Daniel sent us this one — he's asking about small speakers, specifically whether something like the Sony SRS-XB100 can actually deliver decent sound when it fits in a backpack. There's a tension here between physics and marketing that's worth unpacking. A 46-millimeter driver shouldn't be able to fill a room, yet products claim they do. Are they lying, or is our intuition about what makes good sound just incomplete?
That's the question that got me down a rabbit hole this week. Because you look at the SRS-XB100 — 274 grams, 46-millimeter full-range driver, 5-watt Class-D amp, sells for around 58 dollars on Amazon right now — and it's the best-selling sub-60-dollar speaker on the platform. Twelve thousand reviews, four-and-a-half stars. People clearly like it. But the physics says a driver that small can't move enough air to produce real bass. So what's actually happening?
Before you get to what's happening, let's define the category. When we say ultra-portable, what are we actually talking about?
Ultra-portable in this context means under 300 grams, under 10 centimeters in any dimension, battery-powered, and typically a single full-range driver — sometimes with a passive radiator, sometimes without. The SRS-XB100 uses a single 46-millimeter driver with a passive radiator and a 5-watt amp. That's the template. You see variations — the JBL Clip 4 is similar, the Bose SoundLink Micro pushes it to 99 dollars — but the basic recipe is the same.
The basic recipe is also the basic problem. Small driver, small enclosure, limited power. Which brings us to Hoffman's Iron Law.
Hoffman's Iron Law. This is the thing that governs everything we're going to talk about. It's not actually a law of physics in the Newton's Second Law sense — it's an engineering trade-off identified by J. Anthony Hoffman back in the 1960s. For a given driver size, you can optimize for two of three things: deep bass extension, high efficiency, or a small enclosure. You cannot have all three.
The good-fast-cheap triangle, but for speakers.
And ultra-portable speakers pick small enclosure and efficiency every single time. They sacrifice bass extension. That's the trade-off. A 46-millimeter driver has a cone area of about 16.6 square centimeters. To produce a frequency of 80 hertz at 85 decibels SPL at one meter, you need to displace roughly half a cubic centimeter of air. The SRS-XB100's passive radiator helps — it's essentially a weighted diaphragm that resonates at low frequencies, extending the effective bass response — but the driver's linear excursion, the Xmax, is probably less than 3 millimeters.
You don't know the exact number?
Sony doesn't publish the Xmax for this driver. I dug through the spec sheets. It's not there. But based on the driver size and the enclosure volume, the theoretical maximum SPL at 80 hertz works out to about 88 decibels. That's listenable. It's not room-filling. And that's at the limit — in practice, with distortion, you're probably getting usable output at 85, 86 decibels.
When the marketing says "fills a room with rich sound," we're talking about a very small room.
We'll get to room size. But first, let me address the thing that makes this speaker work at all, which is DSP. Digital signal processing. Sony has something they call the Sound Diffusion Processor and a mode called MEGA BASS. These are not just brand names — they're doing real psychoacoustic work.
That word, psychoacoustics — that's where the illusion lives.
The missing fundamental effect. This is one of the most elegant tricks in audio. If you play a harmonic series — say, 120 hertz, 180 hertz, and 240 hertz — our auditory system reconstructs the fundamental frequency of 60 hertz. Your brain literally hears a 60-hertz tone that isn't there. The speaker can't produce 60 hertz. It physically cannot. But it can produce 120 and 180 and 240, and your brain fills in the fundamental. That's how a 46-millimeter driver creates the perception of bass.
The speaker is not lying to you. Your brain is lying to you, and the speaker is just providing the raw materials.
That's one way to put it. The other way is that evolution spent millions of years optimizing our auditory system to extract information from incomplete signals, and DSP engineers figured out how to exploit that. The missing fundamental is why you can hear bass on a smartphone speaker. It's why a tiny Bluetooth speaker doesn't sound like a tin can. It's genuinely clever engineering.
It also explains why the frequency response curve of these speakers looks the way it does.
If you pull the actual measured frequency response of the SRS-XB100 — and Rtings published a full set of measurements for this speaker — you see exactly what the DSP is doing. The response is down about 10 decibels at 100 hertz. Then there's a 5-decibel hump centered around 150 hertz. That hump is the fake bass region. It's where the harmonics live. Then it's reasonably flat from about 500 hertz to 5 kilohertz — that's the midrange where voices and most instruments sit — and then you get a roll-off above 14 kilohertz.
The "rich EQ" claim is essentially a 150-hertz bump that tricks your ear into hearing warmth, and the actual deep bass isn't there at all.
That bump masks some midrange clarity. It's not a neutral speaker. But neutral isn't what most people want from a 58-dollar backpack speaker. They want something that sounds full and pleasant, and the 150-hertz hump does that job. It's the musical equivalent of adding a little sugar to tomato sauce — it's not authentic to the raw ingredients, but most people prefer it.
The tomato sauce of audio engineering. I'm going to let that sit there.
I stand by it. But let's talk about the room-filling claim, because that's where the numbers get concrete. The SRS-XB100 has a 5-watt Class-D amplifier. At 85 decibels efficiency, to reach 95 decibels SPL — which is roughly the threshold for "loud enough to fill a 20-square-meter room" — you need 10 watts. You need double the power. The SRS-XB100 maxes out at about 92 decibels at one meter. That's fine for a bedroom or a small office. In a 40-square-meter living room with ambient noise from a refrigerator or street traffic, you'll hit distortion before you hit satisfying volume.
The real answer to "how much space can this fill" is about 20 square meters, give or take.
For listenable audio at a comfortable volume, yes. You can push it further if you're willing to accept distortion or if the room is quiet. But 20 square meters is a reasonable ceiling. Compare that to a JBL Flip 6 — 30 watts, 97 decibels max SPL. That speaker can fill a 60-square-meter room. But it's three times the weight and roughly double the price. You're trading portability for headroom.
That trade-off is the whole game. The market has settled on a set of compromises that work for most people most of the time. 5 watts, 46 millimeters, 300 grams, 60 dollars. That's the formula.
It's worth noting that watts are probably the most misunderstood spec in portable audio. People see "5 watts" and think it's weak, or they see "20 watts" on a cheap speaker and think it's powerful. Watts determine maximum SPL, but driver size, enclosure design, and DSP determine sound quality. A well-tuned 5-watt speaker with a good passive radiator can sound dramatically better than a 20-watt speaker with a tiny, poorly designed driver and no DSP. The wattage number alone tells you almost nothing.
More watts means more potential volume, but potential is doing a lot of work there.
Continuous SPL at one meter is the spec that actually matters, and almost no manufacturer publishes it. They publish peak power, which is a number that looks good on a box and means nothing in practice. Peak power is the maximum instantaneous power the amplifier can deliver before it melts. Continuous power — sometimes called RMS power — is what the speaker can sustain. The SRS-XB100's 5 watts is continuous. A cheap speaker claiming 20 watts peak might only deliver 3 watts continuous. The number on the box is marketing, not engineering.
We've established that the SRS-XB100 is a master of illusion — it uses DSP to fake bass, it trades absolute volume for portability, and it works within the hard constraints of Hoffman's Iron Law. But here's the twist I want to get to: that same set of limitations makes it surprisingly good for something you might not expect.
This is where the narrative flips. The SRS-XB100's frequency response — boosted at 150 hertz, flat through the midrange, rolled off above 14 kilohertz — is actually close to ideal for human speech. The fundamental frequency of the human voice ranges from about 85 hertz for a bass voice to 255 hertz for a soprano. The formants — the resonant frequencies that give vowels their distinctive character and carry most of the intelligibility — sit in the 500-hertz to 4-kilohertz range. The SRS-XB100 is dead flat through that entire range.
The speaker that can't reproduce deep bass for music is accidentally optimized for the frequency range where voices live.
Not just accidentally optimized — the 150-hertz boost actually adds warmth to male voices without muddying the consonants. Consonants are mostly high-frequency transients that live above 2 kilohertz. The boost is low enough that it doesn't interfere with clarity, but it adds body. A completely flat speaker can sound thin on spoken word. The SRS-XB100's DSP curve adds presence. For podcasts, this 58-dollar speaker is arguably better than a 200-dollar bookshelf speaker paired with a subwoofer, because the subwoofer's 40-hertz rumble is irrelevant for speech and can actually be distracting.
That's a counterintuitive claim. The cheap portable speaker beats the proper hi-fi setup for spoken word.
For spoken word specifically. For music, the bookshelf speaker wins every time — it has actual bass extension, better stereo imaging, more detail. But if your primary use case is podcasts and audiobooks, the SRS-XB100's limitations become features. The natural roll-off above 14 kilohertz doesn't matter because speech has almost no energy up there. The lack of deep bass doesn't matter because voices don't go below 85 hertz. And the midrange-forward tuning makes voices sound present and clear. It's the audio equivalent of a chair designed for one specific posture — uncomfortable for lounging, perfect for sitting at a desk.
The Herman Chair of portable speakers.
I'm not sure if that's a compliment, but I'll take it. The broader point is that "good sound" is not a universal property. It depends on what you're listening to. A speaker that measures poorly on a music-oriented frequency response curve can measure excellently on a speech-intelligibility curve. The SRS-XB100 is a case study in how the same engineering trade-offs that limit music performance can enhance voice performance.
Which means the question we should be asking isn't "is this speaker good" but "is this speaker good for what I actually listen to.
Most people don't ask that question. They look at star ratings and price and maybe wattage, and they assume all speakers are trying to do the same thing. They're not. The SRS-XB100 is optimized for casual listening in small spaces — background music while you cook, podcasts while you get ready in the morning, audiobooks on a nightstand. It's not trying to be a party speaker or a hi-fi reference monitor. It's trying to be pleasant and portable and cheap. And on those terms, it succeeds.
Let's talk about the broader market, because the SRS-XB100 doesn't exist in isolation. What are the best-reviewed ultra-portable speakers right now, and how do they compare?
As of May 2026, there are four products that dominate the category based on aggregated reviews from Wirecutter, Rtings, and Amazon ratings with at least 500 reviews and four stars or above. The Sony SRS-XB100 is 58 dollars, four-and-a-half stars, 12,000 reviews. The JBL Clip 4 is 70 dollars, four-point-four stars, 8,000 reviews. The Bose SoundLink Micro is 99 dollars, four-point-three stars, 5,000 reviews. And the Anker Soundcore 3 is 35 dollars, four-and-a-half stars, 20,000 reviews.
The Anker has more reviews than the others combined.
Anker's distribution is enormous, and at 35 dollars it's an impulse purchase. The Soundcore 3 uses dual full-range drivers and a passive radiator — it's actually a stereo speaker in a single enclosure, which helps with perceived spaciousness. But its DSP is less sophisticated than Sony's. The bass illusion isn't as convincing. For pure value, it's hard to beat. For sound quality in the bass department, the Sony wins.
The Bose at 99 dollars — what do you get for the premium?
The flattest frequency response of the group. The SoundLink Micro has the most accurate midrange and the least DSP coloration. If you're listening to acoustic music — folk, jazz, classical — the Bose is the best choice. It's the most honest speaker. But it costs nearly twice what the Sony costs, and it's heavier. The JBL Clip 4 splits the difference — 70 dollars, good bass for its size, the carabiner clip is useful, and it's built like a tank. JBL's tuning tends to emphasize the upper bass around 100 to 200 hertz, which gives it a punchy sound that works well for pop and rock.
The decision tree is: if you're buying for spoken word, get the Sony or the Anker. If you're buying for music, spend the extra for the JBL or the Bose.
If you're buying for music and you can stretch your budget and your backpack space, skip the ultra-portable category entirely and get a JBL Flip 6 or a UE Boom. Those are in a different weight class — 500 to 600 grams, 20 to 30 watts — but the jump in sound quality is significant. The ultra-portable category is about the best sound you can get while still being able to forget the speaker is in your bag. Once you cross the 300-gram threshold, you're making a different kind of trade-off.
There's a theoretical question I want to explore. What would a perfect ultra-portable speaker look like if you removed the market constraints and just optimized for physics?
I love this question. The constraints are: it has to fit in a 10-centimeter enclosure, it has to be battery-powered, and it has to be portable — let's say under 400 grams. Given those constraints, the theoretical optimum in 2026 is probably a 60-millimeter driver — that's the largest driver that can fit in a 10-centimeter enclosure with room for a passive radiator and electronics — paired with a 10-watt Class-D amplifier, dual passive radiators, and a 2,000 milliamp-hour battery. That configuration could achieve about 95 decibels SPL at 80 hertz at one meter. That's enough to fill a 30-square-meter room with listenable audio.
What would that cost?
Probably around 120 dollars. The driver would be custom, the passive radiators would need precise tuning, the amplifier would need better thermal management. And it would weigh closer to 400 grams than 300. No one makes this speaker because the market has spoken. People want 60 dollars and 300 grams. They don't want 120 dollars and 400 grams, even if the sound is dramatically better. The market has chosen the 46-millimeter, 5-watt compromise.
The market is not always optimizing for the thing you care about. It's optimizing for what the median buyer cares about.
The median buyer cares about price first, weight second, and sound quality a distant third — at least in the sub-100-dollar category. Once you cross 100 dollars, sound quality starts to matter more. But in the ultra-portable space, the 58-dollar Sony is the reference point, and everything else is measured against it.
Let me give you a framework for evaluating these speakers, because I think that's what the prompt is really asking for. Not just "which one is good" but "how do I think about this category.
Four things to check. First, look for an actual measured frequency response — not the manufacturer's claimed response, which is always smoothed and idealized. Rtings and Audio Science Review publish real measurements. For podcasts, you want less than 5 decibels of variation from 200 hertz to 5 kilohertz. For music, you want less than 10 decibels of variation from 100 hertz to 10 kilohertz.
Second thing: ignore peak power ratings. Look for continuous SPL at one meter. If the manufacturer doesn't publish it, that's a red flag. You can estimate it from wattage and driver size, but it's better to find a measurement.
Third: for podcasts, a speaker with a slight boost in the 150 to 200 hertz range is actually better than a perfectly flat speaker. That boost adds warmth to voices without hurting clarity. Don't chase flatness if you're mostly listening to people talking.
Fourth: for music, you need at least 10 watts and a driver larger than 50 millimeters to get satisfying bass. Below that threshold, you're getting DSP-generated bass illusion, not actual bass extension. The illusion can be convincing, but it's not the same thing.
Here's a rule of thumb for room size. A 5-watt speaker like the SRS-XB100 is good for rooms up to about 20 square meters. A 10-watt speaker like the JBL Flip 6 is good for up to 40 square meters. A 20-watt speaker like the UE Boom 3 is good for up to 60 square meters. Beyond that, you need a dedicated subwoofer or a larger form factor. These aren't hard limits — you can push a 5-watt speaker in a 30-square-meter room if it's quiet and you're sitting close — but they're useful heuristics.
The other variable that doesn't get enough attention is the room itself. A hard-surfaced bathroom with tile and glass will make any speaker sound louder and brighter because of reflections. A carpeted bedroom with curtains will absorb high frequencies and make the same speaker sound muffled. The speaker's acoustic environment matters as much as the speaker's specs.
Putting a small speaker in a corner can boost perceived bass by 3 to 6 decibels because of boundary reinforcement. Putting it on a hollow surface like a wooden table can add unwanted resonance. The SRS-XB100 has a rubberized base that helps with this, but it's still a factor. Most people just plunk the speaker down wherever and wonder why it sounds different from room to room.
The speaker is half the system. The room is the other half, and the room is the part nobody thinks about.
That's true of all audio, but it's especially true of small speakers because they have so little headroom. Every decibel of bass reinforcement from room placement is precious when your driver is 46 millimeters.
Let's talk about the future for a moment. Solid-state batteries are supposed to start showing up in consumer electronics in 2027. Higher energy density in the same form factor. What does that do to this category?
It's interesting. A solid-state battery with double the energy density of current lithium-ion cells would let you put a 10-watt amplifier in a 300-gram speaker without sacrificing battery life. You could have SRS-XB100 portability with JBL Flip 6 performance. But the driver size constraint doesn't change. You're still limited to about 50 millimeters in a 10-centimeter enclosure. So you'd get more volume and less distortion at high volumes, but the bass extension wouldn't improve dramatically. Hoffman's Iron Law still applies.
The battery improvement gives you headroom, not extension.
And the question is whether the market wants headroom. The SRS-XB100 already gets 16 hours of battery life. Most people don't need more than that. The extra power would mostly be useful for outdoor use — picnics, camping, the beach — where ambient noise is higher and you need more volume. But for indoor use, 5 watts is sufficient for the rooms these speakers are designed for.
The other thing on the horizon is the blurring line between portable speakers and smart speakers. The next version of the SRS-XB100 might have a voice assistant and Wi-Fi streaming. Does that change the acoustic design?
It shouldn't, but it might. Adding a microphone array for far-field voice pickup creates new constraints on enclosure design. You need to isolate the microphones from the driver's vibration. You need to account for the speaker's own output when doing echo cancellation. And the DSP that handles voice pickup competes for processing power with the DSP that handles sound enhancement. It's a non-trivial engineering challenge. My guess is that the first generation of voice-assistant ultra-portables will have compromised audio quality because the acoustic design will be secondary to the smart features.
The smart speaker that forgot to be a good speaker. We've seen that movie before.
The original Amazon Echo was not a good speaker. It took three generations before the Echo Studio was competitive with dedicated speakers at its price point. The ultra-portable category will probably follow the same trajectory — a dip in audio quality when the smart features arrive, then a gradual recovery as the engineering matures.
The SRS-XB100, right now, might be a sweet spot — it's optimized for the one thing it does, without the distraction of being a smart device.
It's a speaker that knows what it is. No microphone, no Wi-Fi, no voice assistant. Just Bluetooth, a driver, a passive radiator, and some clever DSP. There's an honesty to that. It's not pretending to be anything other than a small speaker that sounds pretty good for 58 dollars.
The lo-fi girl of portable audio.
You've used that one before, but it fits. There's a purity to a single-purpose device. No notifications, no updates, no privacy policy to read. You pair it and it plays sound.
On the privacy point — a Bluetooth speaker with no microphone is a privacy-positive device. It can't listen to you because it literally has no way to listen to you. In a world where everything has a microphone, that's almost a feature worth advertising.
The SRS-XB100 does have a microphone, actually — it supports speakerphone calls. So it's not fully privacy-positive. But the point stands for the category. There are ultra-portables without microphones, and for people who care about that, it's a meaningful differentiator.
I stand corrected. The lo-fi girl has a phone line.
The lo-fi girl takes calls but doesn't initiate them. I can live with that.
Where does this leave us? We've got a framework for evaluating small speakers: check the measured frequency response, ignore peak power, understand that a 150-hertz boost is good for podcasts, and know that 5 watts gets you about 20 square meters of coverage. We've got a market landscape: Sony for DSP sophistication, Anker for value, JBL for durability and punch, Bose for accuracy. And we've got a theoretical ceiling: 60-millimeter driver, 10 watts, 120 dollars, 400 grams — a product nobody makes because the market prefers cheap and light.
The thing I keep coming back to is how much of this category is about managing expectations. The SRS-XB100 is not a great speaker in absolute terms. A pair of 200-dollar bookshelf speakers will destroy it on every objective measure. But it's a great speaker for what it is — a 274-gram device that costs 58 dollars, fits in a jacket pocket, survives being dropped in a puddle, and makes podcasts sound warm and present. If your expectations are calibrated to the form factor, it's excellent. If you're expecting it to replace a home stereo, you'll be disappointed.
That's the central tension of the whole category. The marketing implies "big sound from a small box." The physics says "small box means compromises." The reality is "compromises that are engineered to be as pleasant as possible." And for most people, most of the time, pleasant is enough.
Pleasant is underrated. The audio world is full of people chasing perfect frequency response and sub-bass extension and soundstage width, and meanwhile millions of people are perfectly happy listening to podcasts on a 58-dollar mono speaker while they make breakfast. The SRS-XB100 is not for audiophiles. It's for everyone else.
Everyone else is a much bigger market.
By about a factor of a thousand.
The question I'm left with — and I want to leave our listeners with — is this: as solid-state batteries arrive and smart features creep in, will we see a 10-watt speaker in this form factor that pushes the acoustic envelope? Or will the market keep prioritizing weight and price, and the 46-millimeter, 5-watt template will persist for another decade because it's good enough?
My bet is on good enough. The SRS-XB100 is not the best possible speaker in its category. It's the best compromise for the largest number of buyers. And that's a harder engineering problem than building a perfect speaker. Building a perfect speaker is just physics. Building a compromise that millions of people love — that's design.
On that note — we want to hear from you. What should we cover next? We've been thinking about the physics of noise-cancelling headphones, or the real-world performance of spatial audio. Send us your weird prompts. We read them all, even the ones that make Herman sigh and reach for another research paper.
I don't sigh. I exhale with scholarly intent.
Of course you do.
Now: Hilbert's daily fun fact.
Hilbert: In the 1960s, a linguist documenting the Dyirbal language of northern Queensland discovered that the language has four genders and requires speakers to use entirely different vocabulary when talking to certain relatives — a system so complex that the field notes describing it ran to over 900 pages for kinship terminology alone, and the manuscript is still cited as one of the densest grammatical descriptions ever produced for a single language.
Nine hundred pages on who you're allowed to call what. Thanksgiving must have been intense.
This has been My Weird Prompts. I'm Herman Poppleberry.
I'm Corn. Find us at myweirdprompts.com or wherever you get your podcasts. If you enjoyed this episode, leave us a review — it helps other people find the show.
Until next time.
