Hacker News new | ask | show | jobs
by WhitneyLand 2560 days ago
This measurability nonsense is a red herring.

Think about it, the question can easily be answered by blind A/B testing using real listeners with varying aptitude for discerning any differences and the qualities.

Over the years people have tried and failed often to show much difference, when tests were done with proper controls and with scientific rigor.

To be fair I haven't read up on the subject recently so if there's anything current that shows otherwise I'd be glad to read a link to a paper from a credible source.

2 comments

Yes. Approximately no-one believes that we have definitive measurements of subjective sound, any more than we have definitive measurements of how good a painting is.

The key thing is the equipment owner's self-suggestion, vs. whatever adequately controlled blind testing would reveal.

I've done a small amount of audio restoration work, in which you're often trying to judge (e.g.) how much of a noise-removal process to apply. How much of that high-frequency content is just noise, or is there actually some of the original signal still lurking in there? Apply too much noise-removal and you'll take away some of that signal.

In doing this you frequently bump up against the question, "Can I actually hear a difference?" My rather sobering experience when I actually put it to the test was that I was way beyond the point of being unable to identify different treatments of the same audio, any more than could be explained by random chance, and yet I was still telling myself I could hear a difference.

Which is why some professional audio plugins offer a “blind”-mode where all interesting but visually distracting meters, spectra etc are switched off. This is also the appeal of analog gear, where suggestive metering is uncommon.
False.

Rigorous double blind a/b/c/d tests using real listeners has been performed, and the research has been published in peer reviewed journals since the 80's.

What has happened is that most of the audiophile community saw research as a threat to their revenue streams and actively FUD around it. And the audio engineering community has simply moved on.

Here is a talk that could be very informative. https://www.youtube.com/watch?v=zrpUDuUtxPM

His book is worth a read as well.

What's one thing I said that was false?

I said studies have been done that failed to show any difference, that's absolutely true.

I said I was open to hearing about recent papers published showing otherwise, yet you've provided zero links to a peer reviewed paper. You offer a Youtube video? If it's a video about research can we just skip to a direct link to the science? Citation please?

I said it's not necessary to know if current testing equipment can detect differences, to know if people can tell the difference. True.

Where are the rigorous double blind studies, with full disclosure, conducted with no conflicts of interest in funding or affiliation?

And of course I assume we want to stay on topic and look at research that isolate amplifiers for some tests, not conflate them with moving parts such as speakers.

For personal listening maybe. PA systems for concert halls are sold in blind demos. Several vendors competing for the contract hang their systems behind an acoustically transparent curtain and the buyer decides which one sounds best.