Bach Cello Suites visualized

As a resident at Eyebeam, Alexander Chen visualizes the first Prelude from Bach’s Cello Suites:

Using the mathematics behind string length and pitch, it came from a simple idea: what if all the notes were drawn as strings? Instead of a stream of classical notation on a page, this interactive project highlights the music’s underlying structure and subtle shifts.

Interaction version here. Charming.

[Alexander Chen via @blprnt]


  • Now should spacing of “notes” not to take account the curvature of the plucking paths?

    • If you want exactly equal spacing between notes, then yes. A musician playing it would vary the spacing between notes, though, and the variation here (where each progression speeds up a little in the middle) is a reasonable one.

      The thing I don’t understand is the beginning. Until about 45 seconds in it’s a muddle that doesn’t sound like any Prelude from Bach’s first cello suite that I’m familiar with.

      • A musician will vary the spacing based on their interpretation of the mood and flow of the piece, which is something mathematics can’t [yet?] interpret. Until then, the spacing should be exactly as written. The variation in this piece was clunky and reminded me of an old wind up music box.

  • Lovely and elegant. Thanks!

  • I felt that the visualization detracted from the music. A decent one would enhance enjoyment and knowledge of the piece. A clever one circulated a couple years ago, based on a Bach piece that has the word “crab” in it’s title.

  • I don’t see how this adds anything that isn’t already visible in the sheet music. To me, it shows much less.

  • William Mullen December 12, 2011 at 6:30 am

    Nicethought but nothing that analysts haven’t been doing for centuries to plot the musical lines of a composition. This particular application does contributes nothing new to that knowledge.

  • It’s very simple and it gets the basic premise accomplished. My favorite aspect is the lengthening/shortening of the “strings” to show pitch.

    I think some guys are too quick to say “Oh, this isn’t too great”, or “I’ve seen better.” That’s not at all useful. The guy had an idea and shared it. If someone made a physical model of this, you’d think it were pretty neat. Maybe someone WILL make a real version of it.

    As a music teacher, I can see that this simple demonstration has real classroom potential.

    Alex, ignore these jack-hards.

  • Here’s a series of visualizations of string playing (beginning with this ‘cello suite movement):
    … and here’s one showing hands playing plus structural elements:

  • Stephen –

    I dislike vizualizations like the first, because the horizontal (temporal) spacing is distorted as the note is being played. I find it so distracting to try to sort out this distortion, that I miss the music as well as any insight the viz could have provided.

    While less “exciting” (in terms of localized movement) the second one is clearer in its representation of the music.

    I find these interesting, for example:

    and these distracting:–ykTqoQnqI *this would have been fascinating without the horizontal distortion

    • What I am trying for with the horizontal warping is (a) to reflect the way events in the recent past and future (the “perceptual now,” so to speak) are “bigger,” psychologically, than those further in the past or future, and (b) to reflect the way musicians move in anticipation of (and follow-through after) the performance of notes — not responding at all to notes that are several seconds in the future, then getting slowly ready, and then moving quickly as part of the act of performing the note. I agree that the warping introduces distortions in the meaning of the horizontal axis, and is therefore not an ideal solution. In some videos, I’ve included barlines so that there’s a point of reference.

      However, I think the horizontal warping is something a viewer can get used to. Imagine that you are standing beside train tracks, looking at the wooden ties laid perpendicular to track at equal spacing. When you look to the left and right, the ties seem to be closer together (due to perspective) the further away you look. We don’t consciously perceive this is a distortion — we accommodate to it. I think the same thing happens when looking at my time-warped videos.

      • I’m glad you haven’t taken my interpretations as harsh criticisms. I enjoy visualizations of music, and even the “boring” ones can add insights.

        I get the narrow “perceptual now” window. Events within this band are psychologically bigger, as it were, but that’s attentional as much as anything. They are not necessarily changing in physical size or in speed.

        The analogy of the railroad ties would work if your time axis was very gradually foreshortened at the far ends of the visible screen, but in your viz, the notes are all equidistant except within the narrow “now” in the center, and there is a disconcerting discontinuity of the time axis at the edges of “now”. This visual discontinuity causes dissonance with the smooth aural stream I am hearing, so I lose track of the audio.

        As a musician or as a listener, I am “looking” (“listening” or “feeling” might be better terms) ahead and behind to keep pace and to feel the flow of the music. The jumping of the notes disrupts the flow.

        I’ve actually watched a fair number of these and similar visualizations, and I still can’t get used to the horizontal distortion. Maybe it’s because I’m not a very good musician and I am highly biased towards the visual in my cognitive processing, but aren’t most humans biased in that direction anyway?

  • Jon,

    You wrote “The analogy of the railroad ties would work if your time axis was very gradually foreshortened at the far ends of the visible screen, but in your viz, the notes are all equidistant except within the narrow “now” in the center, and there is a disconcerting discontinuity of the time axis at the edges of “now”.” What do you think about this one: ?

    • Stephen –

      I like it a lot. The time behavior is much more gradual. The fold at t=now is not nearly as distracting as the funny motion of the notes in the other ones. (I would prefer a smooth transition, but the single “fold” is not detrimental.)

  • Hi,

    I was introduced to this because I play cello.

    The use of pitch gradients by size along with a sense of direction set by visual frames to create an overall musical shape in 2 beat pulse per measure is a nice model for foundational didactic purposes. I think it would also be cool to see a sinusoidal wave in 4 beats with alternating of pulse direction (or not?). It might detract from the cyclical theme of the beginning that this model provides, but may provide light for more linear areas and voice displacement.

    Overall, nice visual model. Thanks for sharing!

    -Hyung Joo

  • It was so beautiful – it brought tears to my eyes. I play this – and when I play I “see” the notes a certain way, but now I can see them in a different way – I can see the phrases as Bach may have seen them – little groupings of “eight”. It was beautiful.