and have never looked back. How would I ever be able to present a measurement for this? What I learned was that the potential improvement from tube rolling was much more significant that I expected to be possible. Assuming that I had access to get every available measurement of the Pathos before and after, would there be any way use that information to identify another amplifier that I might like more?
 
One thing that I'll say is that I'd like nothing more (maybe a slight stretch) than to sit down and listen with several of you to better understand your perspective and experiences. I'm fascinated at the broad spectrum of audiophiles and I enjoy trying to understand the different approaches. When it comes to the "measurements rule" perspective, I have yet to understand how the approach translates into evaluating audio equipment. It seems that it should result in way less turnover in equipment compared to the "trust your ears" approach where literally everything matters.
 
When you were taking chemistry in school, did you first go into the lab and start combining things will-nilly? Or did you first study the chemistry text in class, learning principles and cautions, and THEN, and only then, go into the lab and learn, as you say, the "practical" side of chemistry?

Audio isn't quite as potentially dangerous as chemistry (unless you build tube amps) but the idea is the same: learn first, do second.



It sounds as if you expect tests and measurements to justify your subjective ideals. It may be possible to use them that way, but that is not the reason they exist. They exist to tell us what something does, be it a speaker or an electronic device, and how well it does what it does. It doesn't tell anyone how well they're going to like it. That's subjectivism.

"Tests tell you what something is.
Opinion tells you how much you like it.
Two different worlds."

Jim
Extra points for the use of the word "willy-nilly".

I agree that measurements are there to tell us what something does and how well it does it (objective) and not how much I will like it (subjective). I do, however, think that if measurements account for everything that can be heard it should be possible to at least correlate measurements to my preferences.

If I know that I like the Focal Utopia Diablo speakers, for example, then it is unreasonable for someone to look at the measurements for it and then be able to predict which one of two other speakers I will likely prefer based on their measurements?

If you were shopping for a new integrated amplifier, how would you go about it? What measurements would you be looking for?
 
One thing that I'll say is that I'd like nothing more (maybe a slight stretch) than to sit down and listen with several of you to better understand your perspective and experiences. I'm fascinated at the broad spectrum of audiophiles and I enjoy trying to understand the different approaches. When it comes to the "measurements rule" perspective, I have yet to understand how the approach translates into evaluating audio equipment. It seems that it should result in way less turnover in equipment compared to the "trust your ears" approach where literally everything matters.
I can describe another experience which you can share in. People tend to believe many DACs have different sound. Some say all are different. Which is true to the source however? One thing I've done here is run output from a DAC into input on an ADC. Repeated it 8 times. So you'd think if a DAC is not true enough to source in one pass, then certainly it is easy to hear after 8 generations thru it and thru an ADC. Further if my DAC sounds different than the one you are listening to and you play my files surely you can tell when one file is the original digital file and one is an 8th generation copy. The files are available for download. Would you care to try them?

I don't intend to challenge you or show you up or any such thing. It is an experience for yourself if you are interested. One which can be interesting I think.
 
Yes, it is unreasonable. Data is unemotional. "Prefer" is emotional. The two are different functions; they don't cross-index. Specifically, preferences have a very great degree of lattitude, especially through time.

Let's put it this way: tests and measurements on two speakers will tell you how much they differ, but not how much you will like or dislike that difference.

They also cannot account for the tendency of the human brain to lock onto some new (auditory) stimulus, providing the listener with a "new" and "more interesting" or "better" sound. You may like the Focal compared to other box speakers, but then fall head over heels with a Sound Lab electrostat, which has enormously different measured characteristics.

I suspect that what you want is to use objectivity to explain subjectivity. Unfortunately, that's a pipe dream. ;)

Jim
So what is the value in measurements if there's no connection between them and my enjoyment? What should I do with measurements? It seems like you're suggesting that I have to trust my ears when picking gear?
 
So what is the value in measurements if there's no connection between them and my enjoyment? What should I do with measurements? It seems like you're suggesting that I have to trust my ears when picking gear?
The problem is if you are like most who say "trust my ears" what you really mean is my ears, my eyes, my knowledge of how something is supposed to be better or worse or made differently.

Measurements can tell you if something is accurate or not. If it is high fidelity. The truth is almost everything in the electronics end is a solved problem. It is possible to make a different sounding device, but more often various things bias you to hear something which has nothing to do with the physical sound being produced. Transducers are what still matter. Microphones, headphones, and speakers all differ enough to have a character. Measurements can be very informative of those, but they are far enough from perfection there is still some preference involved. There is not much need or room for preference in other parts of the chain. Well other than musical choice.

If you think you hear something to prefer and measurements indicate you likely are not able to hear a difference you can use your ears and only your ears to decide if you really can hear a difference. Even after finding you can hear a difference no one other than you can tell you which you would prefer. For the most part everything not a transducer is high enough fidelity they are interchangeable and usually at modest cost these days.

You mentioned earlier it would seem if measurements are useful there would be greatly reduced equipment turnover. You were right, and that is one benefit of things being so good. Still plenty to enjoy and explore with headphones and speakers. So you still won't get bored.
 
I can describe another experience which you can share in. People tend to believe many DACs have different sound. Some say all are different. Which is true to the source however? One thing I've done here is run output from a DAC into input on an ADC. Repeated it 8 times. So you'd think if a DAC is not true enough to source in one pass, then certainly it is easy to hear after 8 generations thru it and thru an ADC. Further if my DAC sounds different than the one you are listening to and you play my files surely you can tell when one file is the original digital file and one is an 8th generation copy. The files are available for download. Would you care to try them?

I don't intend to challenge you or show you up or any such thing. It is an experience for yourself if you are interested. One which can be interesting I think.
This is an interesting concept and not unlike things that I've ready about by Ethan Winer. I would be interested in checking out your files.
 
You mentioned earlier it would seem if measurements are useful there would be greatly reduced equipment turnover. You were right, and that is one benefit of things being so good. Still plenty to enjoy and explore with headphones and speakers. So you still won't get bored.
Clearly your focus is on speakers when it comes to your stereo rig. I agree that speakers have the most significant impact on the final sound of any system. Assuming you were able to start with you dream speakers (whatever they are), how would you go about selecting the DAC, preamplifier, amplifier, cables, etc. to go with them?
 
Clearly your focus is on speakers when it comes to your stereo rig. I agree that speakers have the most significant impact on the final sound of any system. Assuming you were able to start with you dream speakers (whatever they are), how would you go about selecting the DAC, preamplifier, amplifier, cables, etc. to go with them?
My approach to system building is find a speaker you like and work backwards. The amp needs to be enough to power the speaker. Everything else prior to that is more about features and how you want it to work. Harman has made great strides in what kind of measurements result in at least very good speakers. Not quite enough you could rely on it only to pick your optimum speaker, but those that do well on the Klippel type testing are unlikely to be poor. And though they don't do so well I have a love for electrostats myself.

As for cables, there is not much to do. I currently use balanced cables. I have some Mogami based XLR's with Amphenol connectors. Even that is fancier than needed.

So speaker, with appropriate amp. Any of a large number of DAC/preamps will work. I often currently use a recording audio interface for playback. Like an RME Babyface. Fed from my computer(s) as sources.

PS I started a private conversion with links to the files if you wish to try them.
 
Measurements are great to filter out clearly faulty or badly performing equipment. There is so much to choose from that it's completely reasonable to filter out large majority of them based on performance and you still end up with quite a lot to choose from. One can even filter a lot of stuff that could probably work completely fine in your situation, just because why even consider sub-par equipment if you don't have to.

Still, I think there is currently too much emphasis on arguing what is audible in the best case scenario. Low noise floor environment, resolving speakers/headphones, skillful critical listener, 100% focus on trying to find small details in music... Most of these are not the norm to most of the people, most of the time.

At least it would be good to clearly make it understood that there is this super strict, almost theoretical discussion about equipment performance and it doesn't always have that much to do when choosing equipment for real life situations with space and budget constraints. There are a lot of threads about people asking advice and I'm not sure new people always understand what kind of value they would get from big Genelecs or Neumanns compared to somewhat more reasonable approach.
 
There is this story, associated to Wittgenstein, not sure if it is true or not, but goes something like this:

“Tell me," Wittgenstein's asked a friend, "why do people always say, it was natural for man to assume that the Sun went round the Earth rather than that the Earth was rotating?"

His friend replied, "Well, obviously because it just looks as though the Sun is going round the Earth."

Wittgenstein replied, "Well, what would it have looked like if it had looked as though the Earth was rotating?”


I think it summarizes how I feel about this topic quite well. If the history of how we generate information about the world around us should teach us anything, it must be that our senses and initial instincts can not be trusted as a reliable source of information. I don't know if amps that measure the same sound different or not but I found it very surprising when people insist they hear things so they must be right while almost all of our history points to the fact that more often than not, we are wrong.
 
Last edited:
One thing that I'll say is that I'd like nothing more (maybe a slight stretch) than to sit down and listen with several of you to better understand your perspective and experiences.

Have you read through this thread? There are 6,500 or so posts where you'll see a lot of the same questions answered over and over by often the same people.

Here is another that gives an idea of what the basic thresholds are thought to be.


Between these two threads you can get a better feel for what we are about.
 
Good post @Jim Taylor BUT…

…when blackdiamond says “So what is the value in measurements if there's no connection between them and my enjoyment? What should I do with measurements?”, you needn’t reflect on the deep meaning of the questions, because his next words are, “… It seems like you're suggesting that I have to trust my ears when picking gear?”.

In other words, he is simply angling towards the good ole ‘you are measuring the wrong things if they don’t affect the sound heard’ which takes him back to his comfort zone of ‘I trust my ears’ as being the ultimate way to decide on what sounds best.

Firstly, he should have read enough responses already to know that we are NOT “…suggesting that I have to trust my ears when picking gear?” when we say that well designed and performing DACs and amps are sonically transparent. I mean, which part of sonically transparent means to him that he needs to trust his ears to pick gear? It’s disingenuous. He’s angling back to what he wants to hear, not what we are saying.

Back to your post: a good post, and a reminder that we need the measurements of DACs and amps to be sure that a product is sonically transparent, then we can confidently pick on grounds other than ‘what it sounds like’, be that cost, visual impressiveness, or useful features, connections, and user interface. Some of us ‘engineering types’ also include engineering achievement as one of the grounds for choice, ie aligned with our values, which is a valid reason to gravitate towards units with SOTA measurements.
 
Last edited:
This exchange with @blackdiamond has made me conscious about how I use measurements .... and how others use them. It had always bothered me that there were consumers who used measurements to find the ne plus ultra in their quests. All around us, we are bombarded with that type of thinking. Stereophile has its "best of ..." ratings. The car mags constantly tease us with which car is "best". Firearms publications have articles on the "best" pistol or the "best" shotgun. Even yoga pants are "rated". :rolleyes:

I can see that this competitive way of thinking has become part and parcel of our modern society. Everyone wants to know, "Which one is best, and how do I get it?". The cost of the object at the pinnacle is assumed to be commensurate .... and it usually is.

I don't think that way.

I want something that satisfies my requirements. To get to that point, I use measurements, but I use them to reject what is inferior or inapplicable. The remainder are all "good enough", and further winnowing is based on purely personal or financial criteria.
I have noticed that some people are offended by this way of thinking. They view "good enough" as a heresy or a crime. They call it "settling for second-best", and deride it.

I don't view it as settling for second-best at all. Sometimes I view it as settling for third-best. :p Seriously, though ....... I view this competitive mode of thought with a great deal of discomfort. It breeds dissatisfaction and causes consumers to not use that which is useful and not apply that which is applicable, all because they overlook what is at hand while they gaze wistfully at the horizon for something "better". (Notice I said "consumers". I realize that scientists and engineers constantly search for something "better".) In a nutshell, this means that I reject uncontrolled consumerism.

So when @blackdiamond asked about selecting a DAC, amp, cables and so forth, I was in a quandary. I was at a loss what to tell him. I kept thinking, "Well, just pick one that works. There are hundreds." That's how out-of-touch I am with the competitive (or comparative) way of thinking; I don't know how to explain "good enough".

All I know is that I enjoy listening to music on my "second-best" system, and my sleep is not disturbed by thoughts of "gotta-have-it" or "is-this-better-than-that". If I could give that peace to @blackdiamond , I would consider it a wonderful gift.

And if that's a crime, then I'm happy to be a criminal. ;)

Jim Taylor

p.s. - The biggest reason I liked ASR when I discovered it was that Amir seemed to concentrate on showing which items didn't met specs. That was the sort of information that I desired. I wasn't interested in finding out how many veils were lifted.

I also use the "reject stuff that performs poorly then pick something you can afford that you like sufficiently to bother with" method.

To expand on this a bit, criteria for performance will inevitably be somewhat value-laden and open to interpretation (audibility of various types of distortion, just for example) and I've never put much stock in anyones "best of" rankings (they can be interesting as a suggestion list, however).
 
Wash, rinse, recycle, repeat...
Yes. Like usual, we're being exhorted to do endless experiments to prove the Earth is really round, because some guy lays awake at night not believing it.
 
There is this story, associated to Wittgenstein, not sure if it is true or not, but goes something like this:

“Tell me," Wittgenstein's asked a friend, "why do people always say, it was natural for man to assume that the Sun went round the Earth rather than that the Earth was rotating?"

His friend replied, "Well, obviously because it just looks as though the Sun is going round the Earth."

Wittgenstein replied, "Well, what would it have looked like if it had looked as though the Earth was rotating?”


I think it summarizes how I feel about this topic quite well. If the history of how we generate information about the world around us should teach us anything, it must be that our senses and initial instincts can not be trusted as a reliable source of information. I don't know if amps that measure the same sound different or not but I found it very surprising when people insist they hear things so they must be right while almost all of our history points to the fact that more often than not, we are wrong.

Great anecdote, whether or not Wittgenstein actually had that conversation.

The interesting part for me is not that our senses were misleading in that instance (they weren't) but that one model for interpretation of that sensory input was off and led to incorrect analysis.
 
The interesting part for me is not that our senses were misleading in that instance (they weren't) but that one model for interpretation of that sensory input was off and led to incorrect analysis.
I think it really got to me because I realized I was thinking like the friend in that story, that it is natural to assume the Sun went around the Earth, until I realized the absurdity of that thought. Even though we know how things work, we can not help falling into the trap of our naive observations.
 
Did you know one reason the Copernican model didn't displace the Ptolemaic model is the Ptolemaic model with its various corrections was the more accurate one at predicting planetary motion. Everyone was hung up on circles in the heavens as being perfection. When Kepler showed orbits were elliptical then everything fell into place it was far more accurate and simple.
 
Back
Top Bottom