DYNOMIGHT ABOUT RSS SUBSTACK

My stupid noise journey

My stupid noise journey

Jun 2023

Comments at substack.

Interested in how to be a big dumb idiot and over-complicate things and waste time and money and endure tons of stress and some real physical pain all by thinking that you’re cleverer than you actually are? (No?)


Looking back, I was always sensitive to noise but never realized it was the cause of my little bursts of stress and distraction. That link became clear a couple of years ago when I moved into a new place and discovered that my upstairs neighbors had a lifestyle composed principally of stomping. And that there was noise from the nearby park every day from 6 AM to 2 AM. And that my sanity was retreating.

Some friends, trying to be helpful, asked “why don’t you move?” and “why don’t you get noise-canceling headphones?

Why don’t I? Because I am scientific and rational. Since this was so important—the noise extinguishing what glimmers of joy I might otherwise have wrung from life—I decided to use my big juicy brain and work from first principles.

My research said active noise cancellation wasn’t the answer. That worked well for steady noise like aircraft engines, but I faced low-frequency, intermittent noise. What I needed was passive noise isolation—more molecules between the noise and my eardrums.

So I bought noise-blocking earmuffs like construction workers use, along with soft moldable silicone earplugs. I proudly blogged about how cheap and effective these were when used together.

Time went by. They were very quiet! But:

  1. They made conversation impossible.

  2. The earmuffs look really stupid.

  3. There’s no way to play music or white noise. (Earbuds inside the earmuffs don’t work because they get knocked out of position and then can’t be adjusted.)

  4. After 20-30 minutes, the earmuffs become uncomfortable and then escalate into ever-higher levels of pain.

Because of all this, I didn’t use them much and mostly just suffered the noise. I’d often read in hopes of finding a better solution. I found out about the ABCC1 gene that determines if you have dry or wet earwax. And I rediscovered the precedence effect where a sound that’s mostly coming from one direction can seem to come from a different direction. And I created a mathematical model of bluetooth speaker usage. These did not solve my problem.

Also around this time, I used caulk to seal up my windows, which maybe helped a bit. And I tried noise-canceling headphones on planes a couple times, which were great, but planes aren’t stomping neighbors.

Then I went away for a few months. Some of the places I stayed had moments of actual silence. Upon my return—my neighbors having meanwhile redoubled their commitment to Stomp Life—I realized I needed to move. But in the meantime, I still needed a solution.

I did yet more research. Should I meditate? Build some kind of isolation chamber? Create white noise with a frequency spectrum matched to stomping?

I kept coming back to the sad reality that nothing would work as well as the damn earmuffs. If only they weren’t so painful! But some people apparently wear them all day, so maybe I could get used to them. I spent weeks trying to persist through the discomfort with little result. When that didn’t work, I decided to harden up and endure the earmuffs until the pain became truly unbearable. Sometimes it would be so intense that I’d have a headache for an hour after. Still I couldn’t get past 45 minutes.

Maybe I needed different earmuffs? I learned about earmuffs made for shooting which have headphones inside, along with active electronics to play outside noises for “situational awareness”. I bought a pair, and then two more. All were horrendous. They were more uncomfortable, and repeated all noises, including the ones I wanted to block. The audio was awful and had a constant electronic squeal even more annoying than the original noise.

I tried kits for making custom-molded earplugs. One hardly blocked any noise. Another had a small piece immediately break off deep inside my ear, resulting in legendary good times trying to remove it with a screw.

Then I learned about earmuffs made for pilots. They were two orders of magnitude more expensive, yes, but maybe they would work better? What’s a couple thousand thousand dollars compared to my sanity?

I did yet more research on passive and active noise cancellation. While most people said active noise cancellation only works for steady sounds, a few grumpy people argued it was fine, as long as the sounds were low-frequency. The debate hinged on technical aspects of phase detection and the relative speeds of light and sound. Reading it, I couldn’t tell who was right.

Gradually it dawned on me that I didn’t understand anything. Not the physics of sound, not how ears and brains perceive it, not how active noise cancellation works, and not even how passive noise blocking works. Nothing.

And since I didn’t understand anything, perhaps I should discard theory and just try stuff? In particular, why hadn’t I tried active noise-canceling headphones?

So I bought some. The ones everyone has, that are advertised to be quiet and comfortable.

And they were… incredible. The effect on my low-frequency intermittent sounds was far better than the earmuffs. They were so good that when testing them I had to repeatedly check that the noise didn’t happen to stop the moment I put them on. They have integrated speakers and microphones, can be taken off in a second for conversation, and are comfortable enough to wear for hours.

Never satisfied, I tried some other models. But, no—the model that everyone says are the best were indeed the best.


So let’s review: I had a problem. The ultimate solution to my problem was to do the most obvious possible thing. But I convinced myself that wouldn’t work and spent two years trying everything else.

Or, OK, it’s worse than that.

I dismissed the obvious thing because I thought I was using first-principles thinking. But what were those first principles? Certainly, I don’t understand how sound works beyond some vague idea of waves of molecules smashing into each other. And I don’t understand how active noise cancellation works, beyond the broad concept of using microphones and speakers to create destructive interference.

I guess those vague ideas are correct. But they are extremely superficial, so much so that they barely predict anything. Not only was I too ignorant to work from first principles, I was too ignorant to tell if other people knew what they were talking about.

And at the same time, there was no need for first-principles thinking because testing was very fast and cheap. (At least if you return stuff that doesn’t work.) You know within seconds if noise is being blocked and within hours if you’re comfortable. It’s absurd to spend countless hours theorizing when experiments are so easy.

How? I spent years enduring noise and some weeks borderline torturing myself trying to acclimate to the earmuffs, all while I had no idea what solutions would work, and I hadn’t tried all the easy-to-try stuff. How could I be so stupid?

One answer is that I was using an overly broad heuristic. I generally think most people trust self-experimentation too much for mundane life problems. If you want to gain muscle, doing your own research on different exercise programs is rough going because the feedback loops are long and convoluted. So I generally try to “find the best practices and follow them”. That’s a good heuristic! But this particular problem had no (legible) best practices and had a very short feedback loop.

Another answer is that it’s easy to overlook your ignorance when dealing with familiar things. Sound is complicated, yes, but it’s not obviously complicated. People hear stuff all the time but rarely report being confused by it. Articles on noise blocking don’t have “WARNING: ACOUSTIC ENGINEERING IS A WHOLE FIELD” at the top. It’s easy to read a superficial explanation and not notice that there’s no content in it.

And finally, there’s “cleverness”. I’m naturally drawn to unusual solutions. A world where every problem was best solved by doing the obvious thing would be, to me, a dull world. I want to believe that there’s free utility out there, that you can make different/weirder choices and grab it. So did I let this hope distort my view of what was?

I’m not sure. But imagine an alternate history where I had bought noise-canceling headphones in the first step, but they were terrible and I suffered with them for two years before finding some weird solution that really worked. Would I feel as stupid in that branch of the timeline as I do in this one? Of course not. So if this bias exists, I’ve decided to give myself a pass on it for now.

Comments at substack.

new dynomight every thursday
except when not

(or try substack or rss) ×
OK, I can partly explain the LLM chess weirdness now

("make LLMs play better with one weird trick")

We recently talked about a mystery: All large language models (LLMs) are terrible at chess. All, that is, except for gpt-3.5-turbo-instruct, which for some reason can play at an advanced amateur level. This is despite the fact that this model...

Something weird is happening with LLMs and chess

are they good or bad?

A year ago, there was a lot of talk about large language models (LLMs) playing chess. Word was that if you trained a big enough model on enough text, then you could send it a partially played game, ask it...

The real data wall is billions of years of evolution

Careful with those human analogies

Say you have a time machine. You can only use it once, to send a single idea back to 2005. If you wanted to speed up the development of AI, what would you send back? Many people suggest attention or...

Fahren-height

(celsi-pour?)

The Internet is well into middle-age, and yet doesn’t seem to have answered humanity’s most pressing question: If you pour boiling hot water from various heights, how much does it cool in flight?

Fancy math doesn't make simple math stop being true

on butts and instrumental variables

What are you supposed to do when someone disagrees with you using a bunch of math you can’t understand? I’ve been thinking about that recently because of the NordICC colonoscopy trial. It took 85k Europeans aged 55-64, invited a third...

Are language models good at making predictions?

politics more than science

To get a crude answer to this question, we took 5000 questions from Manifold markets that were resolved after GPT-4’s current knowledge cutoff of Jan 1, 2022. We gave the text of each of them to GPT-4, along with these...

Grug on diet soda and autism

why bad and why so much promote

grug try to not yell about bads in science news too much because why make same yells over and over? and grug have no new learns, just feel people maybe sometimes not use old learns and grug family often plead...

The second system problem

Building a safe AI ≠ preventing all unsafe AI

In The Vulnerable World Hypothesis, Nick Bostrom imagines we found a technological "black ball"—say a way to make a nuclear weapon with just some glass, some metal, and a battery. He concludes that society in our current "semi-archic default condition"—could...

I still think it's very unlikely we're observing alien aircraft

They'd have to be messing with us.

Some suggest there might be alien aircraft on Earth now. The argument goes something like this: A priori, there’s no reason there shouldn’t be alien aircraft. Earth is 4.54 billion years old, but the universe is 13.7 billion years old,...

Why didn't we get GPT-2 in 2005?

We probably could have

The ancient Romans were never great at building ships and never tried to explore the Atlantic. The basic reason seems to be—why bother? The open ocean has no resources and is a vast plane of death. But imagine that in...

First-principles on AI scaling

How likely are we to hit a barrier?

It's hard not to feel blinkered by recent AI progress. Every week there seems to be an AMAZING NEW SYSTEM with UNPRECEDENTED CAPABILITIES. It's impossible not to wonder what the future holds. Until recently, I thought progress was so dizzying...

Your friend the language model

The world is running out of khakis

I originally wrote this as part of a much longer post on LLM scaling laws and possible barriers/trajectories for progress. The idea was to provide the minimal background necessary to understand all that stuff. But in retrospect, many people probably...

Winner take all science

Is it helpful for things to work this way?

By the early 1950s, it was known thanks to people like Miescher, Levene, and Chargaff that genes were carried by long polymers in the cell nucleus. It was also known that those polymers had a sugar-phosphate backbone and were composed...