Some Drum Notation Headaches
Posted on July 23, 2023
I often criticize the majority of drum set sheet music out there as being ugly (or downright illegible), but I must empathize with my fellow drum notation engravers because there are a few components to the drum set that make the instrument uniquely annoying to notate. So let me go over a few.
Two Voices, or Not Two Voices
This has been a source of self–doubt for a while.
There are a few ways to notate a groove on the drums, by way of what parts of the kit get their own isolated rhythm:
I pretty much always go for option “A”. It most closely mirrors how I conceptualize playing the drums, and I find it to be the neatest.
However, most “professionally” published sheet music tends to split up the drum set, usually by separating the kick and the hands:
A while ago I found one of only a few articles considering the three different notation approaches, making the case that option A should be saved for linear patterns; option B should be used if the hands are busy and the kick is simple; and option C is the way to go if the cymbal is keeping a steady rhythm.
Personally, I think changing which style you use throughout a score can be distracting, so I try to just use one and stick with it.
Using independent rhythms (also referred to as “voices” or “layers” by different scorewriters) adds some needless clutter in my opinion. The sheet music now takes up much more space vertically, exacerbated by dynamics and hairpins, not to mention articulations and tuplet numbers — you have to be careful to make sure nobody gets confused and thinks an accent at the bottom of one line is actually anchored at the top of the next line below.
Splitting the kick and hands up doesn’t seem to have origins in making drum set music easier to read, but rather from orchestras; at some point in time, a composer realized the parts for snare drum and bass drum could be combined into one staff:
When the drum set became its own instrument, I think this convention continued, even though the kick and snare no longer play separate parts per se.
Todd Bishop recently wrote a whole article about this topic (I promise I don’t steal all of my ideas from him!) and he did a chronological roundup of drum books to provide a synopsis of how drum notation has evolved over the decades.
Todd starts with Jim Chapin’s Advanced Techniques for the Modern Drummer, published in the 1940s. I think there is an earlier starting place, however, with Gene Krupa’s Drum Method from the 30s — I’ve mused that this might be the oldest book out there to use something resembling modern drum notation (it’s certainly the oldest still in print). Unsurprisingly, the kick and cymbal are split up:
I’m wary of taking cues from professional drum scores since most of the people producing these might not actually be drummers; I’m not sure where the average drummer would even start to get into this line of work. Maybe in college, and I think the style of splitting up kick and snare has sort of snowballed in the university setting. Moreover, someone commented on Todd’s piece that people tend to let notation software do whatever it wants, even if what you get is very ugly.
I’m fairly convinced the way I do things is okay... but sometimes I wonder. Have a look at this 4/4 groove:
I can’t deny that the hi–hat’s triple beat rhythm stands out a bit more with the second option. Is it worth the extra clutter? Not sure; I don’t really expect people to sight–read my scores.
On the flip side, have a look at an excerpt from “Good Times Bad Times” (the top is my original, and the bottom has been re–engraved):
The bottom line is supposed to follow the logic outlined by that Online Drummer article, and I think it looks disgusting.
For one, I absolutely hate the look of tuplets that use a bracket and rests, and I try to avoid using them as much as possible. The spacing of the cowbell at beat 3 also looks ridiculous on its own — I’m reminded of those goofy camera filters that stretch a picture in the middle:
Because the cowbell suddenly starts playing 16th notes in the second measure, I guess the logic is to give it its own voice.
There’s just no comparison. The way I did it originally is the neatest and clearest.
However, sometimes I can’t avoid using two voices. To notate polyrhythms (like “Dancing Men”), it’s the only practical way:
Back when I cared about the hi–hat pedal, I used to give it its own voice, but I eventually got sick of notating what the left foot pedal is doing, so I don’t bother anymore.
This one has been driving me crazy lately.
There are two common ways of notating ghost notes: wrapping the notehead in brackets/parentheses or making the notehead smaller. I actually use both.
Parentheses take up a lot of space, especially because, by default, MuseScore puts them too close to the noteheads for my liking...
...so I have to manually space them farther apart.
I could just use a small notehead, but I don’t find this to be clear enough for me, so combining them seemed like a clever way of getting the best of both.
And yet, even with smaller noteheads, the parenthesis can mess with the look of the score — see how the space between the steams is now inconsistent on beat 4 of the first measure:
Sometimes MuseScore won’t alter the spacing automatically, which leaves parentheses too close to a neighboring note:
Also, note how there’s a bracket on beat 4 of the second measure which is touching the hi–hat note above it.
One option (for the first problem) is to increase the spacing of all notes in that beat to keep them consistent, but a measure can become much wider doing this, which limits how much music you can get onto a single line of the score (also known as a system). A similar problem is found if I decide to push the parentheses even farther away from the noteheads to keep them from colliding with a cymbal note.
Recently, I have been considering ditching the parentheses and just using a smaller notehead — it’s begun to feel like too much of a mess to deal with the positioning of the brackets.
However, MuseScore is now doing something interesting after the latest update: making noteheads smaller now makes the brackets smaller too:
At first, this irritated me and seemed like another misstep on the part of the software. However, I think this might be a usable solution (and perhaps a serendipitous one — I don’t know if it’s a glitch or not). The ghost notes now take up less space, and the brackets no longer bump into the cymbals.
I have yet to produce a score with the new brackets, so I can’t say for sure if it will actually work, but it’s promising, and I’m tempted to re–engrave my transcriptions with this new notation style.
Too Much Music on a Line
Unfortunately, I don’t know if I have a promising fix for this next one.
One of the hardest aspects of drum notation is finding the right balance of how much music you can cram into a system. Too many measures will look ugly and will be hard to read, but too few measures can still be visually disruptive, akin to letters that are spaced t o o f a r a p a r t . Too few measures per system can also lead to scores that are needlessly long.
I blame this on drums specifically because most other instruments can space noteheads closer together when they are spaced farther apart vertically. In the case of drum grooves, the same notes are used over and over again.
This concern of mine came to a fever pitch after I did my “Ebony Jam” transcription about a year ago — after the fact, I can’t help but feel like some of these systems are bursting at the seams:
When in doubt, two measures per system is probably the most readable, but a score can become many more pages longer than it needs to be. And once more, using only two measures on a system can still be ugly if they are sparsely populated with notes.
I try to work in multiples of two (e.g. four measures per system, eight… it’s hard to make 16 work). Three measures may look better (and MuseScore seems to like defaulting to this), but it can still be irritating to read when a phrase ends in the middle of a system. Again, consistency is something I strive for.
Part of the problem is that the page size for my transcriptions is US Letter, 8.5 x 11 inches in size. The sheet music you’d get in an orchestra tends to use a larger sized paper — the standard is around 9 x 12", apparently taken from Tabloid Extra sized paper (12 x 18"), folded in half. Since most people don’t have that sized paper, I don’t consider it; printing it on a normal–sized paper will shrink the music.
To conclude… there isn’t really a resolution to this problem. I just wanted to complain about it (complaining is something apparently I feel like I don’t do enough).