Turtlerock Forum

Here's one dear to all our hearts - interesting article

"ul 26th 2012, 19:28 by L.R.
THE kids these days play their music too loud and it all sounds the same. Old fogies familiar with such sentiments will be happy to hear that maths bears them out. An analysis published in Scientific Reports by Joan Serrà of the Artificial Intelligence Research Institute in Barcelona and his colleagues has found that music has indeed become both more homogeneous and louder over the decades.
Dr Serrà began with the basic premise that music, like language, can evolve over time, often pulled in different directions by opposing forces. Popular music especially has always prized a degree of conformity—witness the enduring popularity of cover songs and remixes—while at the same time being obsessed with the new. To untangle these factors, Dr Serrà's team sifted through the Million Song Dataset, run jointly by Columbia University, in New York, and the Echo Nest, an American company, which contains beat-by-beat data on a million Western songs from a variety of popular genres. The researchers focussed on the primary musical qualities of pitch, timbre and loudness, which were available for nearly 0.5m songs released from 1955 to 2010.
They found that music today relies on the same chords as music from the 1950s. Nearly all melodies are composed of ten most popular chords. They follow a similar pattern to written texts, where the most common word occurs roughly twice as often as the second most common, three times as often as the third most common, and so on, a linguistic regularity known as Zipf's law. What has changed is how the chords are spliced into melodies. In the 1950s many of the less common chords would chime close to one another in the melodic progression. More recently, they have tended to be separated by the more pedestrian chords, leading to a loss of some of the more unusual transitions. Timbre, lent by instrument types and recording techniques, similarly shows signs of narrowing, after peaking in the mid-60s, a phenomenon Dr Serrà attributes to experimentation with electric-guitar sounds by Jimi Hendrix and the like.
What music lost in variety, it has gained in volume. Songs today are on average 9 decibels louder than half a century ago, confirming what industry types have long suspected: that record labels engage in a "loudness race" in order to catch radio listeners’ attention. Since digital audio formats max out at a certain decibel level, as the average loudness inches towards that ceiling, songs will lose dynamic range, becoming ever more uniform.
This homogeneity is not just jarring to melomaniacs. It might confuse the popular algorithms for identifying and recommending tracks, like those used by Spotify and other music services. Many of these rely on timbre measurements to sort songs into genres, for instance. Some musicians are bound to respond by confounding expectations with new sounds. Whether audiences wish to be confounded remains moot."

page link:
http://www.economist.com/blogs/babbage/ ... ence-music

article link:
http://www.nature.com/srep/2012/120726/ ... 00521.html

Sorry, I don't buy academic's claims on chord structure and melody.
McCartney has a funny (sad) story about a disagreement he had with an American academic running a Beatles degree course.
He was teaching a particular interpretation of some McCartney lyric. Paul contacted him and said "sorry, you're wrong", but the academic begged to differ. To which an exasperated Macca asked "who was there, you or me?"

Really? You've heard the shit they play on the radio right?

ChrisW wrote:Sorry, I don't buy academic's claims on chord structure and melody.
McCartney has a funny (sad) story about a disagreement he had with an American academic running a Beatles degree course.
He was teaching a particular interpretation of some McCartney lyric. Paul contacted him and said "sorry, you're wrong", but the academic begged to differ. To which an exasperated Macca asked "who was there, you or me?"

But many people see stuff in art that was not intended yet still powerful to the viewer. All good art has many facets if interpretation and I'm more than happy when someone gives me a new and complentary insight into my own work. How often do we really know what we are expressing anyway? I often think the artist is the last person you should ask about meaning because they made the art so they didn't have to explain...

Amazing.
If the artist doesn't know what they meant, why would you think a US academic knows better?

Kurt wrote:Really? You've heard the shit they play on the radio right?

I don't listen to music radio no.
So the entire repertoire of post 2000 popular music is represented by music radio? I think not.
Personally, I listen to a lot of electronic music. Which is fairly popular and has absolutely zero in common with the chord structure and melody of the 1950's.

Yes, I agree, popular music is maxed to the max in terms of mastering volume, but that has more to do with not sounding weak in comparison to other releases, and virtually nothing to do with a lack of compositional imagination IMHO.
I just don't think an academic can collect chord structures and melody lines from popular songs, add them up and come to any simplistic conclusions.
I'm sure you could add up the verbs in Dickens and New Idea and say New Idea is more articulate.

"Fairly popular" is not pop music. I listen to metal, it's "Fairly popular" too.

I only listen to the radio at work 'coz the boss has it on JJJ

ChrisW wrote:

Kurt wrote:Really? You've heard the shit they play on the radio right?

I don't listen to music radio no.
So the entire repertoire of post 2000 popular music is represented by music radio? I think not.
Personally, I listen to a lot of electronic music. Which is fairly popular and has absolutely zero in common with the chord structure and melody of the 1950's.

Yes, I agree, popular music is maxed to the max in terms of mastering volume, but that has more to do with not sounding weak in comparison to other releases, and virtually nothing to do with a lack of compositional imagination IMHO.
I just don't think an academic can collect chord structures and melody lines from popular songs, add them up and come to any simplistic conclusions.
I'm sure you could add up the verbs in Dickens and New Idea and say New Idea is more articulate.

I was impressed by the number of eminent Uni's involved and the sample set of a million pop songs (i.e. quantifiable research).

It's not just some wally using smoke and mirrors based on a sample set of 2 or 3 local cletus's....

Drumstruck wrote:Nearly all melodies are composed of ten most popular chords.

Whilst I'm not sure I like the language of melodies being composed of chords (reads like non-musicians talking about music, which I guess it is), I'm always struck by the seeming complexity of pop music from the 20's to the 40's compared to today.
Having a happy pop tune move through 4 keys thirds apart in 48 bars, and having it sound seamless is pretty cool. I'm always happy to hear stuff like that happening in music, and it is sad to me that it doesn't seem to be nearly as common today. I still get surprised every now and then though!

Not sure the above quote is fairly representing the study - they took 1,000,000 songs identified through use of the Echo Nest API (this is the thing that powers automatic recognition and recommendations for MOG, Rd.io, E-Music and others and has something like 22,000,000+ tracks analysed in their database). So 'popular' in this instance relates to how the data-set was put together (beginning with most 'familiar' artists to the Echo Nest's API – i.e. the ones most played across several platforms the API is used on).

The key findings weren't 'music is getting dumber' but actually that when harmony, timbre and 'volume' are analysed by year-of-release the data-set reveals that there was little variation in harmonic material between the songs released in the 50s and today, but that Timbre and Volume had changed. The study also found that, taken as a whole, the harmonic ingredients of tracks analysed from the last decade or so were more homogenous than the tracks analysed from the 50s. Their suggestion that a track from the 50s, re-recorded with modern instrumenation / production style and a simplified harmonic palette might be perceived as 'novel' would appear to gel with my experience of the market for popular music.

I reckon if you repeated the same analysis and included a couple of centuries of classical repertoire you might see a more dramatic 'simplification' of harmonic content as you hit the 20th century – but we already knew that. Western harmony has been explored pretty fully over the last couple of centuries and so the innovation that has occured in music over the more recent period has tended towards new timbres / rhythms; playing styles (often improvised) and the adoption of musical ideas from other cultures (I'm grossly generalising of course).

What's interesting is what they didn't analyse – rhythm, harmonic structure (they just counted instances of chords) for example – and the misplaced assumption that 'evolution' in music over the last 50 years might be found in harmonic content. I'd guess this is because the automatic listening / analysis algorhythms used by the Echo Nest might be tailored towards spectral content (timbre / harmony) and dynamics as more reliable ways to cluster music be genre / same-ness than rhythm.

The data-set is available for free so I think there might be more interesting stuff to come out of it in the future.

It's just hard to get academic and scientific about music.
One song with two chords can have much more impact than another with the same two chords. Why?
Probably because the whole thing is a package - the sound design and instrumentation, the particular performance, the taste of the listener.
It's pretty obvious sound and dynamics have changed over the last 50 years. I don't need research to tell me that.

It's hard to be scientific about the impact of music. Music itself is very analysable, as this study shows.

ChrisW wrote:It's just hard to get academic and scientific about music.
One song with two chords can have much more impact than another with the same two chords. Why?
Probably because the whole thing is a package - the sound design and instrumentation, the particular performance, the taste of the listener.
It's pretty obvious sound and dynamics have changed over the last 50 years. I don't need research to tell me that.

ChrisW wrote:It's pretty obvious sound and dynamics have changed over the last 50 years. I don't need research to tell me that.

Yeah the findings are a bit 'well DUH'.

The process and application is pretty interesting though.

And then there's this:
http://www.soundonsound.com/sos/sep11/articles/loudness.htm

The whole chord thing is really irrelevant.
Good writers draw inspiration from anything - including these often used chords...bird sounds..traffic / whatever gets the juju flowing.

So much talk about music being too loud....BUT everyone blames it on mastering?

No-one mentions the use of individual compression - most (inexperienced) engineers and mixers are compressing when they record - then compressing most instruments when they mix + mix bus compression (and then in mastering it happens again).
All of these sheep buying SW compressors and putting them on anything and everything - FOOLS!
I encourage all engineers to try attempting a mix with ONLY 1 or 2 compressors - for eg only bass and snare.
- or NO compression.
If you like the sound that HW compressors impart then run through them without actually compressing.
You now end up with a mix that CAN be compressed in Mastering, still retain dynamics, and be close enough to most releases volume wise.

Try recording a good singer,tabla player,violinist, cellist,pianist with too much compression and they cannot push sections - intimacy is impossible - they dont get feeling from the performance - mission failed.

It makes a massive difference using very little/or no compression - and because so many more subtleties are left in the music, you have to work less to get a great sound, and mixing is then about dynamics/feeling/emotion/vibe (all the good stuff )

The problem commonly arises like this:
You overly compress the kick and snr and toms and bass when recording or mixing - then add a dynamic guitar - what happens?
Because the other individual sounds are compressed - now the engineer feels that the guitar is too dynamic compared with the other sounds - so this is now compressed. then the vocals get the same treatment and every other sound gets squashed to fit in this non dynamic picture.....
Then you find you need to automate dynamics to compensate for not having any...

Ive been making albums since I was 16 - Ive made more loud albums than I can remember, but I do know, when I stopped using a lot of compression, my mixes went to the next level.
Also when I stopped mixing loud (ears under compression is even worse)

In the past I used to see a waveform on the screen and know that it needed compression, and blindly put one on. But when I used tape, and could not see the information, I had to FEEL when i needed compression - and rarely needed to use them much.
Once I decided to stop putting a compressor on anything that had a bit of movement - It made me work with the music - not try to control it.
This came about because I was liking my mixes more when I strictly used tape - and did not see any waveforms.
Every decision was based on feeling the music.
The dynamics were gold!

Even 100% electronic music - it still needs to express emotion.
Music has dynamics - language & expression have dynamics, feeling is communicated better with dynamics.

Why would anyone want to lose this golden information?
Listeners have emotions - why attack them with a cricket bat?

A tip for the newer guys coming up -AVOID USING COMPRESSORS AND USE YOUR TALENT TO MOVE THE LISTENER
The best compressor is a fader .

(now everyone else stop blaming the mastering guys for music being too loud)

my weekly rant.....

It is clearly not irrelevant to a study of such things. If a study want to examine the complexity of material it obviously needs to analyse structure of said material.

The Tasmanian wrote:The whole chord thing is really irrelevant.

I don't really see (or agree with) the connection between complexity and the volume wars.
For a start, it's a mix up of complexity and innovation.
If we are using the same chords as the 1950's, it's because of a failure to innovate, right?

Anyway, the loudness thing is purely about sounding full and 'in your face' compared to similar music.
I get that it's destructive and a negative to craft and artistry, but someone just has to draw a line in the sand and say they wont go there. Trouble is, with dwindling income from record sales, it's a very brave pop artist that wants to sound less punchy on the radio and tv.
Rihanna just had a number one album in the UK with less than 10,000 sales (CD and digital)!!!!

I don't think the two are related either but they are both readily apparent.We're still using the same chords as we were in the 1700s aren't we? But we're using them differently.

The Rihanna example perhaps proves that loud is not the way to sell records? wasn't hers very loud?

ChrisW wrote:I don't really see (or agree with) the connection between complexity and the volume wars.
For a start, it's a mix up of complexity and innovation.
If we are using the same chords as the 1950's, it's because of a failure to innovate, right?

Anyway, the loudness thing is purely about sounding full and 'in your face' compared to similar music.
I get that it's destructive and a negative to craft and artistry, but someone just has to draw a line in the sand and say they wont go there. Trouble is, with dwindling income from record sales, it's a very brave pop artist that wants to sound less punchy on the radio and tv.
Rihanna just had a number one album in the UK with less than 10,000 sales (CD and digital)!!!!

I'm glad I posted this.... it's good to see some debate going on amongst you eminent guys (and I'm not being sarcastic!).

I like the thoughts that have delta'd from this stream - seems there's the emotive side of music that we should leave in the realm of opinion, and the analytical side that is as valuable as statistics are.

One thing that surpised me was the indication of 9dB loudness difference between the 50s and now - I expected considerably more. And I wonder what the new averaged loudness standards will do to this figure if we look back in another 50 years. The audio fraternity seems to be succeeding in the push to reclaim dynamic range, or at least in educating the public that it may be valuable....

Pop music is the (limited) topic but I didn't find a definition of "pop" in the articles, so I'm assuming they mean top 40 in various genres....

In the end the ones I blame are the 3 chord frauds (now I'm in for it......)

I was messing around in my wayback machine the other night .. I ended up on the outskirts of ancient rome which looked a little too much like punchbowl to me .

What I was trying to do was hook up with Nero so I could listen to the sound of his fiddle while we watched rome burning together .. he told me actually rome burned 400 years earlier and it was a just a bit of artistic licence by a local pop musician of the day .. Imagine people using pop music as a guide to history he chuffed !

I told him in my time people where lamenting on the loss of dynamic range and harmonic structure in pop music over the last sixty years .. he smiled and said well yes sixty years sounds like long time doesnt it ! .

I told him the US empire was drawing to a close after 70 years .. he smiled and said well 70 years sounds like a long time doesnt it .. !
'
I tried to explain to him the internet and spotify ..

he asked if we were still flushing the toilets the same way his lot invented

To touch on the chords stuff- yes, sometimes they are irrelevant to writing a good song, but using all those other chords just gives you access to a different world of colour, and I can't imagine the music of Joni Mitchell, The Beatles or Jeff Tweedy being nearly as expressive without access to all those sounds, not to mention the stuff outside pop music.
Rhythmically there is a shitload of cool stuff happening, and melodically also, but I like harmony too!

I agree Alistair - like you say - its all colour.
There are also so many great chords and combo's that we could go for thousands of years and still make great new music - so to me its not a limitation issue.

As far as rhythm - thankfully we are finally moving away from 20 years of 4 on the floor.
I am SO over working to a grid and click mentality - I have (for years) banned them in 99% of every project I work on.
The justification of recording to a click with every track because its easy and quick for engineer to edit between takes is a lame excuse to destroy the natural interplay between musicians.
This has more of an detrimental impact than choice between many great chord combo's.

My hope now is that someone creates a killer SW recording platform with no grid - no clicks - no waveforms.
Just tracks meters and plug-ins.
Music.

The Tasmanian wrote:In the past I used to see a waveform on the screen and know that it needed compression, and blindly put one on. But when I used tape, and could not see the information, I had to FEEL when i needed compression - and rarely needed to use them much.
Once I decided to stop putting a compressor on anything that had a bit of movement - It made me work with the music - not try to control it.
This came about because I was liking my mixes more when I strictly used tape - and did not see any waveforms.
Every decision was based on feeling the music.
The dynamics were gold!

Mixing with our eyes is one of the worst things we can do. I'm a huge advocate of turning the screen off whenever possible... but that is only a very small, temporary solution.

And I love your idea Chris of the ultimate non-visual DAW - in some ways it would be a lot like ADAT or 3324, which were in so many ways enormously great and successful systems (media issues notwithstanding). Could someone invent that please?

Been thinking about the mixing without seeing the waveform. what about just using the mix window and close the edit window in your daw. Use transport controls to navigate. (this is in reaper, been a while since ive used tools. maybe it possible)

my biggest problem is mixing too loud. its such a temptation to just crank it up and rock out. Maybe i should get some of those aurotones or the bringer copies to cure me of that.

Yes Dwane - its great when you crank it - (I still love it !).
But everytime I have monitored loud for extended periods - down the line on upon listening at a normal volume, I hear problem(s). And in this situation - when I listened loud for long periods, confusion set in when attempting to listen at a normal volume... my ears had lost their sensitivity, and it was even harder to make judgements at a low volume.

(edit)....Ive made some BIG mistakes mixing loud in the past.

Now I have this inbuilt self check mode where I allow myself to crank it for only very brief periods - and only to check really low frequencies / noise / clicks - poss problems - NOT for enjoyment, and then back to granny volume.

And lots of breaks, lots of walking into the next room and listening to it from another perspective.

I used to love to crank it up too, but I learnt to work at a much more conservative level when I was just doing mastering, and I've tried to keep that philosphy when i mix now. Having an SPL meter and being aware of where the level is at is a good move, then after a while you get used to it and will know when you start getting too loud (85 dB SPL was always a good reference for me).

Now that I don't really do any mastering, I'm a bit less strict about it... but I always start a session/day off quiet, and keep a bit in reserve for later in the day, when I will turn it up a bit. If you crank things too early, then you're more or less done for the day.

oh and +100 on the getting up and walking around, listening from the hallway, taking breaks etc... all essential stuff.

On another "loudness" note, I am horrified at the number of people that spend every spare moment with earbuds in and loud, over-compressed brittle MP3's pounding their fragile ears. I know some people in their 20's that have them in almost constantly.

I must be in the minority.

I just cant mix loud. I mix quieter than most anybody I have ever met.

Only near the end, do I run the mix up loud-ish to check levels, during that last 5-10% of mixing time, you know, loud, soft, checking kick bass snare vocal and lead instruments level.

I often find that I have to drop the main things in the mix a little when I do this, but then when I do a "clean ears" check (after a few hours break) its always for the better.

I have my system calibrated for 85 dB, but thats what I call loud. I am mixing below that most of the time.

I'm no true engineer, but by necessity I produce my own music.
I work at a lowish, medium level. Never mix above medium (domestic listening) levels.
A few times a day I will listen to one song at a louder level, but that's it.
I have mild tinnitus from years of drumming abuse. I agree with Steve about the iPod earbuds.
I bought myself a pair of (open) earphones for my iPod and iPhone. Those in ear plugs are a disaster (IMHO).

+1 on reducing the visual reference - I use the PT edit screen as a visual cue when I'm recording, and something like Wavelab for one-pass type analysis, but not for mixing etc - for that my ears and console meters are my preference. I like the title audio engineer, not audio-visual engineer. Actually, all I use the DAW for is to replace the tape multitrack - other than that they can keep their computers (hate computers! hate computers! hate computers! hate computers!!!)

Low volume mixing here too - rarely above 85dB on the SPL meter with a reference listen at a screaming level....

A side effect of low monitoring is that I'm now so picky about "noise" - what's that buzzing - doh, the other snare drum? what's that squeak - doh, the dog door? where is that whistle coming from? how did that bird tweet get in there? what's that rumble .... oh, it's the ocean.... have to live with that one....