My brain, at least chooses to hear distortion, or ignore it.
Or roll-off and ignore it.
I have no doubt that some people can hear the difference in cables. I can't. And it may be because I chose not to.
I realize there's optical illusions, and there audio illusions, and things may be fooling my ears, but I can look at an optical illusion and just just to see what I want to see and take it no further.
Or roll-off and ignore it.
I have no doubt that some people can hear the difference in cables. I can't. And it may be because I chose not to.
I realize there's optical illusions, and there audio illusions, and things may be fooling my ears, but I can look at an optical illusion and just just to see what I want to see and take it no further.
So what about reflections? I saw that mentioned in a previous post.
In video and SVGA signals, reflections, or more accurately, ringing is a big problem - bad quality cables,connectors and impedance mismatch cause it. I had never thought about it before for audio frequency signals, but could it be happening? Could it be measured?
Having spent many years as a video engineer and projectionist, I can tell you that cables do make a big difference there, it's plain to see. If something similar is happening to signals on a speaker cable, the results might be much harder to identify, but just as real.
In video and SVGA signals, reflections, or more accurately, ringing is a big problem - bad quality cables,connectors and impedance mismatch cause it. I had never thought about it before for audio frequency signals, but could it be happening? Could it be measured?
Having spent many years as a video engineer and projectionist, I can tell you that cables do make a big difference there, it's plain to see. If something similar is happening to signals on a speaker cable, the results might be much harder to identify, but just as real.
A few people have said that they can't really tell much of a difference between cables used with class d amps but can with other classes of amps. Think about that for a second... What does the output stage of a class d amp look like? LC filter followed by the load. How can a speaker wire be modeled? An LC filter with a resistance in series with the inductor... It just adds another low pass filter with a very high frequency cutoff between the amp's LC output filter and the speaker and that's probably the reason you most likely you won't detect any difference between cables with a class d amp.
How about you hook up an electrostatic speaker with a series resistor. Hmmm, RC network? See how that affects the speaker cable debate!!
😀
-Wes
😀
-Wes
My cv (under my name) will indicate that I am an ee. I have had the priviledge to work in a research organisation, as well as having had limited participation in hearing research. (How I wished I could have continued that!)
Firstly I think I owe it to mention that participation by all to this thread is of a particularly respectful and mature level. Regular members may have noticed that I write for a local magazine, and over the years I have received more than my fair share of flak regarding this topic. How folks love to shoot the messenger!
Perhaps a few further considerations. Indeed, the hearing faculty (ear-brain) is capable of sensing stimuli below the conscious audibility limit as well as above 20KHz, though we do not "hear" those as such. Consider the dog-whistle, which emits a frequency of some 30 Khz or higher. One does not hear it, but some develops a headache after prolonged exposure. (I can personally testify to that.) Further, the ear can "compress" enery levels up to a factor of >1000 and is substantially self-compensating (otherwise life would have been unbearable).
[Many of us will remember the old school experiment demonstrating compensation of feeling: Three bowls of water, one very cold, one hot and one at room temperature. Place left hand in cold bowl, right hand in hot bowl for a minute, then both in the bowl at room temparature. The left hand will experience that water as warm, the right hand as cold - the same water. This is a rough illustration of the compensatory nature of all our senses.]
The hearing faculty (like several other senses) originally evolved as a warning device. In the existance of the human race music as we know it has only been there for less than the last 1% of our existence. It is therefore to be expected that the brain will still have instincts to fill in artifacts not related to music. Thus it can be mislead, and that is not a derogatory statement - it has been shown to follow routes of expectancy. Cowanrg is correct in pointing this out; psychologists will confirm that this phenomenon is one of the most difficult tendencies in human nature to overcome - that regardless of the subject in question. Again, that is a subject on its own, but there is no doubt of its existence.
Panomaniac is correct in stating that cable can definitely make a difference, e.g. in rf. It is catastrophic to ignore this and nobody implies the contrary. But the point is whether this is true in the (limited) frequency spectrum of audio! A lot of arguments especially in promotion is only valid at several 100 KHz to well into the MHz region. With respect, it is often such claims, bordering on the ridiculous, which makes it very difficult for one to accept that they have any knowledge of even basics.
I have only recently joined this excellent site and was not aware of previous similar threads. Though repetitious then, perhaps some could still learn from such thoughts, since relevant acousto-medical research results are very difficult to access - but they do exist.
Regards.
Firstly I think I owe it to mention that participation by all to this thread is of a particularly respectful and mature level. Regular members may have noticed that I write for a local magazine, and over the years I have received more than my fair share of flak regarding this topic. How folks love to shoot the messenger!
Perhaps a few further considerations. Indeed, the hearing faculty (ear-brain) is capable of sensing stimuli below the conscious audibility limit as well as above 20KHz, though we do not "hear" those as such. Consider the dog-whistle, which emits a frequency of some 30 Khz or higher. One does not hear it, but some develops a headache after prolonged exposure. (I can personally testify to that.) Further, the ear can "compress" enery levels up to a factor of >1000 and is substantially self-compensating (otherwise life would have been unbearable).
[Many of us will remember the old school experiment demonstrating compensation of feeling: Three bowls of water, one very cold, one hot and one at room temperature. Place left hand in cold bowl, right hand in hot bowl for a minute, then both in the bowl at room temparature. The left hand will experience that water as warm, the right hand as cold - the same water. This is a rough illustration of the compensatory nature of all our senses.]
The hearing faculty (like several other senses) originally evolved as a warning device. In the existance of the human race music as we know it has only been there for less than the last 1% of our existence. It is therefore to be expected that the brain will still have instincts to fill in artifacts not related to music. Thus it can be mislead, and that is not a derogatory statement - it has been shown to follow routes of expectancy. Cowanrg is correct in pointing this out; psychologists will confirm that this phenomenon is one of the most difficult tendencies in human nature to overcome - that regardless of the subject in question. Again, that is a subject on its own, but there is no doubt of its existence.
Panomaniac is correct in stating that cable can definitely make a difference, e.g. in rf. It is catastrophic to ignore this and nobody implies the contrary. But the point is whether this is true in the (limited) frequency spectrum of audio! A lot of arguments especially in promotion is only valid at several 100 KHz to well into the MHz region. With respect, it is often such claims, bordering on the ridiculous, which makes it very difficult for one to accept that they have any knowledge of even basics.
I have only recently joined this excellent site and was not aware of previous similar threads. Though repetitious then, perhaps some could still learn from such thoughts, since relevant acousto-medical research results are very difficult to access - but they do exist.
Regards.
Right Brian! I've been thinking about this one for quite a while. What does the LC low pass filter do to the amp/speaker interface? Many people like the sound of amps that have a passive components on the output, e.g., transformer or cap. Does the LC filter on the end of a class-d amp have a similar effect? Is the effect due to interaction with the speaker, or the cable, or both?
I agree with Johan that this is a good thread. These sorts of threads are generally hated, but I think there is good information and speculation here.
SO
I’m a mild cable skeptic, but let me flip the tables. Many cable enthusiasts want to prove that cables can make a difference for whatever reason - impedance, phase, materials, voodoo, etc. Proof of some sort is often asked for and presented.
Now I ask those who claim cables make NO difference to prove why. Give us some good solid electrical (or whatever) reason that cables would NOT make a difference. Can you do it?
SO I’m a mild cable skeptic, but let me flip the tables. Many cable enthusiasts want to prove that cables can make a difference for whatever reason - impedance, phase, materials, voodoo, etc. Proof of some sort is often asked for and presented.
Now I ask those who claim cables make NO difference to prove why. Give us some good solid electrical (or whatever) reason that cables would NOT make a difference. Can you do it?
raintalk said:
Its interesting. It could be argued that the ideal system would be one where it was obvious that all the sound was comming from two speakers, and that there was no issue closing your eyes and pointing to them. I like the illusion myself. Its what makes it good.
panomaniac said:
No doubt relections happen, the point is does it matter. In some types of amps, I suspect it does, otherwise speaker wire wouldn't blow them up so easy. I doubt its ever that you could hear it though, its what the reflextion does to the amp perhaps.
BWRX said:
Expounding on this a bit more. Whats the output impedance of a Class D at high frequency> Basically a short circuit, the output cap to ground.
WHat's the output impedance of a solid state A/B amp at high frequencies? Not zero for sure! And acutally the output of such an amp is actually an input with some non-zero gain!
Best Regards,
Mike
Panomaniac,
Indeed a most valid point!
But we have tried that all along, by showing that the values of inductance and capacitance prevalent in cables can have no effect, neither is line impedance a factor at the lengths of cable used domestically, etc. I sometimes get the feeling though, as I have stated in earlier post, that the position of some is that where subjective observations contradict science, the latter may be dismissed. Is that fair?
I am getting too much exposure here for my worth, so let me just sum up what I have established in earlier research. The worst case inductance of a 4m length of speaker cable can have an effect of 1 dB only above some 500 KHz (for 8 ohms). The worst case capacitance similarly will make a difference of 0,013 dB only at 20 KHz, to put it differently. As this includes phase effect, I think the point is made. (But again, excluding the possible effect on the stability of some amplifiers; that is a different point.) Has it been measured? Yes. And I think you know from your background that cable impedance only gets into the picture for lengths of 0.1 wavelength or more. At 20 KHz that length would be in excess of a kilometre. What more is desired?
About blind tests folks are more sensitive since they feel it indicates criticism of their ears. Still, results of several properly organised tests are indisputable. ("Properly organised" as in taking enough samples etc. as I have argued earlier.)
Considering the effect of cables on the passive filter in digital amps (and in effect regarding all loudpeakers with built-in cross-over filters!) there are equally interesting facts...... but I think I must shut up for a while lest a burly moderator instructs me to give others a chance or else...... It is also rather bed-time here, so perhaps later, if folks have patience.
Regards.
Indeed a most valid point!
But we have tried that all along, by showing that the values of inductance and capacitance prevalent in cables can have no effect, neither is line impedance a factor at the lengths of cable used domestically, etc. I sometimes get the feeling though, as I have stated in earlier post, that the position of some is that where subjective observations contradict science, the latter may be dismissed. Is that fair?
I am getting too much exposure here for my worth, so let me just sum up what I have established in earlier research. The worst case inductance of a 4m length of speaker cable can have an effect of 1 dB only above some 500 KHz (for 8 ohms). The worst case capacitance similarly will make a difference of 0,013 dB only at 20 KHz, to put it differently. As this includes phase effect, I think the point is made. (But again, excluding the possible effect on the stability of some amplifiers; that is a different point.) Has it been measured? Yes. And I think you know from your background that cable impedance only gets into the picture for lengths of 0.1 wavelength or more. At 20 KHz that length would be in excess of a kilometre. What more is desired?
About blind tests folks are more sensitive since they feel it indicates criticism of their ears. Still, results of several properly organised tests are indisputable. ("Properly organised" as in taking enough samples etc. as I have argued earlier.)
Considering the effect of cables on the passive filter in digital amps (and in effect regarding all loudpeakers with built-in cross-over filters!) there are equally interesting facts...... but I think I must shut up for a while lest a burly moderator instructs me to give others a chance or else...... It is also rather bed-time here, so perhaps later, if folks have patience.
Regards.
Johan Potgieter said:
And properly designed... people often extend results of blind tests beyond their scope... for instance the popular ABX test can only statistically validate a result where people can tell a difference. If no statistical difference is found, the test tells us nothing, but it is often used to support the argument that there is no difference (the test is not strong enough to do so).
dave
I would not try to say that cables don't make a differrence. I know that they can... at least in extreme cases (RF, bad amp, weanie speakers). What frustrates me is; NO data... time and time again. AND, anyone with a cable to sell is not a source of, despite their vocabulary and writing skills, credible data.
😉
😉
poobah said:
That is the biggest thing with cables... it is a morass out there. Some people trying to honestly make a good product, and at least just as many wanting to pull the wool over our eyes and extract the maximum dollar out of our pockets -- and it is almost impossible to tell who is who... and on top of all that, the amp & speaker cannot be separated out of the mix, so somethings work with one system and not with another. There is a wealth of data points in a much, much larger space, many invalid, with little to correlate them...
In the end... are you enjoying the music?
dave
BWRX,
I will say only this... nowhere... except in audio... can you get away with separating a transducer (loudspeaker) from its power source in the way that we do... open loop.
There was a day, when we knew more, that when you bought a mic... you the bought the matching preamp as well... just because "they told you to".
No one was more interested in your good results than the guy who designed it...
🙁 🙁 🙁
I will say only this... nowhere... except in audio... can you get away with separating a transducer (loudspeaker) from its power source in the way that we do... open loop.
There was a day, when we knew more, that when you bought a mic... you the bought the matching preamp as well... just because "they told you to".
No one was more interested in your good results than the guy who designed it...
🙁 🙁 🙁
Johan Potgieter said:
Hmmm, Yes, its fair. The end result is how it sounds.
The science might not be there yet.
What I do know from experience, is that things that can put very good test equipment to the very limit can actually be hidiously obvous to the ear. Double double double blind even. Stuff that literial "little old ladies" can catch 100% or the time!
Cables are less sever in my experience, but very significant. I'm an engineer in audio, and what really facinates me is how much some stuff matters that really shouldn't. Perhaps I've "sold out" on my education, but I like to think of it as just gotten wise! I don't know why sometimes, and just accept it.
A buddy of mine who self taught himself amp design, and came up with truly amazing sounding designs. He found some stuff that was very audible that I just couldn't figure out why it would be such a major issue. If he were officially trained, he'd never of bothered to even look there.
Our science is quite accurate for things we know about but our hearing isn't all that well understood.
Can't beat this dead horse much longer ;>)
Best Regards,
Mike
BWRX said:The solution to taking speaker cables out of the equation: put them in the feedback loop!😉
I'm trying to get feedback loops out of my system!
:^)
dave
using a bit of negative output impedance
It would be an interesting experiement to take advantage of the instrumentation type input on some of these amps like the UcD to cancel out the speaker impedance.
Just a bit (speaker cable length/Gain) of speaker cable connecting the amp board to the speaker terminal, then connect the positive input to the speaker terminal.
The output impedance will be -Zcable then, fist order approximation.
This does ignore that the cable is a distributed impedance though.
Mike
It would be an interesting experiement to take advantage of the instrumentation type input on some of these amps like the UcD to cancel out the speaker impedance.
Just a bit (speaker cable length/Gain) of speaker cable connecting the amp board to the speaker terminal, then connect the positive input to the speaker terminal.
The output impedance will be -Zcable then, fist order approximation.
This does ignore that the cable is a distributed impedance though.
Mike
Portlandmike said:
Portlandmike,
I am also not one to continue where the road seems to have ended or become unproductive, but with all this post we do not appear to be there quite yet - and points have been raised that beg answers. Let us at least bow out without loose ends.
Your statement puzzles me. We certainly do not know everything, but we do know certain things. We know that the earth is round and revolves round the sun, the laws of gravity - and Ohm's Law! And as an engineer you will know that we are simply talking about Ohm's Law here and nothing more. (Again I am not disputing amplifier reaction and other secondary effects, but those are common factors.) Otherwise folks who dispute this must be found and make their point.
Also granted that we do not know everything about hearing by a long shot. But that is a common factor too in the seemingly opposing two situations: The same hearing that has claimed to detect differences between two conditions (different cables in this case), suddenly cannot do so any more under the same conditions, with the same music, etc. than before, the only difference being that the listener is now unaware of which situation he is listening to. That is what the tests I referred to reduced the scenario to.
In reaction to other points viz. the normalising filter in a digital amplifier, that is a very valid point. (Digital techniques formed part of my thesis.) But such effects in any well-designed amplifier should again be well above the audio region. A better point would be to look at the all-but-zero-phase-angle impedance provided by many loudspeakers containing cross-overs. Loudspeakers are glibly classified as x-ohms, but I have measured some pretty weird impedances over the audio band and above in some loudspeakers. But that would mainly reflect by the ability of the amplifier to handle the same. If anything I would suggest that such effects would mask any slight influence of cable reactance. Again, this is a common factor in the type of tests referred to above.
Regards.
Johan
- Status
- Not open for further replies.