How-To: Selecting and Sequencing Music - "It's all about balance."
Music is the most complex form of communication on the planet. It's simultaneously four or even five-dimensional -- it goes forward, it goes up and down, it gets softer and louder, the timbre (aesthetic sound) can change from happy to angry depending on the different instruments used and how the notes are "attacked" by the performer. It's transient, not lasting -- the sound is only there at that instant and never to be heard or experienced precisely the same way again because that moment in time is gone. Even listening to the same recording, over and over, is heard slightly differently each time because time has elapsed -- time that you can't relive.
I know I'm sticking my neck out here because musical appreciation is, after all, almost the very definition of "subjective material" and a lot of readers will most certainly disagree with what I write. But what do I care? I'm only making suggestions from a musical standpoint as it relates the the practicality of coordinating it to an electronic light display. If you don't like these ideas, then go write your own. But these come from someone who's studied music somewhat extensively (my major field of study was music composition), and my only goal here is to give a musically untrained person some insight into how music is constructed and how to discern that construction because knowing how a song is built can help your light sequencing.
Some of what I've written will be obvious, but perhaps you've never formally considered it. I see that as part of my job -- to open your eyes to things that your ears have heard thousands of times and to make you aware of them and how they work together. Also, bear in mind that I'm writing in generalities, not specifics. When you get into the genre of improvisational jazz of for example, you have to throw the book out the window for a while. But even jazz is based on "classic" ideas because the musicians that go off into their solo riffs always come "back" to something so they can all play together again, and it's the "coming back" that's based on classic music constructs. Also understand that when I refer to "music" or a "piece" or a "song" it means the same thing, whether there are actually lyrics attached to it or not. After all, an orchestral symphony is essentially a long song that just happens to be void of words.
Okay, enough of that. Here goes...
What is the song about? Think about "Silent Night." You know the tune. It goes both up and down in pitch, doesn't it? How far up or down does it go? When you sing the phrase, "Sleep in hea-----venly peace" do you naturally get a bit louder on "peace?" Sure you do because high points of songs are supposed to be important. But (and here's the kicker) should you get louder there? Think about it -- you're singing a lullaby about sleeping and peace -- is that supposed to be loud? You're not singing a rouser for your favorite football team here. Lesson #1. if the piece has lyrics or even if it's an instrumental version of a song that normally has lyrics, LISTEN to what the text is saying and approach it appropriately. If the text is very tender like Silent Night is, then certainly wildly blinking and flashing lights aren't an appropriate interpretation of it, are they? Matching the lighting to the text of the song is a pretty good idea -- it will give the viewer/listener comfort. When the lighting matches the music, the viewer feels good about it. When it doesn't match, the viewer doesn't know why they didn't like it -- they just will say, "Uh, I didn't like that one" when in actuality, it's a subliminal thing -- the attitude of the music and the attitude of the lights didn't work and created a subliminal conflict. If you're trying to make people feel good, creating conflict isn't a good way to start.
How complex does the song "sound?" In other words, is there a simple melody and only a few instruments accompanying it or is it like a big, loud Sousa march that you might hear on the Fourth of July where everybody in the band is honking away at full blast? (Sorry, you non-Americans will have to come up with your own analogy here...) Also consider what instruments are being used. You'd generally associate the brass section (horns) of an orchestra with "power" and if they're blaring away, then certainly interpreting it that way with fully bright lights is a pretty good idea. But if the woodwinds are playing softly (flute, clarinet, oboe, etc.) perhaps a different interpretation is more appropriate. Think about the sound of the attacks on the notes -- do the instruments or voices seem to bounce lightly from note to note or do the players put the hammer down and really HONNNNNKKKKKK on them? Do the notes sound playful and frolicky or forceful and important? Think about articulating lights accordingly -- maybe by decreasing the intensity and flickering them quickly for a playful section and then make them brighter and solid on heavier sections. You might also think about where your lights are located. Notes that are high in pitch might be highlighted by lights strung high in a tree or on the roof; low notes might be simulated by lights on the ground or on bushes. You could also simulate different instruments in the orchestra by color -- maybe red lights for horns, yellow for woodwinds, blue and green for strings... You get the idea -- make the lighting appropriate to the sound.
Sound placement. This probably makes no sense at all to you, but it will after you think about it. Stereo sound attempts to make music more alive by putting you in the middle of it. You usually hear some things in one channel and other things in the other channel, right? If you have a multi-channel home theatre audio system then you know exactly what I mean. Well, for some selections you might use your "light canvas" to simulate the placement of instruments on an orchestra stage, or if you're using some of the traditional "Charlie Brown" tracks from the Vince Guaraldi trio, you might light the piano in a different place in your display than the bass or saxophone. If you're lighting an a capella singing quartet where there are perhaps to women and two men, think about lighting it as simply as it's performed, with maybe only a few elements of your display instead of the whole thing. Or maybe bring all the lights up 20% on the outer sections of your display to "frame the center stage" and then just play with a few of the others more in the middle at full intensity to simulate the performers. By placing the "sound" into your light canvas you can surround the viewer and envelop them visually as well as audibly as they listen to the music in stereo. Considering sound placement is another way to make the lighting match the music and bring balance to the overall effect.
The tempo/speed/pace of a song makes a difference. If done well, a song that engages the listener can actually change his/her heartbeat, either faster or slower. Think about the "beat" of songs that find you tapping your toes, such as "Let It Snow." ("Oh, the weather outside is frightful... but the fire is so delightful...and since we've no place to go, let it snow, let it snow, let it snow.") It's usually a regular, repeating rhythm, isn't it? And more often than not, it usually continues through at least most, if not all of the song from start to finish without change. The "beat" is called a "constant" and it helps move the piece forward. It may be fast or slow, but it's always there to keep the song moving. You could also think of the beat as controlling the pace of the song. Here's where we need to get into a little math...
The beat of a song is based on its "meter." The meter is how many beats the composer has designed to be together in what is called a "measure." Measures are used to organize the music into consistent groups of notes. Measures allow dividing the song up into mathematically equal parts, which then makes it possible to print it onto a piece of paper so that someone else can read it, understand what it's supposed to sound like and therefore reproduce it. Measures can have any number of beats -- one, two, three, four, five -- nine, thirteen, etc. depending on how the composer wants the music to be reproduced by the performer. The most common of these is the duple or triple meter, where the measure is counted as having either two or three beats. A measure that has 4 beats could be called a "quadruple" measure, but because it's divisible by 2, it's still called a duple meter.
Duple meters are generally more common, especially in common rock, pop or rap music because people generally relate to them easier. They're also easier to dance to, mostly because you have two feet and you can move each foot the same number of times in songs that use duple meters. Why is that important? Because people like to be in balance. People like a "left-right-left-right" rhythm, and one move offsets the other. And, if you think about it, a great deal of poetry and lyrics is based on a duple meter:
MAry HAD a LITtle LAMB it's
FLEECE was WHITE as SNOW;
EVery WHERE that MAry WENT the
LAMB was SURE to GO.
Duple meters balance themselves out -- a STRONG beat is usually followed by a weaker beat. Or sometimes you'll have a weak beat followed by a STRONG beat, which is then called the "back beat" (think of Buddy Holly and early rock-n-roll music from the 1950's). Whatever it is, the human ear likes balance. The brain likes that, too. Look at the above nursery rhyme. When you read it, you probably insert some extra time here and there, don't you? You pause after SNOW before you start the 3rd line, right? Sure you do. That's the "beat" talking to you. You automatically insert it even though it's not written -- you insert it to make it balance -- to make it flow evenly. You do it again after the word GO. Come on, admit it. Now tap your toes to the rhyme and say it again. Then try it and count the beats for each line, even the ones that you add in on your own. How many did you get? That's right, four per line. Now you know how duple-metered songs are designed -- something always fills in the unstated beats -- it's either inside your head, or a tapped toe, or a an instrument played by the band -- something fills it in so that each line is balanced. When you add them all up, you'll have 16 beats in this example for one "verse" of the poem. Music works the same way. Remember the number 16 because you'll find it all over the place.
You can use this to your advantage when you sequence your songs. Because duple meters are generally rhymed in quatrain verses such as above, there are normally 4 beats to a line, and two lines make up a phrase. Then there's an answering phrase that's made up the same way, for a total of 16 beats for the whole verse -- two, 8-beat phrases. So when you sequence a duple-metered song, you can usually assume that the song can be broken down into 16-beat groups where the music remains relatively the same for the entire 16 beats -- the same instruments, the same volume, same range of notes, etc. Therefore, the light sequencing you might do can likely be evenly divided between the 16 beats, which visually would match what the listener hears, and the result is a balanced effect. Remember, people like balance in their lives and when they get it, it makes them feel good, secure, and that all is right with the world. If your lights don't match, it won't seem "right" to them -- they'll think something's not working right. And, of course, they won't know why they don't like what they see, they'll just say, "Uhhh, I didn't like that one..."
What's more, usually you'll have two full 16-beat groups in succession in a song because one balances off the other. Then you'll encounter a thing called a "bridge."
A bridge is simply a musical insertion that allows linking one set to the next, but it gives the ear a break. Think about a hymn that you might have in church -- think of one that has seemingly endless verses and you sing all the verses. Boring, isn't it? It's boring because your ear doesn't get a break -- it just plods along in the same key, verse after verse after verse. That's why once in a while, the organist will insert a little interlude or flourish on the organ every two verses or so -- or possibly modulate the song to a different key. That's a "bridge" and it gives you a new start -- a new point of interest in the music. It works for the ear, why not use it to your advantage in sequencing lights, too? You'll recognize a bridge when you hear it. On a song with a vocalist, there will usually be a section where he/she doesn't sing and the orchestra or band takes over on its own. It may change keys or it may have some instrumental riffs in it. It's musical ear candy. So you'd want to light this differently, wouldn't you? Sure you would! And you know what, it will have the same musical construction in terms of numbers of beats/phrases/etc. so you can plan on it being 16 beats (and sometimes 32) before it returns to the original thematic material of the song.
So now you should be able to listen to a song in a different way and pick out the "sections" of the song, and the bridges that link the sections together. Try it. Count the beats. If it's a duple metered song, I'll bet all of the above will be easy to hear.
BUT - what about triple-metered songs? ARGHHH!!! Yes, they exist, too. And songs that have odd-rhythm patterns based on 5, 7, and even 11 beats. Reggae music, for example. You'll also find some Latin American rhythms that will knock your socks off and don't seem to follow any structure. They do, it's just harder to discern, so we won't concern ourselves with that here. But instead, let's talk about triple meters. Think waltz. 1-2-3, 1-2-3, 1-2-3, 1-2-3 etc. "It Came Upon a Midnight Clear" is a triple-meter song. You know that one. Silent Night is another (although it's actually a compound triple meter because it's written in 6/8 time and not 3/4) but it still "sounds" and "sings" in a 3-beat rhythm, doesn't it?
You'll find that triple metered songs still have to "balance" for our ears -- and so the verses in songs like "It Came Upon a Midnight Clear" still are organized in a sort of duple-way with phrases made up of groups of four, 3-beat measures.
It came upon a midnight clear
That glorious song of old;
From angels bending near the earth
To touch their harps of gold.
But it's sung this way:
1 2 3
1 2 3
1 2 3 1
2 3 1 2
It came up-onnn a mi--d night clear that
3 1 2 3
1 2 3 1
glo-ri ous sonnng of old..................... from
3 1 2 3
1 2 3 1
an.......gels be-en-ding near the earth to
3 1 2 3
1 2 3 1
touch their harps of gold..................
Sing the next 4 lines to yourself. "Peace on... the earth...." Notice that it has a hint of a minor key? (the classic III-vii-ii-V7-I sequence for the musicians in the crowd...) This is actually a bridge in the middle of the song. The last 4 lines have the same melody as the first four. The bridge breaks up the song and gives it more interest. That's why this song is such a favorite congregational hymn during Advent.
Another form a triple meter is 6/8, which is counted as two, triple-meters in a row: 1-2-3 4-5-6 but the feeling of a 6/8 song is really counted in two. 1-2-3 is the first "beat" and 4-5-6 is the second "beat." You could think of it sort of like a fast waltz. "Silent Night" is actually written in 6/8 time where "Silent" is the first "triplet" of 1-2-3 and "Night" is the second one, counts 4-5-6. We won't get into why composers bother with different forms of duple and triple meters like this -- suffice to say that it gets into other phrasing and sometimes just practical matters.
The point is that even though the meter is in 3-beat sequence, our ears still hear "balance" in a duple way -- some measure of two, four, eight, etc. So really, regardless of whether the song is a duple or triple meter, our brains treat it more-or-less the same way, and the same concepts apply. Therefore, you can use them in your sequencing as well.
So now we get back to the tempo/speed of the song. Whether it's in duple or triple meter, some songs just seem faster than others. This is usually based on the primary beat but it's modified by the overall number of notes that are being played per beat. The primary beat is usually the first beat of a duple meter or the first beat of a triple meter: ONE-two-ONE-two-ONE-two or ONE-two-three-ONE-two-three, etc. On 4-beat measures, the primary beat is the first one, but the third beat should have more prominence than either beats 2 or 4, so it reads ONE-two-Three-four-ONE-two-Three-four, etc. That's the classic pattern, anyway. Then some rebellious musicians decided to put the emphasis on the opposite notes instead -- kind of reversing the it: one-TWO-one-TWO, or one-TWO-three-FOUR, etc. and they called this the "back beat." The back beat is the basis of rock-n-roll music. You don't find the back beat used much in triple meters -- it always sounds awkward: one-TWO-THREE-one-TWO-THREE... it just doesn't work because it throws off your internal sense of rhythm -- the one that's in your head -- and suddenly, the music doesn't balance anymore.
A song that has 60 beats per minute (one per second) would be pretty slow, wouldn't it? It is. In music, that's a tempo that's called "LARGO" -- the word itself sounds slow. A tempo called "PRESTO" sounds faster, right? And it is -- it'd be three times as fast, or 180 beats/minute. But what if a song were set at 60 beats/minute yet for each beat of the music, the musicians had to play (or sing) 8 notes? It changes the ball game, doesn't it? Those pesky composers sometimes take a nice, leisurely pace and ruin it by adding all those darned notes! The bottom line here is that the meter of the song is just an indicator -- you have to actually listen (or view the musical score) to what's actually happening to decide whether it's really "fast" or "slow."
So how can you use this information in sequencing? You need to find the primary beat. Whether the primary beat is accented or not, find it. You'll often here a drummer use a bass drum (kick bass) as the primary beat and then a tom-tom or snare as the "back beat" to offset it, but it's always there. The primary beat is what keeps the song moving forward. It's the constant. It may not be very loud, and in some cases it may even be implied, but it's always there. It's also something that lends itself well to lighting with a short pulse, and not necessarily one at 100% intensity. If it's a subtle primary beat, treat it that way with appropriate subtle lighting. If it's in-your-face, well then go for it! You'll often find that the back beat (especially in pop music) is quite prevalent and easier to pick up. It can be lit accordingly. Identifying and lighting the primary and back beats can make your lights "dance" and will make your display more fun to view -- it's eye candy!
"DANGER WILL ROBINSON! DANGER!" You must be careful not to overdo the beat with your lights because it will quickly become very tiring to watch. Ever listen to a song and periodically hear a little rat-a-tat by the drummer every so often? Yeah, that's a mini-bridge. It's a break in the action and wakes up the listener's ears again -- as well as keeping the drummer from falling asleep. You'll find that these little rhythm highlights occur on a regular, repeating basis, usually every time the song gets back in "balance." In fact, whenever you hear them, that's generally an indicator of the end of a musical phrase or an indicator that a new one is about to start. Listen for them -- they can help you sequence your lights -- perhaps you may not want to light the primary beat at that time but flash a different set of lights somewhere else in your display highlighting the rat-a-tat instead. EYE CANDY!
Songs use repeatable, predictable elements. Why? Because our ears like that -- our internal "clocks" like balance and predictability. Our brains like that. We like the familiar. We like the feeling of being able to finish what we started. The beat generally stays constant throughout a song -- it's repeatable and keeps the music moving. You may have two sections of music, then a bridge, then two more sections, then another bridge, then the final two sections followed by some sort of abrupt, dazzling, ending such as TSO's Wizards in Winter. But notice the structure I just outlined: 2-bridge-2-bridge-2-end. A section typically is X number of beats in length and X is almost always some factor of 2. A bridge section behaves the same way. Composers design music by linking sections together and therefore, music becomes very predictable. Let me rephrase that... Music that is pleasing to the ear for most people is very predictable. Christmas music certainly is this way. Rock, rap, jazz, and other more eclectic genres of music have a special following, and while you can almost always find the repeating elements in those musical forms too, they can be a bit more difficult to analyze. That's probably why we generally don't hear a lot of Christmas music performed that way -- it's too hard for people to get their ears/arms around it and embrace it because they have trouble finding the "balance" in it. (Dang, that word balance keeps coming up, doesn't it?)
Another reason for having repeatable elements is to extend the song. Writing music that's NEW to the ear isn't an easy thing to do, and it's even harder for the listener to grasp. Listeners want to enjoy the music, they don't want to have to figure it out as if it was some sort of homework assignment. There are some composers who've written music that while luscious and rich, is extremely tiring to listen to. Richard Wagner (pronounced REE-karrd VOGG-ner) is a perfect example. He was a 19th century German composer who's most famous for his operas, and he used a technique called "durch kompaniert" or "through-composing" where the music really is never satisfied -- it seems like it's going to suddenly achieve balance and then it goes somewhere else and you don't quite get satisfied.... and then finally he brings it all together at the very end. It's a bit like watching an Indiana Jones cliffhanger -- where one incident seems to level out and then wham!, he's back in hot water again. It keeps you on the edge of your seat. Wagner's operas are very hard to listen to and they literally wear you out. Johannes Brahms' music is similar in some respects because it's so rich -- it gets more and more complex and just doesn't let up, kind of like a tidal wave -- it keeps coming at you, although Brahms is primarily known for the lush orchestration of his symphonies.
I'm digressing too much... sorry... but from the paragraph you just read, can you see how this sort of thing gets into you and you can't turn it off? That's my point. Predictability connects with the listener by providing balance and comfort. That's why people like to sing Christmas carols. That's why the "oldies" are still so popular and why every generation's favorite performers always put out Christmas albums that have some the the old standards on them. There's a reason why Bing Crosby's song White Christmas is still on the charts.
Oh sure, you say... there's always a catch to it. And yes, sometimes there is. Once in a while, composers will throw in an extra beat or they'll take one out. This is almost always in a bridge section, and while it's not universal, it's a rather commonly used technique that is there by design to throw the listener off with something "new." Often you'll find it's accompanied with a key change, so let that be your guide. If the key changes at the end of a bridge, be aware that the number of beats there may not be exactly the same as what you're expecting to hear. But it's a fantastic opportunity for doing something special with your lights. Maybe some wild, random flickering, or maybe create your own little visual drum roll with some quick pulses to a couple mini-trees, etc. The point is to use these exceptions to your advantage as opportunities to add more eye candy to your display.
Electronic Timing. Hitting the beat marks inside Vixen is only part of the battle. You have to determine what the response rate is of your electronics is -- how quickly a light fires after you actually hear it in the music. And you must use the music as it will be heard from the transmitter (if you use one) because every piece of electronics adds some kind of delay. Here's a good way to test it. Create a timing test sequence in Vixen. Find a song with a very steady beat that's easy to hear and create a "beat track", turning on a single pixel for each primary beat. Then copy the beat track's pixels to, say, four mini trees. Put Vixen into a loop and play it. Go outside with a radio tuned to your frequency and see if the lighted pulses match the beat you hear or whether they lag. If they lag, go back and move the test pixels one cell ahead and try it again. You may find that some parts of your display are more responsive than others; some SSRs may turn on faster, etc. Try to find a happy average and use that for everything. You'll drive yourself crazy if you try to time every possible SSR dead-on. By the way, this is called "anticipation" and it's exactly the same thing that musicians do when they're on stage -- they anticipate the next beat so they're often preparing for it a slight bit in advance so that they make the sound exactly on the beat, when it's supposed to happen.
If you'd rather not make your own audio time track and want some precise timing, I've created three 30-second MP3 files that you can download and use for your timing test instead of using music. These are taken from a crystal controlled metronome, so they're extremely accurate. To download, RIGHT-click on a link and choose save-as, and use them as the "music" for your timing test sequence.
30 seconds of beat clicks @ 60 bpm
30 seconds of beat clicks @ 92 bpm
30 seconds of beat clicks @ 120 bpm
(What you've been waiting for throughout this whole diatribe...)
Understanding how music is constructed by being able to identify the "sections" of the music will help you design and speed up the sequencing process. No longer will you have to look at a gazillion blank pixels on your Vixen screen for a new 3-1/2 minute song and say, "Ohhhhh geeezz......... I gotta do another one....." You'll be able to divide the song up into sections and conquer it. Find the first section musically and light that section. Later in the song, there will be another section virtually identical to it -- you can simply copy the original pixels to the new section. Then do another section, and copy it to it's sister section(s) later in the song. Then sequence the bridge, and copy it over the other bridge section(s) you'll find. Suddenly you're 90% done and all that's left is to finish the song by sequencing the "end" by itself. A "ordeal" suddenly turns into a song that's only 1 minute worth of sequencing magic with a little copy/paste here and there and bingo - it's done!
Return to main page