|   Great link on wires with info on a Blind Test | skidead
Jun 18, 2002 5:13 PM | | I will admit that I am a newbie and before investing any time or money on cables wanted to do my own research.
I have some long runs I need to hook up Nautilus 805 with and wanted to know the best way. So I began some of my own research.
I don't know if the followers of this forum are familiar with this site but heregoes
The following link:
http://www.sundial.net/~rogerr/wire.htm
Is very well written and addresses most of the issues I have seen argued over on this site. It is put together by an ex-executive at McIntosh and mentions a blind test they perfomed on speaker wire. Much like the one proposed by advocates of each side of this issue, on this site.
I appreciate the help from those who have responed to my questions to date including Audioengr and Jon Risch. I still have some questions for this forum but wanted to let everyone know about this site.
Please read the whole passage regarding wires, I believe it is quite informational and the "tone" is business-like and unbiased, in my opinion.
Best
Eric |
| Out of Date, and Off Target | Jon
Jun 19, 2002 9:43 PM | | There are many things that are quite out of date in the information presented at this web site. This is a good example of the old school on cable matters. Blinders and tunnel vision.
The chart is strictly concerned with resistive losses and the resultant amplitude drops. The author states that the table is based on a nominal resistance (resistive load) and approx. 5% (1/2 dB) of signal loss to the speaker.
Now when ABX/DBT proponents talk about doing speaker cable tests, or interconnect tests, the concern for level matching is always raised, and the criteria used is not 1/2 dB, but +/- 0.1 dB.
Quite simply, if the speaker cable can cause a 1/2 dB loss or more into a nominal rated impedance (say, 8 ohms), then what will it do into a loudspeaker? For a resistive load, the loss will be all across the audio band evenly (ignoring speaker cable inductance, we'll look a that later.), and there will be no effect on the frequency response.
But a loudspeaker is NOT a flat resistance, it has a widely varying impedance across the audio band, and typically has a rising impedance in the crossover region, to as high as 30-60 ohms. When we examine the signal from the amp at the speaker terminals, due to the impedance of the speaker dipping down to below the nominal rating (it is not uncommon for an 8 ohm rated speaker to dip down to 6 ohms or even less), and rising up far above the nominal, the voltage divider action of the speaker cables resistance will cause the frequency response to deviate across the band, directly inverse to the speakers impedance. Where the speaker impedance dips low, the losses are greatest. For the nominal 8 ohm system with a loss of 1/2 dB into 8 ohms, the real world impedance dip of 6 ohms will cause approx. 3/4 of a dB loss.
Conversely, for the high impedance region near the crossover frequency, the losses may drop to 1/10 of a dB. The net change across the audio band can easily be as much as 2/3 dB if this wire table is followed.
Interestingly enough, this far exceeds the +/- 0.1 dB criteria held dear by the ABX/DBT proponents.
The above is all predicated on strictly resistive behavior for the speaker cable, but ordinary zip cords also exhibit a significant amount of inductance, sometimes enough to roll-of the HF's by as much as 0.3 dB at 20 kHz for a 10 foot length. If a speaker system also happened to have low impedance at HF's, which is not uncommon for electrostats or high end speakers, then the total deviation within the audio band could easily approach or exceed 1 dB due to speaker cables that meet the table's requirements.
This very aspect of cables was brought up in an article by E. Brad Meyer in Stereo Review, "The Amp/Speaker Interface",Vol. 56, pp. 53-56 June 1991.
So cables that are substantially less resistive than what is shown in the referenced table would be a good thing, and avoid uncontrolled frequency response deviations due to the speaker cable.
Another problem with using this wire table, is that it will also allow the damping factor at the speaker to drop to around 16 or so with real world speakers. This can cause the bass to sound boomy and ill-defined, with reduced punch and tautness.
The issue of cable inductance is ignored as well, oridnary lamp cord/zip cord can have enough inductance to actually roll-off the high frequencies, this additional roll-off gets added to the resistive losses, and the total deviation across the audio band goes up even higher. Into a 4 ohm load, at 20 kHz, as much as 0.3 dB of roll off can occur. Add this into the resistive losses, and the total loss at 20 kHz is now almost 1 dB ! Make the cable longer, and the inductance goes even higher, and this is NOT accounted for in this chart. For a 20 foot run, using wire gauges according to this chart, the losses at 20 kHz could reach 1.3 dB.
My own opinion is that lower resistance AND lower inductance cables than are recommended in this chart are going to offer less fr |
| Out of Date, and Off Target, Part 2 | Jon
Jun 19, 2002 9:45 PM | | Part 2
My own opinion is that lower resistance AND lower inductance cables than are recommended in this chart are going to offer less frequency response deviations, as well as other less obvious improvements, such as loudspeaker damping, and then there is always the issues of the quality of materials used in the cables, the cheap zip cord (AKA lamp cord) cited in the referenced article uses very poor insulating materials, and the quality of the copper is in question.
Some of the other issues raised at this web site:
Oxygen free copper and silver wire. Once again the only issue that is examined is the resistivity. What about how OFC reduces the disruptions to the copper metal's crystal lattice? Many high end cable companies have found that very high purity copper, or some special way of processing it is necessary for best results. It is not just the resistivity issue, but a certain amount of concern for non-linearities as the signal passes through.
Listening tests are brought up.
While the results of this "test" sound impressive, they really have very little meaning in terms of scientific evidence. Listening tests can not be considered to be scientific evidence until they have been published in a peer-reviewed professional journal, so that the details of the test can be examined by others, and the results adequately duplicated.
Neither this listening test anecdote, or that of the ABX web site have been so published. Anyone can post things on an internet website, I have my own website, were I provide listening test results that show that cables sound different. I believe that my tests were done at least as well, and quite probably better, than either the ABX web site tests, or the tests at Mr. Russell's web site. To be perfectly honest, my test results have not been published in a peer-reviewed professional journal either, so they are also in the category of anecdotal evidence. However, I have presented a paper at the AES on how to conduct listening tests, that highlights many of the things I learned during years of testing.
The Stereo Review Cable test article is brought up. This is one of the more controversial articles, as it was so-heavily edited by the Stereo Review editor to read the way he wanted, that the original author challenged the editor regarding the conclusions that the article appeared to make.
On the whole, this web site page is sadly out of date, and provides no hard core data that is relevant, and the wire table does not allow for a very basic level of accuracy in terms of speaker cables.
Jon Risch |
| Out of Date, and Off Target, Part 2 | FLZapped
Jun 20, 2002 8:16 AM | | [Oxygen free copper and silver wire. Once again the only issue that is examined is the resistivity. What about how OFC reduces the disruptions to the copper metal's crystal lattice? Many high end cable companies have found that very high purity copper, or some special way of processing it is necessary for best results. It is not just the resistivity issue, but a certain amount of concern for non-linearities as the signal passes through.
]
Ahh, here we go agin, leaving out part of the story. So you think the puny conductivity difference in various forms of wire is going to matter in terms of audibility? Where is your supporting evidence?
Now, just what are those non-linearities Jon? Can you demonstrate them? You're not trying to mislead by ommision are you?
[Anyone can post things on an internet website, I have my own website, were I provide listening test results that show that cables sound different.]
No you don't, not really, it isn't in any form that anyone can either review, or replicate. It's no better than any other anecdotal evidence.
[wever, I have presented a paper at the AES on how to conduct listening tests, that highlights many of the things I learned during years of testing.]
Which hasn't been peer reviewed either.
[On the whole, this web site page is sadly out of date, and provides no hard core data that is relevant, and the wire table does not allow for a very basic level of accuracy in terms of speaker cables.]
And your website is equally full of misleading half-truths, incomplete information, and lack of hard evidence. There have even been gross errors pointed out that you refuse to either remove, or correct.
-Bruce |
| Out of Date, and Off Target, Part 2 | mtrycrafts
Jun 20, 2002 10:01 PM | | b Anyone can post things on an internet website, I have my own website, were I provide listening test results that show that cables sound different.
No you don't, you just make empty claims. Anyone can make such bogus claims. Credible demonstration is something you have yet to do. LOL.
b I have presented a paper at the AES on how to conduct listening tests, that highlights many of the things I learned during years of testing.
And you left out important points, not validated nor peer reviewed. So, it is an opinion only, especially with all the flawes, it has no merit.
b On the whole, this web site page is sadly out of date, and provides no hard core data that is relevant, and the wire table does not allow for a very basic level of accuracy in terms of speaker cables.
You have no hard core data, Jon. You are out of date. You have not demonstrated that 5% resistance is audible. Actually, you have not demonstrated anything audible, JON. LOL. Empty claims. Bogus. Speculations plenty.
b as it was so-heavily edited by the Stereo Review editor to read the way he wanted,
More bogus claims.
All the necessary data is presented, Jon. You just don't like it as it contradicts your beliefs, LOL.
b I believe that my tests were done at least as well,
Of course you would believe this. Certainly you wouldn't say you did it worse, LOL.
b Listening tests can not be considered to be scientific evidence until they have been published in a peer-reviewed professional journal,
Yet you want everyone to accept claims as having merit when yours is not even published in a rag nor can be found to have been credibly demonstrated to credible parties, LOL, and, you offere testimonials at every opportunity to support your wires. ROTFLMAO.
Get over it, you have nothing. Come back with evidence. |
| I'm going to have to agree with Jon | Norm Strong
Jun 21, 2002 3:46 PM | | insofar as that article being a bit out of date. Furthermore, nowhere did the author justify his choice of 5% as the maximum acceptable cable loss. Nor did he mention that the loss in maximum amplifier output is closer to 10%, which I maintain is a significant drawback.
Nevertheless, a 1.3db variation in frequency response is the worst possible case, so don't bet your life on being able to identify it. |
| I'm going to have to agree with Jon | mtrycrafts
Jun 21, 2002 5:52 PM | | Based on their DBT, 5% was a safe zone. It may be 5.5% or 6%. Certainly they conducted DBT testing.
Similar to level matching to .1dB. In many cases .5 would do just as well and not be audible in a home environment but .1 is a safer bet. Or a different level to follow the hearing thresholds but the .1 is used across the band for expedience. |
| YOU are Out of Date, and Off Target | mtrycrafts
Jun 19, 2002 9:59 PM | | LOad of bs, Jon. You have no evidence that it is oput of date. Citations please.
b the losses at 20 kHz could reach 1.3 dB.
Same old tired BS. YOU CANNOT HEAR 20kHz, Jon. LOL.
Besides, the JNDs is well above this LOL.
You are out of date, out of touch from reality. |
| Calm down Mtrycrafts | Richard Greene
Jun 21, 2002 11:09 AM | | Deal with facts not name calling.
I read of an experiment where even those
adults who could hear 20kHz. could not
discern an SPL change of less than 3dB
at 20kHz. |
| Calm down Mtrycrafts | mtrycrafts
Jun 21, 2002 5:58 PM | | b YOU CANNOT HEAR 20kHz, Jon. LOL.
Besides, the JNDs is well above this
Yes, 3dB is well above the 1.3dB he cited:) Will this impress on him, or change his misleading posts?
Name calling? being out of date? |
| Out of Date, and Off Target | FLZapped
Jun 20, 2002 8:06 AM | | [ Conversely, for the high impedance region near the crossover frequency, the losses may drop to 1/10 of a dB. The net change across the audio band can easily be as much as 2/3 dB if this wire table is followed.]
And you think you can hear this difference when listening to a broadband signal such as music? Gee, what would happen if you turned you head, Jon, think the room responces would cause a far greater change? Would you even notice that? hmmmmm......
[Another problem with using this wire table, is that it will also allow the damping factor at the speaker to drop to around 16 or so with real world speakers. This can cause the bass to sound boomy and ill-defined, with reduced punch and tautness.]
Where is your evidence to support this audibility statement? Sure it isn't the room responses cuasing the majority of the problem here???? Of course you haven't even included the effect of the crossover components in this, so what do you think the damping REALLY is at the speaker with the largest of damping factors(a short at the speaker terminals) applied externally?
[The issue of cable inductance is ignored as well,. . . Into a 4 ohm load, at 20 kHz, as much as 0.3 dB of roll off can occur.]
Ok, but is this audible? What is the JND number at 20kHz? And why are you using 4 ohms, by far the majority of speakers are 8 ohms nominal. Using 4 ohms GREATLY increases the insertion loss. WHy not use what is more commonly available? You're not trying to mislead, are you?
[Add this into the resistive losses, and the total loss at 20 kHz is now almost 1 dB ! Make the cable longer, and the inductance goes even higher, and this is NOT accounted for in this chart. For a 20 foot run, using wire gauges according to this chart, the losses at 20 kHz could reach 1.3 dB.]
Yes, but what your implying is misleading. The differential is still 0.3dB, since resistive losses are constant for all frequencies and that's all we care about in terms of audibility. So why mention the resistive losses in such a way? You're not trying to mislead are you?
BTW - what is the JND at 20kHz, Jon?
-Bruce |
| re: Great link on wires with info on a Blind Test | mtrycrafts
Jun 19, 2002 9:56 PM | | Thanks for the link, but it has made its rounds here many times but been a while.
b Much like the one proposed by advocates of each side of this issue,
Sorry, Eric, the yeasayers don't advocate blind listening. |
| re: Great link on wires with info on a Blind Test | skidead
Jun 21, 2002 5:41 AM | | I never thought I would touch off such a firestorm. I should have paid more attention to the ongoing debate here before offering any input. But hey as far as I am concerned there is always room for learning and I do enjoy monitoring the this lively debate. Unfortunetly, I can offer no scientific input. |
| re: Great link on wires with info on a Blind Test | mtrycrafts
Jun 21, 2002 6:00 PM | | b But hey as far as I am concerned there is always room for learning and I do enjoy monitoring the this lively debate.
Yes, it is about learning, separating the bs from facts.
Besides, this link has not been used recently, so don't feel bad. And you didn't touch off anything:) |
| As for me and my family..... | jamesgeib
Jun 21, 2002 8:08 PM | | We use Analysis Plus Oval 9! |
| As for me and my family..... | mtrycrafts
Jun 22, 2002 10:40 PM | | Good for you.
What do I need to ask your family? I have already talked with that company. They have nothing different to offer from the same old stuff, wrapped in a different, more expensive package. |
| As for me and my family..... | pctower
Jun 23, 2002 2:00 PM | | b I have already talked with that company.
Such unauthorized communication with the enemy could get you in serious hot water. |
| As for me and my family..... | mtrycrafts
Jun 23, 2002 11:15 PM | | It already has. I didn't keep the actual phone bill as proof. Next time. I am psychic now. |
| |
