|
Audio Asylum Thread Printer Get a view of an entire thread on one page |
For Sale Ads |
211.27.151.237
In Reply to: Speaker Measurements vs. Subjective Listening posted by lenw on February 18, 2007 at 11:26:57:
My wife's statistics lecturer once told her class "Statistics are like a bikini. What they reveal is interesting, what they conceal is vital." Test results are like that. They tell you a lot about the parameters that were tested and they tell you nothing about what wasn't tested. Anyone who believes we now can test for everything is a fool. Anyone who believes that any test report includes every test we can currently make is overly optimistic. All test reports, regardless of whether they be for speakers or for any other component, are incomplete.That doesn't mean that test reports are useless. If something comes up well on the test report and the array of tests conducted is well chosen, that's encouraging. If it comes up badly, that's reason for caution,
But in the end we listen to music for enjoyment. We want a system that lets us enjoy the music we play. I have yet to see an objective test for a component's enjoyability and there won't be one because different people enjoy different things.
Don't buy anything you don't enjoy listening to, but use the test reports to help you sort out which components are better designed/built and or more suitable to your needs.
Follow Ups:
I am at heart an objectivist but understand in the real world,measurements done well have significant value but they have limitations so other tools such as ears and a brain are needed to finish the job.
The again I have a friend who has designed speakers both for himself and as a hired consultant, I've watched him design a speaker on a computer using the specs on the drivers(usually checked by his own measurements on the drivers - sometimes necessary)to design the boxes for the drivers and the crossovers. And the speakers for a couple of years now have been 95% done before any listening was done. And the revisions for the final 5% were for subtle balance tuning and checked again on the computer. Now there were mainy years of experience that also went into the computer manipulation. But that's a lot of objective approach in the design of a transducer.
I've got no problem with that approach to design. One of the things that science is really good at/for is doing something better and there's nothing wrong with using whatever knowledge and technology you have for a design. In fact there's something wrong with not using what's available to you.Using knowledge/technology doesn't guarantee that a given design will work in the flesh, but it increases the odds immensely and, as you comment, it is quite capable of reducing the amount of "fine tuning" required.
The problem with tests in particular, however, is that not only are there gaps in what gets tested but also people tend to interpret results in the light of their particular preferences and favourite design styles. Some people like sealed boxes, some like bass reflex designs, some like horns, some like panels. Give 2 people with different preferences the same set of measurents and you're pretty well guaranteed different interpretations. We don't all agree on what is a good speaker and there's an awful lot of options around. There's an awful lot of ways to interpret the same data and they're all valid from their own point of view.
More problematic is the fact that there's no 'law of physics' which says that we have to like whatever measures well, and dislike what doesn't. Measurements tell us a lot about what a speaker can and can't do, but there's no measurement which says that we'll like a particular speaker. We can, of course, often guess that we won't like a speaker which measures badly on some things but we don't always like speakers which measure well.
At the end of the day, as I said, this hobby is about enjoyment and it doesn't matter how well something measures or what other people think of it, it's not a good choice for you if it simply doesn't float your own personal boat. We can't dispense with listening.
You're correct about having fun and liking what you hear. But to have a discussion about the hobby you have to first agree what the hobby is to you since there is at least one or two other approaches, fidelity to live sound(hard to really define in any objective way I know) or fidelity to the software which is my definition.
There's probably a few more approaches than just those 2 you mention, but they are all just personal preferences and personal prefernce plays a huge part in what we like and enjoy.The bottom line is still going to be the fact that if you don't enjoy listening to a particular speaker or component, then you simply shouldn't buy it, no matter how well it measures. Enjoyment is both personal and subjective and there's nothing wrong with passing up something you don't like in favour of something you do. Even within either of those 2 approaches you mention, you will find different items vying for your money and some of them will appeal to you while others don't, even though each may be equally well regarded overall. It's just that we tend to like something that performs well on the parameters that are important to us, and 'hides' its compromises in the parameters that are least important to us. Listen to something else that overall measures as well, but has some of its strengths in the areas that are less important to you and a weakness in something that is important to you and you'll certainly prefer something else.
I'm not sure audio product reviewers make it clear when something doesn't test well or give plausible explanations as to why they still rate the product highly? If a given speaker's measurements reveal it does not have a relatively flat frequency response or other problems, will that speaker be capable of providing a rewarding listening experience over the long term?
s
This post is made possible by the generous support of people like you and our sponsors: