Quote:
I can't believe I missed this gem for three days.
If what you say is true, that only the human ear can detect differences between cables, then surely there would be a successful blind test in the last 30 years? And surely these cable manufacturers would want to help fund a good one, because of all the potential profit they could gain by proving their cables really are superior? Unfortunately none of that's true.
What does a silver cable do (or any cable that makes a difference) that allows the headphone driver to respond faster to the signal? The headphone driver and resonance from the design is the limiting factor when you're talking about this sort of "detail". Does a "cleaner" signal from a better cable improve the driver's response? All a driver is doing is responding to the signal it's fed. If the signals are identical as far as the limits of our hearing are concerned (and they are by far) then the driver isn't responding much different and not audibly. Why can't this be measured, as you claim?
Here's nick_charles's cable test. Strange you couldn't find it, it was the first result from a forum search for "nick_charles cable test".
No, we can't all agree that a $300 DAC is superior to a $5 sound card. Not without specifications. The price tag means
nothing. There is a countless number of crap products out there sold for exuberant prices because they're "musical" or hyped in some other way. Also, there generally isn't a difference in frequency response between DACs of various price ranges, unless one is made to be colored. The differences come in THD, IMD, noise, reconstruction of square waves, etc. Most of these, past the $100 point if you're generalizing based on price, are inaudible. My sound card, for example, cost me $170 and measures well below audibility in all the tests Stereophile put it through. There will be no audible difference between it and the CEntrance DACport, which costs $350. They measure quite similarly, with the DACport earning the win in a couple tests, but both are for all intents and purposes "transparent".
Separation between sounds appears to be the result of driver response time, amount of resonance, and distortion and is also affected by sound stage. Sound stage is a function of driver placement relative to the ear and strategically placed peaks and dips in frequency response to simulate reflections in a room. Detail has about the same requirements as separation, and separation is actually part of detail. Everything in audio is quantifiable and measurable. It's just electronics and physics.
While I certainly can't point to a frequency response graph and say "Yup, that headphone is detailed", I
can reason that there's nothing a cable can do that will improve detail if it measures exactly the same within audible limits. Physics tells us this, graphs and blind tests confirm it. You'll have to supply a lot of good evidence to overturn such a well understood theory, not just "what ifs".
Maybe someone with a really high end system and a wide variety of cables, and the time, could video tape some double blind tests. Perhaps I'll do it when I build a system. I currently only have some ath-a500's, a pimeta and an emu1212m, which is not sufficient to tell any big differences. I had to sell all my really hi-fi stuff a year ago because of the recession, I made this thread to start considering components to rebuild my setup.
But your demand to have someone 'prove' to you what you should believe I think is really silly. This is a big drawback I find in this whole movement of demanding 'verification through scientific expirimentation'. Because what this movement has turned into is not lots of curious, intellectual minds running expiriments, being actively engaged in research. What it has turned into is a bunch of people sitting around, not doing any active research at all, waiting for someone to come along and preach to them the new 'truth' the new 'dogma' that scientific process has found. If you really want to verify this for yourself, then take it upon yourself to inform yourself and research it yourself. Redo your system with solid core silver cables, give it a listen, spend some time with it, have a friend come over and switch them back and forth without you knowing which is which in order to do a blind test. This is what I did, and I found in systems that have the ability to reproduce extremely high levels of detail, solid core silver does produce a change that can warrant the difference in price for DIY solid core silver cables. I really do not believe you will be able empirically measure sound quality through a cable with scientific instruments, and ultimately any test will be relying on someones perception of a difference. So even if you have someone on video tape succesffuly blind testing out silver cables, the question that will still be a mystery is HOW much of a difference is there? Which that answer will not able to be accurately conveyed to you since a person has to subjectively describe, to know that answer you have to do the test yourself.
One thing I notice about nick_charles expiriments was he has no mention of solid silver cable. He only mentions use of silver plated cables in his test. From my experience 'silver plated' cable basically means 'fake silver cable', because just having silver plating on the outside is not enough to alter anything, and the resulting signal still sounds like copper. When people talk of 'silver cable' they are reffering to solid core silver cable 99.9% purity, nick_charles did not appear to use this. His tests were basically done with cables that are all primarily composed of copper. Which interestingly, they still had differences... So cable can affect output of an analog audio signal. I do understand how asbolutely TINY the difference is in the results. But it makes me curious to see what results he would get when using a solid core silver cable.
The distinction between 'silver plated' and 'solid core silver' also makes me wonder, has there been more people who have confused this? Have there been people buying 'silver' cables which are really 'silver plated', expecting a difference, only to be dissapointed with what is really still primarily a copper cable? There reporting silver makes no difference?
There are more flaws with nick_charles expiriments as well. And as I get into this, I don't want anyone, or nick_charles to think I am putting them down. I think the expiriment was a great contibution to the community. But there is room for improvement.
For one there is no amount of difference defined for when a change in sound is clearly audible. He produced all these tests from various, primarily copper, cables and in the end said 'this is what the difference is'. And gave the numbers. Without any context these numbers tell us very little. You can say 'they're very small', but still, that tells us very little. To put all the numbers and graphs he gave into context, we need to also produce a number and a graph that does have a difference that is clearly audible. So then we can say "When the numbers and graphs deviate this much, we have determined that the difference is audible to the average human ear". Because it could be that the tiny deviation that nick_charles recorded is actually audible.
I remember back when people were doing tests of lossy audio files to prove that lossless is better. If you recorded the frequency response of a lossy track in ogg vorbis format, that you could tell was lossy, and then you recorded the frequency response of a lossless track, you had to zoom way way in on that frequency response to see the difference, and we were talking extremely small differences there as well. Just slightly different angles on the tops of peaks and things like that. But it was enough to change the resulting audio. Those tests led the impression on me that In real life recordings it does not take alot of change in a resulting frequency response to produce an audible difference. So nick_charles expiriments producing "Very small numbers", on primarily copper cables, does not signify to me enough evidence to disregard my experiences that have demonstrated to me solid core silver cables can make a difference.
Aside from that I also think nick_charles needs to do his expiriments at a MUCH higher sampling rate, ideally the sampling rate needs to be 4 times the sampling rate of the actual audio file so that you can then pick up any difference a cable might impose on the wave between the samples of the actual audio file. Nick charles said he used 1024 recording samples for a 24 second audio clip. Assuming the clip is CD quality at 44.1 khz, that means the entire audio clip has 44,100 * 24 samples in it! He has only 1024 recording samples to pick up audible differences in over a million samples of audio data. Literally over 90% of the samples from the audio file are not accounted for in his recording. What needs to happen is an audio source needs to be played at 48khz, and then it's data sampled at 192khz through the cable. Because even though the time in between samples at 48khz is REALLY GOD DAMN SMALL (excuse my caps but I understand how small that time gap is). You can STILL hear a difference if there is variation between those 48khz samples. To me, if you play a reverbed saw wave sampled at 48khz, and a reverbed saw wave sampled at 192 khz, the difference is extremely obvious. Even though the frequency response would be --exactly-- the same between the 48khz and 192khz reverbed saw wave. Having more samples in between each 48khz sample produces a clearly audible difference. If your sound card supports 192khz you can actually do that test yourself with a synthesizer program that can output to 192. I say all that note that quite literally, there could be very slight variations happening in the resulting audio signal through a cable that could only be picked up at sampling rates greater than 48khz, and they would still produce a notable audible difference.
Also:
"Separation between sounds appears to be the result of driver response time, amount of resonance, and distortion and is also affected by sound stage. Sound stage is a function of driver placement relative to the ear and strategically placed peaks and dips in frequency response to simulate reflections in a room. Detail has about the same requirements as separation, and separation is actually part of detail. Everything in audio is quantifiable and measurable. It's just electronics and physics."
While I think that is true. I think the subject of what exactly reproduces sound stage is more complex than that. Sound stage is obviously not dependent on the driver's qualities. It is dependent, even more I think, the sound recording. And in that case, it's qualities to the recording that more accurately reproduce sound. And very slight alterations to the reproduction of that sound recording could alter the way it presents the positions, or depth of sounds.