[bigshot]

He uses our forum as an advertisement for his own. He'll post a reference to his own site, and then criticize you if you didn't click on it.

That's what Jude banned him for last time when he was using his other username. His shenanigans go back years.

I don't see any value to responding to him when he is disingenuous and aggressive like that. I'm going to speak past him to the lurkers. Maybe I'll do screen caps of posts related to him in other forums and highlight the relevant parts.

 

[SilverEars]

When you're dealing with heavy speaker wire, the resistances involved are very low, and this leads to several issues when attempting both to measure them and to evaluate the differences.
For example, the differences noted in that table of measurements, which could be due to non-standard diameter, or even differences in alloy, would probably be insiginficant over normal distances.
However, with resistance values that low, the quality of the contact at either end becomes very important.
And this, in turn, reflects both on how you use the product, and how you test it.

For example, if you're purchasing made-up wires with ends, like spade lugs or banana plugs on them......
- If you're using the sort of banana plugs where you screw down a collet, or use set screws, how tight you make those screws will alter the contact resistance, sometimes quite a bit, and may have a major effect on the total resistance.
- But, if the ends are soldered, then the resistance should be quite consistent, and quite low.
- And crimped connections can vary a lot depending on how they're crimped (hand crimp, machine pressure crimp, or ultrasonic weld/crimp).
(With hand crimps, you should always test multiple samples, preferably from multiple lots, because there may be considerable variation between individual cables).

There may also be issues with surface oxidation.....
A wire that tarnishes quickly due to being a poorer alloy is more likely to build up a higher resistance between the wire and the termination over time.
(So, if your wire is really bad, it makes sense to cut off the end, and re-strip it, every few years.)

I would also note that the DC resistance of most typical loudspeakers is usually in the range of a few ohms (often about half of the speaker's rated impedance - which is an AC measurement).
In order to avoid affecting damping, you want the resistance of the speaker cable, and the connections, to "not significantly increase the overall impedance"......
So, for a 4 Ohm speaker, with a 2 Ohm DC resistance, you would want your speaker cable to have a resistance one or two orders of magnitude LESS than that to avoid any possibility of significant interactions.
Anything past that is unlikely to make an audible difference.

Another thing to note is that there are two issues here:
1) power losses and lowered damping due to the resistance of the speaker wire
2) changes in frequency response due to an interaction between the impedance of the speaker and the resistance of the speaker wire

The power losses and differences in damping depend almost solely on the resistance of the wire....
Alterations in frequency response depend both on the resistance of the wire, and on the complex impedance of the speaker....
Therefore the latter depend heavily on the electrical characteristics of the particular speaker you're using (a speaker with a relatively even impedance will be less affected).

Something I'd like to see. Measurement values based on connections you describe. I've seen speaker impedance responses. There are variations and hump like the high ohm dynamic driver headphone.

 

[HotIce]

1) Seems to me that BigShot mentioned "with music" ... and I have not seen anything in the cut&paste you posted, that contradicts that.
2) We are not all the same in hearing. Even if ones does a 100K (and no one went even close to that) individual testing which will result in a 20ns mean, you will still have population spread quite wide around the mean.

You may want to step back one second and reconsider the approach you are using.

 

[Strangelove424]

You didn't click on the link I provided??? Here are the "hard" numbers:

As to you see measured results are worse for Monoprice but better for Belden and such. And it is much more extensive and detailed than what is in your link.

There is also no "control" in the other link. I specifically tested Belden because that is the industry gold standard. That way, if there are measurements errors, we would see it. And at any rate, relative data to Belden cable remains correct even if the specific measurements have errors in them.

I will not follow links to your forum. This is the forum I frequent. Make your point in our forum or don't make it at all.

You do not understand what a control group is. This is not an experiment, there is no treatment or process that varies. It is a measurement. There is no control group.

If that's DC resistance in ohms (I don't know what else it would be, but you failed to indicate a unit) then your numbers don't match up with others who have tested this, and are extraordinarily high. Measure again and put the decimals in the right place. Your capacitance and inductance do seem to correlate with other sources, and they indicate that there isn't nearly as much difference as you state.

With a 50ft cable run, none of this would make 1db of difference.

 

[amirm]

If it's 1/000 then .0048 ohm is the highest resistance? Insignificant. I wouldn't care about deviations from Belden then. That's magnitudes low that it wouldn't matter in audio. Speaker input impedance varies from as low as 4 ohms to about 8 ohms(I think I've heard of ones reaches higher)?

You are close to the right answer. One foot obviously is not an issue with just about every cable. Question is what is the typical length considering that the resistance needs to be computed round trip. That is, a 20 foot cable is actually 40 feet since that is the total resistance (and impedance) seen by the source (plus the speaker).

The above then says we need two pieces of data here: length of cable and speaker impedance. Speaker impedance unfortunately is NOT the marketing number that is provided in spec sheets. We need to know the actual waveform measurements to know the minimum it hits (same is true of headphones). Stereophile is a great source of such measurements. Here are the measurements for my own speakers for example: https://www.stereophile.com/content/revel-ultima-salon2-loudspeaker-measurements

We see that it hits about 3.5 ohms or so below 100 Hz. Let's round that to 3 ohms and put in a pin in it for a second.

The other important bit is audibility. Research/controlled listening tests performed by Dr. Toole/Olive into audibility of resonances shows that for broad resonances in low frequencies, the threshold is as low as 0.5 db.

So now what we have to do is figure out at what cable length we cause a 0.5 db drop in our speaker response. Assuming we have a DC resistance of 4.016, this becomes 44 feet. Dividing that by 2 we get 22 feet of actual length.

22 Feet is pretty good length that should cover most scenarios. For this reason my recommendation is to always use proper 12 gauge wire. If you do, then you are in business.

To the extent some wire is not 12 gauge, then it upsets this shortcut and you have to do the math. And without actual measurements of the wire -- which is non-trivial to do -- this becomes hard.

So use a proper 12 gauge cable and you are good to go. If you use thinner gauge, you risk changing the frequency response of your speaker.

You all are pretty familiar with impact of output impedance of headphones amps doing the same on on headphones. Same is true here with respect to speaker cables and the much lower impedance we could find in speaker.

 

[amirm]

I will not follow links to your forum. This is the forum I frequent. Make your point in our forum or don't make it at all.

I provided the data. You turned that into an accusation that somehow by magic I had figured out the relative impedance of each wire to Belden without "hard data." If you are going to accuse me of something, then you better know your facts and what is being put in front of you. What I post here was perfectly sufficient if you had read and understood it properly.

 

[amirm]

You do not understand what a control group is. This is not an experiment, there is no treatment or process that varies. It is a measurement. There is no control group.

This is a absolutely an experiment. You don't seem to understand the nature of such tests.

Measuring a thousands of an ohm is not a trivial thing. The meter probes or even the tension put between the meter probes and wire changes the measurements. If you just take one wire with some random spec from a chinese company, you have no idea if you are measuring the right things or not. So you go to a respected company like Belden and see how your measurements match theirs. By comparing my results to Belden, you can see the deviation and level of trust you can put in there.

I went one step further and used the Belden number as reference. That results in factoring out fixed inaccuracies to good extent.

The article you pointed at on the other hand, did not at all follow proper protocol. For one thing, their meter was not sensitive like mine. So they used a 10 foot piece of wire to increase its resistance. Problem with that is that you have to coil that wire every which way to get the two ends to meet the short probes of the meter. No two wires would then be coiled the same way. Those long wires also act like antennas for low frequency emissions around you and feed whatever they pick up to the meter.

The proper protocol calls for a short piece clamped flat which is precisely what I did.

Even with everything I did variations existed especially since we are talking about stranded wires and how they compress and make contact varies at such small resistances.

My wife used to be a lab technician at the hospital. Every morning they would run sugar water through the gear to make sure they got the expected results before testing patient blood and such. That is a control.

Here is the definition of control from Wiki: https://en.wikipedia.org/wiki/Scientific_control

"A scientific control is an experiment or observation designed to minimize the effects of variables other than the independent variable.[1] This increases the reliability of the results, often through a comparison between control measurements and the other measurements. Scientific controls are a part of the scientific method."​

 

[amirm]

If that's DC resistance in ohms (I don't know what else it would be, but you failed to indicate a unit) then your numbers don't match up with others who have tested this, and are extraordinarily high.

Nope. Here is the data again:

Look at the Belden cable. Advertised value by Belden is 1.56 milliohms. My measurement was 1.15. So actually lower, not higher. And pretty close given the challenges for measuring such things as I just explained.

In comparison the claimed DCR by Monoprice was 2.19 and I got 2.30. Pretty close actually to what they advertised. Using their number alone against Belden, 2.19 vs 1.56 shows you that it is a thinner wire/higher resistance. You don't even need my measurements for that!

So there is nothing extraordinarily high. You are just not reading the data in front of you.

 

[Strangelove424]

....

The proper protocol calls for a short piece clamped flat which is precisely what I did.

.....

My wife used to be a lab technician at the hospital. Every morning they would run sugar water through the gear to make sure they got the expected results before testing patient blood and such. That is a control.

.....

Your results are inaccurate and poorly documented, so I won't speak anymore to the validity of your findings on these measurements. However, I want to make it clear to you the difference between calibration and experimentation. Your wife was calibrating her glucose gear, not experimenting with it. She knew exactly how much sugar was in the water (probably from measuring it out before hand) and was testing to see if the machine would accurately detect it, so that when the machine was in the field testing a patient's blood level, she could count on the results. This is called calibration. https://en.wikipedia.org/wiki/Calibration

Now, if she then used her calibrated glucose meter to test two groups of patients, one experimental group who ate candy that day, and another control group of people who did not eat candy that day... THAT would be an experiment.

 

[bigshot]

I think he's beginning to break a sweat. "With music" was never acknowledged, but I've seen that several times in the past few weeks. He just shifts to another subject and attacks on that until he gets boxed in a corner again.

I like to keep an eye on the Users Who Are Viewing This Thread at the bottom of the page. It shows me that the audience is out there with their popcorn in hand enjoying the show.

 

[bfreedma]

I think he's beginning to break a sweat. "With music" was never acknowledged, but I've seen that several times in the past few weeks. He just shifts to another subject and attacks on that until he gets boxed in a corner again.

I like to keep an eye on the Users Who Are Viewing This Thread at the bottom of the page. It shows me that the audience is out there with their popcorn in hand enjoying the show.

I'm just surprised that members are still willing to engage knowing the history. I still see some of the posts in replies - exactly as predicted...

 

[amirm]

our results are inaccurate and poorly documented, so I won't speak anymore to the validity of your findings on these measurements.

I had full details of the experiment including measurement gear, setup, etc.

Let's compare that to data/article you put forward: https://thewirecutter.com/reviews/the-best-speaker-cable/

The 14 gauge monoprice cable they measured came out to 11 milliohms. My measurement of the 12 gauge monoprice cable was just 2.03 milliohms or 6 times lower!!! So if you want complain about someone's numbers being high, it is in your own reference. Likely though their measurements in error. This is all they say about it:

In other words, there is no there there. Nothing remotely close to care I put in my measurements.

They also say this:

You stand by what he says here?

 

[amirm]

I think he's beginning to break a sweat. "With music" was never acknowledged, but I've seen that several times in the past few weeks. He just shifts to another subject and attacks on that until he gets boxed in a corner again.

I like to keep an eye on the Users Who Are Viewing This Thread at the bottom of the page. It shows me that the audience is out there with their popcorn in hand enjoying the show.

Sadly hardly anyone is here based on that list.

Anyway, I am trying to get to the music bit but you refuse to answer how many tracks and their nature was used in your first reference. Have you even read that AES paper?

 

[SilverEars]

I like to keep an eye on the Users Who Are Viewing This Thread at the bottom of the page. It shows me that the audience is out there with their popcorn in hand enjoying the show.

I come here for the cross-feed discussions. Get a beer out of the fridge and just enjoy.

 

[Strangelove424]

The 14 gauge monoprice cable they measured came out to 11 milliohms. My measurement of the 12 gauge monoprice cable was just 2.03 milliohms or 6 times lower!!! So if you want complain about someone's numbers being high, it is in your own reference. Likely though their measurements in error. This is all they say about it:

Amirm, I will not continue with you anymore until you put your decimals in the right place, and learn the metric system. The Wire Cutter tested to 11 centi-ohm (.0X) not milli-ohm (.00X)!

I will not address your numbers until you learn the correct words for place values in the metric system AND put your decimals in the right place.