Hacker News new | more | comments | ask | show | jobs | submit login
They Don’t Make Music Like They Used To (nytimes.com)
105 points by djmobley 15 days ago | hide | past | web | favorite | 94 comments

I'm an amateur musician. One of the more common pieces of advice you'll see floating around in the community is to make your music "louder", especially if you're creating electronic music of any form.

Two reasons for this:

1. More and more people are listening to music through equipment that is significantly inferior. Mobile phones, cheap Bluetooth speakers, cheap earphones - they're all subpar in their ability to render sound faithfully.

2. There is a constant "war for loudness". Once you've listened to a David Guetta song that can be ferociously loud, anything not as loud will be perceived as "soft" that "kills the vibe".

More and more people are listening to music through equipment that is significantly inferior.

I cannot imagine that this is true. Even listening to Spotify on a cheap phone using the headphones that came with the phone offers far better sound quality than listening to the nth generation tape copies on the cheap tape players or the bad reception of the tinny radios that we used to have.

You're forgetting that the vast majority of consumers use cheap Android phones and the earphones packaged with them.

This gear usually has very poor mid-range response. Hence, producers crowd tracks in the bottom-end (bass) and treble ranges. Try listening to a track with a cheap $15 earphones and then with a set of studio headphones like ATH-M50x and you'll hear what I mean.

Ha! If even the packaged earphones.

Forget Spotify. People are listening to music on throttled YouTube.

Compression wasn't the antagonist of music. Not even the consumer's general cheapness is the antagonist. The website that pirated it all, and then got bought by a parent company larger than the market value of the entire music industry combined, is the antagonist of music.

It makes my life miserable. I love writing music but mixing is just a boring, unsexy grind. And it's a skill that takes way too much time to develop - you can't "hack it".

I've produced tracks that sound incredible to me on my 6" Yamaha monitors. Then I send them over to friends who listen to them on their tinny laptop speakers and wonder what the big deal is.

Your 6" Yamahas are not a neutral reference.

Studios deliberately use bad speakers (NS10s, Auratones) to check for this.

In the business it's called "translation" - a good mix sounds good on everything from earbuds to a car stereo to $$$$$$ hifi.

It's almost impossible to make a mix that translates without spending a lot of money on neutral monitors and acoustic treatment.

But there are tricks you can use to make a mix translate better. E.g. You can cut the lowest octave or two in the bass, because no one (outside of a club) can hear it anyway, and use bass sounds with plenty of mids, so you can hear the line even when the low bass isn't there.

Every mixer I ever met did 2 things. 1. Played the mix on a cheap cellphone (this used to be a boombox or 80s wire loop headphones), and a bad car stereo or 2. Listened on Car stereos, as they have whack eqs and soundstage, and will bring out any hint of ugly. Tiny speakers will distort readily, and pop music must sitll sound decent when abused, because omg ppl abuse it. For a club track, you optimise differently, but err on the side of sounding good on crappy radio speakers always otherwise.

Yeah, I'm learning this along the way. Mixing is just a completely different skill set than production or composition. I've realized that it's better to let professionals handle it than learning it - it requires too much investment in equipment and far too much time in training to do it by yourself.

Auratones are not bad speakers though. They're single driver and non-ported. That has implications, namely better low end response, and uniformity in the waveform: ie. no notching from cutoffs; sound comes from one place instead of two (better spatial representation), and also a lot of consumer audio is on single driver setups (earbuds, laptop speaker, p much anything that doesn't have a dedicated tweeter or subwoofer).

When mastering, after I'm happy on my high-end near-field monitors and acoustic room, I haul my studio laptop into every other system and environment I have and tweak accordingly. (i.e. Bose Companion 3, an older JVC shelf stereo, my car stereos, various headphones including ATH-M30, cheap iPod, and Beats, etc etc etc) for this very reason... almost nobody is listening on a nice set of nearly-neutral near-field monitors except other musicians. When done, I make sure I'm still happy on the main rig.

Someday I hope to be experienced enough to just know what to do though, without all the extra work. I haven't found a systematic answer yet.

ouch. too real.

Oh you are all so naive. They listen through the built in speakers on their cheap Android phone. In mono. On the bus. And everywhere else.

No, that's not what the vast majority of users does.

Maybe, but its a big enough use case to be optimized for imho.

Yeah but most people had some stack system or separates at home as well as the walkman or tinny radio. Often horrible by hifi standards, almost always better than the tiny bluetooth speaker, or included headphones. When you went out you knew the walkman or car was second best, but enough background noise that it was still OK.

Parents had separates, they wouldn't take kindly to me blasting out whatever in the living room though.

Yes. This. Compress and EQ everything, if you're trying to get a lot of listeners. You'll have to sacrifice dynamic range for it, but sometimes you have to make a trade-off.

Link up with a producer if you don't know how to do it yourself, if you want a fighting chance.

I think that an underrated factor in the dynamic range compression is the way we listen to music. I'm not sure if it's accurate to say that the average quality of listening devices (that is, speakers/headphones) has decreased, but certainly it has become more common to hear music through bad earbuds, tinny smartphone speakers, or laptops simply because those devices are so prevalent.

In these settings, music with lots of dynamic range may legitimately not sound as good. Due to the quality of reproduction, the low-loudness parts are hard to perceive.

For me, there are certainly songs that I enjoy listening to on good headphones in a quiet room that don't have anywhere near the impact when played on laptop speakers.

Meanwhile, it's possible for someone to spend less money than ever before to get really good sound.

Sell the Beats earmuffs and pick up Superlux HD668B headphones for under $50.

Drop the Apple earbuds in the garbage and get TRN V80 in-ear-monitors for under $50. Add a pack of random eartips to figure out which ones are most comfortable for you.

Your MBP's built-in speakers have no bass below 160Hz or so, and bluetooth party speakers have all the sonic charm of a box of Kleenex. Spend $300 on a pair of JBL 305P powered speakers and plug anything into them -- they're a distinct upgrade from basically everything costing less and an awful lot of speakers costing twice as much.

Buy your music on CDs and spend the ten minutes per disc copying them. Do you know you can fit 1300 full CDs in a one terabyte disk without any compression at all? Or buy FLAC or any other non-lossy compression version. Buy directly from artists whenever possible: give them the best margins.

Use wires. Use headphone jacks. Use cheap ground loop isolators when you get hum from power lines -- they used to be really expensive, now they cost $10.

We live in a time of high-quality low-cost devices that play music much better than anything your parents could have bought without spending a month's grocery money.

>We live in a time of high-quality low-cost devices that play music much better than anything your parents could have bought without spending a month's grocery money.

Is that really the case? I've got some late 70s / early 80s Fisher speakers that sound almost as good as the best bookshelf speakers you can buy today. I rather enjoy the sound more than the JBL 305P studio monitors. While the Fishers are less precise, they're more enjoyable to listen to. (The trick is to put them on a desk at an angle the same way you would for studio monitors, and turn them upside down so the treble stays at ear level.)

I can't imagine Fisher costing an arm and a leg back then, and back then Fisher speakers were sold at normal electronic stores, while today high end audio has to be bought at specialty stores like Guitar Center or online. Today if you go to a Best Buy, they'll have some thousand dollar home theater speaker setup that could sound better. The average consumer thinks that is what top of the line is and sounds like, when in fact they're being sold a lie.

High end speakers were once accessible and advertised to the masses, unlike today. Today you have to be a specialist, an "audiophile", or some other niche title.

In an ever increasing mobile world, mobile speakers have taken over. While some of the in ear headphones are amazing, and I hope this trend continues, it will be a while -- if ever -- when we get portable speakers that sound decent. One of the problems is the direction of the sound, and the other is the Bluetooth compression itself leaving manufacturers to make their speakers fuzzy to counter the compression.

Depending on which Fishers you're talking about -- I'm assuming larger bookshelf size -- they could easily have been $400 brand new in 1970. CPI says that's about $2650 now.

Maybe they were $100 speakers in 1980? That would be about the same as a pair of powered JBLs now, except that you still need to get an amp for the Fishers.

$100. I believe I got them for less. They also came with an equalizer and a stereo receiver. Their equalizer technology was one of their key selling points back then. I can only imagine Bose copied this idea when it made its speakers, except with Bose you don't get much control over the EQ and without it the speakers do not sound great. Fisher speakers sound great without the fancy EQ.

edit: And for anyone who doesn't know what old high end supermarket speakers sound like from the early 80s: I have $400 headphones that are less enjoyable to listen to.

> the Beats earmuffs > the Apple earbuds

They're fashion statement, most people don't care about sound quality.

Unless you have a proper sound card / dac and a really nice headset I doubt investing in FLAC will transcend your listening experience. On top of that most people listen from their phone, during commute, at the gym etc ... no need for flac in these conditions.

It's the same thing with photography, you can get a 300$ dslr that has more options than any slr. Yet the vast majority of people will be happy with their phone camera.

Is there somewhere reliable I can go to do this kind of research before I buy speakers and/or headphones in the future?

The Z Reviews YT channel (https://www.youtube.com/channel/UC3XdYJjWliOdKuZMNaTiP8Q) is a great place to get a good look at affordable/midrange equipment. That guy is a fantastic nerd.

Questions are welcome on rec.audio.high-end -- the high-end refers to quality, not price.

You can save even more money if you buy used. I have a couple of Rega amps and pairs of speakers, for music in two rooms. Total spend less than £400. Original cost would have been ridiculous. But they're just as good as when bought new.

Sources are so good these days that unless you're very picky, you can just plug your TV / laptop / phone into an amp using a cable and you're done.

> Drop the Apple earbuds in the garbage and get TRN V80 in-ear-monitors for under $50.

Awesome recommendation. I ordered these ($40 on Amazon) based on your comment and they are great. I have been using 5 year old $50 Sony earbuds which are good, but these TRNs are amazing.

My only complaint is that they are pretty heavy so I'm not sure I could run in them. For my use that doesn't matter since I have BT earbuds for workouts.

Thanks again!

Apple earbuds are incredibly high sound quality. They are just really uncomfortable in my ears. And Spotify 320kbps is as good as CD.

> And Spotify 320kbps is as good as CD

No, it really is not. But it's close enough for most.


Yes, I didn’t mean to an oscilloscope, I meant to a Homo sapiens.

While these good options may be available, average people probably simply don't know about them. How would they?

It turns out that the internet is full of information. Some of it is good, some of it bad; after a while, many people develop skills at interpreting it all.

I'd say cars have destroyed it even more. The amount of noise inside of a car is really high and you lose a lot of the low end of the song. Many times I have showed a friend a song in their car only to find out the song sounds terrible now.

An interesting thing I heard is sometimes vinyls contain a different master that doesn't have the range squashed because people listen to vinyls in a better environment.

I've heard that music mastered for vinyl can't be over optimised for loudness as it will cause the stylus to jump out of the groove on playback.

And this is my first answer when people ask why I buy vinyls.

I think a lot is just the attitude of those who make and master the music. Back in the 80s, CDs sounded great, but they were sold as being the ultimate in sound quality, and the idea really was that you would keep them at home in their jewel cases and play them on a nice stereo system. Car CD players didn't come until later, in the 90s, and MP3s in the late 90s, and music-playing phones in the 00s.

So I think CDs are now made with the idea that the music is going to be converted to MP3 and listed to on mobile devices, not on nice stereo systems, so they mix them with that target listening environment in mind.

For the unbelievers, listen to Metallica's Death Magnetic first from the official version. After that, try the Guitar Hero version and it'll quite literally blow your mind.

A video comparing both: https://www.youtube.com/watch?v=6Nfqpr3ygSg

I tend to get physically and mentally exhausted when I have to listen to music that's maximised for loudness, took me a while to figure that out. A comparison between these two versions was the eye opener for me.

I can't hear a difference in that video. Maybe Youtube broke it?

You wont notice any difference until about a minute in where all the instruments come into play.

I put on my headphones to be sure. The audio quality seems uniform throughout to me. Which isn't to say it sounds good (it doesn't.)

The drums especially are distorted. Bass is non-existent.

Why do they even bother hiring a bassist if they don't anyone to hear him? They even paid the guy $1M to join the band!

.. whaaat ? the difference between the retail and GH3 parts here : https://youtu.be/6Nfqpr3ygSg?t=219 are enough that I can feel them just by the vibrations in my desk

What speakers / headphones are you using ?

Bose QuietComfort plugged into a Macbook Pro.

Switch off the active noise cancelation if you haven't.

It's not the best headset in town but it should be more than enough to hear the difference.

edit: the switch at 6:10 is very noticeable to me (shure se215) but tbh they both sounds bad, and it sounds just as bad on spotify so youtube's compression isn't the culprit

> [Bose QuietComfort] It's not the best headset in town but it should be more than enough to hear the difference.

This is when I suspect the differences must be minuscule even without hearing the comparison. Bose headphones are way better (and pricier) than what most people use. The sound canceling alone should make significant differences evident.

Which model? I have the QC-20s since I spend a lot of time flying, and they win on raw noise cancellation, but I see a lot of people wearing the 35s.

My condolences.

Not even around 3:10?

You'd think the distorted highs/lows of the retail version would be enough to tell the difference.

I listened on my (work issued) Jabra "whatever" headphones, and i could tell right away (without looking) when it switched from retail to GH, just by the distorted highs/lows.

I also get extremely exhausted from listening to music maximized for loudness, while listening to almost anything recorded in the 70's can go on for hours without any sign of exhaustion.

Yeah, it's really sad: the 70s-80s were the peak of recorded music. There was great stuff earlier in the 60s, but the recording tech was crappy, and later stuff all sucks because of the Loudness Wars, so it's nearly impossible to find good quality mixes no matter how great the music is.

It's not just about dynamics. The GH version has significantly more bass.

Compression can do that, but it's also possible multiband compression was used to make the bass louder so it would be more obvious ("more exciting"...) on a typical home audio system.

Agreed, and I'd like to throw Red Hot Chili Peppers' "Californication" album into the mix too. It really surprises me just how hideous both of these albums sound.

Californication isn't just optimized for loudness. It's intentionally distorted (or so I prefer to believe, or the sound studio should be permanently disbarred).

But this isn't a new phenomenon. Most of U2's albums, including those from before Death Magnetic, are also hideous to listen to.

> It's intentionally distorted (or so I prefer to believe, or the sound studio should be permanently disbarred)

I tend to assume the same, but it just occurred to me that this is kind of the opposite of Hanlon's razor...

Probably sounded decent on the radio though.

Both sound like Nickelback to me so I don’t know…

This is interesting. I know that audio compression is also used by talk radio hosts. I know that Rush Limbaugh designed a special compression system that he uses on his show. He mentions it in this quote:

> The incorrect way to use a microphone is to put it right in front of you like this (leaning in close) and speak into it because you pop your P’s real bad, and that could be dangerous on certain words. We broadcast professionals turn it aside and speak across it so that our P’s do not pop. We don’t have to worry about anything like that, and with the compression that is built into our system here, I can move this microphone even further — well, you’re probably hearing me not quite as loud, but the presence is still there. As I move it in closer, it gets a little bit more present, but not really louder. Now, sometimes I do what we professional broadcasters refer to as eat the microphone. That is when I really want to make a point. I am eating this microphone. My lips are touching this microphone right now. There are any number of highly trained specialist techniques that we who are professionals in this business use.

Audio compression is why talk radio sounds like garbage even when streamed compared to something with a wider dynamic range like NPR. It helps give talk radio hosts that loud, booming voice which helps them sound convincing to their audiences.

>The incorrect way to use a microphone is to put it right in front of you like this (leaning in close) and speak into it because you pop your P’s real bad, and that could be dangerous on certain words.

Wouldn't a pop filter stop this?

That's what most vocalists use. They're like $20. I've got one for use with my compressor mic. It works great.

Also, i've been listening to a lot of tunes mastered for vinyl lately that doesn't use high compression(modern music). The same tracks on youtube sound noticeably worse not because of the file format but because of the automatic compression and gain youtube adds.

You have to take this into account when you upload a music track there.

I found this site for comparing my mixes to the requirements of different streaming sites before uploading: https://www.loudnesspenalty.com/

> Wouldn't a pop filter stop this?

Clearly, the real solution is to speak a language that doesn't use aspirated stops. :D

A pencil and rubber band is a great trick that I use all the time. Simply band the pencil so it is positioned at the centre of the condenser diaphragm does a good job of reducing plosives.

You can get a decent pop filter for like $6 online, or just make a simple DIY one (there are YouTube videos about this).

Related, and interesting in my opinion, is that NPR also use the same microphone for everything, to create a 'signature sound': https://current.org/2015/06/a-top-audio-engineer-explains-np...

As I understand it, Limbaugh is also deaf, which adds to the complexity of working with his speech in the studio. He may need special compression if he can't effectively perceive (and thus control) his own volume.

He got cochlear implants at some point, so I don't think he's totally deaf

The home recording revolution is something that just never gets its due, because people want to find a problem with it. Auto-tune is a style now. Smushed dynamics is an intentional choice. You can make a Superbowl halftime show song with a shoestring budget.

I read about the loudness wars, got super upset about the injustice of lost dynamic range, posted something about it on my dorm room door... 15 years ago. NOTHING has changed.

The article notes that Spotify, but originally it was terrestrial radio, that hyper compresses their music -- it makes listening in the car far less annoying. But to see the dynamic range destroyed on the album itself, that is tough.

This quote from the article "It is a super-discouraging situation." summarizes it quite well.

It's a problem when it's baked into the recording.

In Spotify or in audio equipment, compression often makes sense. It's user-adjustable, and there are times where compression is useful, such as listening to music in a car, where the quiet moments would get washed out by road noise.

And how about movies. I am as much a fan as anyone of having the highest quality home theater experience possible, but some movies (generally the ones with the highest quality audio mixes) are impossible to listen to in my apartment (while being a decent neighbor) without using dynamic range compression.

But as the parent says, these can be applied after-the-fact (in most modern movie players it's called "night mode"). This doesn't mean the original recording has to be shipped with deteriorated sound.

In a way, loudness compression is like a hash function: the operation itself is largely mechanical, but destroys information in the process. Shipping with good-quality audio should be the default, and let compression be applied on playback. In fact, most home theatre systems already do sound post-processing in the form of room correction anyway (Audyssey, Dirac), so why not the let player do the compression?

I don’t disagree, although I wish movies would ship with a professional audio mix with lower dynamic range, since that would be a lot better than the often noticeable compression built into receivers and playback devices.

Good point. What works great in the cinema gives a home play needing subtitles for the quiet parts or it'll take the windows out later.

I might be part of the problem but I agree. I find it disconcerting having to turn the volume on my headphones up to 50% or higher just to hear a movie when that amount of amplification on music or standard Youtube content would blow my eardrums out... and occasionally almost does when I forget I had my volume turned up.

You can't really compare volume dynamics across different genres in any meaningful way. There's still plenty of great music out there, don't worry.

When I buy a CD of older music, I avoid the "remastered" ones and look for the original CDs among the used ones for sale.

I suppose it helps that my stereo and speakers are circa 1980 :-)

But everything sounds better with no dynamic range and the volume turned up to 11.

I also seek the original masters to avoid this, and often prefer the vinyl release as they seem to have mostly escaped this game. The technical limits of vinyl are a big help here.

As a counterpoint, I've never liked classical music on vinyl. There just isn't enough dynamic range there. CDs greatly improved the matter.

Disco music on vinyl also had problems, as the base beat was too much for the groove. Hence my preference for the 45 rpm 12" disco singles, where the wider spaces between the grooves and the faster speed made for much more dynamic range.

The K-TEL compilations would reduce the dynamic range so the LP would play much longer, as the grooves were smaller and closer together. That wasn't really what one wanted if one really wanted to get into the music. But I liked the K-TEL compilations if one regarded them as a "sampler".

Other problems with vinyl are the crosstalk between successive grooves, bubbles in the plastic due to poor quality control, and the usual rumble, wow, flutter, hiss, snap, crackle and pop. Though one could enjoy that like one enjoys the primitiveness of an older car :-)

80s and 90s Decca, DG etc CDs greatly improved things. Current classical, outside the smallest labels, often seems to be blessed with the same compression game. It's disappointing, but at least it's easy to work around. Understanding why they do is more of a challenge.

Cheap pressings and compilations are always going to be an issue. Just like those old compilation tapes you used to get in motorway services that would stretch on first play. I suspect KTel would find a way to make CDs self-destruct after 5 plays.

Most of those other problems went away with a decent deck. Mainly leaves the scratches to "enjoy". :p

The article was a play on words (in that it's quite literal) but interestingly enough there was a very recent study [1] indicating that millennials are more capable of recognizing songs from the 1960s-1990s, music created in many cases before they were born, than of recognizing contemporary hits.

Makes one wonder about the notion of nostalgia. Especially in today's world where marketing is playing an ever larger role in society, there's no real reason to assume that the directionality of things is always going to be positive.

[1] - https://www.eurekalert.org/pub_releases/2019-02/nyu-wag02041...

I definitely think the idea of nostalgia with music is complete BS. I've met way too many younger people who prefer music from well before they were born (60s-90s); the music after the 90s really is different from what came before, due to various factors that affected the popular music industry in America (and the west) starting with the rise of the internet. The "loudness wars" are one obvious thing setting newer music apart from older music.

This has been going on for two decades now. Teenagers have grown up used to the sound and aesthetic of compression. Whether you like it or not compression is here to stay.

The "loudness war" articles pop up on HN every three months or so, and they haven't changed for the last 10 years, always with Death Magnetic references. However, what have been changing is actual mastering.

The primary target now is streaming, and streaming services have started to balance the tracks by their "true" loudness (not getting into technical details here), which pretty mush defeats the purpose of making your track louder than competition. Now, if your track has more dynamic range, it will actually sound better, because the streaming service itself bring other tracks down, and push yours up gently. So, this trend is starting to get reversed - also most mastering engineers and musicians are used to "getting it loud", there are many who're taking advantage of the new reality.

Brandi Carlile got 6 nominations and now has a show at MSG. The corporate music machine isn't 100% efficient at limiting exposure to only cookie cutter acts.

The music machine is weaker than ever. The machine is forced to compete on equal terms with independent acts.

Beside the technological aspects, there's a cultural one. And the ratio of the two.

> A blaring television commercial may make us turn down the volume of our sets, but its sonic peaks are no higher than the regular programming preceding it.

I've heard that before. The VU meters on my stereo say otherwise.

VU meters that do some sort of temporal averaging (i.e. all of them), of course. The fact that the peaks are the same is meaningless unless you mention how often the signal hits the peaks.

The VU levels during the commercials were consistently about 20-25% higher than the maximum I ever saw during the show.

It's roughly as useful as saying "well, a microwave that beeps three times technically isn't actually louder than the fire alarm going off in the other room".

Shrug. They used to compress music so it would sound okay coming out of the speakers in a 57 Plymouth. "What's Going On?" changed all that, much to Berry Gordy's chagrin.

But dynamic range has nothing to do with the decline in quality of music. Shit is shit no matter how you mix it.

Applications are open for YC Summer 2019

Guidelines | FAQ | Support | API | Security | Lists | Bookmarklet | Legal | Apply to YC | Contact