If polar response was what makes speakers sound different, they would all sound identical if we put them outdoors then equalised the response on axis. I suspect this wouldn't be true so I believe that polar response is unfortunately not relevant
-
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!
Polar response is not the key
- Thread starter Audiojim
- Start date
Just curious how you wave away the enormous mountain of evidence presented by Toole and Olive? Where's your contrary data?
Well, you would also need to suspend them in the air too to avoid the floor bounce, but sure. An anechoic chamber would work too.
Oh I would very much suspect that it is true (assuming non-linear distortion is kept at reasonable levels). If you believe otherwise, then please explain why.
There are mountains of evidence showing the opposite, most of which is summarized in this book by @Floyd Toole. It's a highly recommended read if you want more clarity on these matters.
- Thread Starter
- #4
I haven't waved away their evidence I haven't looked at it. What evidence is there to refute that if you put different loudspeakers outdoors, there would be differences in sound once on axis response is normalized?
Polar response is not relevant outdoors because there are no reflections. I trust that you agree with that. Indoors it starts to matter. However, that doesn't mean there is nothing else that could be causing differences in what we hear does it? Cabinet resonances, driver materials, even crossover parts and time coherence have the capability of being factors. Has Toole examined these? How old is the research done by them? That can affect the reliability of the results given the advances in technology.
Perhaps it might be worth doing before starting a thread like this.
Polar response isn't relevant outdoors provided the loudspeakers are on a 10m high pole, on soft ground, and you're listening sitting on a similarly high pole. If that's your normal listening environment, then indeed, polar response doesn't matter.
For all other listening environments, Toole's book is relevant.
S.
Last edited:
Get back to us once you’ve read the basic literature.
That's a weird way to approach the question. I like to think that in audio, the burden of proof is on the one making the claim that something is audible, not the opposite. In other words, it's inaudible until proven otherwise. Going at it in the opposite way will not get you very far, because it means I could sell you "magical" acoustic stones (or any other kind of voodoo) and you would be forced to admit that you would buy them because you couldn't find any evidence that they don't improve the sound.
Just a side note: just because polar response cannot account for all perceived differences (which, again, is an unsubstantiated claim) doesn't mean it's irrelevant. Subjective perception of audio is not an exact science. I don't think anyone here would claim that two speakers having identical on-axis and off-axis response would be indistinguishable, always, in all circumstances, all program material, and all listeners, period. However, I would make the claim that such speakers would be pretty difficult to tell apart in most circumstances, and that's really what matters in practice.
"Cabinet resonances, driver materials, and crossover parts" are irrelevant for this discussion, because any effect they have on the output of the loudspeaker will be reflected in the overall speaker measurements (i.e. the end result, the physical characteristics of the sound that comes out of the speaker). In other words, what matters is the sound waves that reach your ears (and can be measured); you don't directly "listen" to a crossover or a driver material, which are only means to an end.
As for time coherence, it would help if you clarified exactly what you mean by this term.
"Nobody would buy Magico or Wilson audio" is a statement that assumes all consumers are perfectly rational, never fall prey to marketing or hype, conduct proper research (including knowing how to find and interpret studies), have perfect information about the product they're buying (e.g. detailed speaker measurements) and have audio quality as their only criteria, excluding aesthetics and other factors. I don't think such assumptions are reasonable, especially in the consumer audio market.
Last edited:
I suspect we’re not being asked a serious question. I would suggest waiting until one is proffered before responding further.
- Thread Starter
- #12
Going the other way, you could sell me a wooden box with cheap drivers in them and claim they are just as good as a Magico, as long they have similar frequency responses.
I tried to explain what I meant by irrelevant. Polar response is constantly being mentioned but what I'm saying is it may be a red herring if there are other factors that are hidden.
The idea that it's inaudible until proven otherwise prejudices the outcome. If Toole's research is based on such reasoning then it can't be trusted.
Yes you just claimed that any supposed differences are inaudible until proven otherwise
Maybe somebody else could as I'm not exactly sure. It's obviously a well known term that is used by some audiophiles.
So you actually believe that all the Magicos and revels would be just as good as a budget priced speaker?
Oh absolutely. In fact, you could sell me such a box, and I would gladly buy it, too. What you're describing matches my decision process for recommending and purchasing loudspeakers fairly well.
Oh, but we do have numbers on this. This study shows it is possible to predict the subjective performance of a loudspeaker to a fairly high degree of accuracy (specifically, r = 0.86 over 70 speakers) using purely on-axis and off-axis frequency response data. That doesn't prove that other factors don't come into play, but it does prove that their combined contribution is small at best, since the performance of a loudspeaker can be reliably predicted without taking them into account.
Well, that's kind of fundamental to the scientific method, isn't it. Something being inaudible is the null hypothesis. You can't prove the null hypothesis. You can only disprove it.
Yes. For example I'm sure that if you search long enough you can find speakers that have horrible THD. According to research they're not commonplace (it wasn't a significant factor in any of the 13 speakers that they tested), but presumably they exist. I don't think anyone would dispute that high enough levels of THD are audible; that's a fairly trivial self-evident claim. Therefore, you could theoretically find a speaker that has unusually high levels of THD, and that speaker might sound different from other speakers with the same on-axis and off-axis response. But the reason why we don't care much about that is because, in practice, most speakers have THD levels low enough for that not to matter much (as that study shows), and in typical scenarios the off-axis response is therefore much more important.
Not all budget speakers, no, of course. But there are some budget speakers that have really good on- and off-axis frequency response. Examples include the JBL LSR305 and Kali LP-6, both of which retail at around $300/pair. More expensive speakers might of course do better in areas like maximum output level, aesthetics, reliability, features, protection circuitry, etc. but when it comes to pure audio quality, even a speaker sold at 10 times the price may only do marginally better (and many will do much worse).
Please keep in mind that, in the general field of consumer audio, price means very little. @amirm regularly shows in his electronics measurements that it's very easy to find grossly overpriced products that perform very poorly, as well as cheap products that measure very well. In this market, the correlation between price and audio performance is extremely broken (typical example: cables that cost the price of a car). There's no reason to believe the same discrepancy wouldn't apply to speakers. That makes your argument based on comparing "Magicos and revels" to "budget priced speakers" very weak in my opinion.
Agreed. The study I've linked above shows that just on-axis and off-axis measurements account for most of the perceived performance of a speaker, so you'll have a hard time convincing me that they're the "wrong measurements" to make. The book that I've already linked above describes this experiment and many related others in much more detail.
Last edited:
- Thread Starter
- #16
Have you heard a Magico or other high end speaker? Speakers are for music lovers not for reproducing tones from 20hz to 20Khz. People hear music not tones. It has often been found that the ones that measure well do not really sound good.
If I play tones, I hear tones. If I play music, I hear music. If the tones measure well, then I know that the music will too.
S
If it's easier, then why not use tones? Music is only a jumble of tones, so using tones is not only easier, but more accurate.
S
Similar threads
