[b0dhi]

Maybe I wasn't clear with the results, but I'll summarise here:

-A. The effect is audible with all types of headphones (ie dynamic, electrostat, piezo) at high volume, and at roughly the same volume (depending on the bass output of that particular headphone)
-B. The effect is audible using multi-way speakers where the cutoff frequency is above the high tone (ie one woofer handles both the low and high tones)
-C. The effect is not audible using speakers where the cutoff frequency is below the high tone but above the low "tone" (ie one speaker handles the low tones and a seperate speaker handles the high tone)
-D. The effect is audible when earplugs are used with a speaker that has the distortion at high volume, thereby producing a low volume at the ears.

C and D clearly show that the ear cannot be the cause, even in part. The cause is almost certain to be intermodulation distortion created when one speaker attempts to play both the low and high tones simultaneously at high volume.

The only curious thing is the consistency of the magnitude of the distortion across different transducer principles.

Being that the test sample models any type of music that has a prominent kick drum, this form of distortion isn't an academic curiosity but something that really reflects potential issues in sound reproduction. I'm not sure to take the poor performance of these headphones (and dynamic loudspeakers) as a good thing indicating that there's still plenty of room for progress in headphones, or to takes it as a sign that manufacturers are not yet close to being able to produce sound fidelity as good as the ear (at least in terms of IMD).

And, wow, what a digression

 

[Pio2001]

Yep, we're completely off-topic

I'll be quick : I did the experiment D with headphones, and my result is the opposite of yours. Reducing the volume at the ear removes completely the distorsion in my case.

 

[Spasticteapot]

IMD is definitely caused by transducers. However, in this case, there's an even more likely solution - nonlinearity. The ratio of voltage to movement starts to change as the excursion of the speaker driver increases, so that when the cone is moved very far in or very far out, it's not very linear at all. If you're playing an 80hz tone and 8khz tone simultaneously, the driver will be making 100 very small cycles for each big one, and many of these cycles will be made outside of the linear part of the motor structure.

This is also why big woofers often make more pleasing bass than small ones (they need less excursion), and why this rule does not necessarily apply to speakers with very sophisticated motors (ScanSpeak Revelators, for example.)

Quote:

Originally Posted by Sherwood /img/forum/go_quote.gif 1) The majority of what makes a cable what it is is determined at the connection, and by the material the cable is constructed of. Assuming the source is capable of driving a long length (and most are), I defy you to consistently recognize the difference between a 20 foot cable and a 3 foot cable. I can tell differences between opamps and cables, to a degree, but it is impossible for me to discern between lengths of cable. I use as short as possible because I build them, and because I am super cheap.

Ohm's Law says you're wrong.

If speakers were a nice, non-reactive inert load, then no, you couldn't. However, even the most well-behaved speakers' impedances vary with frequency, while the impedance of the wire stays the same. Because the speaker is effectively running in series with a resistor - the speaker cables - the varying impedance will result in more or less of the input voltage reaching the speaker as the impedance varies with frequency. It's sort of like having a pot in series with a resistor.

Of course, there's a simple solution to this one - make the ratio of the impedance of the speaker to the impedance of the wire so high that any differences are too small to measure. This can be done by using a transformer to increase the effective impedance of the speaker (like on 70v intercom systems) or by using really really thick wire.

If you don't believe me, go to one of those Monster Cable demos at Best Buy. There's a difference between the same signal running through 50ft of 14ga. cable and 100ft. of 20ga. cable. (Of course, 14ga. lamp cord would work just as well.)

On the subject of interconnects, capacitance and inductance could potentially cause problems, although this would only occur on the sort of cables you might get from Dollar Tree. Also, shielding is worth its' weight in gold.

Quote:

Originally Posted by Champ04 /img/forum/go_quote.gif And for the person who wonders why I think this is questionable...... Irritation is an emotion. Science is ideally without emotion. This shows obvious bias to me. I didnt ask him to admit his emotional bias. I just commented on the fact that he did and as such he loses some degree of credibility with me. Do I now need to include the obligatory....... IMO

Yes, but he wasn't giving a biased review - he was giving repeatable, measured results.

He did miss one small thing, though - if he had gone with some monstrously thick cable - say, 8ga. solid-core copper grounding wire like you can get from Home Depot - most of his issues would have gone away, despite the long . And that stuff is only a few bucks a foot.

I might also point that Nelson Pass, Sigfried Linkwitz, and McIntosh all agree - just use something cheap and thick.

 

[Sherwood]

Quote:

Originally Posted by Spasticteapot /img/forum/go_quote.gif Ohm's Law says you're wrong. If speakers were a nice, non-reactive inert load, then no, you couldn't. However, even the most well-behaved speakers' impedances vary with frequency, while the impedance of the wire stays the same. Because the speaker is effectively running in series with a resistor - the speaker cables - this impedance change will also result in a change in the input levels as more or less of the voltage is bled off by the wires. It's sort of like a resistor ladder. If you don't believe me, go to one of those Monster Cable demos at Best Buy. There's a difference between the same signal running through 50ft of 14ga. cable and 100ft. of 20ga. cable.

I'll acquiesce, with two reservations:

1) Speaker cables are not the only connectors said to benefit from a shorter length. I was speaking of interconnects, as was the post I quoted.

2) The Best Buy demo is anything but accurate. It is used to sell a product, and it has two variables (length and gauge). It is a proven fact that their listening stations are subjected to "beneficial" EQ, and the guts of the monster demo are not on display. There is no way to know that the "experiment" is being carried out accurately, and that the two proposed variables are the only ones in play.

 

[Spasticteapot]

Quote:

Originally Posted by Sherwood /img/forum/go_quote.gif 2) The Best Buy demo is anything but accurate. It is used to sell a product, and it has two variables (length and gauge). It is a proven fact that their listening stations are subjected to "beneficial" EQ, and the guts of the monster demo are not on display. There is no way to know that the "experiment" is being carried out accurately, and that the two proposed variables are the only ones in play.

The best buy demo is VERY accurate. It shows that a short, thick cable will sound much better than a long, thin one. Of course, just about any copper wire will do - I'm tempted to try making some Cat5 cables, if only for visual appeal.

Of course, both of those cables are wound on spools. And a cable wound on a spool is, by nature, an inductor.....

 

[Sherwood]

Quote:

Originally Posted by Spasticteapot /img/forum/go_quote.gif The best buy demo is VERY accurate.

I don't know how you can say this, not knowing what amplifier, source, etc. either cable is hooked up to. You also can't say with certainty (off of just that demo) whether a short, thick speaker cable sounds better than a long, thick speaker cable -- or whether a short, thin speaker cable sounds better than a long, thin speaker cable. That is the problem of utilizing more than one variable. You can only infer.

 

[nick_charles]

Quote:

Originally Posted by Spasticteapot /img/forum/go_quote.gif On the subject of interconnects, capacitance and inductance could potentially cause problems, although this would only occur on the sort of cables you might get from Dollar Tree. Also, shielding is worth its' weight in gold.

Well I am trying to evaluate this proposition, so far in raw terms of signal degradation (general attenuation and roll-off) I have not found any significant measurable dfferences between 77c cable and $60 cable with a few points in between. No two cables I tested ever differed at any frequency by more than 0.02db and the biggest average (across the spectrum) dfference between any two cables was 0.009db. I ran each cable test 10 times to smooth out random variations between trials. Of course none of these are true exotics which may be more problematical than bog-standard cables. In voltage terms a 0.24V reference signal was attenuated to 0.2292V by the worst cable insertion and attenuated to 0.2297V by the best cable or
a 0.2% difference.

However since even the very cheapest cable I tested was insignificantly different from a "decent" cable I am skeptical that any other cable at any price could improve *substantially* in measured terms (I woud be amazed at even a 1% difference) on the worst (functional) cable possible.

But even this is utterly irrelevant as we have these marvellous things called volume controls so that if you could hear the 0.009db drop you just tweak the volume up 1 degree...

 

[rds]

Thanks for the link - I enjoyed what I read of The Audio Critic. I see a lot of comments that this journalist is biased and therefore his writings are meaningless.
Well, I have to tell that everyone who's ever lived is biased. You should judge an argument by its own merits, not how you feel about the person presenting it.
The writer in this case supports his arguments by correctly applying the underlying science (at least in the articles I read).

 

[gregorio]

The subjetivist vs the objectivist is an interesting debate and one I've spent a great deal of time thinging about. I used to be married to a concert soloist who is regarded as one of the leading practitioners of her instrument in the world. What made her unusual is that she has been, from the age of 12, profoundly deaf. What was interesting was that her brain often filled in the gaps of what her ears missed, so much so that she could often 'hear' things which would be impossible for even the keenest of hearing. For example we were standing in our back garden one day and in the distance (a couple of miles away) we could see some trees moving a little in the breeze. My ex-wife could hear these trees quite clearly even though I could not detect any sound from them whatsoever above the traffic noise. Her eyes saw the movement and her brain made up the rest.

I believe a similar thing happens with many subjectivists. Having more bits, a higher sample frequency, a much more expensive bit of gear or $100 on a foot of cable, the brain is screaming out this is obviously better quality and/or a higher specification, therefore it must sound better. What do you know, they listen and low and behold it does sound better, "it's like night and day"! No amount of fact or scientific proof is going to change their mind. In the end they become insulting, dismissive of the laws of physics and ask if we have ever actually listened to anything. They will fight to the death for their position because they have percieved it for themselves. Unfortunately, perception is not reality and I for one have spent over 30 years actively training my hearing and my perception of sound to be objective. For an acoustic musician it's imperative, as what we hear coming out of our instrument from a few inches away is not what an audience hears 50ft away in a concert venue. Likewise with music production and film sound, w

 

[Goit]

http://theaudiocritic.com/plog/index...eId=7&blogId=1

Did everyone see this? Too bad it's discontinued.