I don't think that's true. The reason is that humans jump to judgments - they are seldom rational. Read the book by Prof. Kanneman about "judgment" and how it is dominated by bias - always.
Some people may come to a new acclimation quickly, others may take some time and some may even never do it. Take the human judgment out of the picture and you have a better chance of coming to acclimate to what is accurate.
exactly and its quite normal. its hard to be rational or objective toward one own diy speaker where if your not satisfied will you need to spend money and time to try to fix it. Often im sure many people ends trying to convince themselves that what they built is good enough:" hey, as long as I enjoy listening to those speakers!"
one of a very important aspect of diy'er is to have a known reference speaker. a speaker that is critically acknowledge as good and so you can compare all your diy speaker against those speakers.
my JBL's LSR32 would fit the criteria. you can grab a pair used for 800$.
the only way to be objective is when theres nothing at stake.
hence why many people find that their wife or friends are great judge of their system.
So, it depends on what we're going for. Designing/building things entails its own enjoyment/satisfaction, even if the ultimate result is "suboptimal". It's nice to know that we have these "generally preferred criteria" so we can use that as a goal/target. $800 is more than my entire system's cost so not exactly in the books, and, frankly, would mean I'd work/play with something else.
Wife/friends are highly biased subjects, too. Depending on how your relationship with them is going, of course.
You need a good speaker on which you can compare it to your diy build. once satisfied with your build, you can sell the reference speaker.So, it depends on what we're going for. Designing/building things entails its own enjoyment/satisfaction, even if the ultimate result is "suboptimal". It's nice to know that we have these "generally preferred criteria" so we can use that as a goal/target. $800 is more than my entire system's cost so not exactly in the books, and, frankly, would mean I'd work/play with something else.
Wife/friends are highly biased subjects, too. Depending on how your relationship with them is going, of course.
of course if your building only for fun, no problem with that.
but I like building my 3 way speaker and be able to compare it with a known design. once my built kick the jbl lsr32, I'll know im in the good enough department!
Maybe we should test our sound systems on dogs, with cat sounds... That would be honest and unbiaised: the first speaker to get eaten wins.
(i'm half-serious... but slowly sliding on the serious side.. it's friday.)
I don't think I need to.
If his theory is correct, my judgments of his book would be invalid.
If his theory is wrong, his book is a waste of my time.
You are obviously have no clue who you are talking to. Why do you keep claiming to know what I have and have not done, what I do and don't do? You are so wrong every time, I have to wonder if you keep mistaking me for someone else.
Bottom line, anyone who can't build a well balanced speaker by ear, anyone who has to resort to measurements, is not a real speaker designer. He's a wanna-be. No need to pussyfoot around the issue.
I fall into the wanna-be category. I have to cheat. Less and less all the time, but I do measure.
There are real designers who can do it by ear. They can achieve their targets without the measurement crutch. How do I know?
- I've listened to their work.
- I have measured their work. It's within a dB or two of claimed target response.
I have to agree with Pano on this one.
Back in the 70s my dad worked at Goodmans (during the time they had the rights to produce the LS35a)
He built his own monitors as a side project, using the B110, and T27 chassis and coil.
Many of his colleagues used the mathematical iterations to design based on measures alone, no listening took place.
My dad continually frustrated them by simply guessing an appropriate box size, crossover point and suitable component values, based on rules of thumb; often bettering his colleagues in half the time.
His LS35a was nearly double the volume of the BBC design, no hump at Fb, and he made new diaphragms for T27 as he didn't like the sound of mylar.
Having heard both side by side (only once before he passed) I can say that his beermat calcs and approach actually bettered what was a good design.
Back then plenty of speakers were voiced by ear, and plenty of those speaker were damned good.
Back in the 70s my dad worked at Goodmans (during the time they had the rights to produce the LS35a)
He built his own monitors as a side project, using the B110, and T27 chassis and coil.
Many of his colleagues used the mathematical iterations to design based on measures alone, no listening took place.
My dad continually frustrated them by simply guessing an appropriate box size, crossover point and suitable component values, based on rules of thumb; often bettering his colleagues in half the time.
His LS35a was nearly double the volume of the BBC design, no hump at Fb, and he made new diaphragms for T27 as he didn't like the sound of mylar.
Having heard both side by side (only once before he passed) I can say that his beermat calcs and approach actually bettered what was a good design.
Back then plenty of speakers were voiced by ear, and plenty of those speaker were damned good.
I accept that there might be the rare exception and that someone could design a good loudspeaker by ear. Even my own test results from Ford indicate that. But as those results also indicated 80% of the "golden ears" could not do that as they were not even capable of consistent judgements in controlled tests. I would never trust my ears over a measurement, so why not just use the measurements? It eliminates the uncertainty surrounding judgements.
And no one can deny that subjective judgements of our own designs are highly biased, just as they would be biased against those of say a competitor. These are simply facts of life regarding human judgement.
And no one can deny that subjective judgements of our own designs are highly biased, just as they would be biased against those of say a competitor. These are simply facts of life regarding human judgement.
Ok I shall back track a little.
Golden ears are somewhat of a fantasy.
Trained listening is different. I am somewhat trained, but there are others better trained (loads of them), none of them have golden ears.
I can sing E2 quite well, without hearing the pitch first. But I do not have perfect pitch (more a internal resonance I can hear when I hit the correct note)
So I most certainly don't have very well trained ears, but I can recognise certain deficiencies in sound (I guess experience helps)
I can hear something in a driver and think "there's something not quite right" and usually I can narrow it down to the octave which is affected. Sometimes it can be obvious there is a amplitude issue, or THD, others it isn't obvious other than "something isn't right".
This is where I would use measurement, to add a definition to what I hear as "wrong". Currently I measure nothing, other than TS parameters, but I do plan on tackling REW/ARTA/Holm at some point.
(I state this as someone with plenty of test and measurement experience; none of which involves acoustics)
Golden ears are somewhat of a fantasy.
Trained listening is different. I am somewhat trained, but there are others better trained (loads of them), none of them have golden ears.
I can sing E2 quite well, without hearing the pitch first. But I do not have perfect pitch (more a internal resonance I can hear when I hit the correct note)
So I most certainly don't have very well trained ears, but I can recognise certain deficiencies in sound (I guess experience helps)
I can hear something in a driver and think "there's something not quite right" and usually I can narrow it down to the octave which is affected. Sometimes it can be obvious there is a amplitude issue, or THD, others it isn't obvious other than "something isn't right".
This is where I would use measurement, to add a definition to what I hear as "wrong". Currently I measure nothing, other than TS parameters, but I do plan on tackling REW/ARTA/Holm at some point.
(I state this as someone with plenty of test and measurement experience; none of which involves acoustics
)
Last edited:
The problem is that it is so easy to be wrong and I have seen this so many times that I just don't trust ears anymore.
For example, I spent a decade or more trying to understand nonlinearity in loudspeakers because I thought that it was a significant factor. I heard it just like everyone else. I believed my ears. But yet when it got down to actually doing real scientific tests of audibility, low and behold it isn't really even a factor. (I have written many papers on this topic.) I can tell you everything there is to know about how distortion is generated, where it comes from, etc. but until I actually did the tests to determine how audible it is I was just kidding myself. Now I know better, but I just wish I had not wasted so much effort on a problem that is so insignificant. It did lead to other more significant and more audible problems like diffraction, so it wasn't a complete waste.
For example, I spent a decade or more trying to understand nonlinearity in loudspeakers because I thought that it was a significant factor. I heard it just like everyone else. I believed my ears. But yet when it got down to actually doing real scientific tests of audibility, low and behold it isn't really even a factor. (I have written many papers on this topic.) I can tell you everything there is to know about how distortion is generated, where it comes from, etc. but until I actually did the tests to determine how audible it is I was just kidding myself. Now I know better, but I just wish I had not wasted so much effort on a problem that is so insignificant. It did lead to other more significant and more audible problems like diffraction, so it wasn't a complete waste.
Many people try to change the truth. In the past few years I have happened upon several "objective" audio sites where their local "experts" try to downplay scientific facts to suit their own beliefs and expectations. The objective stances had dissolved into arrogant arguements because they had stopped learning long prior. It's harder than ever to tell the real science from keyboard warriors intent on self aggrandizement.youknowyou said:
Last edited:
Back on topic gentlemen please
If all subjects participating had expereiece in hearing a good high quality sounding reference system would their opinions be different??
the choice offered is A or B one or the other not (no they both sound crap lack bass and dynamics no treble) but statiscs say pick A or B
Asking individuals wether A or B is better in a two horse race is maybe easy by comparison the tester controlls the situation.
I hate statistics
UB40 Sang a song about statistics
I am a one in ten
A number on a list
I am a one in ten
Even though I don't exist
Nobody knows me
But I'm always there
Statistical reminder of a world that doesn't care.
(maybe change the last lyric line for "A WORLD THAT IS JUDGED UPON STATISTICS THAT DONT COMPARE TO REALITY")
try reading about Simpsons paradox
I liked this one Pano ! Sometimes it's even worst than a wannabe ! ... a want to be !
(Un "Je veux être" if you remember your french !)What you say is true, for all the hifi stuffs I believe ! Experiments are just a must to do for Learning... and sure theory is necessary to learn and understand the cons and poors of the try & errors ! (Argh I know it to miss of the tech understanding part !)
Last edited:
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.