Bad and Wrong Music Lessons, Part 5: Sound Bytes

By Shamus Posted Wednesday Dec 10, 2014

Filed under: Music 38 comments

It’s been a while since I did a music post, hasn’t it? Well, nothing good lasts forever so it’s time for another one.

In his famous deconstruction of the techno genre, cultural critic and doctor of musicology expert Strong Bad observed basically everything you need to know about techno music, and electronic music in general: You need the beat, the lead music part, a high part, and some kind of movie quote / sound sample. As someone who listens to a lot of electronic music I can say this is funny because it’s true.

Once again I need to remind you that all of this is either badly wrong, or an over-simplification of something that real musicians have known for centuries. I’m literally just posting observations as I make them, like someone learning to read exclaiming, “Have you noticed that there are two different shapes for each letter!?” It’s sort of true yet incomplete and sort of wrong but not totally. I have no idea how real musicians can stand to read these.

Sooner or later I’m going to need to stop this musical finger-painting and share something useful. But not today!

Music seems to tickle the pattern-recognition parts of our brains. There are only so many different types of patterns you can perceive at once. If I have three pianos all playing different sequences of notes, but they’re all playing staccato quarter notes in the same octave, then it’s probably not going to sound like three different things layered together. It will all blur into one big mishmash of notes.

But if one piano plays long low notes, another plays rapid high notes, and the other one runs up and down the scales like a crazy person, then you’ll be able to hear each of them distinctly.

There’s a sort of soft cap on our aural bandwidth:

  1. Tempo of sounds. Rapid notes are distinct from slow ones.
  2. Duration of notes. Long notes are distinct from brief ones.
  3. Pitch. High notes are distinct from low ones.
  4. Waveform“. This one is complicated and could probably be split into some sub-categories. A pure frequency (like a sine wave) is distinct from a burst of white noise. A sound that is constant in volume is distinct from one that falls off sharply, which is distinct from one that rises and falls gently on the ends.

The warden threw a party in the county jail / The prison band was there and they began to wail / The band was jumpin and the joint began to swing / You shoulda heard them knocked-out jailbirds sing!

In the real world, you’re limited by physical instruments. This song has four parts because there are four people in our band. It doesn’t have a section where three guitars play at the same time, because we don’t have three guitarists. Our music is shaped by the people who perform it and the instruments they play.

But in electronic music we do away with all of those limits. You want half the song to be all keyboards and the other half all guitars? You can do that. In the real world that would be prohibitively impractical, but when you’re mashing notes together on your computer box it’s not any different from creating any other arrangement of instruments.

The problem with unrestrained composition is that limits usually make for more interesting art. I (usually) enjoy a nice rhythmic rhyming verse more than rambling free-verse and I (usually) like paintings of specific things more than random abstract patterns. But electronic music is kind of like composing with no limitsOther than the fact that all your instruments end up sounding “canned” and don’t have the richness a physical music-making apparatus can provide. I don’t have an ear for it myself, but for some people this is a huge loss.. And so – freed of the normal physical limitations – a lot of electronic music feels like an attempt to “max out” our sonic bandwidth. You can just keep cramming more parts in there until you’re covering every part of the tempo / duration / pitch / waveform spectrum and the average ear can’t usefully detect any more patterns. It’s an attempt to maximize the “interesting-ness” of the music by giving the brain lots and lots of sound to process.

Of course, making a song with “lots” of sound isn’t any better than making a book with “lots” of words. We’re looking for mental stimuli, not filler.

But as a composerIs it okay if I call myself a composer? I won’t tell Mozart if you don’t. I find this leaves me rudderless. How do I know when a song is done? With no goal or constraints, there’s nothing to do but add more crap until the song is full of notes. It’s like abstract painting as a kid. “Well, I’ve used all the colors, and I’ve totally covered the canvas, so I guess I’m done?”

My most well-received songGoing by direct feedback, not going by social-media “thumbs-up” ratings on Soundcloud. so far is Best Part of the Storm. I wrote it specifically to be something that made some kind of musical sense. It could, in theory, be played by real musicians. It doesn’t feature a piano with ten octaves, a drum-kick that sounds like a gunshot, a saxophone playing a long series of 64th notes, and eleven bass guitars. And those constraints helped me make something interesting instead of just aimlessly filling up the soundscape with no greater goal.

Which brings us to movie quotes. Thinking about the above, I decided to use a quote from the movie “Amadeus” where Emperor Joseph II tells Mozart that his composition has “Too many notes”.

I still filled up the soundscape, but this time it felt like I was doing it on purpose? Or something? I don’t know. It was really a lot more fun to have some kind of idea to build a song around besides, “Fill time until you run out of ideas.” The music was about something.

Clearly this was an idea that needed to be explored further. So I took the popular AWESOME BUTTON quote and made a song around that. Again, the purpose and direction helped me shape the song. OF COURSE the song would need to be high-tempo and OF COURSE I’d need to follow the quote with the sound of an explosion. The song practically wrote itself.

So I guess that’s why so many electronic songs have old movie quotes of pop-culture preferences in them. Because otherwise it’s just too abstract.

Other things of interest:

“Too Many Notes” really was designed to have too many notes. Here is what the project looks like:

Which notes should I remove?

I’m sure you can’t even hear half the crap playing at any given moment. It just gets lost. Compare that with AWESOME BUTTON:

Which notes should I remove?

A lot simpler, obviously.

Too many notes is my first song with a key change. In the above image, the beige sections are in C major and the blue parts are in… A minorThe cyan and pink sections are where I shift between the two.. Yeah, that’s not much of a change. It’s the same notes, but using minor chords instead of major ones.

So far almost everything I’ve done has been in those two scales, simply because those scales use only the white keys. In the program I’m using, the note grids are color-coded according to white / black keysEven if you’re mapping out notes for a guitar, it still uses “piano” style key coloring., so it’s really easy to see what you’re doing. I wish the program offered a way to change how the grids are shaded to make working in other keys a little easier.

AND THAT’S WHEN WHEN YOU PRESS A BUTTON SOMETHING AWESOME HAS TO HAPPEN.

 

Footnotes:

[1] Other than the fact that all your instruments end up sounding “canned” and don’t have the richness a physical music-making apparatus can provide. I don’t have an ear for it myself, but for some people this is a huge loss.

[2] Is it okay if I call myself a composer? I won’t tell Mozart if you don’t.

[3] Going by direct feedback, not going by social-media “thumbs-up” ratings on Soundcloud.

[4] The cyan and pink sections are where I shift between the two.

[5] Even if you’re mapping out notes for a guitar, it still uses “piano” style key coloring.



From The Archives:
 

38 thoughts on “Bad and Wrong Music Lessons, Part 5: Sound Bytes

  1. Dan Efran says:

    “change how the grids are shaded to make working in other keys a little easier”

    A while back I wondered what a piano keyboard would look like if you repainted the white/black pattern for keys other than the usual C Major/A minor.

    I didn’t repaint any pianos, but I made a printable guide you might find useful: http://dan-efran.com/blog/2014/05/the-other-white-keys/

  2. Dev Null says:

    I don’t really do techno / electronica much, but I have a sudden desire to hear a song sampling Shamus saying “Have you noticed that there are two different shapes for each letter!?”

    …and possibly the StrongBad email printer sound.

  3. Dennis says:

    Pedantic time!

    “Eleven base guitars” should most likely read “Eleven bass guitars”.

    1. Tulgey Logger says:

      Not to be confused with Guà­tars, the Elven base.

    2. LowercaseM says:

      Maybe it’s eleven in a base “guitar” number system… Like base 10 and base 16!

      1. postinternetsyndrome says:

        Well the word “guitar” can be traced back to the same root as “sitar”, which means “three” (as in three-stringed instrument). So it would be base 3.

        1. Decius says:

          It would be eleven, base three, or S(S(S(S(0)))) in Paeno notation.

          Commonly known as four.

        2. Purple Library Guy says:

          So then technically (well, and more occidentally) it should be a sextar.

  4. Henson says:

    “…they're all playing staccato quarter notes in the same octave”

    Who are you, and what have you done with Shamus?! He’d never use such voodoo words!

  5. postinternetsyndrome says:

    I like these articles even though I’ve studied classical composition. It’s really interesting seeing you find out in your own – sometimes a bit roundabout – way the things I’ve learned from books.

    You made me expect something denser from “Too Many Notes”. It’s true that you can muddle things by having too much music at the same time, but I think you have plenty of margin left. It made me think of Ed Harrison’s Neotokyo soundtrack. Check out the second song, “Tin Soldiers”. It gets pretty crazy towards the end: http://ghostiq.com/neotokyo/ I would never call it too much though.

    I also really like the Freelancer soundtrack. Some of the pieces are close to electronic perfection. One of my favourites is the Independent Worlds battle music. It’s pretty chaotic at its high points. Here’s a spotify link: http://open.spotify.com/track/25wTSImWZx5yk4lBLE6fLn (It can probably be found on youtube too.)

    Putting in a sound clip to make the song about something is only one step removed from actual lyrics, I would say. Similarly, it was really popular during the romantic era to make programmatic music; pieces inspired by a particular painting or a book. Famous examples include Scheherazade by Rimsky-Korsakov (based on the well-known tale) and Nocturnes by Debussy (inspired by three paintings by Whistler).

    After a couple of hundred years of abstract baroque and classical music, I guess the composers of the time felt they needed something more substantial. For my own part though, I prefer music that has the confidence to stand on its own. Not everything needs a meaning, and I feel that song in particular is the ketchup of music: It tastes all right, but add too much and you don’t even notice the other stuff.

    Similarly, being told beforehand “this music is about x” contaminates the experience for me. Anecdote: I once was at a concert were a newly written piece was performed. We were told that the song was about this boat on a stormy sea (real original, that). The music then didn’t at all match my expectations for what that would be like, and I just sat there the whole time thinking “this feels nothing like a stupid little boat”. The music was fine, but I couldn’t concentrate on enjoying it because the composer didn’t have trust in my ability to appreciate it for its own sake.

    So I guess my point is: Keep making abstract electronic stuff Shamus, and feel free to add a lot more chaos. :)

    1. Dennis says:

      I love the Neotokyo OST. Came out when I was in high school without a credit card, so I missed my chance to get a physical copy.

    2. Lilith Novale says:

      The reason why Programmatic music became more popular in the Romantic era was for basically the same reason why Shamus (and creative people in general) find it easier to create stuff if they have some kind of stimulus to start them off.

      In the Baroque era, all music was very mathematically composed – if you were writing a Fugue, you would choose some notes, and then basically just apply a lengthy formula to them to get a full piece.

      In the Classical era, social expectations changed – you didn’t need to be formulaic, you just needed to stick within some guidelines. For instance, if you were to write a Symphony, you would compose 4 movements:
      an opening sonata or allegro
      a slow movement, such as adagio
      a minuet or scherzo with trio
      an allegro, rondo, or sonata

      In both these eras, there were strong restraints on the composers’ creativity, which made it a lot easier to create good music.

      And then the Romantic era happened. The biggest change, by far, was in how musicians were employed. Beforehand, composers were beholden to various nobility, who would either have the composer in their full employ, or would regularly commission works from the composer. A noble’s composer was treated with the same respect as your average servant or cook – important, but not one of the nobles. In the Romantic era, musicians broke off and became a lot more independent of the nobility.

      Unfortunately, it’s much harder to compose something from nothing than it is to compose something from a commission, so musicians went halfway and took their inspiration from paintings, folk stories, and the like. Operas and Ballets exploded in popularity, because they already had a plot, you just needed to accompany it.

      1. postinternetsyndrome says:

        That might be a fairer angle than mine. :) All things happen for a reason, of course.

      2. John says:

        I’m no expert, but it seems to me that as we move into the Romantic period, increasing affluence expanded the potential audience for orchestral music (and ballet and opera). A composer could, in principle, make a living by selling tickets to the haute-bourgeoisie rather than by serving as a status symbol for a member of the nobility. So the Romantic period is the point where it first becomes possible for composers to ditch the nobility rather than the point where they just got fed up and said “You know what, screw you guys.”

    3. Steve Online says:

      Ah, i was going to comment on the Neotokyo OST as well. Although, while Tin Soldiers is another highlight in the album, my piece to mention for aural saturation would have been Scrap IO.

    4. PowerGrout says:

      I didn’t even know of Neotokyo till now.
      And I thought all I had to get excited about this Christmas was the splendid seasonal price of sherry.

  6. Cuthalion says:

    “Too Many Notes” was quite fun.

  7. shiroax says:

    I’d like to hear a song sampling Chris dissing Josh… maybe autotuned :D

    1. Daemian Lucifer says:

      Chris and Galaxy Gun doing valey girl voices.Autotuned.

      1. Otters34 says:

        Nonono, even better: Galaxy Gun laughing in the middle of a really ominous song and telling the rest of the Spoiler Warners to call her that. Like during something that sounds like the Imperial March.

        1. Daemian Lucifer says:

          Yes,definitely!Shamoose should totally make that as a christmas present to Galaxy Gun.

  8. Daemian Lucifer says:

    Wait…Why does awesome button sound like Angry Joe?

    Its kind of cool actually.

  9. Jay Allman says:

    “I wish the program offered a way to change how the grids are shaded to make working in other keys a little easier.”

    This is one of many reasons the traditional staff system is superior to the “piano roll” system you’re using, and why the staff system is worth learning. Those key signatures that look so scary? They’re nothing more than the kind of “color coding” you yearn for. They “repaint” the staff into a new key.

    See here for a visual demonstration:

    https://www.youtube.com/watch?v=-s8qYgiB4oo

    Notice how the first four scale/chord patterns are exactly the same. The scales start and end in the same place, and the chords sit on the exact same lines and spaces. So why do they sound different? Because changing the key signature “repaints” the keys from C major to C minor to C-sharp major to C-sharp minor. You’re running up and down from one “C” to another, along the same path, but the key signature reinterprets the sounds into the desired key.

    So when you’re writing in one of the above keys, you only have to think of the “white keys” (C-D-E-F-G-A-B-C) and can completely ignore the black keys. You can leave it up to the key signature to shift the tones onto the black keys where necessary. So write your songs in C major, then insert the C-sharp major key signature at the front. Boom. Now you’ve taken it from a very safe key to a very scary one that uses lots of black notes. Or reverse the procedure: Write the key signature beforehand, then forget about it and just use the “white notes” while composing, leaving it to the key signature to do the dirty work of changing them into black notes where necessary.*

    (*Caveat: This works only so long as you’re not inserting sharps and flats on your own.)

    And in truth, it’s the same thing with keys that aren’t a “C” key. Suppose you want to write in B major (the fifth example in that video)? Just insert the B major key signature (the “key repainter”). Then run your scales from B to B instead of C to C, and shift your chord relationships down a step. Instead of building the chords on C, F, and G, build them on B, E, and F **. You can even still think of the notes you’re using as the “white keys”, running B-C-D-E-F-G-A-B, since the key signature will automatically reinterpret them as B-C#-D#-E-F#-G#-A#-B.

    (** Yes, technically the chords are built on B, E and F#. The point is that when you make an F# major chord, you only have to put notes in the spaces for F, A and C; the key signature will reinterpret them as F#, A# and C# without your having to consciously put the sharps in.)

    Yeah, you might have to daub in an accidental now and then if you’re modulating or otherwise reaching outside your chosen key, but you’d have to do the same thing on a piano roll.

    Look, the piano roll gives you all twelve notes in the octave when you’re probably only going to be using seven of them. So you’ve always got those extra five tones under foot, and they make you nervous because you’re worried about if or when you need to use them. But the traditional staff system has only seven spaces: seven spaces for the notes of the key you’re working in. It doesn’t matter if it’s a C major scale or a B-flat minor scale: It will run up eight steps and back down eight steps. The key signature “paints in” the seven notes you want, and interprets them in the correct way. The other five are invisible, out of sight and out of mind. You have to go find them if you want to get fancy by bringing them in. The traditional staff system, unlike the piano roll, doesn’t FORCE you to look at them.

    So Kaiser Joey’s complaint–“too many notes”–also applies to the piano roll GUI. The system Mozart used is better.

    1. Abnaxis says:

      I had a similar idea reading the post. I like to think Shamus basically just reinvented staff notation while complaining about the over-simplified notations put in place to try to avoid it.

      1. Jay Allman says:

        Exactly. I don’t know how hard it would be to make electronic music in a staff system, but I would like to see Shamus coaxed out of his skittish antipathy for traditional notation. FWIW, then, another video, showing in more detail how an “I only want to use the white keys” approach evolves easily and naturally into the staff system.

        That’s part 14 of a series I’ve been making, so yeah, it kind of plunks down in the middle of things, and I don’t deal with rhythm or note duration, since that would be (at this moment) off-topic to the larger series. I’ll just say here that, IMO, the bars in a “piano roll” stand to quarter, half, eighth, etc. notes in traditional notation as “***”, “*******”, “********************” and “*******************************” stand to “3”, “7”, “20” and “31” in mathematical notation. Yeah, we can have scrollable GUIs on our computers now, but that doesn’t mean we ought to replace the old-fashioned Arabic numerals–so useful when paper was at a premium–with strings of dots when doing sums.

        BTW, I certainly don’t mean to compete with Shamus’s B&WML series with my own. It’s only that four years I ago I knew even less about music than he does (and I still know hardly anything) but I did try reading some books. His B&WML series provoked me into working up my own version, as the kind of “for dummies” approach I wish I’d found when I first started fumbling around.

        (Bet you thought I was cussing good and hard in that second paragraph before you read farther on!)

  10. Paul Spooner says:

    Yay! More music!
    I am totally going to add more notes to “Too Many Notes”
    Standby…

  11. David C says:

    And from the school of “you can never have Too Many Notes”: Black Midi version of “Let It Go”

    (Black Midi is yet another of those “I had no idea this existed, but now I must follow down this rabbit hole for a few hours” things)

    1. Paul Spooner says:

      This actually sounds worse than normal MIDI to me. What is the point of all the scads of notes? Other than drawing pictures with them?

      1. TSi says:

        you said it, drawing pictures.
        Although, some simply try to add complexity to make it impossible to play irl, others produce visuals that can look interesting and sometimes impressive. It takes time to lay out 3 millions notes …. X )

  12. John says:

    It’s funny that your most well-received effort is the one that one that sounds most like music that a band could actually play–or, dare I say, the most conventional. Is it well-received because it really is better than the others or because it most resembles the music they are used to or already listen to and therefore sounds “right” to them? Could be either, could be neither, could be both. (Could also be that your trusted critics listen to nothing but weird avant-garde stuff, and I’m crazy.) I wonder about stuff like this all the time.

    You know, there’s already a whole heck of a lot of pop (and rock!) music that is basically impossible (or at least really inconvenient) to perform live without the aid of some pre-recorded bits played over speakers. I was at the barber’s once when the Rolling Stones’ “Sympathy For the Devil” came on the radio and the repeated, near constant “HOO-HOO!” backup vocals prompted me to ask aloud “How do they play this in concert, anyway?” You’d think it would kill their voices, or that at the very least they would get sick of doing it. The barber gave me a funny look and told me “they use a backing track,” as if it were obvious. (As I suppose it is.) If the Rolling Stones–who seem like such a traditional guitars-and-drums kind of outfit–can, uh, cheat and use tracked music, maybe electronic music isn’t that weird after all.

  13. David F says:

    One terminology point: When you refer to the “waveform” of a note, that’s the Physics term. A musician would refer to that concept as “timbre” (pronounced “tamber” for some reason) and “envelope” (or “shape”). Timbre is the quality of the sound that lets you tell the difference between a trumpet and a violin if they play the same note at the same volume for the same length of time. From a physics standpoint, this would be the structure of the wave as a sum of sine waves of particular frequencies and amplitudes.

    “Envelope” is essentially the shape of the volume of the note over the course of its life. This has nothing to do with whether the note is loud or soft, but rather with the “attack” of the note (hard or soft), and how the note grows and/or fades before ending.

    1. CJ Kerr says:

      It’s also the reason I’m finding “Too Many Notes” quite difficult to listen to – not because it has too many notes, but because the lead part has the timbre of a circular saw cutting hardwood.

    2. Purple Library Guy says:

      ‘”timbre” (pronounced “tamber” for some reason)’

      It’s a sloppy Anglicized pronunciation of a French word. The fact that the French world also means “stamp” is making me rack my brains trying for a pun about the “timbre” going on the “envelope”.

      1. Lachlan the Mad says:

        Just say “tone colour” (the term I learned in high school music) and we can all rest easy.

  14. Dragomok says:

    Wow, Shamus, you’re making a steady progress. I enjoyed all the musical pieces you linked to in this post. Too Many Notes is really good, actually.

    On a related note: does anyone know how to stop Soundcloud from auto-playing other sound files after you finish listening something? I hate that every time I try to listen to something on their site, it immediately is followed by something else, which quite often is something that makes my ears bleed a little.

  15. TSi says:

    I’m sorry, I don’t want to sound bad with this comment and you can edit out/delete any part or it. It’s okay. I just want to give you an honest opinion as someone who studied music (when I was a kid) and played trombone in an orchestra. I enjoy any form of music and am listening this week to some old favourites from Unreal Tournament 99 OST mostly composed by Michiel van den Bos and Alexander Brandon who also both composed some of the original Deux Ex songs (I suggest listening to both OST as reference). One of my favourites would be Razorback from Peter Hajba http://youtu.be/DzwCsd1TVcA that you can probably remake quite easily in MMM or at least reuse the structure for learning purpose.

    I would like to express my opinion as to why I think most people enjoy Best Part of the Storm more than the others.
    Something that struck me when I started listening to your songs in older posts was the lack of a recognisable melody (theme) and a proper phrasing construction. It felt like you wanted to talk about many subjects at the same time but only had a couple minutes to do so. You can hear this struggle in almost all of them but in Best Part of the Storm you actually started pretty well.
    From 0:03 to 0:50, you provide a phrase that looks like a theme. It’s a nice melody that you could have expanded and I was expecting it to return after the “punctuation” from 0:50 to 1:15 but it didn’t as the pause was prolonged with a variation. That was interesting as it doesn’t happen often in electronic music but the theme still didn’t come back afterwards.
    I was constantly expecting the theme to take over again at least in some variation of the phrase but nope.

    I don’t know if what I say makes sense but your music seems to lack structure. I don’t know if you already considered “musical form” but it could be useful : http://en.wikipedia.org/wiki/Musical_form

    In the music I linked at the top (Razorback from UT 99), you can literally feel like you’re listening to someone telling the story of an epic journey. All songs don’t necessarily tell a story (although that is subject to a debate as anyone can pick up anything from them) but their structure are often similar to a narrative schema ( “Schéma narratif” in French) in novels and the likes. In musical terms, these structures would look like : an intro, a refrain, a phrase, the refrain again, a pause, the refrain, another phrase, conclusion. And you could cut out some parts or mix them in some original way.
    Most modern music uses predefine schemas like this one and it’s often difficult to make something original that goes out of these bounds and still feels nice to hear. Well, we all have different tastes anyway. :-)

    Not only that but you forgot to count and group your phrases in divisions of fours or eights. Awesome Button for example is broken in fours, eights and other weird and long divisions. It’s hard to keep track but let me show you what I mean by counting the beats along the phrases.
    From 0:00 to 0:07, if you count the beats and group them by 4, you get a division of 4 groups.
    From 0:07 to 0:15, also ends with a division of 4 groups.
    From 0:15 to 0:44, it’s a division of 16
    From 0:44 to 1:30, it’s an unnatural division of 24
    It goes on with random sized divisions. It’s not often clear where a phrase begins or ends when there are small variations to it along the way but in this case, it feels weird and i’d suggest to stick to small divisions like 4 or 8 for your refrains and phrases which is probably the most pleasing as it’s neither too long nor too short to make an interesting phrase 16 can also be used to some extent but it’s better to not go too far unless you’re aiming for very large bpm techno.

    Hope this helps in some way. Sorry for the length and for any mistake in naming or conventions. I probably got some translations wrong as well as i’m French. Anyway, as long as you like what you do then by all means, feel free to dwtfyw !

Thanks for joining the discussion. Be nice, don't post angry, and enjoy yourself. This is supposed to be fun. Your email address will not be published. Required fields are marked*

You can enclose spoilers in <strike> tags like so:
<strike>Darth Vader is Luke's father!</strike>

You can make things italics like this:
Can you imagine having Darth Vader as your <i>father</i>?

You can make things bold like this:
I'm <b>very</b> glad Darth Vader isn't my father.

You can make links like this:
I'm reading about <a href="http://en.wikipedia.org/wiki/Darth_Vader">Darth Vader</a> on Wikipedia!

You can quote someone like this:
Darth Vader said <blockquote>Luke, I am your father.</blockquote>

Leave a Reply to Tulgey Logger Cancel reply

Your email address will not be published.