TraviaTechPie Review

The Story

Here’s the thing nobody tells you about the number on the front of the box.

A dehumidifier rated at “50 pints per day” is not a machine that removes 50 pints per day. It’s a machine that removed 50 pints per day once, in a lab, in a room held at a specific temperature and a specific humidity, for a test that exists so manufacturers can print a number. Move that machine into a room five degrees colder and the number doesn’t drop a little. It falls off a cliff.

That’s not a scandal. It’s thermodynamics doing exactly what it’s supposed to do. But it means the first question you ask when shopping is almost universally the wrong one. People ask “how many pints do I need?” The question that actually determines whether the machine works is “what temperature is the room I’m putting this in?”

Everything else — capacity, tank size, brand, price — is downstream of that.

Why capacity ratings are really temperature ratings

A compressor dehumidifier is a small refrigerator that never closes its door. It runs refrigerant through an evaporator coil, chills that coil below the dew point of the incoming air, and water condenses on it and drips into a bucket. Simple, effective, and completely dependent on one condition: the coil has to be colder than the dew point of the air.

Warm air holds a lot of water. Cold air holds very little. So the same machine, doing the same mechanical work, extracts wildly different amounts of water depending on what walks into the intake. This is why the rating standard matters so much, and why it changed.

The US Department of Energy used to test portable dehumidifiers at 80°F and 60% relative humidity. It now tests at 65°F and 60% RH — a 15-degree reduction meant to better reflect a basement, which is where most American dehumidifiers actually live. The result was arithmetic, not engineering: machines got worse on paper overnight without a single part changing. A unit sold as a “70-pint” model under the old test may now be rated anywhere from 40 to 50 pints — the Energy Star conversion table puts 70-pint models at roughly 40–45 pints, though individual manufacturer re-ratings vary. A “30-pint” lands near 20.

Same hardware. Same performance. Different thermometer.

That’s the cleanest possible demonstration of the point. The pint number was never a property of the machine. It was a property of the machine plus the test conditions, and when the conditions moved, roughly a third of the advertised capacity evaporated.

And here’s where it gets genuinely interesting if you’re not in America

Korea’s KS C 9317 standard tests dehumidification capacity at a dry-bulb temperature of 27°C with 60% relative humidity. That’s 80.6°F — essentially the old US standard, and far warmer than the current one.

So a Korean-market 20L machine and a US-market machine are not describing the same thing at all, even after you convert litres to pints. The Korean number is measured in conditions close to a warm, humid summer room. The American number is measured in conditions close to a cool basement. Comparing them directly is a category error, and it’s worth knowing before you read a spec sheet from either market and assume it means what the other one means.

There’s a second-order effect here that actually favors Korean buyers, and it’s underappreciated. Because KS tests at 27°C and a Korean apartment during monsoon season sits somewhere around 25–27°C at 70–85% humidity, the rated number is unusually close to honest for its intended use case. The temperature matches. And the real-world humidity is often higher than the 60% test condition, which pushes extraction up, partially offsetting the couple of degrees of temperature shortfall.

That’s a rare thing in appliance marketing — a rated spec measured under conditions you’ll roughly reproduce. I spent the air purifier guide arguing the opposite about coverage area, so it’s only fair to say when a number holds up.

But it holds up for exactly one season. Point that same machine at a winter condensation problem and the rating becomes fiction.

The cliff

Below roughly 18°C, condensate on a compressor’s evaporator coil stops draining and starts freezing. Once frost forms, two bad things happen at once. The ice acts as an insulating layer between the refrigerant and the air, so heat transfer collapses. And the frost fills the gaps between the fins, choking airflow and reducing the volume of air the machine can process. The unit is working just as hard and accomplishing far less.

Manufacturers handle this with an auto-defrost cycle: the compressor shuts off, sometimes a heater kicks in, the ice melts, and the machine resumes. Which means at low temperatures a compressor dehumidifier spends a growing share of its runtime not dehumidifying — paying electricity to melt ice it just made. The colder the room, the higher that share, until the machine is mostly a defrost cycle with occasional dehumidification.

This is why the reference guidance from dehumidifier specialists is temperature-based rather than capacity-based. Meaco’s rule of thumb is: below 10°C, desiccant only. Between 10 and 15°C, desiccant or a large compressor. Above 15°C, either works. EcoAir puts compressor units’ effective range at roughly 16–40°C and desiccant at 1–40°C.

Notice what’s missing from all of that guidance. Nobody mentions pints.

The other machine

A desiccant dehumidifier doesn’t chill anything. It runs air through a slowly rotating wheel packed with zeolite, a mineral that adsorbs water vapour directly out of the airstream. The saturated section of the wheel then rotates past a heater that drives the moisture back out as hot, wet air, which is condensed separately and drained.

Because nothing needs to be cooled below the dew point, ambient temperature barely matters. There’s no coil to freeze, so there’s no defrost cycle, so there’s no cliff.

And there’s published data proving this rather than just claims. Meaco’s DD8L Zambezi lists extraction at two temperatures. At 68°F/60% RH on fan speed one it pulls 8.66 pints/day at 328W. Drop to 50°F/60% RH — an 18-degree fall — and it pulls 8.41 pints/day at 331W. That’s a roughly 3% loss across a temperature drop that would leave a compressor unit cycling through defrost. On speed three, 16.9 pints at 68°F becomes 15.64 pints at 50°F: about 7.5% down.

A compressor machine does not have numbers like that, which is why compressor manufacturers generally publish one rating and stop talking.

So the desiccant wins? No. It wins at low temperature, and it pays for that win at a price that’s easy to miss.

Look at those wattages again. 328W for 8.66 pints. The compressor units in the table below pull 262–333W and extract two to three times that. A desiccant unit is roughly two to three times less efficient at moving water per watt in warm conditions, because it’s regenerating a wheel with an electric heater, and resistive heating is expensive. Meaco quotes around 330W for a desiccant at 20°C/60% versus 150–260W for a compressor of comparable duty.

Which sets up the tradeoff that decides most purchases, and it isn’t a spec at all.

Every dehumidifier is a heater. The desiccant is a very good one.

This is the part I’d underline.

When water vapour condenses into liquid, it releases the energy that was holding it in gas phase — the latent heat of condensation, roughly 950 BTU per pint of water. In an air conditioner, that heat gets pumped outside. In a dehumidifier, there’s no outside. The condenser dumps it straight back into the room, along with the waste heat from the compressor and fan motor and every watt the machine draws from the wall.

The energy balance is unforgiving: a dehumidifier converts 100% of its electrical draw into room heat, and then adds the latent heat of every drop of water it pulls out of the air on top of that. It is a space heater with a bucket attached. Practitioners describe it as heating with an effective COP in the 3–4 range — meaning it’s better at warming a room than a resistive heater is.

Typical figures for a compressor unit are a 2–5°F rise in room temperature, with air leaving the outlet a couple of degrees warmer than it went in. For a desiccant, it’s dramatically more, because you’ve added an electric heater to the process on purpose. Meaco puts desiccant outlet air at 10–12°C warmer than ambient, describing the effect on room temperature as significant; EcoAir puts its own units’ outlet air at 3–5°C warmer than the room — a more conservative figure. The spread between those two claims is wide enough that I’d treat it as model-dependent rather than a fixed property — but both agree on the direction and on the fact that it’s much larger than a compressor’s.

Now put that in a Korean apartment in July. It’s 27°C, 80% humidity, and you run a desiccant dehumidifier because you read that desiccants are better. You’ve just installed a space heater in a hot room during the hottest, most humid weeks of the year.

The saving grace is that dry air feels cooler than humid air at the same temperature, because sweat evaporates faster off your skin. Dropping a room from 80% to 55% RH genuinely improves comfort even as the thermometer ticks upward. But there’s a limit, and a desiccant in a Korean summer blows straight past it. This is the single most common way people buy the wrong machine: they read “works at any temperature,” conclude “better,” and don’t notice that “works at any temperature” is achieved by adding heat.

Flip the season and the same fact becomes the selling point. Winter condensation on window frames, a cold north-facing room, a garage — the extra heat is a bonus, and the compressor alternative barely functions.

The specs that actually decide daily use

Three things matter more than capacity once you’ve picked the right technology, and only one of them is usually on the box.

Drainage. A dehumidifier working hard in monsoon conditions fills its tank fast. Do the arithmetic: a 20L/day machine with a 5L tank needs emptying roughly four times a day at full tilt. Every unit stops when the tank is full, so a machine you can’t empty is a machine that isn’t running — and the hours it’s stopped are the hours humidity climbs back. Continuous drainage via hose turns the tank capacity spec from a constraint into an irrelevance, but it requires somewhere for the water to go by gravity. If your drain is above the machine’s outlet, you need a model with a built-in pump. Check this before you buy, because a hose port on a unit sitting on a bathroom floor next to a raised drain is decoration.

Noise. This category is loud and the marketing numbers are misleading in a specific way. Korea’s consumer agency tested nine 20L-class units at the standard conditions and measured 49 to 57 dB in auto mode at maximum airflow. That’s a real air conditioner, not background hum. Meanwhile manufacturer sheets advertise minimum figures — Winix quotes 33.5 dB, LG 34 dB — measured on low fan, which is the mode where the machine extracts least. Same structural problem as every fan-driven appliance: the headline capacity comes from the fan speed you can’t sleep next to. If it’s going in a bedroom, the spec that matters is what it does on the quiet setting, and almost nobody publishes that pairing.

Efficiency, expressed properly. Watts alone tells you nothing. The useful unit is litres removed per kilowatt-hour. In the Korean testing, top performers cleared 2.6 L/kWh, and the monthly electricity cost across the tested field averaged about 8,000 won, with the most efficient unit at roughly 7,000. That’s a narrow spread — worth noting, because efficiency differences between comparable compressor units are small enough that they shouldn’t drive the decision. Technology choice and drainage matter far more than a few percent of L/kWh.

One more thing, briefly, because it’s sold heavily and deserves a warning: Peltier (thermoelectric) dehumidifiers. These replace the compressor with a solid-state element. They’re quiet and cheap and they barely work. Budget models pull about 0.3L/day, better ones around 0.8L, and the electricity they consume per litre of water removed runs several times a compressor’s. They’re for a closet or a shoe cabinet. Putting one in a room is buying a nightlight to solve a lighting problem.

The Specs

Notes on reading this table: the rated capacities are not comparable across rows unless the test condition matches. The Korean models are rated at 27°C, the US models at 65°F (18.3°C), and the desiccants at their own stated conditions — this is the entire argument of the article expressed as a table. The Samsung row shows independently measured values from Korea’s consumer agency comparison rather than manufacturer claims, which is why its format differs. Manufacturer minimum noise figures are measured on the lowest fan speed; the 49–57 dB range from that same comparison test reflects maximum airflow, and both numbers are true of the same machines. Blank cells mean the figure isn’t officially published, not that it’s zero. The Meaco and EcoAir units are UK-market products; check voltage and local availability.

The Verdict

Monsoon season, normal apartment, 25°C and up: compressor, and it isn’t close. This is the condition compressors were built for, it’s the condition the KS rating actually describes, and it’s two to three times more efficient per watt than the alternative. A 20L-class inverter unit covers a typical apartment. Don’t buy a desiccant for a Korean summer because a review told you desiccants work at any temperature — that’s true and irrelevant here, and you’ll be heating a room you’re trying to make comfortable.

Winter condensation, cold rooms, garages, anything under 15°C: desiccant, and it isn’t close either. A compressor in a 10°C room is a defrost cycle with a fan. The Meaco figures — about 3% capacity loss from 68°F down to 50°F — are what “temperature-independent” looks like when someone publishes real numbers. The heat output is a feature here, not a bug.

If you only buy one and use it year-round, buy the compressor and adjust your expectations. Korea’s dehumidifier problem is overwhelmingly a June-to-August problem, and that’s a compressor’s home turf. Handle winter condensation with ventilation and by moving furniture off exterior walls rather than by buying a second appliance — condensation forms because a surface is cold, not because the room is humid, and no dehumidifier fixes a cold surface.

Whatever you buy, buy for the hose, not the tank. This is the spec I’d weight highest after technology choice, and it’s the one buyers skip. Continuous drainage is the difference between a machine that runs and a machine you keep meaning to empty. Confirm the port exists, confirm gravity works from where it’ll sit, and buy a pump model if it doesn’t.

Bedroom use: ignore the advertised minimum noise figure. A quoted 33.5 dB and a measured 53 dB describe the same machine at different fan speeds. Assume you’ll run it on low overnight, assume low means meaningfully reduced extraction, and size up slightly to compensate. Or run it during the day with the door closed and sleep in a room that’s already dry — usually the better answer.

Closets and shoe cabinets only: Peltier. Anywhere else, it’s the wrong tool.

The pattern here should be familiar by now. With air purifiers, the coverage figure was arithmetic laid over a real measurement. With smart bulbs, the brightness spec was fine and the protocol decided everything. With smart scales, the displayed body fat percentage was an estimate wearing the costume of a measurement.

Dehumidifiers are the same shape with an unusual twist: the headline number isn’t wrong, exactly. It’s conditional, and the condition is hidden in a standards document nobody reads. Learn the condition and the whole category reorganizes itself around one question.

Not how many litres. What temperature.

This article is for informational purposes only and is not investment advice.

Photo: Álvaro Bernal / Unsplash

The Story

Here’s the sentence that should reorganize how you shop for one of these.

관련해서 dehumidifier buying guide도 함께 참고하시면 좋습니다.

An ultrasonic humidifier does not boil water. It does not evaporate water. It shatters water — a piezoelectric disc vibrating a few million times a second rips the surface into a fog of droplets and a fan pushes that fog into the room. And every one of those droplets carries whatever was dissolved or floating in the tank when it left.

That’s not an edge case or a defect. That’s the operating principle. The mist you can see is the tell: real water vapor is invisible, so if you can see a plume coming out of the spout, you’re looking at liquid water suspended in air, not gas. Which means the machine isn’t a humidifier in the way most people picture it. It’s an atomizer that happens to raise humidity as a side effect.

The best evidence on this is a systematic review by Andrea Dietrich, Wenchuo Yao and Daniel Gallagher, published in Indoor Air, pulling together 27 studies spanning more than four decades. The numbers are unusually clean for indoor air research.

Airborne particle mass tracks the total dissolved solids in your fill water almost linearly — the reported slope is 1.14. Meaning if you fill the tank with water carrying twice the mineral load, you get roughly twice the airborne particles. The machine has no opinion about this. It just aerosolizes what you give it.

The particles land in a size range that matters. Count median diameters across the studies ran 115 to 250 nanometers, and about 90 percent of emitted mass was PM1 — under one micron. That’s the fraction that clears the nose and throat entirely and deposits deep in the lung. The review also found that fill water with a higher dissolved-solids load produced larger particles, with an R² of 0.78.

And it’s not only minerals. The reviewed studies documented emissions of calcium, sodium, potassium and magnesium, along with trace arsenic, cadmium, chromium and lead; insoluble iron and aluminum oxide particles; asbestos fibers at rates of 0.03 to 4.7 percent of the fibers present in fill water; bacteria including Legionella, E. coli and Mycobacterium; fungi; and endotoxin.

The bacteria number is the one I’d tape to the tank. Concentrations rose roughly tenfold over a window running from 150 minutes to 48 hours of operation. Not over a season of neglect. Over a weekend.

So the real spec on an ultrasonic humidifier isn’t output or tank size. It’s this: the tank is an inhaler cartridge, and you’re the one loading it.

The EPA has said a version of this for years, in a fact sheet most buyers never see. Ultrasonic and impeller units “can disperse materials, such as microorganisms and minerals, from their water tanks into indoor air.” Evaporative and steam vaporizer models “generally disperse less, if any, of these pollutants” and “are not expected to disperse substantial amounts of minerals.”

That single paragraph is the entire buying guide. Everything below is detail.

So how do the other two types dodge this?

An evaporative humidifier is closer to a wet towel with a fan behind it. Water soaks a wick, air blows through, and molecules leave the surface one at a time as actual gas. Minerals can’t evaporate — they physically cannot leave with the water — so they stay trapped in the wick. Same for most of what’s growing in the tank. The wick is doing double duty as a filter, and the reason it looks crusty and awful after two months is that it caught everything that would otherwise have gone into your lungs. That crust is the product working.

Evaporative also has a property nobody markets well: it is self-limiting. Evaporation rate falls as the air gets more saturated, so the machine naturally slows down as the room approaches equilibrium. It is structurally very hard to over-humidify a room with one. An ultrasonic unit has no such governor — it will happily push mist into an already-damp room until water is beading on your window frames.

A steam vaporizer boils. Boiling leaves minerals behind in the chamber as scale and kills most of what’s living in the water on the way out. Vicks’ V745A is the archetype: no filter, no wick, water heated to a rolling boil in a stainless chamber. The output is genuinely sterile steam. The costs are real, though — a heating element pulls far more power than a fan, and a device producing boiling water and hot steam is a burn hazard the EPA specifically flags around children.

Warm mist is the same thing with the steam cooled slightly before it exits. Still boiled, still clean.

Now, the part where the marketing gets clever.

Some ultrasonic units ship with a “demineralization cartridge” and the box implies the problem is handled. The EPA’s language here is unusually blunt: the ability of these cartridges to remove minerals “may vary widely,” further research is needed on how well and how long they work, and you should watch for white dust as your signal that they’ve stopped working. That’s not a solution, that’s a consumable with an unverified duty cycle.

Distilled water genuinely does work — it addresses the mineral half of the problem at the source. But run the logistics before you commit. A 500 mL/h machine on 10 hours a day is five liters a day, roughly 150 liters a month, which is a lot of jugs to buy and haul. And distilled water does nothing about the second half of the problem: once the water is in a warm plastic tank, biology starts regardless of how pure it went in. Distilled water is not sterile water, and it doesn’t stay sterile.

Which brings me to the spec almost nobody weighs properly.

Cleaning access is a performance spec, not a convenience feature. The EPA’s guidance is to empty, wipe dry and refill portable humidifiers daily, and scrub them out every third day. Read that again, because it’s a chore schedule, not a suggestion, and roughly nobody does it.

So the honest question when you’re standing in front of two machines isn’t which has the higher output. It’s: which one will I actually clean 120 times this winter? A wide-mouth tank you can get a whole hand into, with a simple base and few crevices, will outperform a higher-spec machine with a narrow neck — because the high-spec machine will be dirty by January and the dirty one is the one aerosolizing biofilm into your bedroom.

This is the same shape of argument as the air purifier guide: the headline number is measured under conditions you’ll never reproduce, and the thing that decides real-world performance is a maintenance behavior nobody puts on the box.

Then there’s the target itself, which people get wrong in both directions.

You are not trying to maximize humidity. You’re trying to hit a band and stay in it.

The EPA is explicit: do not humidify above 50 percent relative humidity, because higher levels encourage dust mites and mold. Other bodies put the comfort ceiling at 60 percent, and Venta builds its whole pitch around a 40–60 percent target. I’d treat 40–50 percent as the safe operating window and 60 as a hard ceiling you don’t approach deliberately, because the biology above that line is not subtle. Dust mites absorb water vapor straight out of the air rather than drinking it, so sustained humidity is what feeds a mite population.

There’s a real upside inside the band. Work published in PLOS One on simulated coughs found influenza virus retained 70.6 to 77.3 percent infectivity at relative humidity at or below 23 percent, but only 14.6 to 22.2 percent at 43 percent and above — with most of the drop happening in the first 15 minutes. That’s a meaningful argument for not letting a heated winter room sit at 20 percent.

It is not an argument for pushing to 60. Modeling work in Scientific Reports on five airborne respiratory viruses found ventilation rate dominating infection risk across all five, while humidification showed meaningful benefit only for influenza — and even there the gain came from lifting a very dry room toward 50 percent, not from pushing past it. Above 40-something percent you’re on the flat part of the curve, and every extra point costs you on the mold and mite side.

Which makes the hygrometer the most undervalued item in this category. A humidistat that actually holds a setpoint converts the machine from “device that emits moisture until the tank runs dry” into “device that maintains a target.” Without one, a decent ultrasonic unit in a small closed bedroom will overshoot past 60 percent overnight and you’ll wake up to condensation on the glass — which is mold-growing weather on the window frame. A separate $15 hygrometer placed across the room is worth more than a $50 upgrade in mist output, partly because built-in sensors sit right next to the mist outlet and read the machine’s own plume.

A note on why this argument lands harder in Korea than anywhere else.

Korea traced a wave of severe, often fatal lung injury — including in infants and pregnant women — to antimicrobial additives sold specifically to be poured into humidifier tanks. The chemicals involved were PHMG and PGH. The Korea Centers for Disease Control and Prevention — the agency now known as the KDCA — confirmed humidifier disinfectants as the cause of the acute lung injury after an epidemiological investigation. Government-registered damage reports have run into the thousands, with deaths in the four figures.

That disaster happened because of a specific toxic additive — but the delivery mechanism was the physics described above. The chemical was in the tank, the ultrasonic disc turned it into a sub-micron aerosol, and people inhaled it deep into the lung. Notably, PHMG appears in the Dietrich review as the one organic chemical the literature identified in ultrasonic humidifier emissions, alongside the minerals and bacteria.

The lesson that got institutionalized in Korea — never put anything in the tank but water — is correct and important. But the sharper version is broader: the tank contents become lung contents, whether you added something or the tank grew it on its own. Which is why “I only use tap water” isn’t the clean bill of health it sounds like.

The Specs

Notes: output figures are published only where the manufacturer states them — blank cells mean the number isn’t officially available, not that it’s zero. Coverage claims come from manufacturers and are not measured under a shared standard the way air purifier CADR is, so treat cross-brand coverage comparisons as loose. The Venta’s very low wattage is a genuine consequence of its design: it runs a fan and a rotating disc stack, with no ultrasonic transducer and no heating element. Prices are street ranges and move constantly.

The Verdict

Default recommendation, and it’s not close: evaporative. For a bedroom you sleep in every night with tap water you’ll actually use, an evaporative unit is the structurally safer machine. The minerals stay in the wick, the biology mostly stays in the wick, and the physics make over-humidification hard. The Honeywell HCM-350 is the sensible version of this at around $90; the Venta LW25 is the expensive version, but under 10 watts at top speed and no consumable filter is a real long-run argument, and 24 dB(A) on speed one is quieter than most ultrasonic units at their quiet setting.

If you’re set on ultrasonic, buy for the tank opening. Output and capacity are the specs on the box; cleanability is the spec that decides what you’re breathing in February. Wide top-fill opening, few internal crevices, a base you can reach into. If you can’t get a hand and a brush into it, skip it regardless of everything else.

Nursery or a child’s room. This is where I’d rule out ultrasonic entirely. The Dietrich review includes case reports of hypersensitivity pneumonitis and one infant with non-reversible obstructive lung injury, and infants have higher breathing rates per unit body weight, so identical air means a higher dose. Evaporative is the right answer. Steam is the cleanest output of all, but a boiling-water device in a child’s room is a burn risk the EPA calls out by name — if you go steam, placement is not optional.

Cold and congestion relief. A steam vaporizer, and here the usual objection stops applying. Steam’s disadvantages are power draw and burn risk, and for short-run symptomatic use in an adult bedroom, neither is a big deal. The V745A is about $40, sterile by construction, and needs no filter.

Hard water. Evaporative or steam, and it’s not a preference. Higher dissolved solids means proportionally more airborne particles from an ultrasonic unit — the 1.14 slope again — and it means faster scale buildup, and the EPA notes that scale is itself a breeding ground for microorganisms. If your kettle furs up quickly, an ultrasonic humidifier is the wrong machine for your water. Budget for wick replacements instead; the wick is a consumable precisely because it’s catching all of that.

Large open living space. Bias toward tank capacity and a real humidistat over raw output. A big room reaches equilibrium slowly, and the failure mode is a machine that runs dry at 3 a.m. or overshoots into condensation, not one that’s short on mist. The OasisMist 1000S at 10 litres and 27.5 watts is a reasonable pick if you accept the ultrasonic tradeoff and commit to the cleaning schedule.

Whatever you buy, add a hygrometer and put it across the room. Then set your target at 40–50 percent and stop thinking about output entirely.

The pattern here is one that’s shown up repeatedly in these guides. With the robot lawn mower, the blades were the obvious spec and navigation was the actual purchase. With air purifiers, the coverage figure was arithmetic and CADR was the measurement. Here, the entire category is sold on mist output and tank size — and the thing that decides whether the machine helps you or quietly makes your air worse is a physics question about what leaves the tank, plus a chore you’ll either do or won’t.

Pick the machine that survives the version of you who doesn’t feel like cleaning it.

This article is for informational purposes only and is not investment advice.

Photo: Ulla Shinami / Unsplash

The Story

Here’s the sentence that should reorganize how you shop for an electric toothbrush, and it comes from a peer-reviewed meta-analysis of fourteen randomized trials, not from a marketing department:

“It must be recognised that any results published are likely to favour the supporting company’s toothbrush.”

That’s the authors of a systematic review on interdental plaque removal, writing in their own limitations section, explaining that they could not find industry-independent studies comparing one brand of powered toothbrush against another. Not “the studies were mixed.” Not “more research is needed.” They’re saying the head-to-head brand evidence base is essentially a set of company-funded trials, and the results track the funding.

Now hold that next to the aisle you’re actually standing in. One handle says 8,800 oscillations and 40,000 micro-vibrations. Another says 62,000 bristle movements per minute. A third has an AI coach that maps your mouth in real time and grades you out of 100. Every one of those numbers is real. Almost none of them is the reason one person’s gums improve and another’s don’t.

So let’s separate what the independent evidence actually supports from what the category has been selling on top of it.

Start with the one claim that’s genuinely well established. The Cochrane review on powered versus manual brushing — 56 trials, 51 of them pooled across 4,624 participants — found powered brushes reduced plaque 11% more at one to three months and 21% more beyond three months, with gingivitis down 6% at one to three months and 11% on the Löe & Silness gingival index beyond three months (17% on the Ainamo Bay bleeding index). Cochrane graded that moderate quality evidence, which in their vocabulary is a real endorsement. Powered beats manual. That question is answered.

But read the conclusion line, because it’s the part nobody quotes: “the clinical importance of these findings remains unclear.” A 21% reduction on a plaque index is a measurable change in a number dentists use in trials. Whether it translates into fewer fillings and fewer extractions over a lifetime is a separate question, and the trials weren’t built to answer it. This is not a reason to skip the electric brush. It’s a reason to be suspicious of anyone selling you the fourth incremental improvement on top of it.

Because that’s exactly where the evidence thins out.

Oscillating-rotating versus sonic is a much smaller fight than the packaging implies. Cochrane’s separate review of different powered brush types found rotation-oscillation beat side-to-side in the short term, but described the difference as small and of unclear clinical importance. The interdental meta-analysis quoted above found oscillating-rotating ahead on interproximal plaque — and the effect sizes were 0.09 at eight weeks, 0.04 at twelve weeks, and 0.03 at six months, the last one not even statistically significant (p = 0.09). The authors state plainly that these “may not represent clinically important reductions.”

Put those two magnitudes side by side and the shopping logic falls out immediately. Powered versus manual: 21%, moderate certainty, independently reviewed. Oscillating-rotating versus sonic: a few hundredths of an index point, heterogeneous, and drawn from a literature where nobody is neutral. The gap between using an electric brush and not using one is roughly an order of magnitude larger than the gap between the two mechanisms. Which means the mechanism debate — the thing every comparison article on the internet is organized around — is the least decision-relevant thing on the box.

So what actually differs between a person whose gums get better and a person’s who don’t?

Pressure. This is the feature that has the best claim to changing an outcome, and it’s also the one that’s cheapest to buy.

The mechanism is unglamorous. Brush too hard and you’re not cleaning better, you’re abrading. Force above roughly 100–150 grams starts working against the gingival margin rather than the biofilm on it, and the damage from that is cumulative and mostly irreversible — gums don’t grow back. The problem is that people are terrible at judging their own brushing force, and the failure mode is invisible for years.

A pressure sensor closes that loop. And there’s now a 36-month randomized controlled trial in people already susceptible to gingival recession that tested it: 92 enrolled, 87 completing, comparing a manual brush against an Oral-B iO with a sensor that signals a “green zone” at 0.8–2.5 N. At three years the manual group’s recession worsened by 0.17 mm on average while the powered group improved by 0.10 mm. More usefully, the share of sites that got worse by a full millimetre or more was 25.5% manual versus 10.6% powered. The investigators attributed it largely to sensor-controlled force.

That’s the strongest single result in this entire category — and I have to tell you the rest of it, because leaving it out would be exactly the sin this guide is about. The study was financially supported by Procter & Gamble, which owns Oral-B, and two authors disclosed personal fees from the company. That does not make the finding false. Three-year RCTs in this space are rare and expensive, and P&G paying for one is how the data exists at all. But it means the honest summary is: the mechanism is biologically plausible, the effect direction is consistent, the magnitude comes from a manufacturer-funded trial, and there is no independent replication. Treat it as strong reason to buy a pressure sensor and weak reason to buy a specific brand of pressure sensor.

Here’s the practical consequence, and it’s the money-saving one. Pressure sensors are not a premium feature. The Sonicare 4100 has one and routinely sells for $35–40. The Oral-B iO3 has one. Even Quip’s $50 Rev has one. What you get by climbing the price ladder is not a better sensor in any documented sense — it’s a different notification style. The cheap brushes buzz or stutter. The expensive ones light a ring red. There is no published evidence that a red light produces better gum outcomes than a stutter.

The second feature with a real mechanism behind it is the timer — and specifically the quadrant pacer.

Two minutes is the interval every clinical protocol uses, and self-reported brushing time is famously inflated; people who believe they brush for two minutes typically brush for closer to one. A timer fixes the total. The pacer fixes the distribution, which matters more than it sounds: without a 30-second nudge, most people over-brush the outer surfaces of the front teeth they can see in the mirror and under-brush the inner-lower and rear molars. Plaque doesn’t distribute evenly, so uniform total time with skewed coverage still leaves the same problem areas dirty year after year.

Notice that both of the features I’ve endorsed are behaviour-correcting, not cleaning-enhancing. Neither makes the brush head better at removing biofilm. They make you better at using it. That’s the through-line, and it explains why the spec race has produced so little.

Which brings us to the part the category most wants to sell you.

On AI coaching and app connectivity, the honest answer is that the independent evidence isn’t there yet. Trials do show connected brushes beating manual brushing on plaque and brushing time — but that’s the easy comparison, and the 21% figure above already covers it. The question that decides whether the app is worth paying for is narrower: does a connected brush beat an equivalent non-connected powered brush? On that, app layers have not clearly won, and the reviews note that the trials which do exist frequently have test and control groups differing in app access, which is a moderate risk of bias baked into the design.

There’s also a plain-physics limit worth naming. The position tracking in these brushes is built on accelerometers and gyroscopes in the handle, sometimes fused with the phone’s camera. That stack knows the handle’s orientation. It does not know where your gum line is, how deep the pocket is, or whether bristles are actually contacting the biofilm at the margin. It infers a mouth map from handle motion plus assumptions about jaw geometry. So a “100% coverage” score means the handle moved through the expected pattern for the expected duration. It does not mean your mouth is clean.

If that architecture sounds familiar, it should — it’s the same three-layer structure behind smart scale body-fat readings and smartwatch health sensors: a raw physical signal, an undisclosed vendor model, and a confident clean number on a screen. The number isn’t fabricated. It’s just measuring the proxy, not the thing.

That said, I’d rather be precise than dismissive. There’s one context where the app genuinely earns its place: if the gamification makes you brush the full two minutes when you otherwise wouldn’t, the app has delivered a real clinical benefit — via compliance, not via cleaning. That’s a legitimate reason to buy it. It’s just worth being clear that you’re buying a behavioural nudge, and behavioural nudges have a well-documented tendency to stop working around week six.

Now the number that actually decides five-year cost: refill heads.

This is where the category’s economics resemble the printer business, and it’s the section that flips purchase decisions. Oral-B iO Ultimate Clean heads run roughly $10–13 each at four-pack pricing, depending on the retailer. Standard Oral-B round heads are far cheaper. Sonicare C3 heads land around $10–12. Three-month replacement is the manufacturer-recommended interval, so budget four heads a year.

Run the five-year math and the ranking scrambles:

• Sonicare 4100 at $40 with $10 heads: $40 + (20 × $10) = $240, and 83% of that is refills.
• Oral-B iO5 at $120 with $12 iO heads: $120 + (20 × $12) = $360.
• Oral-B iO10 near $400 with the same $12 heads: $640.

The handle you’re comparing on the shelf is a minority of what you’ll spend. And the iO line’s premium heads are proprietary to the iO magnetic drive — they don’t fit the older Oral-B round-head system, so the cheaper refill ecosystem isn’t available to you once you’ve bought in. That’s a lock-in worth pricing before you commit, not after.

Third-party compatible heads exist at a third of the price, and they’re a genuine coin flip. Bristle end-rounding is the specific thing that varies: properly finished filaments have rounded tips, poorly finished ones are cut square and act like tiny blades at the gum margin. You cannot assess this by looking at the product photo, and it’s precisely the variable a pressure sensor can’t rescue you from — the sensor limits force, not bristle geometry.

One more filter, and it’s free. The ADA Seal of Acceptance requires compliance with ANSI/ADA Standard No. 120 and a submitted safety certification. Clinical efficacy studies are not mandatory — but when a manufacturer does submit them, the ADA expects at least two studies run at different sites and 80% statistical power for any claim carrying the Seal. It doesn’t rank brushes and it won’t tell you which is best. What it tells you is that a manufacturer agreed to be evaluated against a fixed external standard — which, in a category where the head-to-head literature is company-funded, is a meaningful signal about willingness to be measured.

The Specs

Notes: bristle-movement and oscillation figures are manufacturer specifications and are counted differently between sonic and oscillating-rotating designs — they are not directly comparable across mechanisms. Weights are approximate handle-only figures and vary by revision. Prices are US street ranges and move constantly; the Sonicare 4100 in particular is discounted frequently. Head costs assume multi-pack pricing.

The Verdict

If you’re buying your first electric brush. The Sonicare 4100 or an Oral-B Pro 1000. Both have the pressure sensor and the quadrant pacer — the only two features with a defensible mechanism behind them — at the bottom of the price ladder. Buying up from here purchases notification styling and app features, not documented gum outcomes.

If you already have receding gums or a dentist has flagged your brushing force. This is the one case where spending more is genuinely defensible, because the 36-month recession trial used a light-ring sensor with a defined force window rather than a buzz. The iO3 is the cheapest entry to that specific design. Keep the funding caveat in mind, and note that the trial’s comparator was a manual brush — it does not establish that a light ring beats a $40 brush’s buzzer.

If you want the AI coaching. Buy it as a habit tool, not a cleaning tool, and only if you honestly know that gamification works on you. The coverage score is inferred from handle motion, not from your gum line. If you’d have brushed two minutes anyway, you’re paying several hundred dollars for a chart.

Tightest budget. Any ADA-Sealed powered brush with a pressure sensor and a pacer, plus the discipline to replace heads on schedule. A splayed three-year-old brush head on a $400 handle cleans worse than a fresh head on a $35 one — bristles that have lost their shape don’t reach the gum line at the right angle regardless of what’s driving them.

If you travel constantly. Battery life is the spec that varies most and gets discussed least. The iO3 measures around 18–20 days in independent testing, while several premium models sit at roughly 14. Charging standards are not shared across brands, so the handle you buy determines the cable you carry.

If your household shares a base. Look at handle-only pricing and head compatibility before the bundle price. Two people on one Oral-B round-head system with $6 refills costs dramatically less over five years than two on the iO ecosystem at $12, and the clinical difference between those two systems is the few-hundredths-of-an-index-point margin discussed above.

The thread here is the same one from the air purifier guide, where the headline coverage number was an arithmetic choice dressed as a hardware spec, and from the home security camera guide, where the recurring cost was the actual purchase. Consumer health hardware has gotten very good at selling increments on top of a well-established baseline, and at pricing those increments as though they carried the same evidence as the baseline.

With electric toothbrushes the well-established baseline is: powered beats manual, moderate certainty, 21%. Everything after that is either a behaviour fix worth having cheaply, or a number generated by a company that sells the brush.

This article is for informational purposes only and is not medical advice. For gum disease, recession, or persistent bleeding, consult a dentist.

Photo: Giorgi Iremadze / Unsplash

The Story

On August 13, 2026, at NVIDIA’s headquarters in Santa Clara, LG Corp. chairman Kwang Mo Koo and NVIDIA’s Jensen Huang signed a memorandum of understanding. The headline that followed was predictable: LG is building a humanoid robot. A next-generation bipedal machine, targeted for a public unveiling in the first quarter of 2027.

That’s the headline. It’s also the least interesting thing in the announcement.

Read the actual list of what LG is using, and a different story shows up. The robot’s foundation model is NVIDIA Isaac GR00T. The onboard compute is NVIDIA Jetson Thor. The safety stack is NVIDIA Halos for Robotics. Three layers — reasoning, silicon, safety — and all three have the same logo on them.

So here’s the more accurate version of the headline: LG just decided not to build the brain.

Let me unpack those three pieces quickly, because the names are doing a lot of work. Isaac GR00T is what NVIDIA calls an open humanoid foundation model. Think of it as a pre-trained starting point for robot behavior — it takes in camera images and language instructions and outputs motions, and companies post-train it on their own robot’s body and their own tasks. It’s genuinely open, too — the current generation sits on Hugging Face under NVIDIA’s own open model license, downloadable by anyone and usable commercially, with attribution and acceptable-use conditions attached. Jetson Thor is the computer that physically rides on the robot and runs that model in real time, no cloud round-trip. Halos is the safety layer — the part that’s supposed to keep a heavy, fast-moving machine from doing something regrettable near a human. NVIDIA bills it as a full-stack safety system for robotics.

None of that is a knock on LG. Building a humanoid foundation model from scratch is a multi-year, multi-hundred-million-dollar detour, and the companies that tried mostly ended up behind. Renting the brain is a rational call. But it does force an obvious question, and it’s the question worth sitting with: if the model, the chip, and the safety stack all come from Santa Clara, what exactly is LG’s differentiated asset?

The answer is in the part of the announcement almost nobody covered.

Separately from the 2027 humanoid, LG is putting a wheel-based robot called LG CLOiD onto the washing machine line at LG Electronics’ Tennessee plant within 2026. Not a demo floor. Not a lab mockup. A live production line that ships appliances, where the robot handles material transport and assembly support. LG says the validation results will feed back into the robots’ performance, and deployment then widens to global production sites, homes, and commercial spaces.

There’s also a piece of infrastructure attached: a robot data factory built on LG CNS PhysicalWorks, meant to run continuous data collection, synthetic data generation, training, and verification. If you’ve followed the argument that data, not hardware, is robotics’ real bottleneck, this is what acting on it looks like. That’s the tell. LG isn’t just borrowing a robot brain — it’s building the pipeline that feeds one.

And that reframes the whole deal. The competitive question in robotics is quietly shifting from “who builds the best robot AI” to “who owns the places where robots can actually work, and the data that comes out of them.” NVIDIA can sell GR00T to everyone. What it can’t sell anyone is a washing machine line in Tennessee with real throughput targets and real failure modes. Call it the “deployment moat” — the advantage doesn’t come from the model, it comes from having somewhere to put it. LG has factories, appliances in millions of homes, and retail distribution. Those are the assets you can’t download.

The rest of the MOU makes the shape clearer. It goes well past robotics. LG plans an AI factory reference site using NVIDIA’s DSX platform with Vera Rubin in the first half of 2027, scaling to a full 80-megawatt AI factory in Cheonan, South Korea by the first half of 2028. On mobility, LG is developing a computing platform for AI-defined vehicles using NVIDIA DRIVE Hyperion. Robots, data centers, cars. One supplier underneath all three.

The Takeaway

The single most useful thing in this announcement is a timing gap most coverage flattened, so let’s be precise about it.

The wheeled CLOiD goes onto a real production line within 2026. The bipedal humanoid gets unveiled in the first quarter of 2027. Those two words are doing completely different work. One is deployment — a robot doing paid labor in a factory with quotas. The other is a reveal — a robot on a stage, or in a video, doing a demo.

This is the single most useful filter for reading any humanoid robot story. Ask whether the verb is “deploy” or “unveil.” The gap between them is usually measured in years, and sometimes the second date never arrives at all. Wheels are working. Legs are still auditioning.

It’s worth noting CLOiD isn’t a dumb cart, either. It has two arms and five-fingered hands — LG previewed it as a home robot at CES 2026. It just doesn’t walk. And that’s exactly the point: the hard, useful part of a robot is the manipulation, not the gait. Wheels are a solved problem that never falls over, never burns battery on balance, and never needs a safety case for tipping onto a worker. The industry keeps relearning this. I made the same argument when a hospital logistics robot — essentially a cabinet on wheels — turned out to be the most commercially successful Physical AI product around, while bipeds kept doing backflips in videos.

Now the second thing, and it’s the part LG’s press release won’t say out loud. NVIDIA releasing GR00T as an open model isn’t generosity. It’s the same playbook that already worked once in AI: make the upper layer free so everyone builds on it, then sell the layer underneath. CUDA did this. Free software, priced hardware. A foundation model that’s genuinely downloadable and commercially licensed still needs somewhere to run — and the reference implementation runs on Jetson Thor, with a safety stack that NVIDIA also supplies. Every layer given away increases demand for the one layer that isn’t.

So who gets what here? Be cold about it. LG gets speed and de-risking: it skips the multi-year foundation-model detour, gets a safety stack it would otherwise have to certify itself, and can credibly show a humanoid in early 2027 instead of 2029. NVIDIA gets a flagship reference customer, a global manufacturer validating its stack on a real appliance line, and lock-in across three business units at once. Robots, AI factory, vehicles — swapping suppliers later means unwinding all three.

That’s not a bad trade for LG. It’s just a trade, and the price is architectural dependence.

Which brings up the comparison that matters. Hyundai owns Boston Dynamics outright and builds Atlas’s body in-house, with a roadmap to deploy it in North American plants from 2028 — starting with parts sequencing, widening to assembly work from 2030. Even there the intelligence layer is partnered rather than purely internal; Hyundai’s own CES 2026 robotics materials name both NVIDIA and Google DeepMind. But it owns the robot company outright, which LG does not. LG is doing the opposite: rent the intelligence, own the deployment surface. Two Korean conglomerates, two opposite bets on where the value in robotics ends up sitting. If the foundation models commoditize the way language models did, LG’s bet looks smart and Hyundai looks like it overpaid for something it could have licensed. If robot intelligence stays hard and defensible, Hyundai owns the crown jewels and LG is renting from a landlord that also supplies its competitors.

Nobody knows which yet. But note that NVIDIA wins in both scenarios, which tells you who negotiated from strength.

One last thing worth watching, and it’s the metric I’d actually track. Forget the Q1 2027 unveiling — that’s a marketing date and it will happen roughly on schedule because reveals always do. Watch Tennessee instead. If CLOiD is still on that washing machine line a year in, with the line running normal volume and the robot doing work that used to require a person, that’s the real signal. If it quietly disappears after the validation period, that tells you something too, and it’ll tell you before any humanoid takes a stage.

Deployment is a fact. Unveilings are a plan.

This article is for informational purposes only.

Photo: Simon Kadula / Unsplash

The Story

Here’s a number that should change how you shop for a printer.

관련해서 water purifier guide도 함께 참고하시면 좋습니다.

A cheap cartridge inkjet costs about $60. A genuine ink tank model costs about $200. That’s a $140 gap, and it’s the number every store puts in front of you, in big type, right next to the box.

Now print 3,000 pages of black text on each one — a few years of a normal household with kids and forms and boarding passes. On the cheap inkjet, at roughly 10 to 20 cents a page depending on which cartridge you buy, that’s $300 to $600 in ink. On the tank model, at around 0.4 cents a page, that’s about $12.

So the $60 printer costs you $360 to $660. The $200 printer costs you $212.

The body price stopped mattering somewhere around page 1,500. Everything after that is consumables, and consumables are where the entire printer industry actually lives. This is the oldest razor-and-blades business on your desk, older than shaving subscriptions and coffee pods, and it still works because the price on the box is the only number most people compare.

So let’s do this differently. Forget model names for a minute. There are really only four machines in this category, and they’re not four brands — they’re four cost structures. Once you know which structure fits how you actually print, the model almost picks itself.

### The four structures, and the page count where each one wins

Structure one: the cartridge inkjet. Body is cheap — often under $80, sometimes practically given away in a bundle. Ink is brutal. Real numbers: Brother’s LC401 black cartridge runs about $18 for roughly 200 ISO pages, which is around 9 cents a page. HP’s 67 black for a DeskJet runs about $23 for roughly 120 pages, call it 19 cents. High-yield versions improve this — an LC401XL gets you to 500 pages, a 67XXL to around 400 — but you’re still in the 4-to-6-cent neighborhood at best, and that’s for black text. Color is multiples worse.

Structure two: the genuine ink tank. Epson EcoTank, Canon MegaTank, HP Smart Tank. You pour ink from bottles into reservoirs instead of clicking in a chipped plastic box. Body runs roughly $200 to $350. Then the economics invert completely. An Epson 522 black bottle runs about $18 for an ISO-rated 4,500 pages — that’s roughly 0.4 cents per page. Canon’s G3270 ships with enough ink in the box for something like 6,000 black and 7,700 color pages. HP’s Smart Tank 5101 ships with bottles rated around 6,000 black and 8,000 color.

Read those numbers again. The tank printers ship with more ink in the box than a cartridge inkjet will consume in its entire service life.

Structure three: the mono laser. Toner powder, not liquid. A Brother HL-L2460DW with a TN830XL runs about 3,000 ISO pages per cartridge, and street pricing on that toner lands the cost somewhere in the 1-to-2-cent range per page. Body around $150 to $200. It prints black text, fast, forever, and it does not care how long you ignore it.

Structure four: the color laser. Here’s where people get hurt. A Brother MFC-L3780CDW is a genuinely good machine — auto duplex, 50-sheet ADF, Ethernet, the works. But a full set of XXL toners runs roughly $480 or more, and independent reviewers who measured real-world output put color graphics north of 50 cents per page. Black text on the same machine is around 5 to 6 cents. Color laser is a document machine that happens to do color, not a color machine.

Now, the break-even. Take the $140 gap between a cheap cartridge inkjet and a tank model, and divide by the per-page savings. At a conservative 9 cents versus 0.4 cents, you’re saving about 8.6 cents per page, so the tank pays for itself at roughly 1,600 pages. If you bought the 19-cent HP instead, it’s under 800 pages.

Eight hundred pages. That’s two reams of paper. Almost every household that prints at all crosses that line. And it’s not close — once you’re past it, the tank keeps winning by a factor of twenty on every page for the rest of the machine’s life.

### How to actually calculate cost per page (and why the number is optimistic)

The formula is embarrassingly simple, and nobody does it:

Cartridge price ÷ rated yield = cost per page.

The rated yield isn’t a marketing number, which is the useful part. It comes from published ISO standards: ISO/IEC 24711 for inkjet cartridges, ISO/IEC 19752 for monochrome toner, ISO/IEC 19798 for color toner. Every major manufacturer tests to these, which means yields are genuinely comparable across brands. That’s rare and worth exploiting.

Here’s the catch you need to internalize. Those standards print a fixed suite of test pages at roughly 5% coverage per color channel — meaning each ink lays down on about 5% of the page area, so a four-color page tops out near 20% total while a mono page sits at 5%. Five percent of black is a sparse business letter. A double-spaced page with a header and not much else.

Your actual documents are darker. A kid’s worksheet with a big graphic, a recipe, a PDF with a shaded table, anything with a logo — real-world coverage commonly runs 8 to 20%, sometimes more. Which means your true cost per page is often two to four times the number you calculate.

That doesn’t break the comparison, because every printer is inflated by the same factor. But it does break your budget if you plan around ISO numbers literally. My rule: calculate with ISO, then multiply by three before deciding whether you can live with it. If a printer looks affordable at 15 cents a page ISO, ask yourself how you feel about 45.

One more trap in the same family: the starter cartridge. The ink that ships with a cheap cartridge inkjet is frequently a partially filled “setup” cartridge with a fraction of the rated yield. The first replacement often costs more than what you paid for the printer. That’s not an accident. That’s the business model announcing itself.

### Subscription ink: read the exit terms first

HP’s Instant Ink is the big one, and it deserves a fair hearing before the criticism, because for some people it genuinely is the cheapest option.

The structure: you pay monthly for a page allowance, not for ink. HP’s own plan pages list tiers running from a little over a dollar a month for 10 pages up to somewhere near thirty dollars a month for 700 pages, with several steps in between. Subscription pricing gets reworked often enough that the live tier table is the only version worth budgeting against. Cartridges ship automatically when the printer reports it’s low. Critically, a page is a page — a full-bleed color photo and a one-line receipt both count as one. If you print a lot of dense color, that’s a genuinely good deal, and it’s the one case where subscription math beats a tank printer.

Three things the marketing doesn’t lead with.

Rollover is capped. Unused pages carry forward, but only up to three times your monthly plan allowance on the ink plans — the toner plans cap it at two times. On the 50-page ink plan, you can bank 150 and no more. Everything above that evaporates. So the light, bursty printer — nothing for four months, then 200 pages of tax documents — is exactly the user the cap punishes. You paid every month and the pages you paid for expired.

Overages are priced per set, not per page. Print past your allowance and you buy additional pages in blocks — HP describes sets of roughly 10 to 15 pages for about $1.50 depending on plan. Go one page over and you’ve bought a whole set.

And the part that matters most: the cartridges are program cartridges. They’re specially designed for the subscription, and they’re chipped. When you cancel, you keep printing through the end of your current billing cycle, and after that the cartridge stops — even with ink still in it. Accumulated rollover pages die with the subscription too. To keep printing you buy standard retail cartridges.

Sit with that. It means the ink in your printer was never yours. You were renting access to it, and the printer enforces the lease in firmware. That’s a fundamentally different ownership model than an $18 bottle of Epson ink sitting on a shelf, and it’s the single most important thing to understand before enrolling.

I’m not saying don’t subscribe. I’m saying subscribe with your eyes open, on a plan sized to your floor rather than your average, and understand that you’re choosing a relationship, not a purchase.

### The failure mode nobody warns you about: drying out

If you print rarely, this is the section that decides your purchase, and everything above is secondary.

Inkjets — cartridge and tank alike — spray liquid through microscopic nozzles. When the printer sits idle, ink in those nozzles dries and blocks them. You get streaks, gaps, missing colors, or nothing at all. Then you run cleaning cycles, and here’s the cruel part: cleaning cycles consume ink. A badly clogged printer can burn through more ink unclogging itself than it ever spent printing. In the worst cases the printhead is permanently damaged, and on many consumer models the printhead isn’t user-replaceable, which means the printer is finished.

Laser printers have no liquid anywhere in the path. Toner is a dry powder that gets fused to paper with heat. Leave a laser printer alone for eight months, hit print, and it prints. There is no nozzle to dry.

This is why “how often do you print” outranks every spec in this article. A household that prints twice a month should buy a mono laser almost regardless of what else it wants, because a $200 tank printer that clogs into uselessness has a cost per page of infinity. Meanwhile a household that prints weekly can run an inkjet indefinitely — regular use is genuinely the maintenance.

While we’re here: pigment versus dye. Dye ink dissolves in liquid, produces richer photo color, and smears when wet or fades in sunlight. Pigment ink suspends solid particles that sit on the fiber, resisting water and light far better. Most consumer inkjets use a hybrid — Canon’s GI-21 set for the MegaTank line is pigment black with dye cyan, magenta, and yellow, which is a smart split: crisp waterproof black text, vivid photo color. If you’re printing documents that need to survive — anything filed, signed, or archived — pigment black matters. If you’re printing photos, dye color matters. Toner, incidentally, is neither; it’s plastic fused to the sheet, and it’s the most durable of the three.

### Connectivity, drivers, and the reason printers die young

Modern printers mostly connect fine. Wi-Fi setup has gotten dramatically less painful, and the standards did most of that work: AirPrint on Apple devices and Mopria on Android let you print without installing a manufacturer driver at all. If a printer supports both, phone and tablet printing is essentially solved. Check for both. It’s a one-line spec that eliminates an entire category of frustration.

Two things worth paying for. Ethernet, if you can put the printer near a router — Wi-Fi printers dropping off the network is the single most common printer complaint in existence, and a cable ends it permanently. And auto duplex, which is the difference between printing a 40-page document and babysitting a 40-page document. Note that this is exactly where Epson splits its lineup: the ET-2800 has no auto duplex, the ET-2850 does. Same family, and it’s the feature worth the upgrade.

For scanning, the question is whether you need an ADF — the tray on top that pulls in a stack. Flatbed only means one page at a time, by hand. If you scan multi-page documents even occasionally, an ADF changes the experience completely, and a duplex ADF that captures both sides in one pass changes it again.

Then the thing that actually kills printers. Not mechanical failure — driver abandonment. A printer is a fifteen-year mechanical device attached to a software stack the manufacturer supports for maybe five. A macOS or Windows major version lands, the driver isn’t updated, and a perfectly functional machine becomes e-waste. This is a real and underrated cost.

The defense is standards. A printer that speaks AirPrint, Mopria, and standard IPP Everywhere can be driven by the operating system’s built-in support without a vendor driver at all. That’s not a nice-to-have — it’s the thing that determines whether your printer survives the next OS upgrade. Buy for protocol support, not for the quality of the bundled app.

### The most neglected internet-connected device in your house

Quick but important. Your printer is a networked computer running a full operating system, and it is almost certainly the device in your home you have thought about least since the day you set it up.

The common failure points are boring and consistent: default admin credentials nobody changed, firmware never updated, legacy protocols like FTP or Telnet left enabled from the factory, raw print ports such as 9100 or 631 sitting open, and — worst — printers accidentally exposed to the open internet through router misconfiguration or over-eager port forwarding.

Three things, once, and you’re done. Change the admin password. Turn on automatic firmware updates if offered, or check manually once a year. Disable protocols and cloud features you don’t use. If your router supports a guest or IoT network, putting the printer there is a reasonable extra step — the same logic that applies to every other always-on device on your network, a home NAS very much included.

### The comparison table

Prices are US street pricing and move constantly; treat them as ballpark, and note that models on clearance or nearing end of life can sit well below the figures here. Cost per page uses ISO-rated yields at typical consumable pricing — remember the 5% coverage caveat.

Read that table by column, not by row. The Black CPP column ranges from 0.4 cents to 19 cents — a roughly 48x spread — while the body price column ranges from $50 to $500, a factor of ten. That’s the whole argument in two columns.

The Takeaway

Match the structure to your printing pattern, and ignore everything else.

You print almost nothing — a form every couple of months. Buy a mono laser. The Brother HL-L2460DW class of machine, around $180. Not because it’s cheap per page, though it is, but because it’s the only category that survives neglect. Every inkjet you buy for this use case will clog, and the cleaning cycles will cost you more in wasted ink than you’d have spent printing. If you need occasional color, use a print shop or a photo service — genuinely cheaper than owning a color machine you don’t use.

You have kids, worksheets, forms, and print constantly. Ink tank, auto duplex, and don’t overthink it. The ET-2850, the Canon G3270, the HP Smart Tank 5101 — all fine. Frequent use is exactly the condition inkjets want, and at 0.4 cents a page you’ll stop thinking about ink entirely, which is the actual luxury here. Prioritize auto duplex over everything else; it halves your paper and stops you standing at the machine flipping sheets.

You print photos. Ink tank with dye color, and pay attention to how many inks it uses. Four-ink machines are document printers that can do photos; six-ink photo models with extra gray or photo-black tanks are the ones that render skin tones and shadows properly. Also: paper matters more than the printer. Cheap paper on a great printer looks worse than good paper on an average one, every time.

You run a home office with real document volume. Mono laser with an ADF and Ethernet, plus a cheap tank inkjet on the side if you occasionally need color. That combination sounds redundant and costs less than one color laser — because color laser toner is the most expensive consumable in this entire article at over 50 cents a page for graphics. Two right-sized machines beat one machine that’s wrong for half its jobs. That right-sizing argument runs through the rest of the desk too — it’s the same one I made about mini PCs.

The uncomfortable meta-point: the printer aisle is designed around the one number that doesn’t matter. Body price is the headline because it’s the only figure where the cheap machines win, and the industry has spent decades making sure you compare it. Cost per page is printed in the spec sheet footnotes, in a unit — tenths of a cent — deliberately hard to feel.

Do the division. Cartridge price divided by ISO yield, multiplied by three for real coverage, multiplied by however many pages you actually print in a year. That’s the number you’re buying. Everything else is packaging.

This article is for informational purposes only and is not investment advice.

Photo: Mahrous Houses / Unsplash