@ebm
Single line replies are sad.
What covid research can teach us about audio measurements.
Recent studies in Canada for patients with so-called long covid show us on how science and measurements and research actually works.
Patients with long covid suffering from limited ability to exercise passed most "normal" tests but it took a new type of test to positively identify a mechanism that explained why the patients suffered.
Honestly there is a lot of snake oil and charlatanism in our hobby, and I don't claim to discount that fact. What I do want to say is that science doesn't rest with 50 year old measurements. It evolves to measure and explain constantly.
The reason I am personally dissatisfied with audio measurements in the common literature is exactly because of this stagnation, and when these fail us we trust our ears and gut for lack of better tools.
Anyone who runs the same 20 measurements on an amplifier or DAC and claims it is science and that these measurements are all that can be known is fooling themselves into believing that they are scientists or that we have reached the limits of understanding.
And above all, caveat emptor!
- ...
- 59 posts total
| Post removed |
Our brains do a lot of correction, making sound into what we expect to hear, given our experiences. Some day we'll see how hard it works via brain scan. That could be a key measurement in ear fatigue vs relaxation. My analogy is another sense - Vision. I wear a not unsubstantial prescription that includes near sightedness and astigmatism. With a new prescription or a change in lens material or even just a regrind, first tryout is always disorientation. Peripheral vision is warped with straight lines noticeably curved. After a few days everything appears normal again. That's my brain at work "getting used to it". Tests say I'm 20/20 corrected (or so). That's the measurement standard all prescriptions are judged by. No test measures how hard my brain works to "get used to it" or why everything looks more natural, atmospheric and relaxing when I take off my glasses. |
Like others have said, I do not see the relevance at least to the arguments in audio. Few if any of the scientific community doubt the existence of long Covid, and physiological and psychological testing confirms it exists. Knowing it exists, they went looking for the cause. You will have to provide strong proof of something in audio that contradicts current scientific understanding in order to get them to chase the cause. I am sure we could come up with 5 or 10 things, medical related, that many people are convinced are true, that the medical community is not chasing, or has chased and debunked. Are not most measurements in audio a simple electrical signal, no matter the source? How much more complicated does it have to get than X = A-B. A = input, B = output. Hardly as complicated as the human body. The other measurements in audio are speakers, and microphones. No one considers those settled science. If they did I would be out of a job. Fortunately I will retire before we have direct brain plug in. |
I doubt I will make any impact in this discussion as a discussion of measurement takes a level of understanding of several topics that is likely beyond most people participating in this discussion. Stereophile presents a limited set of measurements. Audio Science Review presents a more comprehensive set of measurements but are limited in time. Do engineers and manufacturers do more tests? Some do. Many do not. Many of the tests may not be for sound. For speakers, I expect many small brands don't have the equipment or knowledge for extensive testing, certainly not for speakers. I wrote a long paragraph but deleted it. I would suggest reading this and coming back: https://www.ap.com/technical-library/using-multitones-in-audio-test/ I don't think this existed 20 years ago or even 15. "The characteristics of a multitone waveform (e.g., spectral content, histogram, crest factor, etc.) give it a much closer resemblance to typical audio program material like music or speech, than a single sine wave"
The speaker testing we do is much more comprehensive. Electronics are nice, predictable, and repeatable (maybe not that SET amp). Distortion normally scales with level, frequency response does not change with levels (it may with load). Nice. Easy. Speakers are nasty beasts. Distortion can be low then rise rapid, and is different at every frequency. Impedance can change with level, which can lead to changes in frequency response (and crossover). Different drivers have different dispersion. Distortion is not consistent across angle. Voice coils heat up leading to parameter changes and passive crossover shifts. I would spend far less time worried about whether your electronics are tested adequately and spend a lot more time worrying about whether your speakers are tested properly.
|
- 59 posts total