Explain home audio to me: Is there such a thing as "metrics-driven" audiophilia?

The home audio market has always kinda weirded me out. The showrooms tend to be a lot of leather couches and fancy lighting mixed with gold-plated digital cables, vacuum tubes, and snake oil. What gives? Is there a way to skip the faith-based part of it and just compare two different speakers/systems by their specs?

It seems like most other consumer electronics have relatively sensible specs that are useful for cross-comparison shopping. Monitors have brightness in nits and color coverage, contrast ratios, etc. If you want to buy a new TV, you can plug in a size and detailed specs and find the exact models that fit those criteria. If you want to buy a new digital camera, you can similarly filter by sensor size, megapixels, lens compatibility, etc. If you want a phone or computer, you can combine the specs of their different subsystems.

But when it comes to “how good something sounds”… how come we don’t have a similarly usable set of metrics (or maybe we do, and I just don’t know about them)? Beyond the wattage and the frequency response (which all seem to mostly just say 20 Hz - 20 kHz?), are there no other numbers that are useful in understanding which would sound objectively “better”, for some definition of better — let’s say to “most accurately reproduce a lossless digital recording”? Maybe a standard test panel of them, encompassing several different genres of music, TV shows, movies, games, etc.?

I’m not in the market for a new system, but just curious how it all works. I’m the (somewhat conflicted) owner of a Covid-era Sonos system. It held together my frayed psyche through the pandemic, for which I am eternally grateful, but then subsequent decisions made by the company lost a lot of customer trust, ultimately leading to the ouster of their CEO. This isn’t about that brand in particular, though. I was just wondering what, objectively, makes one speaker system sound better than another (or not). In my case, I got the Sonos because I wanted something that could play Spotify on its own, without needing a Bluetooth connection to a phone/laptop. At the time, that was relatively rare. The speakers sound “good” to my humble peasant ears, but not significantly better than, say, my $60 computer sound bar. It does have more bass, but that’s because it has a dedicated subwoofer. Compared to my old $150 Creative Labs gaming speakers from 30 years ago (which also had a sub), I think they sound pretty similar, even though the Sonos costs like 10x more. They’re definitely not 10x better, and in some ways are much worse (they were unreliable even before the app fiasco, and only got worse after that) — but that has nothing to do with its audio quality. Listening-wise, all I can say is that they generally sound “good”, but classical sounds a bit “dull”, and dialogue with any “s” sounds makes the person sound like a snake (I hear them hissing unnaturally, but nobody else I’ve asked can seem to hear it). But I’d be hard-pressed to actually measure or present any of those observations in an objective, numerical way. Is that even possible?

Anyway, if that system ever breaks (or Sonos goes bankrupt), I’d like to replace them with something much simpler. But is there actually an objective way to shop for them, some measurement of “sound quality per dollar”, beyond “feelings”?

I think I’d sooner step into a timeshare talk or used car dealership than an home audio shop… they all seem set up to fleece you by overwhelming you with mood and ambience rather than giving you the raw data. How would, say, professional sound engineers kitting out a stage for a band shop for speakers, instead? Is it possible to strip away the feel-good mumbo-jumbo and comparison shop by specs alone?

There certainly are specs for audio gear. The problem is that for most of them, we long ago passed the point where even crap gear has specs indistinguishable from great gear. The quality of the source material matters more than the quality of your own playback chain.

I’ll also suggest that your belief that e.g. monitors have useful specs is a bit overstated. You’re just used to buying monitors and thinking about their specs. Not so audio. So you think one product has useful specs and the other does not.

IOW: Yes, there is a brightness spec for monitors measured in nits. But as a user you don’t really care how many nits it has, as long as it has enough. A monitor w a spec of 1500 nits is surely brighter than one specced at 1200. But if you only need 800 for comfortable use by your eyes in your room, the spec becomes a distinction without a difference as the lawyers say. More than that, it becomes a distraction that does nothing to usefully distinguish between the two monitors for your use case.

Whether it’s eyes or ears, when your subjective impression is “good enough”, it’s good enough. Measurements be damned.

When I think of home audio “metrics”, the website that comes to mind is Audio Science Review. It’s has charts like this in the speaker reviews:


Besides the equipment you also have to consider the source/format/dynamic range of the recording you’re listening to. With reissues, remixes and remasterings, those characteristics can vary widely any given album

Measurements are useful, but they will not help the person who likes a bit of distortion, who may need ABX testing to choose the gear they prefer.

I’ve been an audiophile for years. Here’s all you need to know:

20% of the money will buy you 90% of the sound.

Another 30% of the money will buy you another 5% of the sound.

You can’t buy the remaining 5% of the sound because nobody can agree about what it is.

I was recently shopping for headphones and came across a lot of complains about ‘too accurate’ or flat. Accurate and flat were supposed to be the goal! But a lot of headphones do have boosted bass and people have come to expect it.

Though that doesn’t keep people with more money than sense from trying.

I used to work for an “audiophile” manufacturer. The engineer would claim to hear differences in builds that would test (using very expensive, accurate gear) identical.

I got a big kick out of it. I got stories…

Specs are great, but I prefer my own ears. I run a Peavy Q-215 EQ and tune it to exactly what I think sounds good.

With speakers, some specs such as sensitivity, impedance, and power handling are useful for matching with your amplifier or receiver. Less important is frequency range. Yes that will give you some idea of how much bass a speaker can produce, but what really matters is your personal preference and the type of music you like to play. I am looking for some floorstand speakers to replace my bookshelf speakers. I found one speaker from a well regarded brand that I was able to audition. From the specs and reviews of this speaker, they seemed that they were something that I would really like. I went to the audition wanting to like them. But while they did sound good, I ended up deciding that I didn’t like them. I have heard other speakers that are more to my liking, but still need to check out a few more. It’s hard, because not every retailer carries every brand, or stocks every model.

For items such as receivers, preamps, streamers, and amplifiers specs can tell you a lot more, and there is plenty of good affordable equipment out there. Speakers make a huge difference in the sound, and that comes down to personal preference which is why some sellers allow 30 to 60 day trials. How a speaker sound in your home may be different than how it sounds in a showroom. It’s not snake oil. Within your price range, you’ll find something you’ll like.

100 times this.

With regard to speakers, which are an excellent example of components that have a very high impact on overall sound, allow me to offer an example of why people can’t agree.

I have to confess that I like (shudder) direct/reflecting speakers. There…I’ve said it. Most audiophiles would try to have me placed in conversion therapy if they heard me say this. But I like the sound field they produce. I’m currently running Magnepans and Definitive Technologies, both of which are bi-polar and are generally considered to be good speakers. I don’t think that any metrics would change my mind on this.

Similarly, I find most bookshelf speakers to sound small and confined, even though the measured frequency output and SPL might be excellent. I’ve tried several times to find smaller speakers to replace my big space-wasters, but I listen to them for a week or two and always end up switching back.

I get that there’s more to speakers than just their specs, like your room setup and personal preference.

But is there truly no way to find a “good enough” speaker set using specs alone?

Like this point:

Maybe that’s true if you’re comparing $1000 speakers to $10,000 ones… where the $1000 speaker is 90% as good.

But what about when you’re comparing the $10 gas station Bluetooth speaker to the $20 Target one to the $60 Costco one?

In my experience, at the low end, there can be huge differences in sound quality. Is there a way to use reported specs to figure out which one is the best bang for the buck?

I would say, in theory, yes. But the problem is is at that price point, no one is doing the testing and spec reporting that you’d be looking for. The audience in that price range doesn’t demand it, and the margins on the product itself don’t support the additional effort.

Further, a battery operated Bluetooth receiver / D-A converter / preamp / amplifier / speaker module is a lot more than a “speaker”, despite the colloquial (= read “moronic consumer”) name for them.

The details of the Bluetooth protocol as implemented in your sound source and in the amp/speaker device and how well those implementations play together is probably the single biggest source of good or poor sound.

It’s easy to say “everything digital is perfect”. Anyone in the digital anything biz knows that ain’t so.

Consumer Reports used to rate bookshelf speakers according to some objective criteria years ago.

In their more recent ratings they only look at smart speakers and soundbars. They consider a variety of criteria, including ease of use and versatility, but sound quality is the first criteria they look at. They say sound quality “represents tonal accuracy and ability to reproduce fine sonic detail.” Their detailed test results go into quite a bit more detail, but I don’t see any of the detailed frequency/SPL graphs like they used to present years ago for bookshelf speakers.

Incidentally, reading their detailed review of the Apple HomePod, I feel like I’m reading a wine review. :wink:

Wait, what’s wrong with direct/reflecting bipolar speakers? I have a pair of midrange Definitive Technology speakers that I bought used 30 years ago (from a stereo store where they were traded in by some audiophile). They were all I could afford at the time, but still sound great to me.

There’s nothing wrong at all with bi-polar speakers…that’s why I like my own DefTechs so much. But “true audiophiles” very often dismiss them for various reasons. I find them especially ideal for A/V systems, where they give really nice depth. Maybe the more dismissive audiophiles associate them with Bose 901s.

The Martin Logan ESLs are also excellent speakers that are bi-polar.

But I also realize that a test lab can put a mic in front of a standard speaker and a bi-polar speaker, run their test routines, and get identical results. Metrics only go so far.

And I’m happy to run into another DefTech owner! My BP-30s have kept me happy for years.

Ah yes…we used to agonize over THD and wow and flutter! The bad old days…

I’ve also observed that anyone else’s decent stereo usually sounds better than yours. My theory is that it’s because it’s different, so familiar material sounds a bit new. It wears off, of course.

I’ve seen things like frequency response measured in headphone and laptop reviews. They will show a curve, and I have noticed that I can tell when a product will be too bassy for me and need me to use EQ to fix it.

Too bad that there isn’t a commercial audio studio in the world that’s built to specs that exacting, so the owner gets to listen to every stray bit of echo from the studio walls, every !% difference in the mix levels, every flaw in the final pressing in startlingly clear detail.