Scottmoose said:
And you never will. ABX is a statistically invalid methodology. The beta is too high for it to ever give meaningful results.
dave
Hi everybdy,I will try in one post to cover comments from Cal Weldon,Poobah,Soon qsc.My system consists of Krell KSA250(bought secondhand for $1000),Quad QC22(demo for$700),modified Systemdek IIX turntable SME3009II arm and several cartridges(vdH MC ONE super,Ortofon MC20SII,Nagaoka MP50 and MP20 and a vdHul MM1).See?My favorite component is not cables.Phono preamp is a modified ONIX board in aluminium enclosure(Hammond) for MC,and another ONIX board for MM.On theese boards I used 1% metal film resistors,1%polystyrene and polypropylene caps,Panasonic ECA audio electrolytics and Sanyo OSCON caps.Speakers are JM Lab 946.Now the cables:From arm to phono pre vdH D501 silver hybrid,from there to pre and pre to power amp are vdH First ultimate and Second carbon.I use the Breeze speaker cables in short 1m runs to the outboard crossovers and from there to the drive units the CS14hybrid a reasonably 14awg thing.My room is lightly damped and measures 4.5 x 6.5m.I think the cost of cables is not unreasonable as it is a system with very wide bandwidth,high resolution and frightening power and dynamic range.It took years to build the system so dont think that I have money to waste ok?Iforgot the psu for the phono preams which are LM317 based using toroidals to give 24VDC that I made myself using very good components.
soonqsc,dont tempt me coz I wont upgrade anything.Im working on a tonearm and a turntable now.The money goes to music, LPs ofcource.
Panicos K said:
😀
As a matter of fact I have recently seen a locally designed turntable and tone arm that is very good. In a HiEnd exhibition here, the system had only everage onkyo amps, Jamo speakrs with Elac super tweeters. But the sound was shockingly real for the short time I was in the room. The system basically does not use any antiskid balancing because it's the arm is such that the pivote to sylus line is always tangent to the grove, and they are working on a micro controled version to shift the pivote point such that it's even more precise. Over $6000 for the setup.
Cal Weldon,
You are sadly literally correct when you suggest that this sort of thread often (I would not say always) goes nowhere. But I don't believe the subject is to blame. For me, and I daresay most folks knowledgable about electricity, the reason is the one I gave earlier. If I may be so bold as to quote myself:
Yes, one gets a bit tired of trying to explain facts honestly, only to find some folk coming along afterwards as if one has spoken to the clouds, with neither counter proof nor refutation of one's statements. This has often happened to me, and one cannot be blamed to begin suspecting notions of "I have made up my mind, don't confuse me with facts."
Why are deceptive "shows" like that of Drabitt necessary if there was any honest way to expound the advantages of one's product? (And believe me, they are plentiful. Yet I still have to find a single one where blatant deception could not be showed - perhaps not intentional deception, but then the only alternative is that the writer does not know what he is talking about, which is even worse.)
The Drabitt case: With all respect, Mr Drabitt started by saying that he listened to cable effects, thus audio frequencies are the subject, also confirmed by his use of the audio region in his graphs. Then a test that has nothing to do with practical audio conditions follows, concluded by the remark: "Obviously there is significant phase shift well within the 20Hz - 20KHz region"; the only obvious thing being that it is not so.
This test amongst others has repeatedly been quoted in support of cable yay-sayers, thus the deception is proved.
Scottmoose,
You mention that there is "plenty of scientific proof that cables affect the sound of the components they attach." May I be party to these facts please? (I keep an open mind that I may have misunderstood you.)
Planet 10,
Regarding properly controlled blind tests being statistically impossible, exactly what is meant? If so, why would they be used in several of the world's most prestigeous universities as indicative? To be clear there are limitations, but are they not at least more valid than a batch of subjective results (where the subjects know what they are listening to)? How does one explain that subjects are convinced of their preference one moment, only to loose that ability the next moment when they do not know what they are listening to?
Like others said before, one always seems to overstay one's welcome on subjects such as this. Perhaps for once one can bow out with a better balance of arguments than mainly black magic or snake oil on the one side. As said at the beginning, with me it is not about people's preferences (for the umpteenth time it is their money), but the way in which.
Regards.
You are sadly literally correct when you suggest that this sort of thread often (I would not say always) goes nowhere. But I don't believe the subject is to blame. For me, and I daresay most folks knowledgable about electricity, the reason is the one I gave earlier. If I may be so bold as to quote myself:
Johan Potgieter said:
Yes, one gets a bit tired of trying to explain facts honestly, only to find some folk coming along afterwards as if one has spoken to the clouds, with neither counter proof nor refutation of one's statements. This has often happened to me, and one cannot be blamed to begin suspecting notions of "I have made up my mind, don't confuse me with facts."
Why are deceptive "shows" like that of Drabitt necessary if there was any honest way to expound the advantages of one's product? (And believe me, they are plentiful. Yet I still have to find a single one where blatant deception could not be showed - perhaps not intentional deception, but then the only alternative is that the writer does not know what he is talking about, which is even worse.)
The Drabitt case: With all respect, Mr Drabitt started by saying that he listened to cable effects, thus audio frequencies are the subject, also confirmed by his use of the audio region in his graphs. Then a test that has nothing to do with practical audio conditions follows, concluded by the remark: "Obviously there is significant phase shift well within the 20Hz - 20KHz region"; the only obvious thing being that it is not so.
This test amongst others has repeatedly been quoted in support of cable yay-sayers, thus the deception is proved.
Scottmoose,
You mention that there is "plenty of scientific proof that cables affect the sound of the components they attach." May I be party to these facts please? (I keep an open mind that I may have misunderstood you.)
Planet 10,
Regarding properly controlled blind tests being statistically impossible, exactly what is meant? If so, why would they be used in several of the world's most prestigeous universities as indicative? To be clear there are limitations, but are they not at least more valid than a batch of subjective results (where the subjects know what they are listening to)? How does one explain that subjects are convinced of their preference one moment, only to loose that ability the next moment when they do not know what they are listening to?
Like others said before, one always seems to overstay one's welcome on subjects such as this. Perhaps for once one can bow out with a better balance of arguments than mainly black magic or snake oil on the one side. As said at the beginning, with me it is not about people's preferences (for the umpteenth time it is their money), but the way in which.
Regards.
I have do doubt cables to sound different in many cases. We've been though many shoot-outs to see whose cables sounded better. Although the results were different on different systems, it's a fact that there is an audible difference. Now whether scientists and engineers can prove it or not is a different matter. It took centries to prove the world is round and even much longer to prove other things. But's it's just as easy to look at tests that show no differences to find flaws.
I welcome people to come to this neck of the woods with your cables to have a shoot-out. Maybe the nay sayers can drop in and listen for themselves.
I welcome people to come to this neck of the woods with your cables to have a shoot-out. Maybe the nay sayers can drop in and listen for themselves.
soonqsc,as scottmoose said and I agree,cables do not have a sound of their own.In fact I would say that there are no good cables,only less bad ones.It is also correct that if one hears differences between two cables and choses the one,there is a possibility that in another system something else would be prefferable.So if one wants to compare two or more,he better do it in his system.
Hi Folks,
I've been out of the speaker building hobby for a little while, since having a couple kids, so I'll do my best to describe my test of various speaker cable types.
I was alerted by a fellow speaker builder in my area that this test was being discussed, and I had a quick look at the discussion to see what the confusion might be.
First of all, I should say that I'm a skeptic at heart, and especially vocal on issues of an audiophile nature. A few of my comments have been published by James Randi's website in this area. So my test wasn't an attempt to "prove" that different cables actually sound different, so much as to find the best test leads when I'm using my test setup for speaker and x-over design. I was looking for test leads that would offer little or no influence of their own up to about 100KHz (the limit of my test system). I especially didn't want to run the risk of capacitance causing any feedback loops at the higher frequency.
The results of the test surprised me as well, since I didn't expect to see any impedance rise within the audio freq range. I also thought that the resistor would provide significantly higher inductance than the wire would (and it did).
The test was simple. I used LMS with a four-wire impedance measurement (which measures true impedance using a frequency sweep. The system was calibrated to remove it's own output impedance contributions) connected to the DUT and terminated with a .22ohm resistor (chosen to be approx. what I expected the DUT impedance to be).
After I ran the sweep with the DUT, I also ran the sweep across the resistor only. The resistor data was inverted, and applied to the DUT + resistor data using the LMS tools for this purpose. The resulting data (what I called "normalized") essentially has all components other than the DUT removed mathematically. So the resulting test data should contain only the DUT's signature.
Does that help? I welcome any criticisms of the test procedure, but I felt confident, at the time, that the results were valid. However, I did also mention when I sent the data to the fellow who drew up the graphs, that once connected to a reactive load in an amplifier/speaker combination that I felt the amp damping and speaker impedance would easily swamp any significant contribution the speaker wire would have.
Considering the cost, however, of Cat5 (this was teflon dielectric) and it's FT rating, it was a no brainer for using in this way, especially in new home builds for running AV wire. Worst case, you could use it for telephone and data if you didn't use it for audio! I still recommend it to people for this purpose, so they don't get ripped off on Monster cable "in-wall" speaker wire.
I've been out of the speaker building hobby for a little while, since having a couple kids, so I'll do my best to describe my test of various speaker cable types.
I was alerted by a fellow speaker builder in my area that this test was being discussed, and I had a quick look at the discussion to see what the confusion might be.
First of all, I should say that I'm a skeptic at heart, and especially vocal on issues of an audiophile nature. A few of my comments have been published by James Randi's website in this area. So my test wasn't an attempt to "prove" that different cables actually sound different, so much as to find the best test leads when I'm using my test setup for speaker and x-over design. I was looking for test leads that would offer little or no influence of their own up to about 100KHz (the limit of my test system). I especially didn't want to run the risk of capacitance causing any feedback loops at the higher frequency.
The results of the test surprised me as well, since I didn't expect to see any impedance rise within the audio freq range. I also thought that the resistor would provide significantly higher inductance than the wire would (and it did).
The test was simple. I used LMS with a four-wire impedance measurement (which measures true impedance using a frequency sweep. The system was calibrated to remove it's own output impedance contributions) connected to the DUT and terminated with a .22ohm resistor (chosen to be approx. what I expected the DUT impedance to be).
After I ran the sweep with the DUT, I also ran the sweep across the resistor only. The resistor data was inverted, and applied to the DUT + resistor data using the LMS tools for this purpose. The resulting data (what I called "normalized") essentially has all components other than the DUT removed mathematically. So the resulting test data should contain only the DUT's signature.
Does that help? I welcome any criticisms of the test procedure, but I felt confident, at the time, that the results were valid. However, I did also mention when I sent the data to the fellow who drew up the graphs, that once connected to a reactive load in an amplifier/speaker combination that I felt the amp damping and speaker impedance would easily swamp any significant contribution the speaker wire would have.
Considering the cost, however, of Cat5 (this was teflon dielectric) and it's FT rating, it was a no brainer for using in this way, especially in new home builds for running AV wire. Worst case, you could use it for telephone and data if you didn't use it for audio! I still recommend it to people for this purpose, so they don't get ripped off on Monster cable "in-wall" speaker wire.
Hi macgyver10,
Where the cables laid out flat (I assume that you would)?
How where the cables terminated to the load resistor?
How much current were you testing at?
What instrument is "LMS".
I agree fully with you that speaker reactance and amplifier damping should swamp any cable contribution. I will put forth the idea that the cable and some amplifiers may interact. Interaction between the amplifier and speaker cable is a bad thing.
I also have a problem with a CAT-5 cable having a similar resistance to a 12 GA cable. I work with both and would never consider using CAT-5 for speaker wire. I do spec it for telecom as it has very good noise rejection characteristics. Harder to work with than CAT-3 (our old standard).
I was curious to know if you found that your total impedance was dominated by connection resistance.
Finally, thank you for explaining how the test was done.
-Chris
Where the cables laid out flat (I assume that you would)?
How where the cables terminated to the load resistor?
How much current were you testing at?
What instrument is "LMS".
I agree fully with you that speaker reactance and amplifier damping should swamp any cable contribution. I will put forth the idea that the cable and some amplifiers may interact. Interaction between the amplifier and speaker cable is a bad thing.
I also have a problem with a CAT-5 cable having a similar resistance to a 12 GA cable. I work with both and would never consider using CAT-5 for speaker wire. I do spec it for telecom as it has very good noise rejection characteristics. Harder to work with than CAT-3 (our old standard).
I was curious to know if you found that your total impedance was dominated by connection resistance.
Finally, thank you for explaining how the test was done.
-Chris
Johan Potgieter said:
I wasn't talking about properly controlled blind tests. I was talking very specifically about 1 widely used test -- ABX.
And one would have to be very careful designing any proper blind test of a hifi system as the data being collected is one level removed from what you are trying to measure.
dave
Panicos K said:
The original question was "Do speaker cables make any difference?" It seems we just agree they do. If a system is well designed, in my experience, the better cables are always consistently better. In terms of comparison terminology, I always use the positive terms unless there is a fixed reference.😀
Hi soongsc,
I have a different answer.
Below (bigger size) a minimum gauge (assuming copper), then no. Not unless they interact poorly with the amplifier.
I really think it's that simple.
-Chris
Edit: spelling
I have a different answer.
Below (bigger size) a minimum gauge (assuming copper), then no. Not unless they interact poorly with the amplifier.
I really think it's that simple.
-Chris
Edit: spelling
soongsc said:
Do you consider my Belden 10 ga 64 strand copper audio cable in that group?
[Rant hat on]
Like you I'd love to put my cable against other so-called boutique cables. Mine is 50 cents a foot when you buy a big roll. If someone out there feels their $1, $10 or even $100 a foot are beter, I continue to welcome a shoot out.
So far no takers. No surprise.
When where you place your speakers is so much more important than the wire you use, I feel this thread will never get us anywhere. So many of us are restricted to where we are "allowed" to have our speakers AKA the WAF factor, that talking about wires is a waste of time. Even if you are not restricted and place them where you wish, the wires are so low on the priority scale that it amazes me that these threads can go on for as long as they do. Why is the lowest man on the totem pole of such interest?
[Rant hat off]
Cal Weldon said:
In my system, i'd put my single pair of CAT 5 strands up againt your belden and expect it to be better.
Your system. I don't know.
dave
Hi Dave,
Interesting. What are you running? BTW, if it's because the 10 GA will not fit in the terminals - no fair! 😉 Keep in mind that a CAT-5 has less copper than a 16 GA easily.
-Chris
Interesting. What are you running? BTW, if it's because the 10 GA will not fit in the terminals - no fair! 😉 Keep in mind that a CAT-5 has less copper than a 16 GA easily.
-Chris
wire costs money? hmm.
seriously,
-is where I am at as well. Screwin' with the amp is where all the trouble starts.
IMHO
seriously,
-is where I am at as well. Screwin' with the amp is where all the trouble starts.
IMHO
Cable as an audio component
Instead of measuring impedance, let's measure a loaded speaker cable as a blackbox with some frequency and phase response. Audio Precision System 2 can measure FR to 3ppm. Now I doubt anyone can differentiate 3ppm or 0.000026 dB. Few can differentiate 0.1 dB, if any.
Instead of measuring impedance, let's measure a loaded speaker cable as a blackbox with some frequency and phase response. Audio Precision System 2 can measure FR to 3ppm. Now I doubt anyone can differentiate 3ppm or 0.000026 dB. Few can differentiate 0.1 dB, if any.
- Status
- Not open for further replies.