We have another talk with Ole from System Audio. This time we are discussing the use of measurements. How important are they? Can we measure everything? And what about just listening? Is it reliable enough?
The hifi world is mostly split in two ‘segments’: people that want hard data and people that just follow their senses. What is better? Do we need measurements? And if so: when are they usefull?
Former chief acoustic designer of Sonus faber Paolo Tezzon said in an interview: ” science and measurements are unavoidable, it’s super important. However science is not everything. There is a step above which is beyond science, to me.” I completely agree.
Hi Torben,
In addition to what colleague Martijn writes: we do measure cables and although I am not an expert in the measuring realm, I can assure you that many audio brands would wish they had a measuring lab as well equipped and calibrated as the setup colleague Jaap has developed. Have you read, for instance, the test of 64-ish speaker cables and interlinks? This section gives you an idea: https://www.alpha-audio.net/review/interlinks-dont-do-anything-or-do-they-32-rca-cables-analyzed/24/
You say: ‘According to measurements and theory, there is no basis for saying there is a difference, and this can be confirmed with a null test. There is simply no difference in signal transmission between a $3 cable and a $700 cable.’
The measurements in our tests prove otherwise. Just to mention the “Delay and Propagation Variance: A moderate positive correlation (0.53) indicates that cables with higher delay times may also exhibit higher propagation variances. This suggests a relationship between the timing of signal transmission and the consistency of that timing across different signals or conditions.”
This is just an example. We at Alpha Audio like to remain fascinated by audio equipment that carries us away and into the music. That can be a CD player, an amplifier, a speaker, a power filter, a cable, a power supply. Measurements help us to explain why, but some things remain in the unknown. And we think that’s OK.
We do not aim to ‘satisfy’ any platform or magazine. Why should we?
I should have mentioned there are no measurable differences between cables in the audio range that humans should be able to hear.
Here is a measurement and a null test of Belden’s bid for a high-end cable.
https://www.audiosciencereview.com/forum/index.php?threads/belden-iconoclast-xlr-cable-review.33929/
Where the frequency dependent propagation velocity (Vp) is the main argument, also the impedance varies a lot through the audio range.
I haven’t calculated the consequences, but I’m pretty sure it’s meaningless for what we hear.
But together with high frequency pulse measurement, it will give measurable differences between cables as you show.
But it will not change the transfer of the audio signal, something you have also concluded and the null test is the proof.
However, I completely agree with you that we hear sound differences that we shouldn’t according to measurement and theory, even in blind tests, but how?
You ask the questions which we have asked ourselves, but you cannot make the jump across the often made implicit assumption to interpret the results. The assumption being: the effect has to be found in the audible frequency range, being the effect. But it isn’t. It is about the effect of other electrical characteristics or parameters that are the cause which lead to audible effect in the connected equipment, not in the cable. It is what is called ‘a second order effect’.
Couldn’t agree more.
I’m sorry you don’t understand the problem
If you do a Null test and there is no difference between cables will you explain to me what is the null test not picking up.
The only way we can prove that the Null test is not correct is via a blind test, we can’t prove technically.
But with a blind test we can prove that there is something we should research, there is something the electronics theory does not cover
I understand the problem really well. It won’t convince the ASR people. Because they want to believe there is no difference. It is a predisposed conviction. For the rest of us, we already know there are things that cannot be measured, so what would the test attribute?
Thanks for an interesting conversation, agree that measurement does not tell much about the perceived sound.
But the disagreement between hifi enthusiasts is more simple, it’s about whether there is a difference in sound between e.g. cables.
According to measurements and theory, there is no basis for saying there is a difference, and this can be confirmed with a null test. There is simply no difference in signal transmission between a $3 cable and a $700 cable.
The burden of proof lies with those of us who hear sound differences.
You measure differences in the HF range, there’s nothing strange or new about that, the question is, how can it affect what we hear?
But in the first instance, I think we need to make a blind test of cables that can satisfy ASR and that can be pressurised in AES.
Then sound differences will be a reality for everyone, whether you can hear it or not.
My own experience is that what causes the sound difference is independent of voltage, current and the electrical value of the components, plus the physical length of a cable, for example. It simply doesn’t seem to work according to known electronics theory.
It would be obvious to think it was parasitic components that cause sound differences, but it doesn’t seem to work that way. The only common denominator I see is marterials, which also affect parasitic components, the effect of this can be measured in the high frequency range as you have done, but not directly in the audio range.
I think hifi consists of precision and timbre, everything about precision can be measured, it’s the only thing we can measure, precision is mainly determined by the electrical construction/diagram.
Experience shows that timbre and soundstage are mainly determined by the materials used in the components and cables etc.
To get the sound right, it’s all about finding the materials that give the most natural sound, and here listening is the only tool.
For example, a paper in oil capacitor sounds different to a PVC capacitor etc.
In fact, I think electronics design is starting to resemble Ole Klifort’s work with loudspeakers, where you strive for high precision, which is measurable, and natural sound, which is less measurable and where the ear and brain are indispensable.
Timbre and precision naturally have a greater or lesser common denominator and are difficult to separate completely, personally I always emphasise lifelike timbre and soundstage over ultimately good data , it is no problem to make good data , it is just to use enough feedback.
Ps The post should be perceived as criticism, but as a sharing of experience.
You say: “But in the first instance, I think we need to make a blind test of cables that can satisfy ASR and that can be pressurised in AES.”
Why would we _need_ to do that (I read ‘we’ as ‘audio/hifi market in general’, not just Alpha Audio)? I fail to see a need. There are still people out there who persist in believing the earth is flat and have regular world wide conferences about this. Verifiable, scientific validated tests exist, multiple ones, that proof the earth is not flat, but yet, people refuse those tests and maintain that belief.
You will never change the mind of some people. Let them be. This an audio hobby. My pleasure of listening to music is not diminished by the beliefs of anyone contributing to the ASR forum or anyone participating in AES. My listening experience doesn’t change a bit either because of the knowledge I have of the existence of websites like ASR. I simply do not care.