The Monkees: Most Complex Music Ever?
I am not a musician or a composer. I can’t personally hear or judge the complexity of different songs and records. However I recently noticed that maybe I don’t have to. I’ve been reencoding most of my CD library using Lame. Lame uses variable bit rate encoding. I’m sure audiophiles will correct this simplistic explanation, but in brief Lame samples different pieces of a song with more or less frequency as necessary to match the music. A pure tone could probably be reproduced using very limited sampling, whereas a dissonant cacophony of white noise with no predictability would require a very high sampling rate. Lame also takes into count the nature of the human ear. Frequencies humans can’t hear can be thrown away, and frequencies we hear preferentially need to be sampled more frequently.
In compression terms the more random something is the less compressible it is. Purely random data cannot be compressed at all; and this is what allows me to judge the complexity of any given piece of music. The higher a bit rate Lame has to use to compress a piece of music the more complex and less repetitive it is. Most of my songs compress at about 160-180 kbps; but there’s a very wide variation. Tom Petty’s Into the Great Wide Open is the simplest piece in my collection. It sounds good at only 98 kbps.
The most complex piece in my collection: Pleasant Valley Sunday by the Monkees at 247 kbps. I always knew there was reason I liked that song. :-)
January 26th, 2006 at 6:02 PM
I read your blog through its RSS feed. A Yahoo banner has suddenly appeared at the bottom of every post, and each one tries to set a cookie for a slightly different URL (circumventing Firefox’s cookie-avoidance support, and requiring a lot of clicking on cookie-rejection dialogs).
January 26th, 2006 at 7:35 PM
Do you have any Phil Spector recordings? I wonder if his “wall of sound” would produce the same sort of result. Pleasant Valley Sunday’s chorus has a full sound like that.
“Be My Baby” by the Ronettes would be a good example.
January 27th, 2006 at 10:25 AM
I have that greatest hits package, and there’s some junky reunion stuff toward the end, but overall, it’s great. In the sixties, the earnest people who complained that the Monkees didn’t play their own instruments and then went off to listen to their Simon and Garfunkel records didn’t realize that many of the same people were playing on the Simon and Garfunkel records (although the latter used more NY-based session people than the former, who used LA musicians).
The Monkees wrote and played very little on their best stuff, but the people who did were the best in the business, and what makes the greatest hits package shown above so fascinating is that it does list the people who played–people like Carol Kaye, Hal Blaine, Jim Gordon, Louie Shelton, Larry Taylor… with songwriters like Carol King and Neil Diamond (the sixties version, not the over-dramatic balladeer of later years), and yes, with Tork, Nesmith, Dolenz and Jones, who were good at what they did, there was a talented team that put together some music that has stood the test of time much better than a lot of Simon and Garfunkel (speaking of overdramatic ballads). As with any kind of music, there’s good pop and bad pop, and there’s a lot of great pop in a Monkees singles collection.
January 27th, 2006 at 4:49 PM
This MP3 compression thing is something that crossed my mind recently. I don’t think it has to do with “frequencies we don’t hear”. I think the psycho-acoustic model is more about “details” we don’t hear. I want to say: It’s not a technical thing as a low-pass filter. It’s more the art of finding out what texture (for lack of a better word) of a song’s soundscape can be omitted. I don’t know a perfect example, but I think of it as, “hey, there’s this snare in the mix, it’s very muted, and most of the time it goes “unnoticed” (there’s the non-technicality), so I’ll leave it out.”
So from your example it would transfer to: Hey, the Monkees make music where you hear every detail, so lame can’t drop nothing. And “Into the Great Wide Open” must have loads of “background” stuff lame can drop without tampering with the “gestalt” of the song.
Another aspect: Audio CDs sample with 44,1 kHz with no regard to the actual sound. Many people claim that details of music get lost “between” two samples, so they prefer the analog records (12″). But when we use lame or iTunes we often use the CD sound as the source. So I would really like to know if mp3 encoders would do an even better job on source material of a higher sampling rate. Would they find more details to drop or other details to keep?
Sampling to me is a snapshot, like the single frames of a movie. Sampling just sees frame after frame, without recognizing the story of the motion picture.
MP3 encoding then sees the movie’s story unfold and acts like a cutter, but not on a frame-by-frame basis. It rather cuts details from the whole scene, like “removing” the bricks-texture of a distant wall the audience wouldn’t notice, because the thing they’re focused on is Spiderman flying by.
March 10th, 2006 at 11:48 PM
Actually, it doesn’t surprise me that “Pleasant Valley Sunday” was the most complex piece in your collection. The song was written by Carole King and Gerry Goffin, two of the best songwriters at the time. BTW, it was one of the songs produced after Don Kirschner had been fired from the project, so The Monkees (Micky Dolenz on vocals, Mike Nesmith on lead guitar, Peter Tork on piano, Davy Jones on tambourine) did play the song, although they were assisted by Chip Douglas (of “Wild Thing” fame) on bass, Bill Chadwick on rhythm guitar, and Eddie Hoh on drums.
July 3rd, 2006 at 5:20 AM
Why must this music be complex in any way :S
Its just a simple pop music…………………………………..
Listen to More artist and you will know this…………………………….
I prefer frank zappa over the monkeys if i want complex…………………
ARgghisshhh
August 5th, 2006 at 6:16 PM
I see your point. Music is simply notes played in a sequnce of time, repetitively. The more of these you have in any composition called a song(pleasant valley sunday), the more kilobytes per second(kbs) is required to encode and eventually “sample” the music. In your collection, you have pleasant valley sunday or whatever music of the bygone era which had complex music.(required more skill in actual synthesis of MUSIC) But if you want to listen to a modern version skilled music, you can simply search for some techno/trance and similar music. The encoding rate of some of these are in the 300kbs’ range, also Indian carnatic and other music from the east reach these levels. Europe’s greatest composers are also at the top of synthesizing music, yet with the modernization of technology a new brand of music is emerging.
January 9th, 2008 at 11:32 AM
A few things that may add to the complexity:
The quick rolls on the drumsThe quick, pulsating basslineThe fact that Mike N doubled his guitar part (yes, that is him playing lead guitar on that song!)The Monkees’ heavy use of tambourine and maracas, which add complex, random detailThe very hard-to-hear keyboard parts
It’s a great track. Many criticize them for being a “fake band”, but they actually did more than most people realize, and they put out better music than most modern “real bands”.
September 24th, 2008 at 12:07 AM
What’s really funny about this post is that tom petty is famous for saying that the monkees were more of an influence on him than the beatles.
Also,
Into the great wide open sounds very much Jeff Lynne type production.. everything is very “clean” and “compressed” sounding.. go listen to Paul Mccartney’s “flaming pie” album.. same sound.. super clean.. also the Beatles “free as a bird”… totally Jeff Lynne.
The monkees records (I have all of them) are very pop sounding records of their time… but they have alot of dynamic range in them. Alot of that has to do with the way those records were mastered. Mastering a record back in the day meant manually EQ’ing the peaks in the sound so that the needle wouldn’t poke through the vinyl while making the grooves of the “master” record. That’s all it was. So you got alot of dynamic range in those old records that is very hard to find in modern digitally mastered music. Mastering now is just over compressing and post mixing to make everything LOUD.
Sometimes if you listen to early compact discs of old records.. they aren’t re-mastered and they sound really cool.. but you have to turn the volume up on the stereo ALOT to get the sound. anyhoo..
here is a good article that explains it:
http://www.moozeek.de/mirrors/articles/over_the_limit.htm
November 18th, 2008 at 12:24 AM
This doesn’t even come close to measuring the complexity of the music, the closest thing it could account for is dynamic range and the frequency alterations. If a record is mixed properly the engineers will have carefully placed each instrument in a certain dynamic and frequency range. If the engineer had placed say multiple instruments/sounds/whatever in the same frequency or dynamic range it would be harder to analyze. Also analog signals before the digital era were continuous in amplitude and are much more difficult to reproduce as they must be sampled whereas a digital signal will only contain binary. Also, if your library contains anything ripped from a CD or that has been put through an analog to digital converter at any point (this would account for the newer records) it will have passed through an anti-aliasing filter to filter out frequencies above the human hearing range. So basically any record you have in your library that was recorded or manufactured digitally will be a lot easier to read by your program because it simply does not contain any frequencies that exceed the human hearing spectrum.
Either way, come on you can’t expect to try to analyze art using any computer program. This program you have is designed to alter frequencies and amplitudes, it has no idea what the music. I suggest you develop your musical taste so you don’t ever suggest again that the monkees have the most complex music. (Take a listen to Sargeant Pepper’s lonely hearts club band before you post something like this for people to read)
June 27th, 2009 at 12:00 PM
I’m going off on a tangent here, and this has nothing to do with The Monkees. Instead my topic deals with comparisons between analog music and digital music and the relationship to human hearing. First lets compare analog and digital by way of analogy (pun intended!). If analog music were a solid wall, digital music would be a picket fence. Each picket would represent a sample of the analog wall. The idea from the digital perspective is that the narrower and closer together the pickets are, the more accurate the representation of the analog wall will be. When the pickets are so narrow and so close together that gaps cannot be felt when running one of your digits (again pun intended!) over the surface, one may not be able to distinguish between the digital pickets and the analog wall.
Next is the digital practice of eliminating frequencies that we cannot hear. For CDs this upper cut-off frequency is 22.05 kilohertz or half the sampling rate. Originally, this was done to optimize the file size. Why waste data storage space with data that people can’t hear, right? Eliminating this data allows for a longer playing time. To be fair, analog music often eliminated this range of frequencies too, but in a gradual roll-off through a bypass filter or by reaching the technological limits of the recording device (analog tape).
However, from a listener’s perspective this frequency removal can have noticeable consequences and the listening enjoyment may suffer. Just because the frequency itself cannot be heard doesn’t mean that its presence has no musical value. Music is sound and sound travels in waves. Mixing two tones results in “sum tones” and “difference tones.” If two tones are played, say 800 Hz and 600 Hz, a difference tone of 200 Hz–that is NOT played–will manifest through the interaction of the two “played” tones. This is how barbershop quartets can appear to sound like a quintet or sextet. So the frequencies above the range of human hearing have a significant effect on the audible music below. This may also be why some music listeners prefer analog music to digital music or digital music with higher sampling rates to digital music with lower sampling rates. “Sum tones” also occur so the two tones above would also create a sum tone of 1400 Hz. “Sum tones” may be more noticeable at lower and mid-frequencies than at higher frequencies.
Lastly, humans do not hear in a linear fashion but in a logarithmic fashion (geometric as opposed to arithmetic). An A note is 440 Hz. 880 Hz is also an “A” note. But this determination isn’t achieved by adding 440 + 440. Its achieved by multiplying 440 x 2. The next “A” note one octave up would be at 1760 Hz (880 x 2). “A” notes in octave sequence would be 55 Hz, 110 Hz, 220 Hz, 440 Hz, 880 Hz, etc. So beware of how “complexity” is determined. Higher frequencies alone may be inaccurately considered to be more complex just by their shear numbers!