Update: 581 of your fellow adventurers have now enrolled in Submersion, our new 60-day Subjective Reality deep dive. What more becomes possible when you're living in a simulation? Join us for this epic journey!
Yesterday I decided to try inputting a simple melody into GarageBand by playing it via my Akai MIDI controller. I’m still very new to this, so I came up with a very basic C-major tune, not much more complicated than Mary Had a Little Lamb, but it’s a start.
After a few practice runs — till I could stop screwing up and hitting the wrong notes — I was able to play and record the sequence into GarageBand. My timing wasn’t great, but it’s pretty easy to edit the notes in the software once they’re recorded.
Once I had the recording on screen, I spent a few minutes tweaking the timing and velocity of the notes till it sounded right to me. That part went just fine, and I was satisfied with the result.
Then I decided to try adding a few GarageBand loops to make it more interesting, starting with a basic drum track. I wanted to see if I could get my melody to mesh well with the drums. In fact, when I recorded the melody, I had the drums playing at the same time, so I could play along with the same rhythm.
GarageBand has a feature called a groove track. You can specify one track to serve as your groove track, and GarageBand will slightly adjust the timing of the notes in the other tracks to mesh with the groove track. You don’t have to use this feature, but it’s there if you want it. The benefit is that it helps make multiple tracks sound good when played together. Otherwise small differences in timing can make the different instruments sound a bit out of sync.
It seems you can also specify the granularity of this feature — match to nearest 1/4 note, 1/8 note, …, 1/64th note.
Since I’m new to this, my timing is far from perfect, so I’d definitely like to use this feature if possible. When I listen to my melody with the drum loop added without using the groove track feature, I can tell the timing is a bit off. The notes of my melody don’t quite mesh with the drums at some points.
When I set the drums to be the groove track, and I have my recorded melody sync to it, the result is pretty good. It fixes my slight timing problems, and the melody seems to mesh nicely with the drums. I can’t hear any timing errors at all.
If I try specifying my melody as the groove track and then have the drums sync to it, it doesn’t sound nearly as good. That may be partly because my melody has only a fraction of the notes that the drum loop has. Intuitively it seems sensible that I should set the drums to be the groove track and have my melody sync to that. Is that a reasonable choice?
Now for some reason, when I have the groove track feature enabled and then try to do even the most basic copy-and paste work with this sequence, something goes awry. If I take my short melody and copy and paste it to a different location on the track (perfectly aligned with the start of a new measure), GarageBand shifts some of the notes a little. In fact, it not only shifts notes in the pasted copy — it also changes some notes in the original version! Why is it doing this?
This also happens if I try to move my recorded sequence one or more measures to the left or right. Some of the notes get time-shifted, even if I shift it based on measure-aligned boundaries.
Furthermore this also happens if I create a new track with the same or a different instrument and copy or move my recorded sequence to that track, whether to the exact same measure in time or a different measure.
It’s easy to show you what’s happening visually.
Here’s a piece of a screen capture from GarageBand that shows part of the melody that I recorded and edited. This sounds reasonably good timing-wise when it’s played, and it syncs well with the drum loop too.
And here’s what GarageBand does to this same sequence if I try to copy and paste or move it somewhere, either to a different place on the same track or to a new track. It does this not just to the copy but also to the original.
Notice that it changed the timing of notes in the triplets. Notes that were played at different times are suddenly shifted to play at the same time. When this sequence is played, it doesn’t sound right.
The timing issue seems to have to do with the triplets — those are the sequences it keeps corrupting. I can drag the notes around to fix the timing issue after the fact, but it seems I’d have to do that every single time, which would be quite tedious, especially if I’m going to experiment a lot with this sequence.
Why is GarageBand messing up the timing of my notes when I do simple operations like copying and pasting, or moving a sequence to a new location?
What am I doing wrong here? Is there some obvious mistake I’m making that I’m just not seeing?
I tried experimenting with different settings, and I even see some that include the word “triplet”, but no matter which setting I’ve tried, similar problems still occur — they just show up in different ways, but it always seems to affect the triplets. I think this problem is being caused by using the groove track, but if I disable that feature, then I lose the benefits it provides.
I don’t know enough about music or music software yet to discern the best way to remedy this problem. Was I not supposed to play something with triplets if I’m trying to sync it with drums? Honestly I didn’t even know what a triplet was till I saw the number 3 appearing above my notes, and Rachelle informed me that it means I played a triplet. I just played what intuitively sounded okay to me.
Apparently these are 8th note triplets, meaning that three 8th notes are being played in the span of what would be equivalent to a quarter note (one quarter note = two 8th notes).
Perhaps what’s happening here is something of a mathematical problem. It looks like the quantization of notes is based on the division of powers of two (1/2 note, 1/4 note, 1/8 note, 1/16 note, etc), and maybe the software is rounding to the nearest interval, and it can’t keep the correct alignment of those triplets when it does this.
It seems that no matter what settings I use for the groove quantization effect, it still shifts some of my notes enough to make it sound off.
I don’t know enough about music, GarageBand, or groove tracks to know what to do about it, so I thought I’d blog about it, and maybe someone who’s much more musically inclined than I am could explain what’s going on and how I should remedy it.
Perhaps one solution would be to disable the groove track feature till the very end, and then I could just do one final editing pass to correct any notes it messes up. But I like having this feature turned on along the way — it’s easier for me to tell if a combination of instruments sounds good if their timing is nicely synced. Even between different loops in GarageBand, this seems to make a difference.
I guess another option would be to re-engineer my melody so it doesn’t include any triplets, but that seems a bit lame to me. If I can stumble upon this as a newbie in writing my first melody, and if there’s actual notation for it, I have to imagine it’s a valid way to construct music. It sounds reasonably okay to me when the triplets are played, so why not keep them? It even meshes nicely with the drum track.
The part I find the most frustrating is that I can get everything to the point where it sounds good, and I’d love to be able to copy and paste those segments exactly as is (either on the same track or a different track), but whenever I do that, GarageBand corrupts not only the copy but also the original. I can fix those changes after the fact, but that requires a lot of extra checking and editing, which I’d rather not do — surely that isn’t how someone with more experience would solve this problem, is it?
The benefit of running into this problem is that I know I’m going to learn something new as a result. Until I got stuck here, I didn’t even know what a triplet was, nor did I know anything about groove tracks.
Despite this little snag, I’m still enjoying this music trial immensely. Diving in and composing some songs using GarageBand loops was a great place to start, and learning to incorporate basic effects like echo and reverb was a nice step beyond that. It was fun to do this a few times, but I’m already feeling limited by what I can do with the built-in loops in GarageBand. I can create something that sounds okay, but I don’t feel like I’m really able to express myself with those tools. The songs I created thus far were valuable learning experiments, but I feel very constrained creatively. I’d like to explore new avenues and build new skills, so I can have more options for expressing myself through music.
I figured the next step would be to try playing my own sequences of notes. I’m sure I have a lot to learn when it comes to composing sequences that sound good together, but I think if I can figure out how to do that, then I can build out a complete song. I’m not worried about being good at it — I just want to understand the process that I have to go through in order to create a complete song from start to finish that incorporates at least a few segments I composed. I can focus on getting good later. For now I just want to learn the how-to.
One really cool development has happened, one that took me by surprise. When I began this trial, some of my musician friends gave me advice and suggestions based on the assumption that I’d be learning how to take “the music in my head” and input it into a computer. I couldn’t relate to that because I didn’t hear any songs in my head, other than what I’d heard someone else play. I certainly wasn’t hearing any original tunes in my mind.
But I could understand this in the broader sense of having inspired ideas come to me. This happens all the time with my writing. An idea shows up, and I feel compelled to write about it (as already explained in How I Write). With all the experience I have as a writer, it’s very easy for me to take such an inspired idea and express it through words. If it’s a bigger idea, then I know how to express it in the form of a workshop. But I have no experience expressing these inspired ideas through music. I wondered if those musicians were implying that I was somehow going to figure out a way to express these same ideas through music.
Well, a few days ago, I actually started hearing music in my mind — just a small snippet or two, but it seemed to be original — nothing I recall hearing elsewhere. I still don’t know how to get those sequences out of my head and into the computer , but I imagine that with time and practice, I can develop the skills to do that. Even so, I think it’s really neat that musical inspiration is beginning to show up for me already.
These musical idea waves come to me in essentially the same fashion as my article ideas do, but the energy signature is different. I can easily tell that what’s showing up is musical in nature — I wouldn’t confuse it with the inspiration for a new article. This is exciting to me because if I’m able to tap into a similar abundant source of inspiration for musical ideas as I’m able to do with articles, it means that I could potentially write lots of music over time. It’s just a matter of building out my skills.
I’m well aware that my skill set isn’t yet a good vehicle for taking those musical ideas that show up and getting them into the computer, but this is a long road. With persistence I can get there eventually, just as I did with learning to program, to write, and to speak. I love the idea of building the skill set to express inspired ideas through yet another medium, so when the inspiration for a new song shows up, I could serve as a proper channel for getting that idea into the physical world.
There’s a part of me that wants to keep tacking on qualifiers to everything I do in this beginner phase with words like, “I know I suck at this, so…” or “Don’t expect this to be any good, but…” I think it’s time for me to stop doing that. It’s true that I’m a beginner, and my current skill level surely reflects that, but deep down I know I’m going to become really good at music. It’s only a matter of time.
I’ve worked hard to gain the ability to express inspired ideas through other forms over a period of many years. I started learning computer programming in 1981, but I didn’t see my first computer game published till 1994. I began writing articles in 1999, but I didn’t really start making a good living as a writer till 2006. I joined Toastmasters in 2004 and invested a solid 5 years in gaining comfort with public speaking, but I didn’t start doing my own workshops till 2009. At least these cycles are getting shorter, partly because I’ve gotten better at learning how to learn quickly. Sharing my journey along the way, for instance, helps me learn faster.
Even though I’m only on Day 11 of my music trial, I’m enjoying it enough that I already feel committed to this path of skill building for the long haul. I expect to stick with this line of development well beyond the 30 days. It doesn’t mean I’ll be blogging about it forever, but I do intend to keep working on it.
The benefit of having a long-term ambition is that it accelerates skill-building, especially during the beginner phase. We can acquire new skills much more quickly if we expect to still be using them 5, 10, or 20 years down the road.
I genuinely expect that 5 years from now, I’ll be writing music in some fashion. It’s possible that some new interest might grab my attention between now and then, but if I hold this expectation now, I’m able to learn a lot faster. The most important factor in learning is to have a strong purpose. I feel there are some ideas I can express through music that I can’t express through other media, and I wouldn’t feel good about leaving that channel forever unexplored.
I see new skills as adding to each other, not replacing each other. So I don’t expect to un-become a writer or speaker and shift to being a musician. I want to explore and integrate a new avenue for creative expression while continuing to use and further develop my other skills. Learning music won’t be the end either. There will be other forms to explore beyond that as well.
So I’m going to do my best to stop playing the helpless newbie card. It doesn’t mesh with how I see myself. I may be new at this, but internally I’m already feeling the mastery vibe calling to me. It’s like I’ve already done it, and I’m just waiting for physical reality to narrate the story of how I got from A to B.
So if you catch me doing the qualifying thing when I share my musical progress henceforth, please give me a good smack upside the head. If I’m going to master the ability to express inspired ideas through music, however long that takes, I might as well start getting comfortable where this path will lead. I want to get used to feeling what it will feel like to be able to competently express myself through music. I don’t have to wait 5 years to start practicing that aspect.
And if GarageBand can’t handle my triplets, then its days in my simulated reality are numbered! 😉