Audio Misconceptions?

[tbewick]

Audiosouse, I think you'll find the Stereophile article I referred to earlier interesting, as it explains how various speaker tests are done and how they can be interpreted.

http://stereophile.com/features/99/

The impulse response test tell you how good a speaker's timing is, which has a considerable effect on its sound quality. A speaker with poor timing will have a residual response in certain areas of the signal, like the treble or mid-range. This can result in the speaker having a harshness to its sound, quite disturbing when playing back a difficult signals, such as those from a violin. An example of a speaker with excellent transient response is the classic Quad electrostatic speaker.

The reason for a speaker having poor transient response is that there is a delay in between the voice coil being energised and the speaker cone moving. Also the cone has mass associated with it, which further delays its response. Thirdly, the cone will distort (flex) when it moves. The delay would not be a problem if it occurs evenly across all frequenices, but this is not the case.

These distortions are not imaginary, and can be verified experimentally. The Stereophile reviewer did not have access to an anechoic room when running his tests, and had to model his data on a computer. I have seen speaker response traces taken in an anechoic chamber, which clearly show the difference between speakers with good transient characteristics (they were more expensive) and those with poor transient characteristics.

 

[John S]

Let me just throw this in the mix....if it's been discussed before I apologize. This stirs up a hornet's nest on some forums:

The Ten Biggest Lies In Audio

 

[Nick250]

Let me just throw this in the mix....if it's been discussed before I apologize. This stirs up a hornet's nest on some forums:

The Ten Biggest Lies In Audio

What a great article, I even book marked it.

 

[hopjohn]

I've read this article before, and would agree in principle with most of what is said there. Though some of Aczel's literary tactics go to such extremes (coat hangers, etc.) to make a point that the article loses some credibility.

 

[cornelius]

Good ol' Peter Aczel - one of the original founders of my favorites speakers to this day.

 

[WmAx]

The impulse response test tell you how good a speaker's timing is, which has a considerable effect on its sound quality. A speaker with poor timing will have a residual response in certain areas of the signal, like the treble or mid-range.

Poor timing? Please clarify what you mean by "poor timing".

The reason for a speaker having poor transient response is that there is a delay in between the voice coil being energised and the speaker cone moving. Also the cone has mass associated with it, which further delays its response.

Mass has nothing to do with delay. Refer to Newton's 3rd law.

Thirdly, the cone will distort (flex) when it moves. The delay would not be a problem if it occurs evenly across all frequenices, but this is not the case.

Almost no driver is perfectly pistonic within it's entire passband. This behaviour does not cause substantial problems for well-designed modern drivers. The problem(s) that do occur are related to resonance(s), not delay(s). The only delay(s) as a consequence of the diaphragm, would be ones related directlty to the frequency response[E.G.; a high Q peak takes longer to build up energy].

I have seen speaker response traces taken in an anechoic chamber, which clearly show the difference between speakers with good transient characteristics (they were more expensive) and those with poor transient characteristics.

I am not sure what you mean by "good transient characteristics". But if you are referring to so-called transient-perfect designs that have minimal phase error[by use of a single full band driver or low order butterworth crossover slopes in a multi-way design, for example] throughout the passband, as compared to multi-way speakers with higher order crossovers, then this is not something that has been demonstrated[in credible perceptual research] to be important for hi-fidelity loudspeaker reproduction.

-Chris

 

[tbewick]

John S,

I haven't fully digested everything in the link you referred to, but I will comment on a few of the points made.

I am reading a book about digital audio and have not finished it yet. No one questions the correctness of Nyquist/Shannon's theorems, but the perfect bandwidth/filters which are required to reproduce perfectly the original signal are not available. If you further research the topic I think you'll find that there are limitations forced upon digital audio just as there are in analog audio.

The book I'm reading (Introduction to Digital Audio, John Wilkinson) does say that a well designed DAC is unaffected by the interconnect used to provide the digital data. Unfortunately some consumer external DAC's (like those used in an A/V receiver) do not have the clock loops to prevent jitter which audibly degrades the sound quality.

WinAx,

If you want to know more about speaker 'timing', I'd refer you on to the web site I gave earlier (http://stereophile.com/features/99/). I'll try and scan on the trace I have of the two loudspeakers that have different transient characteristics and upload it onto this thread when I can get round to it.

If you're happy with your current system and do not believe in buying shielded cables/high quality DAC's/vinyl recordings etc. then that's great - good for you. I am not entirely satisfied with the sound from CD's/DVD's etc. so I'll go on researching audio topics to see where my next upgrade is best spent.

The article 'The Ten Biggest Lies In Audio', in my opinion, does not look like it has been written by someone with a detailed knowledge of the subjects he is commenting on, esp. digital audio. The book I referred to earlier is written by someone who has given a thorough explanation of the subject, rather than trying to explain everything in a few paragraphs. 'The Art of Digital Audio', which I haven't read, is written by the same author, and is an even more rigorous discussion of the subject.

 

[WmAx]

WinAx,

If you want to know more about speaker 'timing', I'd refer you on to the web site I gave earlier (http://stereophile.com/features/99/). I'll try and scan on the trace I have of the two loudspeakers that have different transient characteristics and upload it onto this thread when I can get round to it.

There seems to be some confusion. I don't have any questions about speaker performance(s) and behaviour(s), so far as time-related phenomena[and if I did, Stereophile is just about the last place I would look for real information on perceptually relevant data, since they have basicly zero credibility in relation to such]. I asked you to explain how you were using the term 'timing'.

-Chris

 

[MBauer]

Almost no driver is perfectly pistonic within it's entire passband.

-Chris

What does Pistonics mean? I am not familiar with the word?

 

[mustang_steve]

I've been on a bit of a hiatus from teh board for a while...I was working on a custom bicycle build project that was taking all my time and mental power (sometimes vendors can be a real PITA).

I think he means "moves in a piston like manner" by the term pistonic.

In a car, a piston moves perfectly up and down, suspended from the cylinder wall by a thin layer of viscous oil (motor oil). A piston by nature is designed for zero flex during the combustion phase, so that maximum power is transferred to the crankshaft of the engine.

In a loudspeaker under an ideal world, a cone would work like this as well, perfectly linear, with no flex whatsoever...however we do have nonlinearities, and we do have flex, each of varying degrees basedon design and product quality metrics.

It's just a necessary evil until we figure out a way to solve it....however the problem is a cone/coil assembly is "floating" as in there is nothing but dampers holding it in place, and those dampers are very close together compared to the area of the cone. For linearity, the distance between dampers has to be significantly greater than the area of the object going through those dampers. Look at a automobile shock absorber for an example of this.

The problem with a shock absorber design in speakers is: lack of effective cone surface area, therefore manufacturers make up all sorts of compromises to try to reach their goals of linearity along with enough cone area to get the job done right. After all, a loudspeaker works by moving air, so it has to move that air as efficiently as possible.

 

[MBauer]

Mustang, Yes, but

You described the ideal state of a piston in operation, internal combustion engine, pump, etc but is that what he meant by "pistonics"? Like a piston in the real world, a loudspeaker also often functions differently in actual operation, it is subject to many forces and its design attempts to deal with them as best it can.

 

[sploo]

Hey boy, looks like we can have that fight after all!

Gotcha!

Erm...

Non capisco
No entiendo
Je ne comprends pas
Ich verstehe nicht
Nope. I don't understand .

 

[Buckle-meister]

Erm...

Non capisco
No entiendo
Je ne comprends pas
Ich verstehe nicht
Nope. I don't understand .

Ok, you wrote (on a separate thread):

As this thread has the attention of both mtrycrafts and WmAx, I'd like to ask my own question (something of a thread highjacking

I wrote:

Ok, here's another one for you (sorry Sploo, but I need to borrow the thread again ).

You wrote:

I'll fight you for it

So I wrote (on this thread):

Hey boy, looks like we can have that fight after all!

After having realized from the following

...a couple of hundred $$ (or ££ in my case)...

that you lived on the same island as myself. True?

Alas, any humour that once existed has now been lost in translation.

Regards

 

[Audiosouse]

...Stereophile is just about the last place I would look for real information on perceptually relevant data, since they have basicly zero credibility in relation to such]. I asked you to explain how you were using the term 'timing'.

-Chris

I'd have to agree on that one, it can be fun to read about the equipment they test, but I would't heed much in relation to scientific fact from them.

They're basically attempting to justify their lack of controlled testing with pseudo science. They kinda sound like a cable vendor.

 

[sploo]

After having realized from the following...

...that you lived on the same island as myself. True?

Ahhh. Yes. Sorry, going senile in my old age.

I'm not going to pick a fight with a bloke from Glasgow though. I work with a fella from that area.

Imagine (in a thick Glasgi accent) being asked:

"Does your mother sew?" (the reference being, "well stitch this pal").

No wonder the Romans built a bloody big wall. Ahem.

 

[Buckle-meister]

Ahhh. Yes. Sorry, going senile in my old age.

I'm not going to pick a fight with a bloke from Glasgow though. I work with a fella from that area.

Imagine (in a thick Glasgi accent) being asked:

"Does your mother sew?" (the reference being, "well stitch this pal").

No wonder the Romans built a bloody big wall. Ahem.

No worries mate. I feel obliged to correct you though: I only stay in Glasgow because of work; I'm actually from the Highlands.

Big difference.

It's somewhat similar to that of a Scouser being mistaken for a Manc., although in my case I'm not offended . Still, just to be safe, better not tell your Weegie colleague I said that or he'll probably knock my teeth in, in which case I'll not look like this , but rather this !

Cheers

P.S. for others not from the UK reading this, don't worry; it really is English I've written

 

[jaxvon]

I might have friends from the UK (Burton on Trent), but I still have trouble with English English...

 

[Buckle-meister]

I might have friends from the UK (Burton on Trent), but I still have trouble with English English...

Yeah, me too; they never pronounce their 'r's

Regards

 

[Audiosouse]

Let me just throw this in the mix....if it's been discussed before I apologize. This stirs up a hornet's nest on some forums:

The Ten Biggest Lies In Audio

Boy did he nail that article! That's a good one. Yeah the coat hanger was a bit of a stretch, but he's trying to slam his point home.

He's right on about idiophiles behavior when asked to verify their superhuman hearing powers with DBT or ABX testing. What's even funnier (or more disturbing) is that when confronted with the truth of their inability to differentiate a Walmart power cord from brand X's $10,000 power cord, they still don't care! Either genuinely retarded or have more money than brains (I've found the two directly related).

I don't understand how that stirs up a hornet's nest. Are audiophiles really THAT dumb? Is the search for truth and knowledge through research and science dwindling? Are ignorance and stupidity cool?

I know how audio reveiwers do it too. As an undergrad, most of the mentally challenged majored in what they could pass, namely English (excluding those who actually enjoy English). Upon completing their eight year undergrad, they figured they were smart and started reviewing audio using thier newfound vocabulary. I say this because no self respecting engineer would write such dribble or believe such voodoo, hence Audioholics. And if I've never said it, thanks.

 

[John S]

The book I'm reading (Introduction to Digital Audio, John Wilkinson) does say that a well designed DAC is unaffected by the interconnect used to provide the digital data. Unfortunately some consumer external DAC's (like those used in an A/V receiver) do not have the clock loops to prevent jitter which audibly degrades the sound quality.

Thanks for the reference to the Wilkinson book. I will read this. Does your above quote imply that an expensive interconnect will correct the problems with poorly designed DACs?