What is going on in this Noname beat?

Hip-hop in the post-Dilla era has been pushing the boundaries of rhythmic dissonance. The coolest and most mysterious groove I’ve heard in a rap song lately is “Sunny Duet” by Noname.

The rhythms here are bananas and I struggled for quite a while to figure out what was going on. I got very excited for a minute when I thought I realized that the hi-hats are playing a septuplet grid.

I was wrong, though, it’s not the hi-hats doing that rhythm, it’s the “doot doot doot” backing vocals. But I went to the trouble of learning how to do tuplets in Dorico and made the graphic, so you might as well enjoy it.

BTW, here’s how to make septuplets in Live if you want to experiment with them.

You can also explore them on the Groove Pizzeria.

Anyway, I asked the Hip-Hop Education group on Facebook what they thought was happening in the Noname track. They had a lot of opinions!

Will Kuhn:

Triplets in the hi hat pattern, and the piano part is intentionally way behind the beat. Your brain makes it work because of the strong backbeat snaps

Robert Komaniecki:

What a cool beat. I’m hearing a hi-hat based on triplets, but often with the second attack in a group of three omitted, giving it a heavily-swung feel. And as Will mentioned, lots of offset elements elsewhere in the beat.

Mike Sundt:

The hi-hat is in septuplets, while the snaps are giving that “lunge forward” feeling in quintuplets.

Will again:

OK septuplets is a fun thing for us nerds to throw around, but I think I figured out where the discrepancy is. If you set a drum machine to 1/16 note swing, and then alternate between that and a triplet grid you’ll get a slightly different value than a full-on triplet swing. Here is a video I wisely spent my time making on the topic. The snaps are a red herring because they’re sampled off beat. Toward the end of the loop one of them is especially loose. Two swing 16ths and then a 16th note triplet is not quite the same as a perfect tuplet, but it could still be notated that way depending on how precise you wanted someone to replicate it.

Kyle Adams:

Before listening: “how complicated can this be?” After: “AAAAAAGH WTF” I’m less concerned with the triplets vs. septuplets thing and more with what the hell meter this is in, how we can tell, and where the downbeats are. But thanks for sharing this, I love her music! “Self” is another great one. The finger snap sound must be from a sample bank b/c I swear it’s the same on as this Dirty Projectors song.

Robert Roche:

I hear a bigger grouping/groove in 5 followed by a group of 3.

Allen Murabayashi:

just for fun: i extracted the vocal and added straight 1/8th drum/piano tracks. first 16 bars. I found two live performances on youtube, and interestingly, both use standard backbeat groove.

Robert Collazo:

Sounds like someone was having fun lining up specific parts and leaving others loose. It sounds to my ear like someone was trying to replicate a live musician cypher jam feel… I do think it’s important to note, from a theoretical standpoint, the difference between mathematically lining up and rhythm/meter. It does seem clear that the mathematical symmetry was achieved using the grid in a DAW, which leaves accents that don’t line up with the meter that you can apply to an analysis from a mathematical perspective.

Is this type of analysis as useful if the mathematical accuracy of the rhythms don’t line up with meter and accents that the math suggests? I once got into an argument with a friend when we were writing a horn chart for “Move On Up” by Curtis Mayfield. He pointed out (correctly) that the horn part mathematically lines up on beat one (assuming that all measures are 4 beats). But, I received confirmation from the elders we were playing with (Chicago soul musicians close to Curtis Mayfield’s generation) that they heard it as beat 3. Another similar example of would be “Movin” by Brass Construction, where beats 1 and 3 don’t agree with regards to math and accent.

I don’t mean to say that myself, or the musicians I consulted with have the definitive word on this, but in my experience as a musician I can say that I will always place accent above math in cases where there is a disagreement of some kind.

Prince Charles Alexander:

The first kick of a bar is in a quantized position. The snaps and backbeat (rim) are ahead of time. Focus on the kick and hi hat. They are totally in time (with the hats playing lazy 8th note swing). The vocals and bass are also slightly late.

Syreeta Neal:

I’m very into how much this very common type of beat is breaking all the white folx’s brains lmao. [When asked for examples] J Dilla/Slum Village/Jaylib, D’Angelo, Nujabes

After giving the matter some more study, here’s what I’ve got for the drums.

Part of the confusion here is that there are two different closed hi-hat patterns, one panned left, the other panned right. The left one is playing straight eighth notes, but they are way behind the beat. The right one is playing triplets in uneven bursts. The rim shots are exactly on beats two and four. In the first half of each bar, the kicks are squarely on the downbeats and on the sixteenth note after beat two. However, in the back half of each bar, the kicks are seemingly unquantized; if there’s some deeper order to their placement, I’m not sure what it is. The first conga hit in the pattern is also placed randomly, while the other two are exactly on the sixteenth note grid. Finally, the two open hi-hats at the end of the pattern exactly line up with the last two kicks, I assume because they’re part of a single sample.

Like I said, I’m confident that the backing vocals are using septuplet swing, or some organic feel that happens to align very closely with septuplet swing. I’m not sure what’s going on in the piano or bass. Incredible track!