12-Mar-2020, 23:22
@thumb5 ,
Ian,
I've got an ER and it does make an audible difference compared to my previous Cisco switch, at least to my ears. Short of test results showing the effect the ER has and some kind of listening tests demonstrating that the effect is audible, there's always the possibility that I'm mistaken about hearing a difference. I'd be more prone to give some weight to that possibility if what I was noticing was simply a difference in some sonic characteristic such as bass, detail, or the like but the big difference for me is in how I respond to the sound. I've always liked attending to the contribution of individual musicians in the music I listen to (mostly jazz) so I would often focus on a single musician and attend to what they were doing against a backdrop of the sound of the other musicians. Since installing the ER I find I'm doing that much less frequently and instead focussing on the flow of the music and how different elements of the music work together as a whole and that has changed my impression of a lot of the music I listen to. That's a significant change in my listening approach and the only thing I can attribute it to is the ER. That's not the sort of thing you notice when you're doing A/B comparisons. It's a shift in the way I actually attend to the music and while it is caused by differences in some aspects of the sound, it is also quite a different thing to the differences in the sound. Since I installed the ER, the things that attract me in the music I listen to have changed. Something is different and it's not the sort of thing you notice by doing A/B comparisons, it's something that affects what things attract me in music rather than in how much things attract me. That's a qualitative change in what attracts me in the music and how I listen rather than a quantitative change in any characteristic of what I hear.
I too wondered about how why the jitter tests in the ASR don't show a difference and I went back and reread the white paper. On rereading I noticed a point I'd missed on first reading. That is that the jitter side bands cause a "burp" of current between the power and ground pins of the circuit which results in ground plane noise and the ER is designed to reduce ground plane noise. Perhaps what it does has little effect on the jitter measured in the output of the DAC and it's the reduction in ground plane noise that produces the audible benefit.
As well as jitter measurements, the ASR review also included jitter noise measurements but ground plane noise may not be revealed in the noise spectrum of jitter, it may be revealed in the signal to noise ratio of the analog output of the DAC.
We know from 50 years ago in the disputes about the whether amps with the same THD measurement could sound different that amps with the same THD measurement can have quite different harmonic distortion spectra, and from the early days of CD before jitter measurements were common that CD players with similar THD and IMD performance could sound different as a result of different jitter performance. Measurements tell you a lot about what you measure but they tell you absolutely nothing about what you don't measure. One of the big mistakes in the ASR review in my opinion is that while the reviewer measured things he felt should reveal the differences that the ER was claimed to produce, he makes a very strong assumption that he's measuring the right things and that the lack of difference in the measurements indicates that the ER isn't making a difference. If he's measuring the wrong things then his measurements are of no value. I think he's measuring the wrong things. There is, of course, his listening test but if those of us who hear a difference may be mistaken because we expect to hear a difference, then it's also possible that he didn't hear a difference because he expected not to hear a difference. Expectation bias certainly exists but anyone with an expectation is susceptible to it and people who don't expect to hear a difference are no less prone to it than people who do expect to hear a difference. Expectation bias doesn't give a damn about what your expectation is, all you need is an expectation of some kind and you're susceptible.
His listening report has no more intrinsic validity than anyone else's report and it's quite disingenuous of him to demand that anyone who doesn't hear the same thing as he did on a similar sighted listening test to the one he conducted should then conduct a blind listening test with multiple trials and achieve a particular success rate. You validate scientific tests with independent reproductions of the identical test procedure which achieve identical results. Sighted tests have lower validity than blind and double blind tests because they tend to yield more inconsistent results than blind and double blind tests. In my view he's being dishonest by implicitly claiming that his sighted listening test is intrinsically valid while the conclusion that there is a difference that others have reached on the basis of sighted listening tests are universally mistaken.
I've got no technical background in audio or IT so I'm guessing when I outlined above what I think may be going on in relation to the effect of clock phase noise but I do know that if there is an audible result then there must be a measurable difference in something, and I also know that you're only going to find that measurable difference by measuring the right thing. That means that if the very strong preponderance of listening reports stating that the ER does make a difference are correct are correct, then the measurements in the ASR review are measurements of the wrong things.
Alex and/or John may reply and say my guess about what's going on is wrong and give a different answer but the one thing I strongly believe is that the ER does make an audible difference and the reason I am believe that so strongly is because it's a lot harder for me to be mistaken about how the way I listen has changed than it would be for me to be mistaken about whether there is a difference in the characteristics of the sound I hear.
Ian,
I've got an ER and it does make an audible difference compared to my previous Cisco switch, at least to my ears. Short of test results showing the effect the ER has and some kind of listening tests demonstrating that the effect is audible, there's always the possibility that I'm mistaken about hearing a difference. I'd be more prone to give some weight to that possibility if what I was noticing was simply a difference in some sonic characteristic such as bass, detail, or the like but the big difference for me is in how I respond to the sound. I've always liked attending to the contribution of individual musicians in the music I listen to (mostly jazz) so I would often focus on a single musician and attend to what they were doing against a backdrop of the sound of the other musicians. Since installing the ER I find I'm doing that much less frequently and instead focussing on the flow of the music and how different elements of the music work together as a whole and that has changed my impression of a lot of the music I listen to. That's a significant change in my listening approach and the only thing I can attribute it to is the ER. That's not the sort of thing you notice when you're doing A/B comparisons. It's a shift in the way I actually attend to the music and while it is caused by differences in some aspects of the sound, it is also quite a different thing to the differences in the sound. Since I installed the ER, the things that attract me in the music I listen to have changed. Something is different and it's not the sort of thing you notice by doing A/B comparisons, it's something that affects what things attract me in music rather than in how much things attract me. That's a qualitative change in what attracts me in the music and how I listen rather than a quantitative change in any characteristic of what I hear.
I too wondered about how why the jitter tests in the ASR don't show a difference and I went back and reread the white paper. On rereading I noticed a point I'd missed on first reading. That is that the jitter side bands cause a "burp" of current between the power and ground pins of the circuit which results in ground plane noise and the ER is designed to reduce ground plane noise. Perhaps what it does has little effect on the jitter measured in the output of the DAC and it's the reduction in ground plane noise that produces the audible benefit.
As well as jitter measurements, the ASR review also included jitter noise measurements but ground plane noise may not be revealed in the noise spectrum of jitter, it may be revealed in the signal to noise ratio of the analog output of the DAC.
We know from 50 years ago in the disputes about the whether amps with the same THD measurement could sound different that amps with the same THD measurement can have quite different harmonic distortion spectra, and from the early days of CD before jitter measurements were common that CD players with similar THD and IMD performance could sound different as a result of different jitter performance. Measurements tell you a lot about what you measure but they tell you absolutely nothing about what you don't measure. One of the big mistakes in the ASR review in my opinion is that while the reviewer measured things he felt should reveal the differences that the ER was claimed to produce, he makes a very strong assumption that he's measuring the right things and that the lack of difference in the measurements indicates that the ER isn't making a difference. If he's measuring the wrong things then his measurements are of no value. I think he's measuring the wrong things. There is, of course, his listening test but if those of us who hear a difference may be mistaken because we expect to hear a difference, then it's also possible that he didn't hear a difference because he expected not to hear a difference. Expectation bias certainly exists but anyone with an expectation is susceptible to it and people who don't expect to hear a difference are no less prone to it than people who do expect to hear a difference. Expectation bias doesn't give a damn about what your expectation is, all you need is an expectation of some kind and you're susceptible.
His listening report has no more intrinsic validity than anyone else's report and it's quite disingenuous of him to demand that anyone who doesn't hear the same thing as he did on a similar sighted listening test to the one he conducted should then conduct a blind listening test with multiple trials and achieve a particular success rate. You validate scientific tests with independent reproductions of the identical test procedure which achieve identical results. Sighted tests have lower validity than blind and double blind tests because they tend to yield more inconsistent results than blind and double blind tests. In my view he's being dishonest by implicitly claiming that his sighted listening test is intrinsically valid while the conclusion that there is a difference that others have reached on the basis of sighted listening tests are universally mistaken.
I've got no technical background in audio or IT so I'm guessing when I outlined above what I think may be going on in relation to the effect of clock phase noise but I do know that if there is an audible result then there must be a measurable difference in something, and I also know that you're only going to find that measurable difference by measuring the right thing. That means that if the very strong preponderance of listening reports stating that the ER does make a difference are correct are correct, then the measurements in the ASR review are measurements of the wrong things.
Alex and/or John may reply and say my guess about what's going on is wrong and give a different answer but the one thing I strongly believe is that the ER does make an audible difference and the reason I am believe that so strongly is because it's a lot harder for me to be mistaken about how the way I listen has changed than it would be for me to be mistaken about whether there is a difference in the characteristics of the sound I hear.
Roon Nucleus+, Devilalet Expert 140 Pro CI, Focal Sopra 2, PS Audio P12, Keces P8 LPS, Uptone Audio EtherREGEN with optical fibre link to my router, Shunyata Alpha NR and Sigma NR power cables, Shunyata Sigma ethernet cables, Shunyata Alpha V2 speaker cables, Grand Prix Audio Monaco rack, RealTRAPS acoustic treatment.
Brisbane, Qld, Australia
