Subjective and Objective

[HarryHWombat]

Not everything that can be counted counts, and not everything that counts can be counted.

-Albert Einstein

 

The pure subjectivist will damn all science if their ears tell them there exists a difference even if this would then require a rewriting of the fundamental and incontrovertible rules of existence. The arrogance one finds in this camp is at times frankly astonishing. Hearing (and of course any sense) is a complex series of electrical and chemical interactions in the brain caused by an external stimulus and is heavily dependent on context and influence from sources independent of the external stimulus (be this an individuals memories, suggestion or other). All this is proved beyond argument. Moreover it should be stressed that to be human is to be extremely susceptible to suggestion overt or otherwise. So in an experiment where one hears something that contradicts what the laws of the universe demand should one throw out these laws or should one be at least amenable to the fact that the sense of hearing has been fooled in some way? I am sure many of you have experienced the experiment in the link below (I may even have originally found it here) but the first time I did this it kind of opened my eyes as to how flawed a subjective comparison can be as it proves ... well, give it a go and hear for yourselves. For any or all of the sentences listen to the SWS first then the Original then the SWS again.

 

http://www.lifesci.sussex.ac.uk/home/Chris_Darwin/SWS/

 

There are more things in heaven and earth, Horatio,

Than are dreamt of in your philosophy.

- William Shakespeare

 

The pure objectivist view does not take into account that there are some things that count that cannot be counted (measured). More importantly, that there may be things that we do not yet understand that are important. I have read several times about the Nyquist frequency being half the sampling rate and that for a CD at 44.1kHz that the maximum frequency that can be re-constructed is 20 kHz which is above normal humans perceptive abilities. Fairly straightforward one would have assumed but googling around I found this paper below which shows a change in the brains electrical activity producing an emotional response to frequencies above this range.

 

http://www.av123.com/download/aps_auditory_hypersonic_effect.pdf

 

So I sit somewhere in the middle but positioned safely toward the objectivist end of the spectrum but always amenable to the fact that there is stuff I do not understand and maybe that nobody has yet understood or discovered. I am interested to "hear" where other contributors sit on that spectrum.

 

To close, I had a discussion once with a non hi-fi friend of mine basically about soundstage produced from stereo speakers and how a 3D sound (up down forward back) can be produced from an essentially 2D source. He was skeptical but I had heard it. So I became skeptical and started googling around doing my own research. Again, many will have heard of and/or heard the link below but it is very ... well, give it a go. You will need headphones as it does not work otherwise. There are few others like this just google Audio Illusions if you are interested.

 

http://www.moillusions.com/2006/05/shaking-matchbox-holophonic-audio.html

 

So I applaud the work of Peter and Daniel in starting the process of measuring and despair of others who just claim "it sounds better" when "sounding better" is impossible.

 

 

 

[BobH]

I'll get the ball rolling by saying that I count myself as a SOBjective - mainly because the two terms, when applied to listening to music, make me want to cry!

 

Science is completely objective. Art is completely subjective. But neither will have great accomplishment without the influence of the other. Black and white. Left and right. Light and dark. Two sides of the same coin.

 

The two terms, when used to define a stance or belief, are destructive, divisive, pointless and self-defeating.

 

IMHO, of course!

 

edit - that reads a bit 'bolshy' - it's not supposed to. Put your cuddly hat on when you read it, please!

 

[PeterSt]

So Harry, here is the challenge; my challenge to have a nice response to a beautiful topic. Let's try :

 

I think - with audio - everyone wants to be objective, but the first thing you'll run into is recognizing whether you really are, or whether you "fall back" to the more easy being subjective.

I think being objective can't be done otherwise than recognizing the elements being in there, and judging them on their own. Wow, what's that ?

 

Well, to start with a highly suspicious one (and someone mentioned that yesterday, it could even be you, Harry) ... ABXing brings nothing to me, because once a detail is heard in one session, you will hear it in the other just the same. Useless. So, when I listen and judge for a "setting" I always give myself 5 days, and these 5 days seem to be sufficient for gathering the elements which the setting may show. Mind you, I already need those 5 days to play a sufficient amount of different types of music, because they all have - and show their own elements.

On a side note : listening for details is a most dangeous thing to do, because it is not difficult at all to bring forward details. Usually, though, the rest goes down with that (try an opamp and you will know).

 

5 days of listening may bring "more detail" as an element throughout. If 5 subsequent days of listening over and over makes me think I hear something I did not hear before, it is just so. Ha, easy !

But the other way around is not so easy at all. If detail is missing, will I know that I am missing it ? Now, with my example of the ABX, even then that would be hard.

However, from this a kind of mathematically may follow that when I did not perceive any better from it, it may just as well have been worse. And, thinking about this twice, chances are fair that the previous setting (session) was the better one, and if not, it's always equal at least. So, back to that setting ...

 

This sounds a kind of stupid and liveless perhaps, but this is how I work.

I don't work with candles, I don't even recide on sweet spots, and instead I let the music come to me, and actually never compare two tracks at different settings.

If, after 5 days of listening nothing disturbs, that must be a good setting.

 

From here a next dimension springs;

If something *does* disturb, I must know what it is what is doing that. Here the real elements to work with come in, because I will memorize it, and use it for infinit future. Btw, this is not something everybody can do, and it requires a high degree of relative thinking.

As a (far out) example I may refer to the haze Diana Krall may show in her voice. Usually this is a kind of annoying, and somehow only she has it. However, this "element" doesn't seem natural to me, and sounds more like some processing in there, applied to her voice only. Hmm ...

(and keep in mind that you might bump into these kind of things in 5 days time)

Now, one day a setting may occur that removes this haze, but other things are wrong ...

 

This will be the time that I *must* know what is causing the haze to disappear, and now all becomes rather technical. So, this is not a matter of cables or preamps or whatever at the higher level, no, this is about something which filters, smears, and possibly makes a haze out of fine detail. This is such micro level stuff that ... well, it can become a controllable element (changing a cable or preamp is major influencing stuff, and it can't be controlled at all, except for exchanging at that level).

 

The fun is, that such an element can be discovered just by exchanging a preamp, and this would be the typicle situation that the haze has gone, but the bass too. So, what in the new preamp is causing the haze to go, and never mind the disappeared bass (although that can help in finding the answer to the haze question).

 

Without elaborating further, in the end I will know exactly what is causing the haze, and from there on it will be in each equation ever to come.

The more (controllable) elements found, the more easy it becomes to find others.

 

Once a bunch of these elements is known, it becomes the most easy to point them out to others. So, in that case you're not listening to music, but to their comprising elements, and they can be pinpointed easily.

I always have the example of feeling tight bass, while you can't hear it per se. That is, not in the track you're currently listening to, which does not say it is not there; when tight bass is in order, a good vibrating surface will let it feel. You feel nothing (this is just a bass wobbling) tone, or you feel the (large) surface just vibrating at the frequency (like 60Hz); keep in mind, 60Hz can be perceived as a 60Hz tone, but it also can be perceived as a vibrating string at 60Hz; the latter would only occur when it's a nice undisturbed sine, which for that matter is much more powerpul than a disturbed sine.

Of course this can be felt easily at the loudspeaker diaphragm as well.

So this is an element which can be visualized so to say, or is at least prone to better objectiveness.

 

While these elements and many more will help at determination "what is good and what is not", there's the part of : what we hear we must be able to measure.

Well, my standpoint is known I think, and what we can't measure - is ... well ...

WRONGLY MEASURED.

 

Of course this is preceeded by objective listening as how I at least try to do it, but you can bet if a very much audible haze is there or not, we must just be able to measure that. If not, we're only incorrectly measuring. That's all there is to it.

 

If it were that easy ...

 

The longthreaded story of the elements is more important than you think, because when seeking for measurements which may show something, we won't be looking for foottapping music, right ? maybe in the next century we can do that, but that will be beyond my times.

So, we *can* look for elements. I mean, once we found that a haze will be a smear of higher frequencies, we must be able to detect *that*. And so it can ...

 

Now things quickly start to backfire in a positive sense;

Once you can see what causes such a smear, it can be solved by various means. Software means as how I apply them are very indirect and hardly predictable (while in the end it sure can solve things like the haze), but we must recognize that in the end it is hardware doing this to us.

Now things change drastically, because hardware can be influenced easily. Just for that element.

 

As far as I am now all is related to how DACs enormeously flaw. It is good that this is so bad, because it allows for improvement so easily.

So, the first already sprung from this measuring, and I improved my own DAC so hugely ... it is beyond imagination.

 

Well, to keep it a kind of shortish, subjective listening brings nothing (for problem solving or improvement). Only objective listening can, but it needs a strategy first and a line to follow strictly (if you can't - start all over, which in my case would mean a fall back of 3 years or so). You must be honest to yourself too and never think you just "like" something. IMHO it is allowed to be the other way around only : have all those elements there so you like them individually, and *then* lit the candles and have a good evening of listening.

Lit the candles first, a good glass of wine, and find yourself the next day not to like it at all, also is a measurement : you were influenced by mood. This too is a reason for doing it 5 days or so. It just won't take less. Not for me.

 

Thanks for listening,

Peter

 

 

 

[souptin]

That one with the shaking matchbox is amazing! I swear I shivered when it felt like the person shaking the box was close enough to touch me, and I was convinced that they were behind and below (no sniggering please) not in front. Interestingly I have also witnessed a demonstration of the opposite effect, where a blindfolded listener was unable to locate whether sounds made an equal distance from both ears were in front of or behind them.

 

As you folks are no doubt aware, Albert Einstein, as quoted in HarryH's very eloquent opening post, was himself a keen and able musician as well as scientific genius. There are varying reports of his musical ability and I recall reading some rather snooty put-downs of Einsteins written thoughts on music (dare I suggest some professional jealousy?). But one of his close friends described him as playing with great feeling and sincerity. Fascinating, and quite the reverse of the analytical stereotype you might have expected.

 

It's very true that listening to and playing music can be a very emotional and instinctive experience. If you hear some music you love then you still respond to it whether it's a perfect live performace in a beautiful venue or a mangled mp3 heard in passing through an open car window. Well, within certain limits *praise be that my band's 30 minute version of Bela Lugosi's Dead (with only one line of the lyrics) was long before the temptation of youtube fame became possible*. Which reminds me that with art and music it is rarely sufficient to have enthusiasm and emotion alone, without dull, mechanical, repetitive practice. Why the one thing doesn't always cancel out the other I have no idea. Magic perhaps?

 

[HarryHWombat]

Which reminds me that with art and music it is rarely sufficient to have enthusiasm and emotion alone, without dull, mechanical, repetitive practice. Why the one thing doesn't always cancel out the other I have no idea. Magic perhaps?

-souptin

 

Agreed - hard graft, ability and emotion need to go hand in hand. Whoever wrote write something like: writing a book is 10% inspiration and 90% perspiration had it spot on. So the subjective and objective need to go hand in hand, I would propose.

 

 

[HarryHWombat]

Peter hi!

 

Your lengthy reply deserves more time than I have at the moment - it is too lovely a day in London! One thing though:

 

Well, to start with a highly suspicious one (and someone mentioned that yesterday, it could even be you, Harry) ... ABXing brings nothing to me, because once a detail is heard in one session, you will hear it in the other just the same. Useless. So, when I listen and judge for a "setting" I always give myself 5 days, and these 5 days seem to be sufficient for gathering the elements which the setting may show. Mind you, I already need those 5 days to play a sufficient amount of different types of music, because they all have - and show their own elements.

-PeterSt

 

Fully agree! It wasn't me who mentioned the ABX but I was considering a reply in the same manner as yours above. The "audio illusion" I linked to in the original post above I think should demonstrate the flaws to everyone. The first time you listen to the sine-wave modulated version one cannot make anything out. After listening to the unmodulated version then going back to the SW version you can now make out what the guy is saying. So, clearly, that is your brain knowing what to expect and "interpolating" the sound for you to make sense. In the same way, once you hear a cymbal ticking in the background of a track, for instance, you will always hear it through almost any replay mechanism and with any quality of recording. In fact, it would probably be difficult not to hear it.

 

I like your method of listening (the 5 days thing) and have heard others on other forums put this forward as the best way to do things.

 

More later but need to read your post more carefully.

 

Ah - one more thing - and to hijack my own thread. I am trying to mentally figure out what your "Q" settings actually do working forward from what I think is possible. Very interesting.

 

[Abstraction]

I have no interest in audio equipment. I am interested in music. Audio equipment is a tool. It works by way of measurable voltages, measurable sound waves in the air, and so forth. I want it to be the purest possible signal. I want it to reproduce not only the sounds of the instruments but the sounds of the space in which they were played. I want Peter and Gordon and all of the engineers and designers doing everything they can, including measuring and dreaming. Designers, after all, are not primarily scientists, they are also dreamers, not trying to understand Nature, but trying to figure out how to get the nature of things--which likes to make noise--to make highly structured sound.

 

But, of course, the reason for all of this is to produce the illusion that we are hearing music. The important thing is always the illusion. The sound stage is an illusion. It is precisely that illusion that audio equipment tries to produce. We listen to the results of these varying voltages and experience music to which we attribute meaning--the greatest and most immeasurable illusion of them all.

 

Hallucinogenic drugs are entirely chemical in their production and their results, but one might reasonably ask about the quality of the hallucination.

 

[usernaim250]

I have been wading into a few debates lately along these lines. Here's what I see. There are some "objectivists" such as Arny Kruger, Peter Aczel, and Brad Meyer who state that pretty much all differences in modern solid state electronics are below the level of audibility, because they haven't yet stood up to ABX testing. There are many "subjectivists" who believe that if they hear a difference then there is a difference, and any questioning of that is a threat to their ego.

 

Of course the first group could be wrong. You can't prove a negative. The ABX test, as noted above, has an awfully high threshold. Of course if we want objective knowledge, then we prefer too high a threshold to too low. But just because an ABX test has a null result does not mean there is no difference. If the test is competent then a null result does mean that if there is a difference, it is reasonably small.

 

But the subjectivists are just as wrong. Sighted listening is extremely prone to false positives, because expectation sets a filter on our perception. When you hear something in a sighted situation, it is at best suggestive. Particularly among audiophiles, among whom "I hear distinctions, therefore I am" describes their self-image.

 

I too hear differences. But since I have learned more of the history of psychoacoustics, I hear them in a much different light. And when people talk as if their subjective experiences are absolute, it makes me question their judgment. Particularly when those experiences fly in the face of scientific understanding.

 

[Audio_ELF]

At a fundamental level, the whole purpose of HiFi (I use the term HiFi loosely to mean any equipment for the reproduction of music in the home) is to reproduce music. From the start therefore we are looking at something that due to it's very nature is subjective - though even within these choices there are objective points an expert can measure.

 

Do you like Jazz or Rock; Classical or Dance? If you like Classical, do you prefer the compositions of Mozart or Beethoven? Within the Beethoven catalogue, is your favorite his Piano Sonatas or his Symphonies? His 5th of 9th Symphony? Conducted by Herbert von Karajan or Simon Rattle? Performed by the Birmingham Symphony Orchestra, or the Berlin Symphony Orchestra? When listening to music we have so many choices. Even our mood and the environment change the way we perceive music. Most of these are purely subjective things.

 

Before I go on, I'd like to take a step back and define objective and subjective too ... the Oxford English Dictionary defines objective as (1) not influenced by personal feelings or opinions; (2) not dependent on the mind for existence; actual; (3) Grammar relating to a case of nouns and pronouns used for the object of a transitive verb or a preposition. There is also a definition for the word objective when used as a noun but this is not relevant here and in our use of the word we are not interested in the 3rd definition. The definition of subjective is (1) based on or influenced by personal feelings, tastes, or opinions; (2) dependent on the mind for existence; (3) [relates to grammar]. So they are the complete opposites of each other - thought of course when writing we can make some comments that are objective along side others that are subjective.

 

Anyway - back to what I was planning on writing... One thing I see happening is a big change in how we buy audiophile equipment. 10 years ago, you would decide that you were interested in upgrading your HiFi, work out how much you could afford to spend and start planning. For many our first port of call would be special interest magazines where people who's background was a combination of testing of audio equipment (both subjectively and objectively); in many cases some engineering knowledge to examine good design and journalistic experience or at least experience in being able to convert their thoughts into the written word - wordsmiths for want of a better term. The press was full of articles giving a subjective view of the author, usually backed up with some objective evidence from the test suites. You may supplement the written work with advise and recommendations of friends, the emerging internet and other sources. Once you'd identified potential products from written articles, you would most likely pay a visit to your local HiFi dealer or be prepared to travel if there was something less mainstream you were interested in. Once at the dealer, you could at your leisure compare the products you were interested in and maybe find something else they recommended to you, listening to the music you were interested in and making a personal subjective decision. If you were lucky, then you could borrow the equipment and listen in your own home before making such a choice.

 

Fast forward to the current situation. HiFi dealers are dying out. Many people are suspicious that the specialist press is just interested in getting the most advertising revenue. The HiFi market has undergone a transformation. At the low end companies such as Sony and Panasonic aren't interested in the upgradable component systems. The specialist manufacturers have gone from CD with specialist transports; to using cheep computer based transports saying there is no quality loss and (in many cases) back to specialist transport mechanisms ... or they've split from the CD completely providing us with music servers. Many of us (all of us on this forum I suspect) are moving away from physical medium, except as a delivery method, and we're looking for ways to improve our experience with our computer's as the music source.

 

No longer are we concerned with CD player to amp to speakers. Now it's iTunes to USB interface to SPDIF interface to DAC to pre amp to power-amp to speakers...

Or would my system sound better if I bypassed the pre-amp with it's analogue volume control?
Or should I get a soundcard and go direct from SPDIF to the DAC
Or would a high resolution capable FireWire device suit me better
Or should I use MediaMonkey? ...or Foobar? ...or XXHighEnd? ...or Amarra?
Do I need to re-rip everything with dbPowerAmp rather than using EAC?
Am I hurting sound quality using Apple Lossless compression?
Should I install ASIO4All?
Or (in the extreme) should I use a desktop rather than my laptop?
Or (even more extreme) should I use a Maxtor 1TB drive, or a WesternDigital 500GB drive, or a 32GB SSD?

These are all questions we must all have seen asked recently and many, many, more that are just as difficult to answer. Objectively every hard disk has the capability to read the data in a way that will NOT affect the playback quality; yet there are people who have subjectively experienced differences playing the same track from different disks. Objectively a bit-perfect output from any piece of software will give the same waveform from the DAC and therefore sound the same, yet subjectively there is difference using different software.

 

And where do we turn to to get information? The Internet! The place where anyone can write anything. The place where anything, however outrageous, will be believed by at least one person who reads it. No longer is it possible to rely on subjective opinions, though we are needing to because it's just not possible to test the equipment we are wanting to use. Most is just not possible to get a demonstration of, and even if you can demonstrate it, then any changes you experience are also dependent on the exact system combination. Thats where we get to wanting objective measurements. No they don't tell the whole story, but can help us to decide if it's worth trying to beg / borrow / steel one to try out.

 

Of course a set of measurements, or a graph, mean nothing out of their context. Testing, if presented as a means to help us buy something (i.e. marketing) needs to be very robustly done. The people who carry out the testing need to be open. When you ask a question of a measuring technique, in my opinion, that technique starts to sound suspicious if the answer that comes back is along lines of "sorry I can't answer because it's a proprietary technique". (Note ... there is less suspicion on my part if the testing is being carried out by independent tester rather than the person trying to sell you the product). Anyway my point is that objective measurements are pointless without examining how they have been taken.

 

As a final point (maybe or I may think of more to write) I acknowledge myself as being skeptical of claims such as computer components making a difference in sound quality; that there is a difference (let alone a BIG difference that some report) between different bit-perfect software - however I am open minded enough to realize that (a) I don't know everything and that (b) things do sometimes run counter to logic. As way of reference to put my thoughts into context: I am an experienced IT professional with around 15 years of administering networks, assembling computer components, fixing software issues for users, specifying networking components, etc. Prior to that I studied (but acknowledge I didn't graduate) Electronic Engineering and Computer Science at university where I learned about putting together robust testing methods. One thing we were taught from the beginning, is that the final results are not as important as recording how you did the testing so that someone else can later repeat the test and also acknowledging any assumptions you make regarding testing conditions, and acknowledging any external influence. I also learned how minor changes in circuit design / layout can have a major effect on the quality of the end result / signal. Several small / minute changes can end up giving a large variance of results, but independent of each other are unmeasurable.

 

Well I've written far more than I actually think anyone is interested in so will finish up at this point.

 

Thanks for reading

Eloise

 

 

 

 

[Audio_ELF]

We need both objective and subjective testing.

 

Subjective is personal - if you like the results with a piece of equipment then who cares what anyone else said. That highly distorting valve pre-amp just might give your system that warmth you want.

 

Objective is without bias - if you're trying to decide if it's worth the cost (time, money, etc) of trying a piece of equipment or software, it's useful to know if it actually IS going to be different from your previous thing. Test figures don't mean everything, but CAN give you some idea if interpreted correctly.

 

Eloise

 

[PeterSt]

But, of course, the reason for all of this is to produce the illusion that we are hearing music. The important thing is always the illusion. The sound stage is an illusion.

 

No, this is not true, although you won't find many that say/claim what I say;

 

A.o. Harry referred to the 3D thing, or IOW, music through two loudspeakers can be perceived not only left/right, but also up/down, back/forward. He said (or whoever says it actually -> I did not follow the links) that this works with headphones only. This is not true ...

 

There's a high chance I will be talking abacadabra, so sorry then ...

 

All is about phase. Or better : phase difference;

We don't hear left/right because ther'e a loudspeaker on the left and on the right, but because the distance to our ears the music travels, makes the phase between left and right to differ. This is how we hear 3D, and we do that with 2 ears only. The latter is key, because you'll have to believe at least that (because you know it is so, never mind why yet).

 

Our GPS receivers do their job by the grace of at least 4 satellites all being at a different distance, and from there your exact position can be calculated. Yes, 4 and not 2 satellites ...

 

The technical way the satellites calculate the position, goes by phase shift measurement. The waves used are rather long (11m IIRC), therefore wave angles are not steep, and that is the reason the accuracy is so so. However, add satellites (transmitters) and the accuracy gets better (never mind why please).

 

Our ears behave the same, but it kind of works the other way around. There is one point source transmitting (a bird) and with our two ears we can hear where it is. This is because one ear has another distance to the point source than the other ear, and no matter how small the differences are, by means of phase shift "calculation" we know the direction. If we turn our head a bit, that's taken into the equation, and we still know where the bird is.

But, roughly, and the further away the bird is, the more imprecise we know where it is. And this is not because of wind, air molecules or smearing otherwise ... this is just because of physics ...

 

Before I proceed I'll try to give some confidence on my knowledge about the subject, because I have been working for two years or so on a GPS system for 300m2 which btw would be called a Local Positioning System, LPS. I stopped when all theories and simulations were worked out, and the implementation had to come (which is a million+ dollar investment), but I can tell you the system has an accuracy of 0.1mm which is a 100 times better than anything existing (which gets to around 10cm).

The mathematical formulas to do it sit on over 10 A4 pages, and as we found (I did this with another guy, he making the formulas) only 3 people in the world could do these hyperboliod calculations needed for this job.

On a side note, this system works with lightspeed waves (and very short wavelengths), meaning very steep slopes of the waves, allowing for the precision.

 

Back to the subject, such a system works with receivers with one antenna (the receive is a single source point -> could be you with TWO ears, see below) and transmitters with any number of antennas you like.

In our case we have two antennas ... (two loudspeakers)

 

When there are two transmitters (loudspeakers) and one receiver, and the receiver is able to calculate the phase shift between those two received waves, you can graph the possible locationS of the object with the receiver antenna.

 

I said locationS because this won't give one unique point. However, it will give the possible points, and they are on a hyperbole for 2D. Here you can see that :

 

 

 

Notice the white circle, which shows the real object position, and which is on one of the lines.

 

From this follows that when one frequency would be sent from both the loudspeakers, you would be able to "see" that frequency at the positions indicated by the lines you see above you see above.

 

In between the lines something some of you can try yourself :

If you have fairly directional loudspeakers, like a horn would be, you will find that being on a proper distance like 7 meters or so, there's a sweet spot in the middle, but also some 3 meters at each side of the middle. Same principle ! (just look at the lines above).

What happens here is that at those 3 meters aside, all waves (within a frequency range) have shifted 360 degrees (180 for the left speaker and 180 for the right) and all is okay again, and the only difference is that you will be perceiving the sound from the farthest speaker 1 wavelength later opposed to te closest.

 

Now, working with two wavelengths instead of one in my project is better, because one gives less unique points. Mind you, the wavelengths used are very specific, and not all two random wavelengths produce the same "thin" lines. Here you see the result of that :

 

 

 

But now we add a third wavelength ...

 

I'm afraid that things are too complicated to explain, but take it from me that a third wavelength again brings more accuracy, just as an additional transmitter would. So, talking about additional transmitters :

3 transmitters are needed to create a unique 2D space, and 4 transmitters are needed to create a unique 3D space (now you know why GPS needs 4 satellites at least, btw knowing that GPS also shows you the altitude at 5M accuracy).

 

What can be done with an additional transmitter (and I mean including its location opposed to the others) can not be achieved with additional wavelengths. To get this a little, think like this :

A wavelength of 10cm is half of a wavelength of 5cm. This means that when both are transmitted from the same antenna location, each two "cycles" of the 5cm wave, the 10cm wave is exactly met (on whatever position of the slope). When this would be used, the math wouldn't know the difference between each exact 5 and 10 cm (in practice this is a bit different, but never mind).

Look again at above picture, where you see the opposite of this graphed : the empty spaces in between the lines you can see here and there, are more unique than the shown line positions, and the shown line positions will be the matching 5 cm points (note that a line is wrong and ONE dot is the thing to achieve).

 

Now, think upside down a bit, and try to imagine that our object in the picture is not a thing with an antenna, put is the repositioned instrument by two loudspeakers, and on that one location only, all the frequencies carried by the instruments are in exactly the same relative phase differences as how they were captured by the microphone, and for easy thinking your ears being at the position of the microphone.

Because at that point only there's a cross point of frequencies originally springing from that (now virtual) location, the sound will be louder there, equally to the working of standing waves.

 

For fun, this is a picture of one frequency through two loudspeakers, and were it all can be according to phase shift calculation :

 

 

 

When you're all still there, now it is time to really pay attention ...

 

Each additional wavelength used at the transmitter brings more accuracy. But, it also brings a progressively larger amount of formula to work with, so no practical solution for the LPS there.

But the formula is in our brain ...

 

Wavelength is the same as frequency. Aha, many frequencies are transmitted from our loudspeakers at playing music through them. Hmm ...

It looks like we are getting somewhere.

 

This is what comes from it when a sufficient amount of frequencies is used (and try to see the 3D) and in this case the object is uniquely identified by crossing lines, each line representing a frequency of which its position can be calculated from two antennas. In practice : there is no other position to be found where all frequencies meet with the same phase shift.

 

 

 

Now think further : the playing instrument will let spring all its frequencies from the same point (but see below) and only at that point the phase shift is exactly equal (namely : 0).

 

So far, I left one "dimension" out of the story, and it is of utmost importance : the size of the object.

This is related to what I just said "all its frequencies spring from the same point". This is not true !

A violin not only makes sound through the strings, but also through its cabinet. Many frequencies spring from it, including (and mostly) non-sines. But one thing remains : from wherever the sound springs, its phase is 0 when it springs.

 

Looking at the last picture above, we see rather thick lines. This is the accuracy inherent to the measuring system and which is also related to impossibility of measuring phase difference when the wave crosses full plus or minus, when the angle of the wave is "near horizontal" for a longer period, and phase shift rounds to zero for a short period of time. However, try to see that our object is as large as the white circle, but the lines of accyracy being thinner, and as long as that is so, the object can be localized.

 

Now try to get the both dimensions "accuracy" and "object size" together for audio ...

And first imagine someone singing with a 2meter wide mouth, vs. the normal smaller size ...

(I'm sure many will have (had) the bad experiences of music making objects being too large)

 

I know by experience of the project, that at using so many frequencies (springing from one instrument) it can easily very accurately be pinpointed in the 3D space with two transmitters only. But in our audio practice it is not so, or otherwise it is less or more, but far from how it should be in theory. Why ?

 

The frequencies sent by the instrument through loudspeakers are way too vague, or better : way too much off from how it should be. The wave, which should be an accurate longitudinal 3D wave travelling through the air in your room, is not accurate at all. It is a wobbling thing.

Btw, this is *not* because of reflections, because your ears are perfectly able to deal with "early reflections" and they count. The other reflections just create the room (size), and there is nothing wrong with that.

 

For the remainder of the story I'm afraid you have to take a couple of things from me, already because I am myself in the first stage of this "audio measuring" only.

 

a. That indeed waves can be thicker and thinner is proved by more accurate playback avoids standing waves (and I mean completely, and in all frequencies). -> The waves get thinner, and can't touch (add up) as easy.

 

b. That waves are off by miles in the first place, can be derived from my latest DAC measurements, the analogue output being nowhere near to what goes in.

More importantly, it is not only miles off, it is also "slow". Slow means the real transients (hence waves) cannot be followed, and before a peak is reached, the wave has to go down already because the next sample tells it to go down, and it distorts. In the end nothing is left from the reality as intended. This is "wobbling".

 

It is important to combine a. and b., because where a. already works in the bit perfect domain (so nothing is changed to the bits, but output can get more accurate because of "influences"), I can easily see that this is maybe 1 promille of the work to do only.

 

So ... assuming that the recording was accurate and put through all the way to the disk (CD), once we are able to play back more accurately opposed to the lousy job we're doing right now, we will be able to hear in 3D space from two loudspeakers.

Keep in mind, this already happens, but the line of accuracy currently is too thick. For left/right and the width of an instrument we can fairly easily get there, which is because the prerequisites we put to ourselves at listening distance, distance of the loudspeakers and some more stuff, and of course the two antennas far better create the 2D space in a horizontal plane just because of the physics and phase calculation thing as explained. The other way around ... might you think you have a huge depth ... close your eyes and try to locate the one single instrument *not* related to the others, in the depth dimension; you can't and it is everywhere (from the top of your head to the end of the room). Why ? well, same reason as the sweetwpot is allowed to move further and closer ... at all those locations the line in the middle is in, are in proper phase. This works if you move, but it also works (thus not) when the instrument plays. On that line in front of you, it is located everywhere (look at the first picture above).

 

To visualize what I just said ... earlier pictures assume an inherent accuracy (of, say, the DAC) of 100%. Below picture however, shows reality when the phase shift compared to the actual waves is 1 degree off (which comes down to 1/180 btw) :

 

 

 

This is a width (depth for our audio purpose) of 5 meters ...

... and not much is left from localization possibilities at all.

But with current DACs this is our practice from today.

 

Concluded, not so much illusion at all but just physics, but loads of imagination needed to get some illusion of how things maybe later will be possible.

 

Peter

 

 

[Audio_ELF]

Abstraction wrote...

But, of course, the reason for all of this is to produce the illusion that we are hearing music. The important thing is always the illusion. The sound stage is an illusion. It is precisely that illusion that audio equipment tries to produce. We listen to the results of these varying voltages and experience music to which we attribute meaning--the greatest and most immeasurable illusion of them all.

and Peter disagrees, but I think Peter has missed the point...

 

In his post, Peter starts giving technical reasons as to why sound can be placed within a space (soundfield) when you are dealing with waves coming from two fixed points. All very interesting, but ultimately irrelevant. Most of us, listen to music for the enjoyment ... we want the illusion that Abstraction referred to to of the musicians actually playing for us. We enjoy trying to perfect that illusion, but don't care how that illusion is created (I think).

 

We ask for objective measurements, not because we want to listen to what an output with less jitter sounds like, but because we are trying to reassure ourselves that we are spending money wisely. Ultimately (for most) the music is the enjoyment, not understanding the hardware.

 

Eloise

 

P.S. on another note ... it is my understanding that GPS only needs 3 satellites because it is giving something a position in 3 dimensions - X, Y (on the ground) and Z (height above sea level). Additional satellites are then used to help to improve the accuracy of that fix.

 

[HarryHWombat]

A.o. Harry referred to the 3D thing, or IOW, music through two loudspeakers can be perceived not only left/right, but also up/down, back/forward. He said (or whoever says it actually -> I did not follow the links) that this works with headphones only. This is not true

-PeterSt

 

Yes I did say that but only because that is what the site itself says - I have tried it without headphones and it didn't work so well if at all.

 

Peter - you cram so much information into your posts they usually need three or four reads to get a grip on what you are saying

 

I did some google based research on how two ears can locate sound in three dimensions. Two interesting things came out - firstly, directional 3D hearing also uses the skull in that the sound waves travel through the skull and are recognised and processed by the brain. For example, a sound source directly above your head also passes through the top and centre of your skull - this information is used by the brain to locate the sound.

 

Secondly, those wiggly bits of cartilage inside your ears also help to locate sound by slightly frequency modulating any received sound waves - can't find the link for this. A researcher was able to cast someone's ears and then use these as stereo microphones. When replayed through headphones the person was able to accurately pinpoint sound in three dimensions.

 

 

[HarryHWombat]

And where do we turn to to get information? The Internet! The place where anyone can write anything. The place where anything, however outrageous, will be believed by at least one person who reads it. No longer is it possible to rely on subjective opinions, though we are needing to because it's just not possible to test the equipment we are wanting to use. Most is just not possible to get a demonstration of, and even if you can demonstrate it, then any changes you experience are also dependent on the exact system combination. Thats where we get to wanting objective measurements. No they don't tell the whole story, but can help us to decide if it's worth trying to beg / borrow / steel one to try out.

-Eloise

 

Very much agreed! Internet forums are also extremely powerful marketing tools for people trying to sell us equipment - we have seen it here and I have also seen it on every other forum except for manufacturer-based forums which, to an extent, exist to help the community or market for that product-set. However much one trusts the manufacturers and retailers on forums one must remain skeptical as DAC manufacturers are trying to sell you DACs, software guys software and jitter-reducers, jitter-reducers. "Buy my product because it sounds better" just does not wash any more and any manufacturer or retailer who doesn't offer some sort of proper audition simply has to come up with objective backing for their "sounds better" claims. Also note that I do not think anybody is being a nasty money-grabber it is simply that they cannot all be right (maybe they can in some cases but in many there are competing claims which are mutually exclusive).

 

 

 

[One and a half]

A gymnast, to perfect their two and half somersault twists, needs to practice, over and over the same sequence, so that the brain begins to learn the shortcuts needed to make this happen. It never happens first go, but over repeated trials, the desired goal can be achieved. The limiting factor is the brain.

 

If we were to apply the same anlogy to hearing and making comaprisons, and make a subjective opinion on it, we would have to listen to the same music piece over and over until we are probably sick of it (and then it's not much fun). Can the brain be so perfect as to retain a piece of music, with correct frequency repsonse, direction and gasp *jitter detection*?

 

Human beings are designed to detect changes very well, it's most likely where we are today because of it. When it comes to long term memory (ie more than a minute for some us...?) the recalling of information does take some time, but is it bit accurate? Very difficult to quantify. It's easy to recite poetry, it's fixed bits like a word processor, but an audio file like an AIF is very complex to recall bit by bit. Mind you, I had an LP of the first ELP album, and in "Take a Pebble" there was a scratch in a very quiet part in the guitar solo which irritaded me for years. To this day I expect to hear the loud scratch when playing the same piece of music on noiseless CD.

 

From what I read of speech for example, if a young girl says "hello", we can distinguish the same meaning when an elderly persons says the same thing. For us the *implication* of the word is the final result to get the meaning because our audio perception is all about getting the message quickly, and with some fuzzy memory recognition working in the background (the brain), yes that does mean a hello to us.

 

Subjective audio critique is just that, an experience that is personal, difficult to prove otherwise, since the device responsible for the objectivity isn't made from copper and silicon, but rather synapses and neurons. Somehow there needs to be a balance, and luckily this forum manages the two "opposing" viewpoints with a degree of positive result. No one is right either way, and as many posters put forward, there's needs to be a balance without reducing the enjoyment of music. This is possible , never easy though, with healthy debate as a prime foundation.

 

More reading : http://brainconnection.positscience.com/topics/?main=anat/auditory-phys

 

 

[omahapianist]

Interesting thread. I would like to throw a knuckle ball into this arena and discuss "the absolute sound" (the concept, not the magazine). In theory, there is what we know as "the absolute sound", that is, an acoustic sound (single I assume) that exist in an open space, the parameters of which have never been fully disclosed to my knowledge. This is not to be confused with "an absolute sound" which would indicate more than one possible acoustic sound within more than one type of open space (which actually defines our existence as audiophiles in the real world).

 

Now, here is where things gets a little fuzzy: because of our inherent inability to fully understand the enormously complicated, yet deceptively simple concept of what truly makes up the elements of "the" absolute sound, we cannot even begin to understand the process of reproducing it accurately. Sure, we can break down what we hear in so-called audiophile terms but the thing that distorts it all is our varied (and flawed to a point) ways of describing what we are hearing. Our hearing mechanisms pretty much work the same, biologically, but explaining what we hear and our ways of re-creating what we hear are many.

 

This leads me to believe that manufacturers of audiophile equipment, as well as those who try to capture a live event (in a venue or recording studio), can only attempt to re-create what they believe is "the absolute sound" as it registers on their psychoacoustic mechanisms. In other words, when we listen to a piece of gear, or a recording as realized by its' engineers, we are only listening to their re-creation of the event which is as far removed from the actual event as you can possibly imagine. Whether you have a $300.00 mini system or a take-no-prisoners $300,000.00 "I-Have-More-Money-Than-Sense" set-up, it can't possibly re-create the actual event! I know, big news, right? However, the illusion of that event that a system can provide is powerful, albeit totally inaccurate.

 

We could very well agree that an acoustic guitarist playing in an open space (Carnegie Hall, my living room, or a corn field) is a representation of "the absolute sound" for that environment at that moment regardless of who the guitarist is or the type of guitar that's being played. Or maybe we, in our current state of evolution, are incapable of universally agreeing on what "the absolute sound" is. If we all agreed on what it is, we would in theory, have a single manufacturer of a single component capable of re-creating the event that would garner universal acceptance. This concept would make discussions of "tube vs. solid state", "analog vs. digital", "objective vs. subjective", etc., completely and absolutely moot. Now, I realize that the pursuit of an actual re-creation of a music event is the holy grail of our hobby and that's what manufacturers are trying to strive for but with so many different interpretations and applications of this "holy grail", it's impossible to realize this vision.

 

So what's the default in this case? Illusion. Illusion is everything. This is one reason I believe a blind evaluation between two or more musical components is pointless because it can't possibly tell you which is better or worse than another because "better or worse" cannot exist when there is no universal agreement on a true re-creation. All this does is teach you how to evaluate sonic differences between similar products. That's it. It ultimately boils down to what is individually appealing and what's not. You like component 'A' and not component 'B'. Component 'B' over 'A'. "Hey, I can't tell any difference!" Introduce opinion and conjecture and we've opened up a Pandora's box (or maybe that's a DAC box to see if it has the "right parts" in it!) :-)

 

Everyone is right and everyone is wrong. Whatever you believe, enjoy the illusion!

 

Randall