kartino said:
Just curious, What is proof?
If you heard the difference would the be proof enough?
I'm not sure I would hear it, but still.
Why do we consider seeing a better proof than hearing?
Did the apple fall to the ground before Newton put forward a formula to describe it?
I've seen some where that any reasonably well designed cable will do. And that is good enough for me.
I also strongly believe (as stated here) that any money spent on cables could probably be better spent on speakers.
But that doesn't mean that cables makes no difference, it just means that it is not the best place to spent cash if you want to improve sound.
/Erland
Exactly, and as Johan pointed out, that's what the data shows. No spooky magic, just plain ol' LCR. Has anyone done any speaker cable test at sub-VHF which shows any effect that's NOT simple lumped impedances?
Just as a side note, if your concept of localization is correct, how closely would the diameter of left and right speaker cable (assume 3m) have to be matched?
kartino said:
As you can see, there is no fighting, but merely a discussion.
kartino said:
Spectral analysis will prove useless, as it has for many decades now.
It does not see that which we hear. It is single dimensional and does not retain any temporal characteristics.
We no longer hear in one dimension, we hear in stereo. Things which alter our perception of perceived direction, as in soundstage, are not tested using FFT.
It sounds nice, but it is not a method which will see that which we hear.
Cheers, John
SY said:
To the best of my recollection, nobody. Buuuut, this level of test is certainly beyond the capabilities of normal resistive loads.
Honestly, I do not think it is necessary to go beyond lumped analysis. However, when one considers the transfer function of a second order lowpassusing lumped elements, then cuts the lumps into smaller and smaller distributed ones, I do not know what the final delay tally will be. So I cannot blindly toss distributed analysis away..
SY said:
Just close enough..😀
Don't worry bout the diameter, worry bout the match in "lumps" first.
Then worry bout how the system reacts to the lumps. Non bi-wire situations with inductive storage is even more difficult to model than the resistive component. I am crawling at this point..
Cheers, John
Hi John,
Now, as far as I am concerned, the only things that occur about 30 KHz (as a number) are things that upset an amplifiers normal operation. This then affects the operation in the audio band. Most dynamic high frequency drivers have "broken up" past 25 KHz. So essentially, most of what is up there is noise and distortion. We are better off not to excite those frequencies in the first place. Fido will be happier.
John, are there any changes between the "A" and "B" models that you know of?
-Chris
Oh yeah!
Well, if the user reads the manual (don't say it! 😉 ), it clearly indicates that the 1m cable is at it's limit at 100 KHz. When you then see readings that are hard to reproduce, there should be no surprise involved. Most things that operate at those frequencies have carefully designed test sets.
Now, as far as I am concerned, the only things that occur about 30 KHz (as a number) are things that upset an amplifiers normal operation. This then affects the operation in the audio band. Most dynamic high frequency drivers have "broken up" past 25 KHz. So essentially, most of what is up there is noise and distortion. We are better off not to excite those frequencies in the first place. Fido will be happier.
John, are there any changes between the "A" and "B" models that you know of?
-Chris
anatech said:
Ah, but I cannot help think it..
The explanation in the measurement basics section of the manual is absolutely priceless. It is such a good reference, it should be required reading..
anatech said:
I also do not worry about sounds that I cannot hear.
The thing that set me "back" is our ability to distinguish right to left delay shifts at the 1.5 to 2 uSec level..that is a measured and proven capability that defies our understanding of how we hear sounds up to 20Khz. We are a system that cannot hear beyond 20K, but yet we are sensitive to an inverted bandwidth approaching 500 Khz for right to left issues.
anatech said:
I do not know what the difference is. Perhaps the B model uses single crystal cryo'd near superconducting negative dielectric helium impregnated quantum dot filled conductors..😉
To be honest, if I had both of the specification pages in front of me, I would not be able to notice a difference. That specification page is not written in any language I understand..
Cheers, John
Kartino,can you tell us a good reason why do I have to prove to anyone that I can hear sound differences between cables when he believes only numbers and formulas?For those who want a cable test I wrote one on page 32.It is a good challenge for anyone interrested.It's like asking you to prove that you don't hear them.The only way to convince me that you don't hear cable differences is to tell me so.Simple as that and as many said'you either hear them or not.As for numbers and formulas I believe they are right,although I cannot see how this can limit my haring.
Holy chit. 35 pages of wind, religious misunderstanding of reality and word salad.
Getting back to the posted question. Do speaker cables make any difference?
The answer is simple Grasshopper. Yes, if you do not use speaker cables you will be disappointed by the total lack of sound.
Getting back to the posted question. Do speaker cables make any difference?
The answer is simple Grasshopper. Yes, if you do not use speaker cables you will be disappointed by the total lack of sound.
Hi Rob,
Hi John,
-Chris
It's very hard to argue with that statement. Very zen.
Hi John,
Phase. I'm guessing that phase discrimination is very sensitive. Not very surprising since we are an evolved fearsome hunting machine. Phase differences leading to source localization was life and death. Therefore the internal timer we each must posses must be finer in resolution than anyone knew. Possibly timing interrupts, a more passive system. What's cool is that the system is continuously auto calibrating.
I read it over from time to time. Definitely should be required reading.
That's got to be it! Thanks John! 😀
-Chris
jneutron said:
No need to appologize...we never really leave "school" do we?
I agree with your above statement entirely, which is why I didn't use a speaker in the first place. I can't completely control the loudspeaker for variations due to air pressure or subsonic interference either..
As for the test, if I terminate with a 4 to 8 ohm resistor and try this again, can you forsee any issues that you'll have if I obtain the same results? Do you have a recommendation for resistor type?
I don't mind doing this again, if only for the sake of burying it, but I'd like to know ahead of time where the criticisms of the test procedure will be, so I can control for it.
Personally I don't like having this out on the 'net with my name attached if it's completely bogus....
anatech said:
I've often thought that we might be responding to an audible difference frequency due to non-linear mixing of supersonic frequencies...
Maybe there's some sort of acoustic demodulation occuring? Even within our own ears?
Hi macgyver10,
You know, the fact is that you even bothered to test these cables with real test equipment and a scientific procedure. So arm chair critics can be quiet unless they have something constructive to add.
I'm thinking that a standard dummy load, 4 or 8 ohm. For the purposes of your test you could run both if a shop will loan you them. Generally they should be a low inductance type. Driving them through an amplifier may be helpful. You will have to null out the amp if it has any effect. So run the test with and without. I would think that any level between 1 and 5 watts would pretty much cover things. That would be loud with most speakers in most rooms.
John,
You have more experience. What would you say?
-Chris
You know, the fact is that you even bothered to test these cables with real test equipment and a scientific procedure. So arm chair critics can be quiet unless they have something constructive to add.
I'm thinking that a standard dummy load, 4 or 8 ohm. For the purposes of your test you could run both if a shop will loan you them. Generally they should be a low inductance type. Driving them through an amplifier may be helpful. You will have to null out the amp if it has any effect. So run the test with and without. I would think that any level between 1 and 5 watts would pretty much cover things. That would be loud with most speakers in most rooms.
John,
You have more experience. What would you say?
-Chris
macgyver10 said:
Issues? Well, if you follow my recommendations but yet come up blank, then I can forsee at least two possibilities...there is an error we are missing, or there is no error to find..
I believe it is necessary to drive the wire with zero ohm impedance, and terminate with 4 or 8. Using a source with Z=500 ohms forces the system to pull current at the terms of the source. Using a power amp forces the system to draw at the load's whim.
macgyver10 said:
First...performing specific tests is not bogus. First, you start with a model, then use it to provide a prediction..a criteria for what is expected of the test given a model. Either you will have results which are consistent with the prediction, or you will not. In any case, the tests themselves, when rigorous and repeatable, are never bogus. There may be errors in models, analysis for prediction, or test methods, but full disclosure and discussion of such is never bogus.
I would go 4 ohm, but the resistor inductance and loop pickup must be understood and removed to eliminate the possibility that it is confounding the results.
I am aware of only one such design resistor, that is very cheap to make, but does require some skill at soldering.
There may be others out there, but I have not searched for them, simply because I made what I need. I will admit, I went just a tad overboard, as my resistor is about 60 picohenry (tested to below 250 pH), no loop coupling, and 50 watts at 4 ohms, which will give resistive operation at frequencies significantly above the requirements. My guess is above a Ghz, but I do not have any test equipment capable of square wave operation at a gig into 4 ohms. But it will meet requirements at a 100 Mhz down, so should not be an issue at audio.😉
Microwave BeO cylindrical resistors are great w/r to load impedance, but they do not provide a way to connect across it without making a loop.
What level of power dissipation do you believe will be adequate for your needs?
Cheers, John😉
anatech said:
Standard dummy loads are a good start. They can be used to debug the system.
The primary problem with them is their physical size, and the inability to connect the voltage measuring device to the terminals of the resistor without forming an inductive loop.
NI style resistors eliminate the solenoidal inductance, but do not remove the physical inductance created by the mean current path through the device.
At the 1 to 10 watt level, it will not be too difficult to make some more fancy shmancy resistor loads, I will visit my parts bin to determine what I have available to make them.
It would be fun to have my zistors used with some neat audio equipment.
Cheers, John
Chris and John,
All the input is welcomed, however both of you want me to deviate from using my test suite. I'll check my manuals, but I don't think that I can insert a power amp with 0 output Z and have the LMS system accurately measure.
At this point I'll have to see if I can get Linear X to tell me if I am attempting to measure outside the design parameters for their LMS equipment.
I'm also still skeptical that the voltage divider method employed by the LMS system is somehow flawed.
All the input is welcomed, however both of you want me to deviate from using my test suite. I'll check my manuals, but I don't think that I can insert a power amp with 0 output Z and have the LMS system accurately measure.
At this point I'll have to see if I can get Linear X to tell me if I am attempting to measure outside the design parameters for their LMS equipment.
I'm also still skeptical that the voltage divider method employed by the LMS system is somehow flawed.
QED
I disconnected both my speaker cables tonight, and there was almost no sound.
They do make a difference!
I disconnected both my speaker cables tonight, and there was almost no sound.
They do make a difference!
Re: QED
Hardee Har!
On this forum I'm finding out that it's not the cable, but whether or not you have something connected to the ends that matters....
guffaw...
dhaen said:
Hardee Har!
On this forum I'm finding out that it's not the cable, but whether or not you have something connected to the ends that matters....
guffaw...
dhaen,I think that with your speaker cables disconnected if you measure their open end you will see something there.If yes,maybe your speaker's terminals,or even worse your speakers may have a problem...
dfdye,and others interrested,through a wide bandwidth system, play some pink noise signal and take some measurements of the speaker's fr.response.Then repeat the same with another set of speaker cables,then another and another.Who knows you may be the first to repeatedly be able to measure what he cannot hear.There is a prize on post 279
- Status
- Not open for further replies.