Hacker News Reader

Comment by calichoochoo | Hacker News Reader

I predict a lot of wrong conclusions will be drawn from this. This paper does not preclude the possibility that there exist high-priced headphones with better-than-average or even spectacularly good frequency response. It only says that if you bin together all of the high priced items, their aggregate quality is no better than any other price bin.

View on HN

Yeah. The authors misinterpreted their own data. Go to figure 3. Take the best headphones at each price point (the bottom-most points in the scatter plots), and there is an obvious correlation between price and quality.

A better conclusion would be "headphones with poor audio quality exist at all price points"

It's also the case that people pay for features other than audio quality. Look and feel, comfort, connectivity all influence price. This feels a little like releasing a study saying, "No correlation between car top speed and retail price."

My underpowered very cheap car would like to say otherwise...

But a $75,000 model T Ford from the 1920's wouldn't. When there's no correlation at all you can find a datapoint anywhere you want.

That's true, but I would imagine people think that all the expensive headphones have the sound quality part covered.

This is noted at the beginning of the piece:

> Research suggests that factors influencing consumers' choice as to which model to purchase are mostly based on wireless functionality (Iyer and Jelisejeva, 2016) and attributes such as shape, design, and comfort (Jensen et al., 2016). Interestingly, sound quality does not seem to be a major attribute for purchase decisions.

This isn't surprising: the base level sound quality is probably good enough for the vast majority of users, so they don't care about minor differences there.

I worked for an outfit that had a similar problem: we tried differentiating ourselves on quality, only to find out that all customers expected that the vendors in that space already had high quality as the price of entry into the market (a fairly accurate assumption). As a result, they didn't care about "we're better quality than those guys." They cared about "those guys make equipment that's easier to use than yours."

Yes. The amp and speakers in the living room sound much better than the Logitech UE Boom. However the boom just works, avoids the flakey airplay interface, can be carried where I go and sounds good enough. So I bought another one.

What makes a frequency response "good" anyway? Sound engineers wanting to produce the best experience for the most people should tune their music for the average frequency response, meaning that a "flat" or otherwise non-average frequency response would distort music in an unintended way.

What makes a frequency response "good" is subjective. A lot of audiophiles prefer flat, but many like e.g. tube amps, which aren't exactly flat, or headphones with V-shaped frequency response etc.

> meaning that a "flat" or otherwise non-average frequency response would distort music in an unintended way.

Not sure how to make sense of this. A flat frequency response is by definition the one that does not distort the recorded music.

If for the sake of the argument bass-heavy headphones and speakers were all the rage and a sound engineer set out to record music compensating for such devices' frequency response, then they wouldn't sound "bass heavy" anymore.

That's already happening, deliberately, in the music production stage. Arguably the most popular speaker in the world for professional music mixing is the Yamaha NS-10 (you may have seen pictures of them in recording studios, with their distinctive white woofer cones). They are terrible in terms of frequency response! They are aggressive to the point of nasty right in the 2khz range, where a: our ears are most sensitive (ears do not have a linear frequency response!), and b: human vocals are most present.

For vocals and vocal-like instruments such as saxophones and lead electric guitar, the NS-10 is downright evil. And that's why it's popular. If you can make a mix sound good on the NS-10, it'll sound good on almost anything else.

Reply on news.ycombinator.com