Comment by lblack00 | Hacker News Reader

Comment by lblack00 | Hacker News Reader

Building an iOS app for metronome sequencing to get faster at playing guitar and reaching "shred" speeds at different subdivisions/time signatures in a single sequence. Planning on adding accuracy indicators and scoring so rushing or dragging can be easily identified when finishing a saved routine. I.e., some post-routine metrics.

I've been playing guitar for a little under 6 years and ran into the common problem among many intermediate guitarists fall into, which is stagnating into a plateau at a certain BPM.

The most effective solution I've found is to take the top speed hit playing a chunk of a lick and simply increase it 20-50 BPM past that limit, attempting one's best to stay in tempo. Regardless of how sloppy it sounds. Then roughly halve that increased addition of BPM, it will become relatively easier to play. For example, if you are stuck at 120 BPM, upping it to 150 BPM with sloppy attempts, then dropping it back down to 130-140 BPM.

I've gone cleanly from alternate picked 140 BPM triplets to 220 BPM triplets in two months after being stuck at 140 BPM for over a year with this method. Sometimes even hitting 280 BPM triplets when I have the focus and time for it.

Even then, I want a more consistent, and variable way of customizing a practice session using a metronome from a hobbyist perspective without using a DAW. With a simpler interface for doing so. As well as encourage with said method above for other guitarists in the pursuit of speed.

When I was 14 I would have wanted to "shred at 220bpm" but today I wouldn't get my wallet out for that. What I really would pay for is help getting into the pocket.

Anyone who can read a guitar tab can play the notes of "Superstition" by Stevie Wonder. But simply playing the notes against a metronome sounds mechanical -- the song only comes to life when you get the timings right (both the note attack and decay have to be timed for a "swing.") A good swing will practically force your audience to start dancing to your music -- it's magical! But it's very difficult to learn because regular metronome practice won't achieve it.

If you're measuring "rush and drag" against a straight metronome, could you also measure against a swung time, perhaps against timings extracted from in-the-pocket songs we know and love?

> When I was 14 I would have wanted to "shred at 220bpm" but today I wouldn't get my wallet out for that.

That's fair, essentially why I put "shred" in quotes originally is that shredding guitar isn't necessarily playing fast. You laid out a nice example with Superstition for that.

I don't see why that couldn't be implemented in some way (accenting specific notes and different sustain times).

What would be difficult is quantifying note attack exactly for XYZ's riff sections. I.e., what constitutes a relative baseline pick attack and the target pick attack. If we are using a float and define the "normal" attack as 0.5, then how do we know, for example, the first or fifth note in the iconic Superstition riff is 0.85? Is it empirical?

Either way, that is a lovely insight I will consider. Matching another guitarist's intonation down to a tee can be extremely difficult, but very rewarding.

> What would be difficult is quantifying note attack exactly for XYZ's riff sections.

At 60 BPM, mathematical quarter notes are 1000ms apart. But in a pocket groove, you may notice that every other quarter note is "late", or "swung":

  |o---o---o---o---|  (mathematical)
  |o----o--o----o--|  (swung)

If you load a groovy song in Audacity you should be able to see these inter-note delays.

Another factor that affects rhythm is note duration relative to the tempo -- you'll want to measure that too.

> Matching another guitarist's intonation down to a tee

I suspect if you study a few groovy songs you'll find there's just a slightly different note grid that's common to these songs (there could be more than one grid!) Teaching this grid (rather than teaching one specific song) will help the student learn to shift notes away from the the mathematical metronome placement. This skill will equip many of them to mimic the feel of their favorite artist by ear.

I see what you mean now. I interpreted note attack as pick attack, which is traditionally defined as how hard a given note is played with a pick.

Yeah, swing usually has uneven subdivisions. Funk is almost always syncopated, and depending on the style, can be a mix of syncopation and swing.

This goes back to how you noted simply using a metronome will give that mechanical, or even a soulless characteristic, to playing a piece which inherently has a soulful quality about it. And with respect to intonation, there's a lot more that goes into that than just timing it right (how hard fingers are pressed on the fretboard, the pressure between fingers holding the pick, the angle of the pick, where the pick strikes relative to pickups, the pick attack, accenting notes, etc.)

I do love this idea of being able to apply some "in the pocket"/"swing" deviation to a metronome sequence. I agree with you that it adds that magical musical quality that people would instinctively dance to.

Accuracy indicators for rushing and dragging are very useful, but equally interesting would be an indicator that checks for a consistent peaks, so every note is played with equal volume. A dream would be if your app can detect differences in sound (fingernails occasionally scratching the string, fretbuzz).

Is there any way to get notified when your app is done, or do you have a name for it already so we can search for it in a couple weeks?

Love those ideas, would be awesome for really nailing down consistent tones. Latter one would be fairly ambitious for me but I'll keep it in mind for future improvements.

Just published the waitlist[1], the app will be called ShredBlocks.

[1]: https://shredblocks.app/

My AI suggests a few existing algorithms to tackle the problem of comparing two sound curves - maybe this one is not too hard?

I'll check out that app when it's ready, good luck!

Thank you!

Reminds me of John Petrucci's Rock Discipline instructional video where he outlines exactly this technique on how to build up speed.

Great instructional video. First place I learned E natural minor with his scale fragments section.

Yes, not a new technique by any stretch of the means. AFAIK John Petrucci takes a less aggressive approach with raising BPM. Funnily, Shawn Lane goes into a very similar methodology >30 years ago[1].

[1]: https://youtu.be/dpLDN1QCkQM?t=112

https://news.ycombinator.com/item?id=45450565

Beautifully designed IMO.

Unfortunately not a sequencer, which would be a programmable set of a series of blocks with different tempos, time signatures, subdivisions, etc., played in single sequence.

When will it be done? Would like to try.

At least a couple more weeks. Hopefully less than a month out from now.

I have most of the UI done for sequencing. Workflows for speed building and metronome sequencing will be completely free, which is also a top priority for me to get out the door first.

Reply on news.ycombinator.com