When I was in high school our amateur jazz band went into a recording studio to make a fun keepsake album. My memory is fairly hazy, but the whole experience was a lot of fun! Everybody got their microphones and a headset. Maybe we recorded by sections and likely with a metronome track.
When we finished we got to go watch the sound guys at work. They meticulously gridified and autotuned every note. After it was all over I got the cd, I listened to my favorite tracks that I got to spend the semester playing with my friends, and it had no resemblance to what I knew our group to be. I think I threw the album away.
When I've talked to people about modern music in the past, it's only ever people who play music, or who have an intense interest in it, that seem to even be aware that the vast majority of modern music is autotuned (and "gridified" as this author puts it).
You hear but you don't recognize it. Rick Beato has a clip where he takes Jogn Bonham's drums from Led Zeppelin and gridifies them. As a non-musician, I can feel the soul being sucked out of the song in the process.
Pfft - this doesn't make a general point about quantisation - he completely altered the drum part. He didn't just shift things by minute amounts to hit the "exact" timing while maintaining the swing. Of course if you remove swing then it's completely changed - it could also be quantized while maintaining swing.
It's easy to take one of the greatest drummer's performances, quantize things, make it worse, and say "quantization sucks the soul out of music."
But what about the non-greatest drummers? The other 99.9% (or higher) of musicians who are tiers below Bonham? Quantization is a net improvement. It takes unlistenable performances and makes them listenable.
One of my absolute favorite bands is a random street band I stumbled across in New Orleans some years ago, called Yes Ma'am. They had one album at the time, and it is... let's call it "loose". They must have been half in the bag when recording it.
Their subsequent albums got progressively more polished. Still have the street band energy, but they obviously put a lot more effort into the production.
My favorite album BY FAR is the first one. It's just FUN!
You do not need to be the greatest musician to make great music.
This is a strange argument. It's not like there was a huge unfulfilled demand for good musicians before quantization / autotune.
I don't think anyone is saying music should all be recorded "raw" without any post-production. The argument is that the workflows used by most big record producers today encourages "sanding away all of the edges" of music, thru quantization, pitch correction, comping etc.
I'd argue that's only one problem (algorithmic music discovery, and how that removes gatekeeping and incentivizes bland "genre" music being a far bigger one). But the two go hand-in-hand. Lower barrier to entry (of "listenability") + lower barrier to distribution = a music landscape dominated by bland, samey bands.
It is also part the relatively new (say, 20-30 years) distributed aspects of music production.
In the 1970s, you began to see the emergence of the drummer "playing to the click" and the main motivation behind this was that some number of the rest of the performers were not in the room and/or not recording at the same time as the drummer. If you let the drummer just do their human thing, the timing will wander a bit in ways that are hard to follow and respond to if you were not in the room. So producers and engineers began to use click tracks to ensure that now matter when/where the differ performers actually recorded their work, everyone would be locked together.
If you record all together in a single space at a single time, you generally don't need this (unless the musicians are not very good and the drummer still wanders a lot).
So it's a phenomenon that didn't begin with DAWs and computers - that technology just took the overall approach even further.
It’s just cheaper. It has nothing to do with it being more listenable. Same with auto tune. It’s cheaper to have some mediocre singer do the track and fix it later. You don’t have to pay the great singer. And if you do get a great singer, you don’t keep them in the studio for as long. It’s just a cost savings. It doesn’t improve the music.
It's cheaper to do what? The thing they're getting for less money is precisely the improved quality or listenability. Their definitions of quality and good enough might be fucked up, but the intent and purpose is clearly to improve it up to some minimal level of quality, otherwise how are they judging a mediocre vs. great singer? How do they know when the great singer is done singing (sang a good enough take), or when they've tweaked the mediocre vocal track enough? The technological interventions change the ratio of cost vs. quality, they don't make quality irrelevant. If they did, you could just record the crappy singer as-is and save even more money.
This post's point is that musical "quality" is not a unidimensional feature and that mechanically optimizing for these "quality" metrics can kill something interesting in music.
>It takes unlistenable performances and makes them listenable.
How is that a good thing?
Humans have limited attention and time, so most of us don't like to waste them on bad things. Human artistic history is suvivorship in practice. The best things rise to the top and are kept, and the worst fall away. This is a natural mechanism that produces a shared culture of good things. When everything is good, nothing is, which is exactly the place we find ourselves now.
Not everybody gets to be an astronaut when they grow up. Our culture celebrates mediocrity because we're afraid of the implications that certain people are naturally better than others. The result is we have no more heroes or aspirations, and computers can fix lack of talent and lack of effort at improvement.
Yeah. If you can’t do something that’s fine. I draw as a hobby but I suck at it. When I see some of the pre-post autotune performances I can’t feel that people are just feeling entitled to a paycheck.
Lack of awareness does not mean that it doesn't impact how the music is received. Non-musician not having the knowledge and skills for analysis does not mean it makes no difference to them.
It's like a non-cook not realising the role of MSG in a dish: despite the lack of awareness, they wouldn't enjoy the dish as much without the MSG
To jump on that analogy: Try telling normal people that all (almost all?) restaurant food uses loads of salt. They will tell you "That's fine; I don't want bland food!"
Man fuck that. Audio engineers lost their minds snapping all these live performances. One thing to make dance music on the grid but a live high school band?
As the “exception that proves the rule”, there’s a fair bit of dance music that intentionally goes off grid to either build tension or sound more organic. There are even entire subgenres dedicated to sampling more acoustic and traditional instruments (usually percussion) while embracing underengineered sound.
Yeah, especially for a high school band. I'm sure budget was limited and they probably didn't have time for retakes. Playing to a click isn't easy, and playing to backing tracks not perfectly in time with a click is even harder. These are high school kids and professionals will tell you this isn't easy. Given that they probably didn't have time to go through and manually clean things up, you snap it to a grid so it doesn't sound like a mess.
Famous studio and touring bassist Lee Sklar has a great youtube channel [1] where he tells war stories and plays along to a few of the thousands of songs he has recorded on over the past 50 years.
In one (which I can't find) he mentions a Japanese musician coming to LA to record with the top LA studio guys. It is very common to play to a click track, and one song just wasn't gelling. The studio guys said, "Let's try it one time without the click track." The Japanese musician said, "We're allowed to do that?" Anyway, they went ahead and the song came to life. The musician literally cried after because it felt so good to not be focused on tracking the click instead of being in the moment making music with a bunch of other humans.
I just find a lot of modern music... boring. It feels perfect but just isn't interesting.
In contrast, when I check out older music from less successful musicians, it's often very imperfect; both in that it's boring, but in that there's little talent there.
Or, to translate: Gridification is modern style that doesn't make up for lack of talent; and ironically masks the talent that used to be there.
Related: I suggest checking out some of the newer Lemon Twigs albums. They're decent songwriters and excellent musicians. They went 100% analog on their most recent album, which I initially thought was a gimmick, until I read this article.
Another recommendation is jazz music. Modern music, and especially modern pop music, is incredibly harmonically simple. You rarely hear anything beyond a 7th chord. But step into jazz and all of sudden you're hearing augmented, diminished, and other more complicated chords and sounds that in just a raw way sound more complex and interesting than a lot of modern music. (Classical is even more out there, but that's a harder sell in 2026 unfortunately)
There is a level of melodious incoherence with Jazz that I can't get onboard with. I think that's the entire point of it but parts of it start to sound like a Sam Ash music store.
There is definitely some free-jazz that is too much for me as well. But, some of the incoherence is just chromaticism that purposely approaches or plays with our expectations for tension and release.
An incredible player is someone like Chris Potter on the sax ( https://www.youtube.com/watch?v=4Fvsz8KuD-g ), who has a truly virtuosic control on tension and release. In the video I linked, he (and everyone in the band too) plays with the ballad form over and over, never seeming to resolve perfectly, until they transform into a wild funk sound that is an ecstatic release that is also just wildly dance-able. The incoherence, even some of the screeching, that emerges is like the cows that are finally let out of the barn after a long winter and jumping for joy in the sun.
I'll take it all a step further: even better if you can find a place to take in some live jazz. Also, on your point about classical: live classical music is also incredibly moving.
While Kansas City isn't necessarily "special" in this regard, it's where I presently call home, and we have an amazing live jazz scene here, and the symphony at Kauffman Center has amazing classical music performances - both of which I cannot stress enough to check out to break apart from the "boring" of the gridified music this essay discussed. Find some live music in whichever locale you're based.
Granted, I'm biased... as I was already jamming out to some jazz when I hit your comment.
I agree, but I prefer old jazz from before I was born. Modern jazz doesn't appeal to me, it sounds too clean and unoriginal to me (maybe gridification, I don't know, I'm not a musician), with some exceptions. But maybe I didn't find the good modern jazz.
Some good modern jazz I'd recommend (I'm into sax and guitar):
- Chris Potter, I already linked him in another comment but he is a virtuosic post-bop player with incredible improvisational skill. Many people consider him to be one of the best living sax players in that style.
- Shabaka Hutchings, another sax player (he also plays clarinet and flute) with incredible songwriting ability and ability to play over varying genres and styles. It's hard to keep up with all of his projects. A very powerful player and a very gentle player when he chooses. https://www.youtube.com/watch?v=gpfpYTmohAk
Hell, just throwing in some syncopation into a pop song would make it wildly more interesting.
"Triplet-flow" in rap took the genre by storm after Migos and Future popularized it a while ago, and the initial appeal was the rhythmic novelty of it. (It obviously got stale after the novelty wore off.) Pop music in many of the big genres is in general, however, quite reticent to move away from perfectly riding a 4|4. Even using a 6|8 would be interesting.
There is a decent amount of pop music in 6/8. I always joke that it's like a hack to making a song better because 6/8 is rare enough that it makes the song sound interesting but common enough to not alienate the listener. It ends up sounding mostly like a hard swung 4/4 but then fills and some melodic lines are just a little bit more rhythmically spicy.
I found that surprisingly many k-pop songs do microtonal shenanigans. Nowhere near most, but it's there. Even something as popular as Boombayah (check the 4th note, it's not an on-tune semitone).
There's also more niche stuff like Flying Microtonal Banana.
I've always found stuff about how technological/sociological advances beget changes in music really interesting. David Byrne's book 'How Music Works' covers this exact point about quantisation and much more.
I would argue that there's probably a lot more nuance here when you look at how economics of music have changed in the last 50 years - it's now so much cheaper and easier to produce and listen to music. There are millions of bedroom producers experimenting with production techniques, and the market for music is completely saturated with more good music than anyone could possible listen to. Record labels just can't compete when it comes to making interesting, creative music, however they can still win at marketing. I think the modern charts now are less determined by what resonated well with audiences, but rather what has been calculated to be the most 'marketable' kind of music, and I suppose a metronomic tempo is an easier sell in this regard.
Also worth noting: a perfect tempo makes a song a lot easier to remix, which can often gain the song a lot more exposure than the original edit.
I didn't read David Byrne's book but I heard about it on the radio. Doesn't it also mention that drummers started playing with less tempo variation since drum computers are used?
I just looked at my copy... Yes, Chapter 4: How Technology Shapes Music. He also discusses how he decides which "flubs and mistakes" he wanst to fix, remove, replace or keep.
> the kick and bass do not line up exactly. You never hear that in a contemporary pop mix anymore. That is not because musicians lost the ability to play together.
Why would losing the ability to play together cause the kick and bass to line up? This is the most forced claudism I've seen in a long time.
Looks like a standard editing mistake to me. The previous sentence probably used to say something like ‘musicians used to not be able to play together’ but when it was edited the editor forgot to update the next sentence for coherence. Editing long-form text is hard :)
>Why would losing the ability to play together cause the kick and bass to line up?
Because if they lost the ability to play together, then they would be forced to either overdub tracks played individually with a click track, OR use midi, in both of which cases the kick and bass typically lines up.
So what this say, is we get this exact "perfect line up" now, but not because musicians lost the ability to play together and sound naturally, but because we prefer these other two production techniques and audio processing that removes all natural nuance.
It was an awkward conclusion to this paragraph. When I read it, I assumed it to mean that the kick and bass may intentionally be misaligned (slightly) to make each distinct. (But I have little knowledge of music production.)
I can only make sense of it as "musicians are so bad at playing together that we don't try". If say the bass player cannot play with others I'm going to record them playing each chord/note in isolation, and then use that as midi samples to track. Do that to everything and you get perfectly lined up music. If you can play together "good enough" then I'm not going to do that.
I think its the difference between playing live, playing in parts and playing to a click track.
again these are not exclusive, Even though I fucking hate it, I can drum live to a click track. I hate it because everytime I am slightly off I can fucking hear it like being hit over the head with a frying pan.
> It is because modern music production now treats human temporal particularity as error in need of correction.
I was thinking about this the other day as I was playing guitar adding slight pauses and swings to what I was playing. The imperfection makes things so much more interesting and fun imo.
Battery 3 from Native Instruments is a drum sampler from around 2013. This is in the manual:
Remember, machines don’t kill music, people do—and quantization is one of the main weapons.
...
Quantizing everything is the quickest way to suck the life out of a piece of music.
I remember in the late 00s when making fun of autotune was a common joke, and a way to attack "bad" or "lazy" singers / musicians. Probably just because some bands were using it stylistically (T-Pain most notoriously).
It was around the same time there was a backlash against the (rather egregious) use of Photoshop in advertising. But for better or worse, in both cases, the use of these and similar tools have become ubiquitous, just in a way that's more subtle.
Nowadays, hearing a band that isn't pitch corrected and quantized to all hell is refreshing, similar to seeing an actor with imperfect teeth. It feels like, for probably a whole host of reasons, almost every art form has been trending toward removing all the imperfections that make human art feel "human" and interesting. I don't expect this will get better now that we have AI.
> I remember in the late 00s when making fun of autotune was a common joke, and a way to attack "bad" or "lazy" singers / musicians. Probably just because some bands were using it stylistically (T-Pain most notoriously).
I always thought T-Pain and Cher did it as a way to help normies hear the autotune that was everywhere in that moment.
Donald Fagen is/was well known for doing take after take after take when recording looking for what he found perfect. He would sample notes, replace notes, replace drums and cut and paste until he made the mix he wanted. I was in the music software/hardware industry at the time and the Wendel and Wendel Jr. computers/samplers/drum replacers along with Garfield Doctor Click rhythm controllers were heavily used. I am sure he also did a ton of old school punch in/out and used other mechanisms to get things just the way he wanted.
The point I am trying to make is that the processes that used to take a maniacal amount of time, resources and attention to detail can now be done quickly and automatically. Does Steely Dan's Aja sound better or worse than a modern album produced using modern techniques?
This comes down to taste and who am I to determine what someone else finds appealing in their music? Producers have been "producing" songs and moving them from their originally recorded state into something entirely different for decades.
I personally miss singers being slightly out of tune and beats being slight out of time in songs. I love hearing imperfections in older music. But, would Trevor Horn have pulled some of the singers he produced back into tune before pressing the vinyl? Someone should ask him. I am inclined to say yes.
The analysis is interesting, but I think the premise that human error is a defect to be corrected is flawed in that bands have consistently always strived to play in time together. Humans tolerate a certain amount of variance, and may even like it, but if you have a drummer who can't keep time, you don't have a band. Big Band music made it a spectacle of keeping so many instruments in time and tune with each other.
I think the thing that has changed is that the tools to "fix" the bugs are better and more commoditized, so they're used more. And similarly, production costs have gone down as hardware is commoditized, so there's more music to choose from. The last piece is that social media has made it popular to all like the same thing and make dance videos to the same songs, so already perfect little pop pieces climb up faster.
There's a bit of an artistic treadmill here; I don't know if there's a name for it.
For ages, making painting realistic was the gold standard. An assumed skill for an artist was producing photorealistic art, with composition and light and all of those other characteristics being secondary (though they most likely affected how the art was perceived).
When photography appeared, the ability to render a photorealistic scene was no longer the sole domain of painters; and in fact much of art at that time shifted towards more abstraction.
So too with music and audio; technical perfection was once a prerequisite skill for a musicisian, and now is no longer primary because it can be papered over in the editing booth.
Its about musical production. This isn't the first time someone has done analysis of timing in modern music. I think I first saw an article on slashdot in around 2010.
Recording "live" can be hard. it requires a bigger studio, more equipment, better engineers and crucially, everyone to be there at the same time.
you can, if you have a click track, record everything separately, and then marry it up later. once you have the drums lined up, you can just replay the tracks you have without the click track.
this means that you can record separately, at different times, in different locations. or if its your bedroom, you can move stuff around to the microphone you have.
But if you record a track as parts, you can then edit it. You want it to sound good right? laptop power has increased significantly, so any plugin can be run in realtime on multiple tracks.
SOme music suits it, others doesn't.
I _love_ the sound of a live band who are playing well. I don't love the sound of a live band who aren't.
Its hard to play together well all the time, so click track gives you a better bang for buck.
I prefer keeping the concepts of "playing in time together" and "playing in time together on a strict grid" separate. For example, in a classical orchestra you typically very much strive to play in time together, but you do that according to the conductor's interpretation of the tempo marks of the piece, which often calls for tempo variation.
In jazz, you also typically do your best to play in time together, but for long sections timing is often made an open-ended problem where the rhythm section doesn't interfere too much with stylistic timing variations that IMO often separate the simply skillful solo from the deeply felt one. The idea of an accent in modern popular music seems largely have practically narrowed to dynamic articulation, but agogic accents are a thing that skilled musicians can perform in a band setting without it seeming like an "imperfection".
> The last piece is that social media has made it popular to all like the same thing
Social media definitely didn't invent this. I'm not overly familiar with what is popular on social media, but it seems like the internet has greatly expanded the viability of different styles and niche music. When I get together with friends, we all have tons of music than most of us haven't heard before.
I don't think we are talking being able to keep time, that is a given for any musician playing to be recorded.
What the article is talking about is the teeny tiny imperfections of hitting the notes/drums at the beat (or "grid").
I agree that the imperfections are overfixed to an inhuman level.
Interestingly, there is a "swing" or randomize feature available in some hardware and I guess in every DAW, so if the producers wanted, they would be able to "fix" the notes to a grid, and then add a perfectly calculated random touch to make it sound human.
The first example the article uses of the kick and bass is not a subtle imperfection. It's massive, almost a full 16th. That's middle school levels of imperfection. With pro musicians it's obviously a deliberate stylistic choice.
> Big Band music made it a spectacle of keeping so many instruments in time and tune with each other.
"tight" big band music "in the pocket" is deliberately and consistently behind the beat.
Peggy Lee and other great singers are more in tune than auto-tune or any perfectly tuned instrument is. Modern tuning, aka equal temperament, is a compromise. Every key is perfectly identically subtly out of tune. A great singer will sing with natural tuning, adjusted on the fly for each key change. Peggy Lee would sing a third slightly flatter than an auto-tune will. Unless an instrument is also playing the third, in which case she would sing in unison with the instrument.
Most of us aren't Peggy Lee or the incredible studio musicians that came out of New Orleans. We need to learn the rules before we can break them.
I’m a lifelong musician and producer who has been experimenting a lot with Suno for music creation this year.
It’s been interesting to notice that Suno-generated tracks have all kinds of subtle tempo drift, and the tempo maps look very similar to the irregularities shown in the article, at least from a numerical perspective. I have actually spent a lot of time quantizing AI-generated tracks to the grid (to aid chopping/editing), even creating automations that do it for me now when I ingest Suno tracks into Ableton for follow-up production work.
Unknown to me whether this is
A) an emergent “imperfection” fundamental to the model and its training dataset, or
B) something Suno intentionally injects to “humanize” their output,
but knowing that low-effort Suno-generated tracks are currently flooding music distributors, I think it’s ironic that the same factor the author describes as exemplifying the very humanity of recorded music may also be one of the biggest tells that a song was AI-generated, for music being released today.
Beautiful use of sound analysis tooling, and pretty much incontrovertible evidence. Love it! Maybe that's the thing that makes most pop songs completely uninteresting to me. It just makes me want to go watch live music even more.
I dabble in creating mashups from time to time [1] and I see that first hand. One of my first was built around Let's Groove by Earth Wind and Fire. As a funk act, they are as tight as a group of musicians can ever hope to get, and still, the tempo would subtly speed up and down from bar to bar.
It makes things much more challenging to sync up! You can't just speed up the slower song (or vice versa), because even if the averages match, it will still be all over the place from bar to bar. I now have much better workflows to deal with this, but for that first one, I spent a looong time manually adjusting things to get them to fit.
Working with tracks that were recorded to a click (or digitally gridified, as per the article), are much easier to work with. Match the tempos and everything automatically works. Definitely helps creating my mashups...
But I do agree with the post, in that it loses some indescribable human aspect. I love the push and pull of musicians trying to follow each other rather than concentrating on the click. Or even just pushing and pulling against oneself with things like rubato in classical music, sometimes accelerating, and sometimes slowing, at times almost to a dead stop, so as to build a sense of antecip.... pation.
Often I get a lot more out of some YouTuber/streamers's slightly scuffed cover of a song than the original studio recording. "Perfect" music often makes me feel nothing. It's the tiny individual mistakes and flourishes that make it easier to relate to the performer and music as a human - and that's what puts a smile on my face listening to it.
The article doesn't seem to say this specifically, but my take at defining "gridification", in the sense of "why is this term in particular used to refer to the thing they are talking about". The article says "Gridification is when a human performance is treated as error in need of correction and turned into a machine-perfect representation of itself" but they could have used "blorficoration" for this; why "grid"?
Griddification is trying to take each note and align it to a grid, the two axes being time and pitch. All notes (spoken or sung or struck) start aligned and are tonally compatible. In that sense it is like the grid you see on a piano roll describing music. Basically a de-jittered midi-style reduction of the actual music being player.
I may be missing something here but most of the 2025 songs sound like they are clearly programmed drums.
Die with a Smile sounds like the exception but I wouldn't discard the idea that those drums were entirely programmed / painted either. It's not like it's difficult to make a flam in a piano roll.
Why not pick from genres that don't predominantly use drum machines as a better comparison?
I watched the other day Rick Beato's "What If Stairway To Heaven Came Out Today?" and what stroke me is that his granddaughter remarked that the song didn't have auto-tune. First, she has great musical ear. Second, she destroyed my illusion that "true" musical producers were in the business of zapping the singer whenever they went out of tune, not of using auto-tune, which is something more becoming of the plebeian musicians who make songs from their bedroom.
With that said, a silver lining of AI learning to imitate mainstream producers is that the (big players of the) music industry may finally lose their grip on people's musical taste. Maybe people will start searching for alternative styles, and going to their (minor artists with a unique voice) performances in person.
Led Zeppelin was influenced by musical traditions that placed an emphasis on playing with intonation and rhythm as a technique for conveying emotion and creating musical tension. Producers back then would have known that, and known knot to mess with it even if they did have the tools back then.
But also, a perfected, highly produced sound is part of the aesthetic values of modern popular music. So producers who do it now aren’t necessarily wrong, either. They’re just following taste. If people start loving a loosey goosey White Stripes sound again, producers will go back to leaving it in.
> With that said, a silver lining of AI learning to imitate mainstream producers is that the (big players of the) music industry may finally lose their grip on people's musical taste. Maybe people will start searching for alternative styles, and going to their (minor artists with a unique voice) performances in person.
I share the belief that this accelerated the rise of Angine de Poitrine, who use microtonal stringed instruments and uncommon time signatures. Both of those things have been around for a while, but never went viral until now.
Some people get really disheartened when they see examples of young people with perfect pitch like this, but they really shouldn't. Perfect pitch (or absolute pitch) is one of those things that is obviously incredible on its own, but it's not really necessary to make amazing music. Presumably it might make music making easier, but the idea that perfect pitch destines someone for musical virtuosity is somewhat more a marketing pitch than a thing. Mastery of relative pitch and ear training in the context of songs is, for the most part, far more important. So much so that I'd argue perfect pitch is kind of irrelevant in the grand scheme of things.
Is this arguing that, in order to produce commercially valuable music, all music is conforming to a set of standards, or traits and conform to "a grid"? That's how I read it.
Does this (2019) article propose the opposite? That hit music usually contains a sort of pleasing surprise, and that over time our hits have to become more and more unpredictable musically?
It's definitely extreme these days but it's been going on for quite a long time. Apart from live music, I really enjoy some recordings which were raw and largely untouched. Lilywhite Sessions is a perfect example (the original unmastered leak), and there are some really cool rehearsal tapes from Deep Purple floating around. I wish I could get the "as it was recorded" versions of some modern albums.
That's partially why Burial's first album brought such a fresh air to the music scene. He didn't use a DAW, but a plain, boring audio editor Sony Sound Forge and adjusted his samples until the groove and the tones felt right.
I've also seen produces complain that the default BPM is 120 so they start there and later realize that it should have been changed. Minor annoyance, but it reveals the importance of defaults.
> I was always scared of people who had studios. Heroes of mine like Photek suddenly became Rupert Parkes in his studio, telling everyone how he did it. The magic got a bit lost.
> So I thought to myself fuckit I’m going to stick to this shitty little computer program, Soundforge. I don’t know any other programs. Once I change something, I can never un-change it. I can only see the waves. So I know when I’m happy with my drums because they look like a nice fishbone. When they look just skeletal as fuck in front of me, and so I know they’ll sound good.
[..]
>> So does that mean your drums are not necessarily in time?
> My drums are definitely not necessarily in time. When I try and do drums that are too regimented, they lose something. But the moment I put drums where I think they sound good, rather than in time, they seem to have that roll, the swing of the jungle and garage tunes I love.
I think that’s definitely part of the sauce that his many imitators are lacking. It’s really hard to find anything that hits that high bar - have you had any success finding other artists releases that measure up? Moderat has a few - rusty nails, versions maybe
When I was in highschool in the 90s, our school spent a ton of money on a set of midi keyboards hooked up to workstations, new technology at the time, and offered a class on digital music production. Our teacher was the school music director and I remember he was really excited.
The software had a quantization feature which he showed us how to use but he warned us not to use it on everything, because it makes music sound mechanical and less human. I guess that philosophy did not survive in the music industry.
Something seems strange about this publication. They're doing a lot of talking about machine perfection in music, but some of the articles have big chunks that read as 100% AI on Pangram, mainly “The Shame Machine” which is itself primarily about generative AI. However, I didn't observe that in the OP here, nor on the most recent post “The Climate Crisis Has A Music Problem”. Was that an early approach that's since been abandoned, or a dubious meta joke for that specific article, or what? And then the rest… clearly they have a thesis they want to keep harping on, and as someone with musical experience I'm empathetic with what seems to be the main throughline of it, but the Substack as a whole reads as being heavy propaganda for their particular certification, which is a scammy feel.
Also, an existing creative who understood and knew the topic would just call it what the rest of the industry calls it; quantization and autotune, not "gridification", that's some term they've seemingly made up together with Claude, and it sure sounds like it.
"we like music in the abstract as a series of numbers" seemed like a nod to AI or quantization general. Now, it might have just been genuine instead of an ironic statement. Also explains an unreadable-to-humans giant table in the middle of the article.
That tempo analysis graph is cool to see, showing the subtle fluctuation of "local tempo", rushing or leaning back intentionally. It's a sophisticated technique though, and I bet outside of pop music, like jazz musicians - especially drummers - still know how to do it; and seasoned listeners also know how to feel and appreciate it.
I bet the same is happening with writing, at least with technical and academic. Automated grammar checkers and AI correctors bland the language by enforcing strict grammar. Whereas a slight departure from standard grammar and punctuation is something that could give writing character and personality.
Man, the '75 hit just grabs me from the very first note. The funky rhythm, those fun synthy parps... everything is just so much more compelling, more human, more catchy. Apart from anything else, the Mars-Gaga thing sounds mushy in comparison.
A great book tangential to this subject matter is Jaron Lanier's "you are not a gadget" which spends a substantial amount of time on the development of music in parallel to computers. My recollection is that midi basically became the dominant format because it was first, and it was first because it was more or less the most obvious way to represent music in a simple programmable format. Other formats that allow for more flexibility didn't crack through due to the critical mass of adoption of midi, so it persisted until today.
Well shoot, I snap-to-grid because while I make music that I love[1], my actual playing skills and timekeeping are poor -- I'm always a bit fast, hitting notes before the beat, even with a click track -- and without either automatic quantification or at least the ability to edit the timing to be more reasonable in post, my tracks would be worse.
Where in the binary between "you're a real musician who uses human feel/timing to create art" OR "you're a hopeless commercial grifter chasing perfection" do I fall?
Maybe controversial, but this is why I don't like live performances. If tickets were cheap, maybe, but paying a high price to hear a mediocre reproduction of the digital copy I already have is a pass for me.
I literally do not care that someone can "hit high notes" or that "unlike this other artist, my favorite artist can actually sing." I'm listening to music, not grading music class students.
Well you’re not just listening to music, you’re also watching a performance. Seeing someone’s fingers on the guitar, touching the keyboard, gripping the drumsticks, watching the expressions on their face and their body posture - none of that adds to the art for you? It’s performance art, and seeing it live gives you something visceral that you can at best imagine or recieve second-hand from a recording.
So yeah I guess a little controversial, haha
Honestly if you’re not into it you’re not into it and that’s fine - but for sure, if you’re going to a concert expecting a pitch-perfect recital of studio-produced tracks, then… you’re kind of just setting yourself up for disappointment. That’s the wrong expectation to go into it with, imo.
Some artists would consider the gridified recording as what they strive to achieve in practice.
Some artists would gridify recording so people who come to their concerts enjoy their plays with "errors" on purpose.
As humans, we don't want to be seen what we spent our entire life to be considered err'ed. I understand the pitcher perfectness in that regard, though I'd still like unpolished "rock/metal" music.
Now I can only enjoy those raw moments in concerts.
> Some artists would consider the gridified recording as what they strive to achieve in practice.
Sure. We would also consider it impossible to achieve in reality, and so the art is in the trying, and gridification removes the art. Hell, it removes the artist.
They say perfection is enemy of the good. But in a context of "art", it does not apply. Without striving to achieve such perfection, an art is not an art.
Sure, it wont be achievable to microsecond level but I prop the artists who strive to be perfect (as well as who don't)
---
I sometimes go to concerts and sometimes wishes they sing/play like they recorded. Some really suck live, with some diff gig as they cannot pull studio quality performance.
I want to see what I paid for. They can change their plays but still want some solid work.
I've been wondering if we're going to end up with some kind of backlash against all the autotuning, 'gridification' as this article explains, etc etc., and AI slop as well as deepfakes filling up the interwebs. You can't know what's real anymore. So people will start preferring live performances. Not only live music, but, say, theater/musicals/opera instead of going to the movies, and so forth?
Not saying this as some kind of general "anti-AI" argument or anything. I'm not advocating for the Butlerian Jihad. I find AI a very useful tool, and I used it at work pretty much every day. But for art, which in a sense is a celebration of human creativity, something gets lost if we use AI to generate it, doesn't it?
I'm kinda hoping we get a revival of playing music (e.g. at the piano, on the guitar) at home. As Gillian Welch said "If there's something that you want to hear | You can sing it yourself".
This should be at the top of the thread. Metal might be the most obvious exemplar to an untrained ear.
When I hear a drum machine execute with microsecond level precision, I feel like I want to throw my computer through a window. It is instantaneous revulsion. There isn't a ponderous "gee I wonder if this is real/fake" phase of discovery. It's obviously fake and trash. Immediately. Every time.
Orchestral arrangements done the same way might be the only thing that is worse to my ear.
There are situations where having good timing is important, but you cannot put the entire thing on an atomic clock and call it "art".
This is largely pretty subgenre specific, though. For tech-death the robotic sound of quantized perfection is often a stylistic choice, it's literally a part of the appeal of the music for some people. (An obvious example is Archspire, not my cup of tea, but some people love that production style.)
I don't really hear the same thing as much in doom, black, or more extreme death variants.
Yeah I'd love to see/hear some examples here; I mostly listen to small-to-mid-level black/doom/trad/power bands and haven't noticed this type of standardization of sound.
It's possible I don't have the ear for it though, so now I wish the article had posted how it made this type of analysis so I could run some of my albums through it myself.
> I wish the article had posted how it made this type of analysis so I could run some of my albums through it myself.
I very simple way is to throw a track into a DAW and see if the drum hits line up perfectly with the "grid". Most DAW's have a grid background so you can see. If it's perfect for the whole song, then it's definitely quantized to death (pun unintended, though does sound quite kvlt).
The entire premise of this essay rests on the idea that musical performances are necessarily played live (particularly, with a live drummer), in the tradition of rock music. This just qualitatively does not apply to much modern music, where drums are played by a machine or computer. Top on the '2025' list is two good examples from totally different genres: Kendrick Lamar (hip hop), and Billie Eilish (vocal-forward electronica).
Are electronica and hip hop worse for this? I suppose that's up for debate - but we now have multiple full generations of music listeners and critics who obviously, passionately love those genres, and they've been growing for decades. This is easy to see not just via charts but online discourse and the zeitgeist.
Do those genres 'groove' worse and feel less human? This is more subjective, and more debatable, but I would argue that hip hop has redefined what 'groove' can even mean, with its very off-the-grid, loose feel despite the technically very precise machine looping the beats (because producers nudge each beat off the grid, usually in very intentional ways, or play the beat in live then adjust it). And it's hard to argue that electronic dance music in its many forms hasn't made people move and connect with the music's rhythm in powerful ways, despite almost complete quantization in most dimensions.
I think purely looking at tempo stddev and not much else papers over these aspects a little much.
I agree purely looking at tempo variation isn't everything, but I think perhaps "Zero tempo variation is a necessary but not sufficient indicator of "gridifcation" is closer to the message?
More specifically, they define gridification as: recording multiple takes, nudging every single note onto the grid for each one, then selecting the best take section by section (however small those sections are) so that every single note is absolutely perfectly on time, in tune etc, loses... something - it removes some of the humanity. The author also explicitly cites many other kinds of recording and editing as NOT gridification - e.g. playing to a click, using a drum machine.
>The entire premise of this essay rests on the idea that musical performances are necessarily played live (particularly, with a live drummer), in the tradition of rock music. This just qualitatively does not apply to much modern music, where drums are played by a machine or computer
One could say modern music is not "musical performances" then. A sequencer or drum maching playing isn't "a performance", it's a programmed loop.
>Do those genres 'groove' worse and feel less human?
Drum machines are instruments. Even if sequenced, drum machines are live instruments and they require skill. It requires time to master. It's not only "programmed loop".
When I play the drum machine, I program it live. I never play the same thing twice. It IS a performance. It applies for both vintage drum machines (such as E-mu drumulator or tr808) and new gen ones or grooveboxes (such as Electribes or Octatracks).
On top of that, not all drum machines are created equal. There are whole genres based on 808s, 909s, 303s combined with these… using DINsync, which involves retriggering internal hardware clocks on the machines at trigger rise-time speed. rather than sending a succession of MIDI notes. That stuff is still 'electronic clocks' but not in the sense of 'lives on a computer that triggers notes', in the sense of 'sequencers firing notes very simultaneously'.
That's a timing lock that's tighter than the usual sort in recent generations. There are DAWs (Renoise) that pride themselves on being able to send MIDI with less slop and irregularity (nice trick since MIDI is a serial bus, but you'd think it'd be more common to get right). There are whole genres (see: psytrance) that do very well in trackers which can fire sample-accurate notes simultaneously without resorting to MIDI.
Even though it's of the machine, the groove of an 808/303 pair synced by DINsync is a different beast from 'snap to grid and autotune'. There's subtleties to everything.
> One could say modern music is not "musical performances" then. A sequencer or drum maching playing isn't "a performance", it's a programmed loop.
The drum machine is not all of the performance. In pop you'll have at the very least a singer. Even if all of the instrumental is machine-made, having a live singer definitely makes it a "performance".
I didn't feel like the article is being in any way judgemental of music created using perfect digital instruments. The author seems to be only against over-correcting human performances. So it was about all of the new ones being on the grid, rather than about each of them.
> And what about dance music? Isn’t that entirely gridified? No. Not by the definition being used by the Musical Form Institute.
>
> What!?
>
> The next essay will cover dance music.
>
> To recap: gridification is not playing to a timing reference. It is correcting human performances to a fixed grid after the fact.
I felt that the essay acknowledged differences in production between genres. Tempo is used as the main example, but they also mention vocal tuning as an example of gridification.
> What modern production calls “correction,” and mistakes for care, is really a form of epistemic violence against the record of human beings objectifying felt time for our understanding.
I hope this was written as tongue in cheek, otherwise wow, you are taking yourself too seriously.
To err is human; to be flawless, artificial. I think people will appreciate this more as they become saturated by AI art.
What I want to know is what the author thinks of drum machines, with their mechanical perfection. Roland's very first product, the TR-77 drum machine, was released in 1972, in time for his 1975 comparison. They have been popular ever since.
> What hits me immediately is how perfect the vocals are. There are melodic leaps in the verse that are incredibly difficult to nail, and Bruno Mars is hitting every single note pitch-perfect and in absolutely perfect time.
Isn't every song recorded as every word singularly at a certain pitch? And timing done completely in the DAW instead of while recording.
Barely was singing when a friend of mine did ``professional'' recording, and had to know each pitch of a song to perfection, since no overlap was ever sung.
Now was there autotune? dunno. its not like some people who've obviously been tuned. There are some people who are really that good. Michael bauble is one of them
Charitable of him to exclude drum machines, MIDI and so on, which have been quantizing and automating time for 40 years. We've been on a slippery slope that whole time. And to me there's not a lot of difference between recording yourself in the act of conforming to an unvarying mechanical beat (essentially gridifying yourself in real time) vs. fixing it in post.
When we finished we got to go watch the sound guys at work. They meticulously gridified and autotuned every note. After it was all over I got the cd, I listened to my favorite tracks that I got to spend the semester playing with my friends, and it had no resemblance to what I knew our group to be. I think I threw the album away.
https://www.instagram.com/reels/C9h9xpNOoEN/
But what about the non-greatest drummers? The other 99.9% (or higher) of musicians who are tiers below Bonham? Quantization is a net improvement. It takes unlistenable performances and makes them listenable.
Their subsequent albums got progressively more polished. Still have the street band energy, but they obviously put a lot more effort into the production.
My favorite album BY FAR is the first one. It's just FUN!
You do not need to be the greatest musician to make great music.
I don't think anyone is saying music should all be recorded "raw" without any post-production. The argument is that the workflows used by most big record producers today encourages "sanding away all of the edges" of music, thru quantization, pitch correction, comping etc.
I'd argue that's only one problem (algorithmic music discovery, and how that removes gatekeeping and incentivizes bland "genre" music being a far bigger one). But the two go hand-in-hand. Lower barrier to entry (of "listenability") + lower barrier to distribution = a music landscape dominated by bland, samey bands.
In the 1970s, you began to see the emergence of the drummer "playing to the click" and the main motivation behind this was that some number of the rest of the performers were not in the room and/or not recording at the same time as the drummer. If you let the drummer just do their human thing, the timing will wander a bit in ways that are hard to follow and respond to if you were not in the room. So producers and engineers began to use click tracks to ensure that now matter when/where the differ performers actually recorded their work, everyone would be locked together.
If you record all together in a single space at a single time, you generally don't need this (unless the musicians are not very good and the drummer still wanders a lot).
So it's a phenomenon that didn't begin with DAWs and computers - that technology just took the overall approach even further.
How is that a good thing?
Humans have limited attention and time, so most of us don't like to waste them on bad things. Human artistic history is suvivorship in practice. The best things rise to the top and are kept, and the worst fall away. This is a natural mechanism that produces a shared culture of good things. When everything is good, nothing is, which is exactly the place we find ourselves now.
Not everybody gets to be an astronaut when they grow up. Our culture celebrates mediocrity because we're afraid of the implications that certain people are naturally better than others. The result is we have no more heroes or aspirations, and computers can fix lack of talent and lack of effort at improvement.
It's like a non-cook not realising the role of MSG in a dish: despite the lack of awareness, they wouldn't enjoy the dish as much without the MSG
In one (which I can't find) he mentions a Japanese musician coming to LA to record with the top LA studio guys. It is very common to play to a click track, and one song just wasn't gelling. The studio guys said, "Let's try it one time without the click track." The Japanese musician said, "We're allowed to do that?" Anyway, they went ahead and the song came to life. The musician literally cried after because it felt so good to not be focused on tracking the click instead of being in the moment making music with a bunch of other humans.
[1] https://www.youtube.com/@lelandsklar6363/videos
In contrast, when I check out older music from less successful musicians, it's often very imperfect; both in that it's boring, but in that there's little talent there.
Or, to translate: Gridification is modern style that doesn't make up for lack of talent; and ironically masks the talent that used to be there.
Related: I suggest checking out some of the newer Lemon Twigs albums. They're decent songwriters and excellent musicians. They went 100% analog on their most recent album, which I initially thought was a gimmick, until I read this article.
An incredible player is someone like Chris Potter on the sax ( https://www.youtube.com/watch?v=4Fvsz8KuD-g ), who has a truly virtuosic control on tension and release. In the video I linked, he (and everyone in the band too) plays with the ballad form over and over, never seeming to resolve perfectly, until they transform into a wild funk sound that is an ecstatic release that is also just wildly dance-able. The incoherence, even some of the screeching, that emerges is like the cows that are finally let out of the barn after a long winter and jumping for joy in the sun.
I'll take it all a step further: even better if you can find a place to take in some live jazz. Also, on your point about classical: live classical music is also incredibly moving.
While Kansas City isn't necessarily "special" in this regard, it's where I presently call home, and we have an amazing live jazz scene here, and the symphony at Kauffman Center has amazing classical music performances - both of which I cannot stress enough to check out to break apart from the "boring" of the gridified music this essay discussed. Find some live music in whichever locale you're based.
Granted, I'm biased... as I was already jamming out to some jazz when I hit your comment.
- Chris Potter, I already linked him in another comment but he is a virtuosic post-bop player with incredible improvisational skill. Many people consider him to be one of the best living sax players in that style.
- Shabaka Hutchings, another sax player (he also plays clarinet and flute) with incredible songwriting ability and ability to play over varying genres and styles. It's hard to keep up with all of his projects. A very powerful player and a very gentle player when he chooses. https://www.youtube.com/watch?v=gpfpYTmohAk
- Kamasi Washington makes big band style jazz that mixes old school and modern sounds. https://www.youtube.com/watch?v=x8WTPgeVPjg
- Some great guitarists:
- Julian Lage, exists somewhat in an overlap between jazz/classical/blues https://www.youtube.com/watch?v=bFRVK_1o2cU
- Bill Frisell, no one else sounds like him and his deeply romantic sound https://www.youtube.com/watch?v=Xpu33a_Y8VU
- Mary Holvorsen, she's much more avant garde and loves using whacky pedals for effects but still has incredible chops https://www.youtube.com/watch?v=zlozC3nHisI
"Triplet-flow" in rap took the genre by storm after Migos and Future popularized it a while ago, and the initial appeal was the rhythmic novelty of it. (It obviously got stale after the novelty wore off.) Pop music in many of the big genres is in general, however, quite reticent to move away from perfectly riding a 4|4. Even using a 6|8 would be interesting.
There is a decent amount of pop music in 6/8. I always joke that it's like a hack to making a song better because 6/8 is rare enough that it makes the song sound interesting but common enough to not alienate the listener. It ends up sounding mostly like a hard swung 4/4 but then fills and some melodic lines are just a little bit more rhythmically spicy.
Some examples:
* The Doors - People are Strange
* T. Rex - Hot Love
* Fleetwood Mac - Don't Stop
* Depeche Mode - Personal Jesus
* Marilyn Manson - The Beautiful People
* Green Day - Longview
* Oasis - Lyla
* Blur - Parklife
* Living Things - Bom Bom Bom
* Britney Spears - Womanizer
* Goldfrapp - Strict Machine
* Feist - My Moon My Man
* Gotye - Heart's a Mess
* The Black Keys - Gold on the Ceiling
* Paul Kalkbrenner - The Grouch
There's also more niche stuff like Flying Microtonal Banana.
I would argue that there's probably a lot more nuance here when you look at how economics of music have changed in the last 50 years - it's now so much cheaper and easier to produce and listen to music. There are millions of bedroom producers experimenting with production techniques, and the market for music is completely saturated with more good music than anyone could possible listen to. Record labels just can't compete when it comes to making interesting, creative music, however they can still win at marketing. I think the modern charts now are less determined by what resonated well with audiences, but rather what has been calculated to be the most 'marketable' kind of music, and I suppose a metronomic tempo is an easier sell in this regard.
Also worth noting: a perfect tempo makes a song a lot easier to remix, which can often gain the song a lot more exposure than the original edit.
Why would losing the ability to play together cause the kick and bass to line up? This is the most forced claudism I've seen in a long time.
Because if they lost the ability to play together, then they would be forced to either overdub tracks played individually with a click track, OR use midi, in both of which cases the kick and bass typically lines up.
So what this say, is we get this exact "perfect line up" now, but not because musicians lost the ability to play together and sound naturally, but because we prefer these other two production techniques and audio processing that removes all natural nuance.
I can only make sense of it as "musicians are so bad at playing together that we don't try". If say the bass player cannot play with others I'm going to record them playing each chord/note in isolation, and then use that as midi samples to track. Do that to everything and you get perfectly lined up music. If you can play together "good enough" then I'm not going to do that.
again these are not exclusive, Even though I fucking hate it, I can drum live to a click track. I hate it because everytime I am slightly off I can fucking hear it like being hit over the head with a frying pan.
I was thinking about this the other day as I was playing guitar adding slight pauses and swings to what I was playing. The imperfection makes things so much more interesting and fun imo.
Remember, machines don’t kill music, people do—and quantization is one of the main weapons. ... Quantizing everything is the quickest way to suck the life out of a piece of music.
https://docs.native-instruments.com/pdf-guides/Battery_3_Man...
It was around the same time there was a backlash against the (rather egregious) use of Photoshop in advertising. But for better or worse, in both cases, the use of these and similar tools have become ubiquitous, just in a way that's more subtle.
Nowadays, hearing a band that isn't pitch corrected and quantized to all hell is refreshing, similar to seeing an actor with imperfect teeth. It feels like, for probably a whole host of reasons, almost every art form has been trending toward removing all the imperfections that make human art feel "human" and interesting. I don't expect this will get better now that we have AI.
I always thought T-Pain and Cher did it as a way to help normies hear the autotune that was everywhere in that moment.
The point I am trying to make is that the processes that used to take a maniacal amount of time, resources and attention to detail can now be done quickly and automatically. Does Steely Dan's Aja sound better or worse than a modern album produced using modern techniques?
This comes down to taste and who am I to determine what someone else finds appealing in their music? Producers have been "producing" songs and moving them from their originally recorded state into something entirely different for decades.
I personally miss singers being slightly out of tune and beats being slight out of time in songs. I love hearing imperfections in older music. But, would Trevor Horn have pulled some of the singers he produced back into tune before pressing the vinyl? Someone should ask him. I am inclined to say yes.
I think the thing that has changed is that the tools to "fix" the bugs are better and more commoditized, so they're used more. And similarly, production costs have gone down as hardware is commoditized, so there's more music to choose from. The last piece is that social media has made it popular to all like the same thing and make dance videos to the same songs, so already perfect little pop pieces climb up faster.
For ages, making painting realistic was the gold standard. An assumed skill for an artist was producing photorealistic art, with composition and light and all of those other characteristics being secondary (though they most likely affected how the art was perceived).
When photography appeared, the ability to render a photorealistic scene was no longer the sole domain of painters; and in fact much of art at that time shifted towards more abstraction.
So too with music and audio; technical perfection was once a prerequisite skill for a musicisian, and now is no longer primary because it can be papered over in the editing booth.
Recording "live" can be hard. it requires a bigger studio, more equipment, better engineers and crucially, everyone to be there at the same time.
you can, if you have a click track, record everything separately, and then marry it up later. once you have the drums lined up, you can just replay the tracks you have without the click track.
this means that you can record separately, at different times, in different locations. or if its your bedroom, you can move stuff around to the microphone you have.
But if you record a track as parts, you can then edit it. You want it to sound good right? laptop power has increased significantly, so any plugin can be run in realtime on multiple tracks.
SOme music suits it, others doesn't.
I _love_ the sound of a live band who are playing well. I don't love the sound of a live band who aren't.
Its hard to play together well all the time, so click track gives you a better bang for buck.
In jazz, you also typically do your best to play in time together, but for long sections timing is often made an open-ended problem where the rhythm section doesn't interfere too much with stylistic timing variations that IMO often separate the simply skillful solo from the deeply felt one. The idea of an accent in modern popular music seems largely have practically narrowed to dynamic articulation, but agogic accents are a thing that skilled musicians can perform in a band setting without it seeming like an "imperfection".
Social media definitely didn't invent this. I'm not overly familiar with what is popular on social media, but it seems like the internet has greatly expanded the viability of different styles and niche music. When I get together with friends, we all have tons of music than most of us haven't heard before.
What the article is talking about is the teeny tiny imperfections of hitting the notes/drums at the beat (or "grid").
I agree that the imperfections are overfixed to an inhuman level.
Interestingly, there is a "swing" or randomize feature available in some hardware and I guess in every DAW, so if the producers wanted, they would be able to "fix" the notes to a grid, and then add a perfectly calculated random touch to make it sound human.
But yep, for the chorus the offset is definitely deliberate, since it's glaring and repeated every time.
Listening to the lyrics, they're clearly making a heartbeat :)
"tight" big band music "in the pocket" is deliberately and consistently behind the beat.
Peggy Lee and other great singers are more in tune than auto-tune or any perfectly tuned instrument is. Modern tuning, aka equal temperament, is a compromise. Every key is perfectly identically subtly out of tune. A great singer will sing with natural tuning, adjusted on the fly for each key change. Peggy Lee would sing a third slightly flatter than an auto-tune will. Unless an instrument is also playing the third, in which case she would sing in unison with the instrument.
Most of us aren't Peggy Lee or the incredible studio musicians that came out of New Orleans. We need to learn the rules before we can break them.
It’s been interesting to notice that Suno-generated tracks have all kinds of subtle tempo drift, and the tempo maps look very similar to the irregularities shown in the article, at least from a numerical perspective. I have actually spent a lot of time quantizing AI-generated tracks to the grid (to aid chopping/editing), even creating automations that do it for me now when I ingest Suno tracks into Ableton for follow-up production work.
Unknown to me whether this is A) an emergent “imperfection” fundamental to the model and its training dataset, or B) something Suno intentionally injects to “humanize” their output, but knowing that low-effort Suno-generated tracks are currently flooding music distributors, I think it’s ironic that the same factor the author describes as exemplifying the very humanity of recorded music may also be one of the biggest tells that a song was AI-generated, for music being released today.
Support your local bands!
It makes things much more challenging to sync up! You can't just speed up the slower song (or vice versa), because even if the averages match, it will still be all over the place from bar to bar. I now have much better workflows to deal with this, but for that first one, I spent a looong time manually adjusting things to get them to fit.
Working with tracks that were recorded to a click (or digitally gridified, as per the article), are much easier to work with. Match the tempos and everything automatically works. Definitely helps creating my mashups...
But I do agree with the post, in that it loses some indescribable human aspect. I love the push and pull of musicians trying to follow each other rather than concentrating on the click. Or even just pushing and pulling against oneself with things like rubato in classical music, sometimes accelerating, and sometimes slowing, at times almost to a dead stop, so as to build a sense of antecip.... pation.
[1] Link if anyone is interested in checking it out: https://www.youtube.com/watch?v=fP74tMiXBU8&list=PLsj2XVeWwU...
Griddification is trying to take each note and align it to a grid, the two axes being time and pitch. All notes (spoken or sung or struck) start aligned and are tonally compatible. In that sense it is like the grid you see on a piano roll describing music. Basically a de-jittered midi-style reduction of the actual music being player.
Die with a Smile sounds like the exception but I wouldn't discard the idea that those drums were entirely programmed / painted either. It's not like it's difficult to make a flam in a piano roll.
Why not pick from genres that don't predominantly use drum machines as a better comparison?
With that said, a silver lining of AI learning to imitate mainstream producers is that the (big players of the) music industry may finally lose their grip on people's musical taste. Maybe people will start searching for alternative styles, and going to their (minor artists with a unique voice) performances in person.
But also, a perfected, highly produced sound is part of the aesthetic values of modern popular music. So producers who do it now aren’t necessarily wrong, either. They’re just following taste. If people start loving a loosey goosey White Stripes sound again, producers will go back to leaving it in.
I share the belief that this accelerated the rise of Angine de Poitrine, who use microtonal stringed instruments and uncommon time signatures. Both of those things have been around for a while, but never went viral until now.
Does this (2019) article propose the opposite? That hit music usually contains a sort of pleasing surprise, and that over time our hits have to become more and more unpredictable musically?
https://www.independent.co.uk/news/science/pop-music-good-br...
That's now I'm reading it. But I'm a music consumer, not a music creator, so I may be missing some nuance or conflating related ideas.
I've also seen produces complain that the default BPM is 120 so they start there and later realize that it should have been changed. Minor annoyance, but it reveals the importance of defaults.
> So I thought to myself fuckit I’m going to stick to this shitty little computer program, Soundforge. I don’t know any other programs. Once I change something, I can never un-change it. I can only see the waves. So I know when I’m happy with my drums because they look like a nice fishbone. When they look just skeletal as fuck in front of me, and so I know they’ll sound good.
[..]
>> So does that mean your drums are not necessarily in time?
> My drums are definitely not necessarily in time. When I try and do drums that are too regimented, they lose something. But the moment I put drums where I think they sound good, rather than in time, they seem to have that roll, the swing of the jungle and garage tunes I love.
06 https://blackdownsoundboy.blogspot.com/2006/03/soundboy-buri...
The software had a quantization feature which he showed us how to use but he warned us not to use it on everything, because it makes music sound mechanical and less human. I guess that philosophy did not survive in the music industry.
> Something happened to recorded music over the last twenty-five years that most people can hear, and feel, but almost nobody has the language for.
That is 100% Claude
Where in the binary between "you're a real musician who uses human feel/timing to create art" OR "you're a hopeless commercial grifter chasing perfection" do I fall?
[1] I hope all musicians feel this way
I literally do not care that someone can "hit high notes" or that "unlike this other artist, my favorite artist can actually sing." I'm listening to music, not grading music class students.
So yeah I guess a little controversial, haha
Honestly if you’re not into it you’re not into it and that’s fine - but for sure, if you’re going to a concert expecting a pitch-perfect recital of studio-produced tracks, then… you’re kind of just setting yourself up for disappointment. That’s the wrong expectation to go into it with, imo.
Some artists would gridify recording so people who come to their concerts enjoy their plays with "errors" on purpose.
As humans, we don't want to be seen what we spent our entire life to be considered err'ed. I understand the pitcher perfectness in that regard, though I'd still like unpolished "rock/metal" music.
Now I can only enjoy those raw moments in concerts.
Sure. We would also consider it impossible to achieve in reality, and so the art is in the trying, and gridification removes the art. Hell, it removes the artist.
Sure, it wont be achievable to microsecond level but I prop the artists who strive to be perfect (as well as who don't)
---
I sometimes go to concerts and sometimes wishes they sing/play like they recorded. Some really suck live, with some diff gig as they cannot pull studio quality performance.
I want to see what I paid for. They can change their plays but still want some solid work.
Not saying this as some kind of general "anti-AI" argument or anything. I'm not advocating for the Butlerian Jihad. I find AI a very useful tool, and I used it at work pretty much every day. But for art, which in a sense is a celebration of human creativity, something gets lost if we use AI to generate it, doesn't it?
When I hear a drum machine execute with microsecond level precision, I feel like I want to throw my computer through a window. It is instantaneous revulsion. There isn't a ponderous "gee I wonder if this is real/fake" phase of discovery. It's obviously fake and trash. Immediately. Every time.
Orchestral arrangements done the same way might be the only thing that is worse to my ear.
There are situations where having good timing is important, but you cannot put the entire thing on an atomic clock and call it "art".
I don't really hear the same thing as much in doom, black, or more extreme death variants.
It's possible I don't have the ear for it though, so now I wish the article had posted how it made this type of analysis so I could run some of my albums through it myself.
I very simple way is to throw a track into a DAW and see if the drum hits line up perfectly with the "grid". Most DAW's have a grid background so you can see. If it's perfect for the whole song, then it's definitely quantized to death (pun unintended, though does sound quite kvlt).
https://www.youtube.com/watch?v=7ZWcbQys6sw
Are electronica and hip hop worse for this? I suppose that's up for debate - but we now have multiple full generations of music listeners and critics who obviously, passionately love those genres, and they've been growing for decades. This is easy to see not just via charts but online discourse and the zeitgeist.
Do those genres 'groove' worse and feel less human? This is more subjective, and more debatable, but I would argue that hip hop has redefined what 'groove' can even mean, with its very off-the-grid, loose feel despite the technically very precise machine looping the beats (because producers nudge each beat off the grid, usually in very intentional ways, or play the beat in live then adjust it). And it's hard to argue that electronic dance music in its many forms hasn't made people move and connect with the music's rhythm in powerful ways, despite almost complete quantization in most dimensions.
I think purely looking at tempo stddev and not much else papers over these aspects a little much.
More specifically, they define gridification as: recording multiple takes, nudging every single note onto the grid for each one, then selecting the best take section by section (however small those sections are) so that every single note is absolutely perfectly on time, in tune etc, loses... something - it removes some of the humanity. The author also explicitly cites many other kinds of recording and editing as NOT gridification - e.g. playing to a click, using a drum machine.
One could say modern music is not "musical performances" then. A sequencer or drum maching playing isn't "a performance", it's a programmed loop.
>Do those genres 'groove' worse and feel less human?
Yes.
When I play the drum machine, I program it live. I never play the same thing twice. It IS a performance. It applies for both vintage drum machines (such as E-mu drumulator or tr808) and new gen ones or grooveboxes (such as Electribes or Octatracks).
That's a timing lock that's tighter than the usual sort in recent generations. There are DAWs (Renoise) that pride themselves on being able to send MIDI with less slop and irregularity (nice trick since MIDI is a serial bus, but you'd think it'd be more common to get right). There are whole genres (see: psytrance) that do very well in trackers which can fire sample-accurate notes simultaneously without resorting to MIDI.
Even though it's of the machine, the groove of an 808/303 pair synced by DINsync is a different beast from 'snap to grid and autotune'. There's subtleties to everything.
The drum machine is not all of the performance. In pop you'll have at the very least a singer. Even if all of the instrumental is machine-made, having a live singer definitely makes it a "performance".
https://musicalform.substack.com/p/why-dance-music-is-not-gr...
They're comparing the tempo deviation not just in the drums but in literally everything.
Billie Eilish may be singing over a drum machine rather than a live drummer but that wouldn't make _her_ perfectly on time.
I hope this was written as tongue in cheek, otherwise wow, you are taking yourself too seriously.
It is utterly baffling to me that this is a thing someone felt needed expressing. Surely, intuitively knowing this is part of being human?
How else do we explain the blues, love songs... anything?
What I want to know is what the author thinks of drum machines, with their mechanical perfection. Roland's very first product, the TR-77 drum machine, was released in 1972, in time for his 1975 comparison. They have been popular ever since.
Isn't every song recorded as every word singularly at a certain pitch? And timing done completely in the DAW instead of while recording.
Barely was singing when a friend of mine did ``professional'' recording, and had to know each pitch of a song to perfection, since no overlap was ever sung.
Now was there autotune? dunno. its not like some people who've obviously been tuned. There are some people who are really that good. Michael bauble is one of them
What? No? What are you saying?