Blog

Guinea Pig Bedding: Paper vs Fleece vs Aspen Compared

Some links on our site may earn us a commission at no extra cost to you. Learn more.

Bedding choice touches a guinea pig’s respiratory and skin health. This is general information, not veterinary advice — ask an exotics vet what suits your animal.

Updated August 2026: every dollar in the cost table below traces to a maker’s own published price, a published bedding depth from Humane World for Animals, and a published cage area, with the arithmetic shown. Two findings shape the page. Absorbency measured by weight and absorbency measured by volume rank the same materials in opposite orders. The cedar warning everyone repeats also rests on liver-enzyme research rather than on anything measured in a lung.

The verdict on this guinea pig bedding comparison. Small Pet Select’s paper bedding takes the tally at 59.0 out of 100. It is the only product here that publishes its volume on a stated basis. GuineaDad’s fleece liners win the money axis outright — about $138 once against $333 to $999 a year in paper. They finish second at 54.5 because almost nothing else about them can be checked. Small Pet Select’s kiln-dried aspen is the named loser at 52.5. Its own bag states two volumes that cannot both be true.

I’ve kept cats through humid Gulf Coast summers, and damp litter turns to ammonia fast once real heat sets in. That’s why I check how a bedding claim was measured before I check the price. A number that holds up dry can still fail once heat and moisture get involved. That gap is exactly what this page is chasing.

Prices are approximate, current as of publication, and change frequently — confirm the latest price with the retailer before buying.

Which Guinea Pig Bedding Absorbency Claims Are Real?

Every bag on the shelf makes an absorbency claim. Not one of them tells you how it was measured.

The claim	Whose page it is on	What sits behind it	Verdict
“ABSORBS UP TO 6X TIMES its weight in liquid”	Kaytee, Clean & Cozy	No test method stated. The basis is weight — the basis that inverts, see below	Unmethodized, and on the wrong basis for a buyer
“2x more absorbant than shavings”	Carefresh brand page	No test method, no basis, and no statement of which shavings	Not checkable at all
“able to hold up to three times its weight in water”	GuineaDad, Original Liner	No test method stated. Weight basis again	Unmethodized
“super absorbent”	Small Pet Select, kiln-dried aspen	No number of any kind	No claim to check
“141 L expands to 4 cu ft”	Small Pet Select, kiln-dried aspen, same page	4 cubic feet is 113.3 litres. The two figures differ by a factor of 1.24	The page disagrees with itself
Pack sizes 24.6 L, 49.2 L, 85 L	Kaytee, Clean & Cozy	Kaytee publishes no price anywhere on its own product page	Cannot be priced from the maker at all
Bamboo “reduce[s] the chance of guinea pigs acquiring bacteria-related illnesses, ranging from respiratory diseases to conjunctivitis”	GuineaDad	No study, no measurement, no citation offered	Unsupported

Claims check, run against each maker’s own current pages on 31 August 2026.

Why do compressed and expanded litre figures differ?

Small Pet Select’s own paper bedding proves the point. It lists 178 litres expanded against 52 litres compressed for the same bag, a ratio of about 3.4. Carefresh and Kaytee each publish one litre figure and never say which kind it is. Compare a Carefresh 60 to a Small Pet Select 178. The two may differ by three and a half times, and you would not know it before you start.

Oregon State’s Wet Shavings and the Ranking That Flips

Here’s why “six times its weight” tells you little. You buy bedding by the litre and fill a cage by volume, and the two bases do not agree.

E. G. Pearson, S. Leavengood and J. E. Reeb published the study at Oregon State University’s College of Forestry in September 1999. It compared the absorptive capacity of western juniper, western redcedar and Douglas-fir. Twenty-gram samples sat in mesh bags, submerged for 30 minutes in horse urine at a specific gravity of 1.015. Each was then drained and reweighed to saturation.

Material (mixed particle size)	Absorbed, by weight	Rank by weight	Absorbed, by volume	Rank by volume
Western redcedar	328%	1st	8.4%	3rd
Douglas-fir	215%	2nd	8.1%	2nd
Western juniper	214%	3rd	16.0%	1st

Oregon State University College of Forestry, September 1999. Same samples, same experiment, opposite winners depending on the basis.

Redcedar takes up 1.53 times as much liquid as juniper per gram. Juniper, in turn, takes up 1.90 times as much as redcedar per litre. Both numbers are correct, and the first one is the one bedding makers print.

Does the same weight-vs-volume problem show up elsewhere?

Yes. Kathryn A. O’Connell and Diego Celdran-Bonafonte published a 2026 review in the Journal of the American Association for Laboratory Animal Science, volume 65, issue 2. It makes the same point about the wider lab-animal literature: “When compared on a volume-to-volume basis, corncob bedding tends to exhibit higher absorbency than cellulose or wood chip bedding. In contrast, when comparisons are made on a mass-equivalent basis, cellulose and wood-chip beddings outperform corncob in absorbency.”

Two honest limits. Oregon State tested wood for horses, not paper for guinea pigs. I could also find no independent or peer-reviewed absorbency measurement of Carefresh, Kaytee Clean & Cozy, Small Pet Select paper, Small Pet Select aspen or a GuineaDad liner. Every absorbency figure in this market traces back to the company selling the bag.

Two guinea pigs resting on a bedded surface — Two guinea pigs on bedding. How much of it a cage swallows per change is the number the bag never prints.

The Guide’s Cedar Sentence and the Study It Cites

Never use cedar, because it causes respiratory problems. Every rescue says it, we said it, and it is worth knowing what the primary literature actually measured.

The National Research Council’s Guide for the Care and Use of Laboratory Animals, eighth edition, is the key reference here. Both AAALAC and NIH’s Office of Laboratory Animal Welfare defer to it. Page 69: “Cedar shavings are not recommended because they emit aromatic hydrocarbons that induce hepatic micro-somal enzymes and cytotoxicity (Torronen et al. 1989; Weichbrod et al. 1986, 1988) and have been reported to increase the incidence of cancer (Jacobs and Dieter 1978; Vlahakis 1977).”

Read the endpoints. Liver enzymes, cytotoxicity, cancer. Lungs are not in the sentence.

The 2026 review by O’Connell and Celdran-Bonafonte, available in full through PubMed Central, says the respiratory part out loud: “Although airborne particulate matter from bedding is known to accumulate in microisolation cages and may theoretically affect rodent respiratory health, no direct evidence currently links inhaled bedding dust to adverse health outcomes in rodents.” On the softwood question it is equally direct: “Softwood beddings contain aromatic hydrocarbons that induce hepatic enzyme activity in rodents, leading to altered drug metabolism.”

Does the Guide’s cancer citation hold up?

No. George Vlahakis’s “Brief Communication: Possible Carcinogenic Effects of Cedar Shavings in Bedding of C3H-Avy fB Mice,” JNCI volume 58, issue 1, pages 149 to 150, 1977, is the paper the Guide cites. Its title promises a positive finding. The paper’s own conclusion says otherwise: “From these results, the high incidence of cancer in the C3H-AvyfB strain could not be attributed to the routine use of cedar shavings in the bedding material.”

The most authoritative document in the field cites a null result in support of a positive claim. I am reporting the two texts side by side rather than telling you the Guide is wrong overall. The enzyme-induction citations are a separate matter, and the effect there is real and repeatedly measured.

What this changes for you: nothing about the advice, quite a lot about the reason. Skip cedar. Skip untreated pine. Not because a lung study says so — there is not one. It is because there is measured metabolic disruption and no benefit on the other side of the trade. Aspen is a hardwood and the softwood literature does not reach it at all. Paper is not wood. Neither one is a compromise you are making.

One more thing worth knowing before you pay extra for a kiln-dried bag. The Guide hedges on whether heat treatment removes the compounds: it “may, depending on the material and the concentration of aromatic hydrocarbon constituents, reduce the concentration of volatile organic compounds.” Small Pet Select says on its own aspen page that “the process of being heat-treated is to help eliminate bacteria, yeast, and mold” — which is a different purpose entirely. I found no published measurement of aromatic content in commercially kiln-dried bedding either way.

How Much Does Guinea Pig Bedding Cost for a 2×4 Cage?

Here’s the arithmetic for a 2×4 cage’s real bedding bill. Three sourced inputs, one multiplication.

Humane World for Animals publishes both numbers you need: “Provide 2 to 3 inches of bedding so absorption is maximized,” and for two guinea pigs “at least 10.5 square feet (30″ x 50″) is preferred.” A 2×4 C&C cage, the size GuineaDad sells a liner for, is 27 by 56 inches — 1,512 square inches, which is exactly that 10.5 square feet.

How many litres does a full bedding change take?

1,512 square inches at 2 inches deep is 3,024 cubic inches, which is 49.55 litres. At 3 inches it is 4,536 cubic inches, or 74.33 litres. That is one full change.

Then the price per litre, from each maker’s own page. Small Pet Select’s 178 L paper bedding is $45.99, so $0.2584 a litre. Its 141 L aspen is $34.99, so $0.2482 a litre. That figure becomes $0.3089 if you use the 4 cubic feet the same page also claims.

Route	Cost per litre, from the maker’s own page	Per full change (2″ to 3″ deep)	A year at weekly changes	A year at fortnightly changes
GuineaDad fleece liners, two 2×4	Not sold by volume	$137.90 to $139.98, once	Laundry only after that	Laundry only after that
Small Pet Select aspen, 141 L basis	$0.2482	$12.30 to $18.45	$639 to $959	$320 to $480
Small Pet Select paper, 178 L	$0.2584	$12.80 to $19.21	$666 to $999	$333 to $499
Small Pet Select aspen, its own 4 cu ft figure	$0.3089	$15.31 to $22.96	$796 to $1,194	$398 to $597
Carefresh Natural 60 L	$0.4332	Excluded — see below	Excluded	Excluded
Kaytee Clean & Cozy	Kaytee publishes no price	Cannot be computed	Cannot be computed	Cannot be computed

Bedding only. Cage area and depth from Humane World for Animals; prices from each maker’s own product page on 31 August 2026; the multiplication is mine.

Why are Carefresh and Kaytee excluded from the cost table?

Neither publishes a checkable basis. Carefresh’s page gives a litre figure and never says whether it is compressed or expanded. Small Pet Select’s own products differ by 3.4 times across those two bases, so putting Carefresh’s $0.4332 next to Small Pet Select’s $0.2584 would look like a price comparison and would not be one. Kaytee is excluded because its product page carries no price at all.

The fleece gap is real and it is large. Two liners cost roughly what eleven weeks of paper costs. After that, they stop costing anything but detergent and a wash cycle.

The Change Interval’s Fourteen-Fold Problem

In a hot, humid climate, litter turns ammonia-sharp fast. Credible sources still put the change interval fourteen times apart.

John Gardiner writes the guinea pig care page for the UC Davis School of Veterinary Medicine. Published 11 April 2019 and updated 28 April 2026, it says: “Bedding should be changed daily to prevent high levels of ammonia from urine, and large quantities of feces that may contaminate food and water.” The Merck Veterinary Manual says to thoroughly clean the cage weekly. The federal Animal Welfare Act regulations for guinea pigs require that enclosures “shall be sanitized at least once every 2 weeks,” on top of cleaning often enough to prevent an accumulation of excreta or debris. The NRC Guide splits the difference and says bedding change “typically varies from daily to weekly.”

Take UC Davis literally at 49.55 litres a change and Small Pet Select’s paper price, and you get $4,673 a year. That is obviously not what UC Davis means. What UC Davis does not do is distinguish a full change from a spot clean. That ambiguity is in the source, not in my arithmetic.

So the table above runs weekly and fortnightly and stops there. If a page gives you one confident annual bedding cost, ask which interval it used and where it got it.

Which Guinea Pig Bedding Wins Overall?

Five axes, weighted, scored out of 100. Cost carries the most because it is the only line that recurs. Comfort is not an axis. No credible body publishes a measurement of it. A number I invented would be the weakest thing on this page.

Axis (weight, and the reason for it)	Small Pet Select paper	GuineaDad fleece	Small Pet Select aspen
Cost per year on a disclosed volume basis (30%) — the recurring bill, and the only one you pay again	40	95	35
Evidence behind the health story (20%) — what the literature actually measured	65	35	75
Figures the maker publishes for checking (20%) — a claim with no basis is not a spec	80	55	40
Labour, and what the routine sanitises (20%) — the chore you will actually do or skip	75	30	75
Volume-basis absorbency evidence (10%) — small weight, because none of it exists for these products	30	20	40
Weighted total	59.0	54.5	52.5

Weights are mine and are printed so you can change them. Push cost to 50% and fleece takes it comfortably.

Six and a half points separate first from last, which is the real finding: on evidence that exists, these three are close. Paper wins at 59.0 on the strength of one unglamorous thing — Small Pet Select is the only brand here that tells you which litre it is selling you.

How risky is fleece’s laundering routine?

Fleece’s laundering is the real risk, because the product wins the money axis outright but loses almost everything else that can be audited. GuineaDad’s own care instructions say to set the water temperature to cold to avoid shrinkage, and warn that too much heat “can even affect the absorbency of the liner.” The CDC’s laundry guidance puts thermal disinfection at “at least 160°F (71°C) for a minimum of 25 minutes.” The maker’s own instruction sits well below that temperature, so detergent, agitation and dilution end up doing the sanitising work instead. The CDC’s cold-water route also assumes a chlorine residual of 50 to 150 ppm that no fleece owner is running.

Metropolitan Guinea Pig Rescue, meanwhile, tells owners to wash fleece in hot water. Those two instructions are not compatible, and neither GuineaDad nor the rescue publishes a measurement of what either routine achieves.

Humid summers taught me how fast damp bedding turns to ammonia; GuineaDad Original Fleece Liner is the pick I’d point a friend to: $138 beats $320 to $999 a year in paper. Fleece at 54.5 and aspen at 52.5 is a tie in any honest reading, so that cost gap is the tie-breaker. The laundering worry is a real unknown rather than a documented harm, and you can also wash hotter than GuineaDad recommends and accept faster wear on a $69 item.

Aspen is the named loser at 52.5. Its health position is the cleanest of the three, and its own bag states two volumes that differ by 24 percent, which means the cheapest-looking option on the shelf cannot be priced to better than a quarter.

Each Route’s Wrong Buyer

Skip paper bedding if you are cost-driven and have a washing machine. It is the most expensive route on the table over any twelve-month window and the gap is not close. Buy it if you have no laundry access, or if a hamper with used fleece in it is not happening in your home.
Skip fleece liners if you cannot commit to sweeping droppings daily and running a wash once or twice a week. Fleece does not absorb; it passes urine to a layer underneath. Miss the schedule and you are not saving money, you are sitting your animals on a wet pad. Also skip it if you bought it for the bamboo health claim, which has nothing behind it.
Skip kiln-dried aspen if you want to know what you are paying per litre. Small Pet Select’s own page gives two volumes for the same bag. It is a fine material — a hardwood, outside the cedar literature entirely — sold with arithmetic you cannot check.

Frequently Asked Questions About Guinea Pig Bedding

How Much Guinea Pig Bedding Does a 2×4 C&C Cage Take?

About 49.55 litres at 2 inches deep and 74.33 litres at 3 inches, for a cage of 27 by 56 inches. That is one full change, and it means a single 60-litre bag is roughly one change, not a month of them.

Is cedar bedding dangerous for guinea pigs?

Avoid it, but the reason usually given is not the one in the literature. The NRC Guide’s objection is that cedar emits aromatic hydrocarbons that induce hepatic microsomal enzymes and cytotoxicity, and a 2026 review in the Journal of the American Association for Laboratory Animal Science states that no direct evidence links inhaled bedding dust to adverse respiratory outcomes in rodents. Use paper or a hardwood such as aspen, and ask an exotics vet about your own animal.

Setting up the habitat around all this? Our comparison of wire cages, glass tanks and C&C enclosures covers what the 10.5 square feet above actually costs to build, and the small pet cage space test measures three ready-made habitats against the same floor-area standard. Everything else lives in Birds & Small Pets.

July 16, 2026

Water Table vs. Splash Pad vs. Sprinkler: Which Backyard Water Toy Wins July?

Some links on our site may earn us a commission at no extra cost to you. Learn more.

Water play always needs an adult actively watching, whatever gear you choose. Follow the manufacturer’s age guidance and supervision instructions for any product here.

Choosing between a water table vs splash pad vs sprinkler mostly comes down to age: water tables own the 18-month-to-4-year window, splash pads are the cheap, zero-assembly pick for new walkers, and sprinklers are for runners — kids 4 and up with room to sprint. If you’re buying one thing for a toddler this summer, get the water table; it’s the only option of the three that holds attention for a full hour and doesn’t need the hose running the whole time.

Our picks at a glance

Best overall (toddlers): Step2 Rain Showers Splash Pond water table — two tiers, 13 accessories, and a “rainstorm” gimmick that keeps 2-year-olds busy longer than anything else we compared.
Best budget: SplashEZ 3-in-1 splash pad — around the price of a pizza night, folds flat, and doubles as a shallow wading pool for babies who can sit.
Best for big kids and groups: a simple oscillating sprinkler — the only option that scales to four sprinting seven-year-olds at once.

What actually separates the three?

The real differentiators are age fit, water consumption, and attention span — not price. A water table is a contained basin: you fill it once and the water stays put. Splash pads and sprinklers are hose-fed: they only work while water is actively flowing, which changes both the supervision picture and your water bill.

The dealbreakers are simple. No hose spigot within reach of the play area rules out splash pads and sprinklers entirely. A crawling baby who can’t stand yet rules out the water table. And a postage-stamp patio rules out the sprinkler, which throws water in a wide arc whether you like it or not.

The water math nobody runs

Here’s the comparison we couldn’t find anywhere else. A water table is a one-time fill: per Step2’s spec sheet, the Rain Showers Splash Pond holds 19 liters — about 5 gallons — and per Little Tikes’ specs, the Spiralin’ Seas table holds 26.5 liters, about 7 gallons. That’s the entire afternoon’s water budget. A splash pad or sprinkler, by contrast, consumes water every minute it runs, at whatever rate your hose delivers. Even at a conservative household hose flow, a running sprinkler passes a water table’s full 5–7 gallon fill in a handful of minutes — which means an hour of sprinkler play can use more water than a month of daily water-table refills.

Option	Typical price	Water per session (our estimate)	Setup each time	Best ages	Storage footprint
Water table (Step2, Little Tikes)	$45–$110	One 5–7 gal fill (manufacturer capacity specs)	2–3 min fill, dump after	18 mo–4 yr	Bulky — lives outside
Splash pad (SplashEZ and similar)	$15–$35	Continuous low hose flow while running	Under 1 min — unroll, connect	1–4 yr	Folds flat, shoebox-size
Sprinkler (oscillating or spinning)	$10–$40	Continuous full hose flow while running	Under 1 min	4 yr and up	Tiny

Prices reflect the ranges we found across current major-retailer listings in July 2026; water-use estimates are our own math from manufacturer capacity specs.

If you have a toddler: the water table is the whole ballgame

Between roughly 18 months and 4 years, nothing in this comparison touches a water table. The child stands, the toys spin, the water stays at hand height, and — critically — you’re not soaking the lawn to keep the fun going. The Step2 Rain Showers Splash Pond (about $80–$90 at major retailers, rated 18 months and up) is the standout: a two-tier design where kids fill a bucket up top and unleash a “rainstorm” on the pond below, with 13 included accessories. Reviewers at The Bump named it their top overall water table pick, with sturdy construction as the recurring praise and assembly as the only common gripe.

The budget alternative is the Little Tikes Spiralin’ Seas Waterpark table at about $45 — a bigger 7-gallon basin with a water wheel and a spiral tower kids drop balls through. Its textured base collects grime faster, so expect to scrub it. Pros for the category: contained water, long attention spans, no hose required after the fill. Cons: they’re bulky to store, algae builds if you don’t dump the water, and by age five the magic is gone.

If your kid is under two — or you have no storage: splash pad

A splash pad is a flat ring that sprays gentle fountains inward, and it’s the right call for the youngest water players. The SplashEZ 3-in-1 (a 60-inch pad, typically under $25) has a low rim so it doubles as a shallow wading pool for sitters, and the spray height adjusts with hose pressure — turn it down to a burble for a nervous one-year-old. When summer ends it folds to the size of a folded towel, which no water table can claim.

The trade-offs: it needs the hose running the entire time, thin vinyl can puncture on rough concrete, and the novelty fades faster than a table’s — there’s nothing to do but stomp the jets. It’s a strong second toy, and the best first one.

If your kids are 4+: the humble sprinkler still wins

Once kids can run, dodge, and invent games, a $15 oscillating sprinkler beats both purpose-built toys. It handles a whole birthday party at once, doubles as actual lawn irrigation, and stores in a drawer. The catch is water use — it’s the thirstiest option here by far, since it runs at full hose flow for as long as the shrieking continues — and it’s genuinely boring for kids under three, who mostly just get cold and cry. Grass gets slick, so shoes off and no sprinting on pavers.

How we compared

We compared current, widely stocked products in each category on price, age fit, water consumption, setup effort, durability patterns, and storage. We drew on manufacturer spec sheets for capacities and age ratings, editorial testing from parenting publications including The Bump, and recurring themes in verified owner reviews — praise and complaints that show up across many buyers, not one-off anecdotes. We don’t claim hands-on testing, and we computed the water-use comparison ourselves from published capacity specs. Read more about our approach on our editorial standards page. Prices are approximate, current as of publication, and change frequently — confirm the latest price with the retailer before buying.

FAQ

What age is a water table best for?

Roughly 18 months to 4 years. Most manufacturers, including Step2, rate their tables from 18 months; by around age 5 most kids lose interest and do better with a sprinkler or pool.

Do splash pads use a lot of water?

More than a water table, less than a sprinkler. A splash pad runs on continuous low hose flow, so total use depends on how long it runs — while a water table uses one 5–7 gallon fill for a whole session, based on manufacturer capacity specs.

Should I empty a water table after each use?

Yes. Standing water grows algae and attracts mosquitoes within a day or two, and manufacturers advise dumping and rinsing after each session. Look for a model with a drain plug to make this painless.

Keep the fun going

More from our Games & Activities guides: when the weather turns, see our comparison of board games vs. card games vs. puzzles for family game night, our breakdown of play kitchens vs. workbenches vs. sensory tables for indoor pretend play, and — for the grown-ups holding the towels — which backyard shade actually survives a breezy afternoon.

July 16, 2026

Robot Lawn Mowers Cost $4,000 to $9,000 an Acre

Some links on our site may earn us a commission at no extra cost to you. Learn more.

The verdict. On price alone the Ecovacs GOAT O1000 RTK at $999.99 is the cheapest capacity here, at $4,000 an acre. On the full ledger it finishes last at 38.3 of 100, behind the Segway Navimow i210 AWD at 51.0, the Mammotion LUBA mini 2 AWD at 48.8 and the eufy Robot Lawn Mower E15 at 46.8. The reason is a twelve-month warranty that names batteries as an excluded consumable, and a documented critical vulnerability on Ecovacs’ own previous mower generation. The cheapest acre on this page is not the cheapest mower to own.

Updated August 31, 2026: prices, coverage claims and slope figures re-read from each maker’s own store. The per-acre price table carries the footnote that undermines a simple ranking, the boundary-slope finding, the Ecovacs CVE record, and the disagreement between Consumer Reports and Stiftung Warentest on child detection. Segway has quietly dropped the X315 from its own collection page while still selling it.

Published July 16, 2026 · Updated August 31, 2026 · By Nate Colburn

Pricing Four Mowers From $4,000 to $9,000 an Acre

Every one of these machines is sold on a coverage number. Divide the maker’s own price by the maker’s own acreage and the spread is wider than the sticker prices suggest.

Model	Maker’s own price	Maker’s own maximum area	Dollars per claimed acre
Ecovacs GOAT O1000 RTK	$999.99	1/4 acre	$4,000
Segway Navimow i210 AWD	$1,149.00	0.25 acre	$4,596
Mammotion LUBA mini 2 AWD 1500	$1,799.00	0.37 acre	$4,862
eufy Robot Lawn Mower E15	$1,799.99 on the product page, $1,299.99 on eufy’s own collection page	0.2 acre (8,700 sq ft)	$9,000, or $6,500

The arithmetic, shown: $999.99 ÷ 0.25 = $3,999.96. $1,149.00 ÷ 0.25 = $4,596.00. $1,799.00 ÷ 0.37 = $4,862.16. $1,799.99 ÷ 0.2 = $8,999.95, and $1,299.99 ÷ 0.2 = $6,499.95.

Two things fall out before you reach any spec. eufy is the most expensive capacity here at either of its own prices, between 1.6 and 2.25 times the Ecovacs. And eufy publishes two prices for one machine on its own website on the same day: the product page and the E-series landing page both say $1,799.99, the lawn mower collection page says $1,299.99 with $500.00 marked as saved.

Prices are approximate, current as of publication, and change frequently — confirm the latest price with the retailer before buying.

Reading Segway’s 1.2x Map Footnote Before Trusting the Acreage

Now the part that makes the table above less decisive than it looks, and I would rather undercut my own number than let you use it wrong.

Segway attaches this footnote to its mower pages: “Supports map storage up to 1.2× the recommended mowing area (e.g. up to 1.2 acres for a 1-acre model). Actual mappable area may vary depending on yard layout and complexity.” The advertised acreage is a map budget, not a lawn. It is shared across every zone you draw, so a front, a back and a side strip all draw from the same pool.

Segway publishes the upward band on several models in the same breath: the X450 “may support a maximum mowing area of up to 1.8 acres” against a 1.5-acre headline, the H220 up to 0.6 against 0.5, the i215 up to 0.44 against 0.37, and the i206 up to 0.18 against 0.15. That is a consistent 20 percent of headroom, stated by the maker.

Apply it to the i210. At 0.25 acre the machine works out at $4,596 an acre; at the 0.30 acre its own footnote allows, $1,149.00 ÷ 0.30 = $3,830.00. Segway’s own published band runs $3,830 to $4,596, and that band contains the Ecovacs figure of $4,000 outright and comes within $266 of the Mammotion.

So the honest reading of the first table is this: three of these four mowers sit inside one maker’s own stated uncertainty, and the ordering between them is not a real ordering. Only eufy is far enough out to be a genuine result. And the downside is worse than the upside, because Segway publishes a ceiling and no floor — “actual mappable area may vary” with complexity, and it does not say by how much.

Counting the Slope Twice, Because Two Makers Publish Two Numbers

The slope rating is the second number people buy on, and on two of these brands it comes in a pair.

Model	Headline slope	The figure that governs the edge	Conditions the maker states
Mammotion LUBA mini 2 AWD	“80% (38.6°)”	Not published	None stated
Segway Navimow i210 AWD	“up to 45%(24°)”	Not published for this model	“Measured under test conditions with 1.97 inch grass height, dry grass surface, and a slope free of obstacles.”
Segway Navimow i110	“30% (17°)”	“10%(6°)” at boundary	None stated
Ecovacs GOAT O1000 LiDAR PRO	“45%(24°)”	17% for virtual boundary	“Experimental result operated by ECOVACS Laboratory… Actual results may vary”
eufy E15 / E18	“18°”	Not published	“For grass under 3.5 inches and flat lawns (height difference under 1.2 inches). Not suitable for St. Augustine and dense Zoysia.”

The boundary figure is the one that decides whether a machine works on your property. A mower that climbs a 30 percent grade mid-lawn but holds a virtual boundary only at 10 percent will not finish a sloped yard, because on most sloped yards the slope runs to the edge. Segway prints 30 and 10 for the i110. Ecovacs prints 45 and 17 for the O1000 LiDAR PRO. Both sit on the maker’s own page, and the small one never makes the headline.

I could not find a published boundary figure for the Mammotion, the eufy, or the Segway i210 AWD. That may mean they do not have the limitation. It may mean they do not print it. Read the second number carefully on any model that has one, and ask about it on any model that does not.

Segway’s test setup is worth reading twice as well. Its 45 percent was measured on grass just under two inches, dry, with no obstacles on the slope. Wet grass on a hill with a tree root in it is not that test.

Weighing Safety When Consumer Reports and Stiftung Warentest Disagree

This is the most important section on the page and it does not resolve.

Consumer Reports published its robotic mower testing on July 14, 2026, covering six machines: the Segway Navimow i105N and X315, the Mammotion Luba 2 AWD 3000X and Luba Mini AWD 800, the eufy E18 and the Ecovacs Goat O1000 RTK. CR states that “We assess whether the robotic mower stops when a person gets close to it and whether its blades stop automatically if someone attempts to lift it,” and reports top marks for safety among the models it recommends.

Stiftung Warentest published its own test on March 23, 2026, covering 14 mowers including Anker Eufy, Ecovacs and Mammotion. Its protocol goes further: alongside adult and child foot dummies, testers used a wooden replica of a crawling child’s arm, 4 cm across and 55 cm long, and hedgehog dummies. On whether the machines detect the foot replicas in time, Warentest’s own free preview says, in German, that many do not.

Two of the most credible consumer testing organisations in the world, same product category, same year, and they do not agree on whether these machines detect a child. I am not going to pretend one of them is right. The protocols differ, the model samples differ, and Warentest’s full results sit behind a €4.90 paywall. A German outlet summarising the paywalled issue reports that ten of fourteen machines drove over the wooden arm; I could not verify that count at test.de itself, so treat it as secondary reporting rather than as Warentest’s published figure.

What is not in dispute: no US safety certification is required. The CPSC lists ANSI/OPEI 60335-2-107-2020, “Particular Requirements for Robotic Battery Powered Electrical Lawnmowers,” under voluntary standards. And the one mandatory federal mower rule does not reach these machines at all: 16 CFR Part 1205 covers “A grass cutting machine either pushed or self-propelled… normally controlled by an operator walking behind the mower,” and requires the blade to “come to a complete stop within 3.0 seconds after release of the control.” A robot mower has no operator walking behind it and no control to release. The three-second figure you see quoted for robot mowers comes from a rule that does not apply to them.

None of the four publishes a certification number against the voluntary standard on its US product pages. Every one of them necessarily holds an FCC equipment authorisation for its radios, and those are publicly searchable. The transmitter is regulated and the blade is not. Run yours when the yard is empty and follow the manufacturer’s own guidance on children and pets, whichever machine you buy.

Scoring Four Robot Lawn Mowers on Five Axes, With the Tally Shown

Weights are mine and the reasoning is beside each one. The first axis scores how much independent testing exists on the exact model you would buy, not how safe the machine is; the published evidence does not support scoring the second thing.

Axis and weight	Why it carries that weight	Segway i210 AWD	Mammotion LUBA mini 2	eufy E15	Ecovacs O1000 RTK
Independent testing of this exact model — 30%	Unattended blades in a yard. Somebody outside the company should have looked.	45 — CR tested the i105N and X315, not this one	45 — CR tested the previous Luba Mini AWD 800	40 — CR tested the E18 sibling; eufy publishes a 5 cm detection floor	65 — CR tested this exact SKU; Warentest tested the line
Warranty and what it excludes — 20%	Three years of grass is the test that matters, and one of these is a one-year machine.	55 — “3-year warranty,” but “Up to 3-year” on the H2 and no US terms document found	75 — 3 years on components and battery; blades, discs and cables excluded as wearing parts	70 — 2 years on mower, base station, adapter and battery pack, published as a table	20 — 12 months, with batteries named as an excluded consumable
Running-cost disclosure — 25%	Blades, batteries and connectivity are the bill after the box.	70 — blades $23.99 at 80 hours, battery prices for six SKUs, $159.99 to $449.99	45 — blades $49.00 with no interval; 4G called lifetime on one page, 3 years on another	50 — blades $11.99 at 240 hours; no replacement battery sold	35 — blades $29.99 on a 4-6 week calendar; no battery price
Published vulnerability record — 15%	These are internet-connected machines with cameras and radios.	50 — no NVD record, which means no published research	50 — no NVD record	45 — 14 NVD records, all cameras and HomeBase, none a mower	30 — a CVSS 9.6 critical on its own previous mower generation
Certification you can look up — 10%	Low weight, because none of them separates here.	15	15	15	15
Weighted total		51.0	48.8	46.8	38.3

Ecovacs’ arithmetic, since it is the surprising one: 0.30 × 65 = 19.5, plus 0.20 × 20 = 4.0, plus 0.25 × 35 = 8.75, plus 0.15 × 30 = 4.5, plus 0.10 × 15 = 1.5. Total 38.25, rounded to 38.3.

That is the named loser, and it is the best-tested machine on the page. It still finishes last, because a twelve-month warranty excluding the battery on a $1,000 machine is a short leash, and because CVE-2024-52325, scored 9.6 critical, records that “ECOVACS robot lawnmowers and vacuums are vulnerable to command injection via SetNetPin() over an unauthenticated BLE connection.” That CVE names the GOAT G1 family and the GX-600, not the O1000 you would buy today, and Ecovacs issued advisories. It is the vendor’s record, not this model’s defect, and I scored it that way.

Disagree with 15 percent on vulnerability history and drop it to 5, moving the difference to independent testing, and Ecovacs climbs past eufy. The axes are printed so you can do exactly that.

Budgeting Blades on Three Clocks That Do Not Convert

Four makers, four different units for one consumable.

Segway sells a 12-piece Blade Assembly Plus for $23.99 and says “Typically the blades should be replaced after 80 hours of use.” eufy sells blades and screws for $11.99 and says to “replace all three blades and screws at the same time” after 240 hours. Ecovacs sells a 12-blade kit for $29.99 and gives the interval as “About 4-6 weeks.” Mammotion sells Endurance Replacement Blades for $49.00 and publishes no interval I could find.

Eighty hours against 240 is a threefold difference in stated blade life between two makers of the same kind of machine, and Ecovacs’ figure is a calendar window that will not convert without knowing your run time. None of the four publishes how many blades one change consumes out of the pack it sells you, so a cost per mowing hour is not computable for any of them. I am not going to invent the missing number to make a tidier table.

The connectivity bill is where the real money is, and Segway is the one to watch. Its i105 page footnotes that “Anti-theft requires separate purchase of Navimow Access+ 4G module,” listed at $95.99 on sale against $119.99 regular. Segway also rents mowers outright: Navimow Flex charges $100 a month for the i105 tier covering an eighth of an acre. Twelve months of that is $1,200, against Segway’s own $669 sale price for the same i105. The subscription costs 1.79 times the machine in the first year. Mammotion goes the other way and states that its iNavi service and 4G data are “free for the entire lifecycle of the product” — though its own product pages say 3 years, and I am printing both because Mammotion does.

Positioning has a physical prerequisite too. NOAA’s National Geodetic Survey lists sites unsuitable for a real-time GNSS base station as including “locations under or very near tree canopy.” Segway’s own support article asks for at least 110 degrees of unobstructed sky and no obstacles within two metres. If your lawn is under trees, a government agency and the manufacturer are telling you the same thing.

Naming the Yard Each of These Four Is Wrong For

Segway Navimow i210 AWD, $1,149. Wins this ledger on disclosure rather than on hardware: Segway publishes battery prices for six separate SKUs down to the cent, an 80-hour blade interval, and the fine print on its own coverage claim. It also charges roughly $96 to $120 extra for the 4G module that makes anti-theft work on the entry models. Skip this if your lawn is under mature trees, because the whole platform leans on satellite positioning that NOAA says canopy degrades.

Mammotion LUBA mini 2 AWD, $1,799. The best warranty here at three years including the battery, all-wheel drive rated to an 80 percent grade, and free connectivity for as long as the machine lives — or three years, depending which Mammotion page you open. Note also a pricing inversion on Mammotion’s own store: the LUBA 3 AWD 1500 covering 0.37 acre lists at $2,399 while the LUBA 3 AWD 3000 covering 0.75 acre lists at $2,239. More mower, less money, same page. Skip this if your lawn is flat and rectangular, because the AWD premium buys you nothing there.

eufy Robot Lawn Mower E15, $1,299.99 or $1,799.99. Camera navigation with no RTK antenna and nothing to mount, and a warranty table that covers the mower, the base station, the adapter and the battery pack for two years apiece. Then it charges between 1.6 and 2.25 times the Ecovacs per acre and states in its own footnote that objects under 5 cm tall and smaller than 10 by 10 cm “need to be cleared or marked as No-Go Zone.” Skip this if your yard has a dog toy problem, or if you want the maker to agree with itself about the price.

Ecovacs GOAT O1000 RTK, $999.99. The cheapest acre on the page, the only model here that Consumer Reports has tested by exact SKU, and a published 0.8-inch positioning precision figure that Segway does not match with any numeral of its own. Twelve months of warranty, batteries excluded. Skip this if you expect to still be mowing with it in year three, because at that point you own every component in it outright.

A note on recalls: a search of CPSC’s published recalls found none for Segway, Anker or eufy, Ecovacs or Mammotion robot mowers. The Segway recalls that surface are personal transporters, UTVs and a kick scooter, none of them a mower.

Settling a 2.2-Point Gap Between Segway and Mammotion

Segway 51.0, Mammotion 48.8. Two points on a hundred-point scale is not a verdict, so here is the tie-breaker.

My own money goes to the Mammotion, and it costs me the top of my own ledger to say so. Segway wins on paperwork. Mammotion wins on the two things that are hardest to buy back later: three years of warranty that includes the battery, and connectivity that does not carry a meter. Segway’s answer to anti-theft is a $96 module and its answer to convenience is $100 a month, and both of those are decisions the company can revisit after you have bought the mower. A warranty is not.

Move running-cost disclosure from 25 percent down to 15 and put the difference on warranty, and Mammotion takes the ledger outright at 51.8 against Segway’s 49.5. That is how close this is.

We have gone at the rest of the automated backyard the same way: robotic pool cleaners, what a $1,200 robot vacuum does that a $200 one cannot, and grill formats for when the mowing is finished. The rest lives in beach and backyard.

If you own one of these four and your yard has a slope at the boundary, tell me what it actually managed. That is the number none of these companies will print.

July 16, 2026

Cosori vs Typhur: Real Air Fryer Capacity Compared

Some links on our site may earn us a commission at no extra cost to you. Learn more.

In the cosori vs typhur comparison, Consumer Reports measured just 4.5 quarts of capacity in the Cosori TurboBlaze CAF-DC601-KUS — the widest gap of any fryer here. Cosori sells it as 6 quarts. That is the widest gap of the four machines on this page, but it is not the only one. The Instant Pot Vortex Plus 6QT (140-3089-01) measured 5.2 against an advertised 6. The Ninja DoubleStack XL SL401 measured 9 against an advertised 10, and the Typhur Dome 2 AF03 measured 5 against an advertised 5.6. Every one came in under. Only Typhur publishes a usable cooking area on its own spec sheet. It is also the most expensive machine here, by a factor of nearly three.

After years of buying ordinary kitchen gear, I have learned the number on the box and the number you get rarely match. So a machine that measures short of its advertised capacity doesn’t surprise me. It just tells me which brand rounds up the most.

Updated August 31, 2026: capacities re-verified against Consumer Reports’ measured figures for all four machines. Every price traces to each maker’s own current listing — Typhur’s Dome 2 lists at $339.99. Added the February 2023 Cosori recall and its twenty-one model numbers, none of which is the TurboBlaze.

Published July 16, 2026. Last updated September 8, 2026. By Nate Colburn. See our editorial standards.

Cosori’s 6-Quart Basket Measured 4.5 Quarts in Consumer Reports’ Test

Consumer Reports measured just 4.5 quarts inside the Cosori TurboBlaze CAF-DC601-KUS, which Cosori sells as a 6-quart basket. They are also the number nothing forces a maker to define. A basket’s raw volume, measured to the rim, is not the volume you can cook in. No US rule says which of the two goes on the box.

Consumer Reports publishes both a Claimed capacity field and a Measured Capacity field on its air fryer pages. On all four of these machines the claimed field is blank. The unit “qt.” prints with no numeral in front of it. The measured field is filled in on all four. So the comparison below is mine. It sets Consumer Reports’ measured figure against the number each maker prints on its own product page, which I read on the same day.

Machine	What the maker prints	Consumer Reports measured	Short by	Measured share of advertised
Cosori TurboBlaze CAF-DC601-KUS	“6 qt / 5.7 L (serves 3–5 people)”	4.5 qt.	1.5 qt	75.0%
Instant Pot Vortex Plus 6QT 140-3089-01	“6.0 Quart”	5.2 qt.	0.8 qt	86.7%
Ninja DoubleStack XL SL401	“10 QT capacity lets you cook two, 5-lb chickens at once”	9 qt.	1 qt	90.0%
Typhur Dome 2 AF03	“5.6 qt”	5 qt.	0.6 qt	89.3%

Verdict on the claim: all four advertised figures are higher than what an independent lab measured. Cosori’s is the one that changes a buying decision. Three quarters of the advertised number is a real difference. It’s the gap between a fryer that holds dinner for four and one that holds dinner for three.

How Much Do These Air Fryers Fall Short of Their Advertised Capacity?

Across all four machines, the measured shortfall runs from 0.6 quarts on the Typhur to 1.5 quarts on the Cosori, below each one's advertised capacity. A rim-full measurement is a real measurement. It is just not the one you cook with. Every maker gets to pick which one it prints.

What none of the four publishes is a maximum fill line. I read Cosori’s manual for the TurboBlaze and Typhur’s manual for the Dome 2. Cosori’s says food that touches the heating coils is a fire hazard and tells you not to overfill. Typhur’s says not to exceed “the maximum level indicated in the basket.” That’s a mark moulded into the plastic, with no number attached to it anywhere in the document. Instant’s manual for the 140-3089-01 carries the same generic instruction and no figure.

So the honest position is that the usable number is not published by anyone selling these machines. It exists in one place, and that place is a lab you have to pay to read.

Typhur Dome 2: 139.5 Square Inches Under a 2.3-Inch Ceiling

Typhur is the only maker that publishes a usable cooking area. That area is 139.5 square inches under a 2.3-inch ceiling. Typhur prints “Cooking Area: 139.5 sq in” and “Basket Height: 2.3″” directly on the Dome 2’s spec table. Those two numbers tell you more about what fits than 5.6 quarts does. That’s because an air fryer cooks in a layer, not a volume.

For the Cosori, the equivalent figure comes from outside. TechGearLab lists a measured cooking area of 85.6 square inches for the TurboBlaze. Typhur’s published area is 1.63 times that. Ninja and Instant publish no area figure at all. I could not find an independently measured one for either, so those two cells stay empty rather than get filled with a guess.

Ninja does publish a food-weight claim instead: “10 QT capacity lets you cook two, 5-lb chickens at once.” That’s one chicken per drawer. It is a more useful sentence than a quart figure. It is also the only one of its kind on any of these four pages.

Cosori vs Typhur: Which Air Fryer Buys the Cheapest Volume?

The Ninja DoubleStack XL buys the cheapest volume at $24.44 per measured quart. That’s once price is divided by measured capacity instead of the advertised one. Prices below are what each maker’s own store displayed on August 31, 2026.

Rank	Machine	Maker’s own price	Measured quarts	Dollars per measured quart
1	Ninja DoubleStack XL SL401	$219.99 (from $249.99)	9	$24.44
2	Cosori TurboBlaze	$119.99	4.5	$26.66
3	Typhur Dome 2	$339.99 (from $399.99)	5	$68.00
—	Instant Pot Vortex Plus 6QT	no price rendered	5.2	not computed

The Typhur costs 2.78 times what the Ninja costs per measured quart. That is not an argument against it. A wide shallow layer under two elements is a different machine from a deep drawer, and 139.5 square inches of single-layer space is the thing you are paying for. It is an argument against buying it for capacity.

Prices are approximate, current as of publication, and change frequently — confirm the latest price with the retailer before buying.

Instant Vortex Plus’s Price Doesn’t Render on Instant’s Page

The Instant Vortex Plus is missing from that ranking. Its current price could not be read from Instant's own product page. That page, for the 140-3089-01, publishes the capacity (“6.0 Quart”) and the wattage (“1700 watts, 120 V~ 60 Hz”). It renders no price numeral at all. The price loads in the browser afterwards.

That is a narrower claim than “Instant publishes no price.” It publishes one. I could not retrieve it from the page. I am also not going to substitute a retailer’s number for a maker’s and then rank on it. The cell stays empty, and the ledger below charges Instant for it. A price you cannot read is a price you cannot compare.

Cosori’s February 2023 Recall Covered Twenty-One Model Numbers

On February 23, 2023 the Consumer Product Safety Commission announced a recall of about 2 million Cosori air fryers, with a further 250,000 in Canada and 21,000 in Mexico. The hazard: “A wire connection in the air fryers can overheat, posing fire and burn hazards.” The CPSC notice records 205 reports of air fryers catching fire, burning, melting, overheating and smoking, including 10 reports of minor superficial burn injuries and 23 of minor property damage.

Twenty-one model numbers are named. They are CP158, CP137, CS158, CO137, CO158, CP258 and CAF-P581 variants. CAF-DC601-KUS — the TurboBlaze — is not among them, and the recall predates the line. I am telling you this because a two-million-unit fire recall on a brand is worth knowing before you buy that brand. It’s not because this machine is subject to it. Instant Brands’ only CPSC recall I could find is a 2015 pressure cooker action. I found no CPSC air fryer recall for SharkNinja or for Typhur.

Cosori vs Typhur: Which Air Fryer Wins on the Scorecard?

On a five-axis weighted scorecard, the Ninja DoubleStack XL edges out the Cosori TurboBlaze, 79.8 to 78.5. The Typhur Dome 2 sits at 66.8, and the Instant Pot Vortex Plus is last at 36.8.

These four air fryers were scored across five weighted axes. Capacity carries the heaviest weight, because it is the thing every one of these is sold on and the thing all four overstate. Price per measured quart comes second, because it is the only price comparison here that survives contact with the measurements. The three 15% axes are disclosure axes: what the maker tells you before you pay.

Axis (weight)	Ninja SL401	Cosori TurboBlaze	Typhur Dome 2	Instant Vortex Plus
Usable capacity you can verify (30%)	75	40	95	65
Dollars per measured quart (25%)	100	92	36	0
Warranty as written (15%)	50	100	65	50
Speed claim with a named baseline (15%)	70	90	35	45
Price and stock on the maker’s store (15%)	95	100	95	20
Weighted tally	79.8	78.5	66.8	36.8

Cosori vs Typhur: How Is Each Axis Scored?

Three of those axes need their scoring explained. Warranty: Cosori gives two years. Typhur gives one that stretches to eighteen months if you register within 30 days. Ninja and Instant give one. Speed claims: Cosori’s US page says “cooks meals up to 46% faster*” and the asterisk names the comparator — “COSORI’s CP168 (or Pro Gen 2) air fryer.” A named baseline is rare, and it scores well. Typhur’s “up to 30%” is against “conventional models,” which is not a comparator. Instant’s “50% faster” is against deep frying and a conventional oven, a category rather than a machine. Ninja’s US page made no speed claim I could find. That’s why it scores above the two vague ones: a claim not made cannot mislead.

Disagree with a weight? Move it. Push capacity to 45% and the Typhur closes on the leaders. Push price per measured quart to 40% and the Ninja walks away with it. That is the point of showing the arithmetic.

Cosori vs Typhur: Which Air Fryer Is Wrong for Your Kitchen?

Skip the Ninja DoubleStack XL if you cook one thing at a time. Nine measured quarts split across two stacked drawers is not one big basket. Each drawer is also shallower than a standard 6-quart. Its whole case is two dishes finishing together.

Skip the Cosori TurboBlaze if you are buying on the quart figure. It is the best-value single basket here. It is also the machine with the largest gap between the box and the lab. At 4.5 measured quarts, it is a two-to-three-person fryer wearing a three-to-five-person label.

Skip the Typhur Dome 2 if your reason is volume. $68 a measured quart is 2.78 times the Ninja’s rate. Buy it for the 139.5-square-inch single layer and the 12-inch pizza that fits in it, or don’t buy it.

Skip the Instant Pot Vortex Plus if you want to price it before you shop. It is the named loser here at 36.8. Most of that deficit is disclosure rather than hardware: 5.2 measured quarts is the second-best result on this page, on a machine whose own store would not show me what it costs.

Ninja vs Cosori: 1.3 Points Apart on the Scorecard

After watching kitchen gear fail right when the warranty ends, the Cosori TurboBlaze is the one I would point a friend to, because two years of coverage beats a bigger drawer. The Ninja and the Cosori finish 1.3 points apart. That’s 79.8 to 78.5 — effectively a tie on the final tally. The tally will not settle it, so I will.

I would spend my own $119.99 on the Cosori TurboBlaze. Two years of warranty against one, a named baseline behind its speed claim, and a hundred dollars kept. The Ninja wins on dollars per measured quart because it is a bigger machine. If you are actually feeding four people twice a week, that is the right trade. If you are two people, you are paying $100 more for a second drawer. You’d also have a taller footprint to measure against your cabinets first.

One more thing, since you are trusting my arithmetic. Every capacity figure above is a single lab’s measurement of a single unit. Consumer Reports does not publish its fill protocol on these pages. I also have no second lab to check it against. That is a real limit on the four numbers this whole comparison rests on, and you should know it before you use them.

Already sold on the category and want the machine, not the audit? Our five-axis ledger on five current air fryers scores the field on cooking performance instead. The same advertised-versus-usable problem turns up one appliance over in electric pressure cookers, where Breville’s own manual caps a 6-quart at 4.5, and if you are still deciding between appliance categories, start with the countertop oven comparison. More in Small Appliances.

July 16, 2026

Razor Bumps: Stubble Length Beats Any Shaver You Buy

Some links on our site may earn us a commission at no extra cost to you. Learn more.

The verdict. Scored on skin rather than closeness, the Merkur 34C double-edge safety razor takes this ledger at 66.3 of 100. Behind it: the Gillette SkinGuard cartridge at 49.3, the Panasonic ARC6 ES-LS8A-K at 44.3, the Braun Series 9 PRO+ 9677cc at 35.3, and the Philips Norelco i9000 Prestige Ultra XP9405/84 at 15.3. Not because a bare blade is gentle. Because every clinical source I could read points at the same variable for razor bumps, and that variable is how much stubble you leave behind, not which machine removes it.

Updated August 31, 2026: every price and part number re-checked against each maker’s own store. Philips now lists the SH98/72 replacement head as discontinued and says its successor does not fit the rounded S9000 Prestige bodies. The published evidence behind Gillette’s razor-bump claim is cited below: twenty men, no control arm, and Gillette’s UK page describes that study differently than the paper does. Scored throughout on skin outcomes rather than closeness.

Published July 16, 2026 · Updated August 31, 2026 · By Nate Colburn

If Braun’s 31 Percent Sounds Like Skin Protection, Read the Footnote

Five shaving systems. Four of them print a number about your skin. Not one of those numbers is a measured irritation rate, and the comparator, where a comparator exists at all, is the maker’s own older product.

System	The claim, as printed	The number	What that number measures	Measured against
Braun Series 9 PRO+ 9677cc	Blades that “shave more efficiently than any previous Series 9 electric shaver, resulting in exceptional skin protection”	31% higher efficiency	Shaving efficiency. The skin claim is drawn as a consequence of it.	The Braun Series 3, per the footnote on Braun’s own product page
Panasonic ARC6 ES-LS8A-K	“reduces pressure on the skin by 10% to help minimize irritation”	10%	Contact pressure. A mechanical load, not a clinical outcome.	Panasonic’s own 5-blade LV97, per the footnote on Panasonic UK
Philips i9000 Prestige Ultra XP9405/84	“50% smoother gliding on skin”; cuts “up to -0.08 mm root-level”	50%	Coating friction	Not stated anywhere beside the claim
Gillette SkinGuard	“Clinically proven and dermatologist tested for sensitive skin”	None on Gillette’s US pages	—	—
Merkur 34C	No skin claim of any kind	—	—	—

Two rows deserve a second look. Panasonic’s sentence does the work in one move: a 10 percent drop in contact pressure arrives, and the phrase “to help minimize irritation” arrives attached to it. Pressure is a force. Irritation is a skin response. Panasonic publishes the first and implies the second.

Philips is the odd one. Its buying page sells “ultimate skin comfort” and a cut 0.08 mm below the surface. Its own support article, titled around irritation after using a Philips shaver, says this: “Allow yourself an adjustment period of two to three weeks with your new Philips shaver system.” The marketing and the help desk are describing different products.

If You Read One Razor Bump Number, Read 59.6 Percent

Gillette’s razor-bump claim does rest on published work. Moran and colleagues, writing in Cutis in 2022, report a mean papule reduction at week 12 “estimated to be 59.6% (P<.0001)” among men with pseudofolliculitis barbae who switched to the SkinGuard regimen.

Read the rest of the paper and the number gets smaller. The design is described by the authors as “A prospective, open-label study … at a single site in the United States.” Twenty-one men enrolled. Twenty finished. The authors write plainly that “A lack of a control arm could be considered a limitation of the study.” The funding line reads: “The clinical study was funded by the Procter & Gamble Company.”

Now the part that matters more than any of it. Gillette’s UK page about that same 20-man study calls it “The randomized, investigator-blinded study.” The published paper calls it open-label with no control arm. Those two sentences cannot both describe one trial. I am not picking between them. I am telling you they disagree, and that the more favourable description is the one printed by the company selling the razor.

Twenty men improving against their own baseline over twelve weeks is a real result. It is not evidence that a SkinGuard beats a foil shaver, a rotary, or a safety razor, because none of those was in the study.

The Cutis paper is public. Read the limitations section before you read the 60 percent.

If Dermatologists Agree on a Mechanism, It Is Stubble Length

Every guidance source converges on the same physical story: a hair cut short enough retracts below the skin surface, and the trouble starts there. The fix is to stop cutting it that short.

The NHS puts it in its don’t list: “do not shave too close – leaving some stubble can stop bacteria getting in,” alongside “do not shave every day, if possible.” DermNet’s pseudofolliculitis barbae page says to leave approximately 1 mm of stubble. A 2023 Cutis review of US military PFB policy puts the working figure at 0.125 to 0.25 inches, maintained with clippers.

Those two numbers are not close. One millimetre against roughly 3.2 to 6.4 millimetres is a factor of three to six, and it is the difference between “shave carefully” and “stop shaving and start clipping.” Both are dermatology sources. Both are current. I am printing both.

Frequency is contested too. The American Academy of Dermatology says “To see fewer razor bumps, shave daily — or at least every 2 to 3 days.” The NHS says the opposite. And the only randomised trial I found on the question sides with the NHS on one endpoint.

If You Want a Trial of Electric Against Blade, I Could Not Find One

Daniel and colleagues randomised 90 African American men across three shaving regimens for twelve weeks and published the result in the Journal of Drugs in Dermatology in 2013. The finding: “There were no significant differences noted between the 3 groups for papule (P=.32) or pustule (P=.46) count for the 12-week study.” On ingrown hairs the less-frequent-shaving control group did better, at P=.005 against daily standard shaving.

Ninety men. Twelve weeks. Randomised. And what it varied was frequency and products, not device type.

That is the strongest evidence in this whole category, and it does not name a shaver. I looked for a controlled comparison of foil against rotary against blade on skin outcomes and did not find one. DermNet does state that PFB “is more common with blade razor users compared to electric shavers,” but attributes that sentence to no study, and I am not building a verdict on it.

The stakes are not small. Military Health System surveillance records “The increase in reported cases from 50 in 2000 to 2,404 in 2022,” with non-Hispanic Black service members constituting “a majority (63.5%) of PFB cases.” A condition that documented, that concentrated, and the trial that would settle foil against blade has not been run.

Close-up of a man shaving his jawline with a razor and lather — One pass, with the grain, is the technique every guidance source above agrees on.

If You Score These Five on Skin, Merkur Wins by 17 Points

Five axes, weighted for a skin question rather than a closeness question. Weights are mine and the reasoning is printed beside each one, so you can move them and re-run the tally yourself.

Axis and weight	Why it carries that weight	Merkur 34C	Gillette SkinGuard	Panasonic ARC6	Braun Series 9 PRO+	Philips i9000
Support behind the skin claim — 30%	The claim is what you are buying on. A number with no comparator is not a number.	60 — makes no claim; AAD names a single-blade razor as a recommended option	45 — a real published study, weakly designed and described two ways	25 — 10% against its own LV97, and it measures pressure	20 — 31% is efficiency against a Series 3	15 — 50% gliding, no comparator printed
Control over stubble length — 25%	This is the only mechanism every guidance source names.	55 — one pass, one blade, no lift-and-cut	65 — the two-blade guard exists to cut at skin level rather than under it	30 — close, with no length you can set	30 — same	10 — Philips advertises cutting 0.08 mm below the surface
Running cost you can compute — 15%	A cost you cannot work out from published figures is a cost you cannot budget.	90 — price and an independent blade-life figure both exist	40 — price published, no cartridge life	70 — $125.99 and a 1–2 year interval, both from Panasonic	35 — 18-month interval, no price on Braun’s site	20 — interval contradicts itself, no price
What the maker discloses — 15%	Naming your comparator is the cheapest honesty there is.	50 — nothing claimed, nothing to disclose	25 — “clinically proven” names nothing on the US page	55 — names the LV97 and states the quantity	40 — names the Series 3, then bridges to skin	15 — no footnote at all
Will the part exist in five years — 15%	A shaver you cannot re-blade is a disposable shaver.	90 — the DE blade is a commodity standard	65 — refills in stock; the obvious product URL 404s	70 — one part number, WES9600P, currently sold	70 — head 94M currently sold	20 — SH98/72 discontinued, successor does not fit the rounded bodies
Weighted total		66.3	49.3	44.3	35.3	15.3

Worked example, so the arithmetic is visible. Merkur: 0.30 × 60 = 18.0, plus 0.25 × 55 = 13.75, plus 0.15 × 90 = 13.5, plus 0.15 × 50 = 7.5, plus 0.15 × 90 = 13.5. Total 66.25, rounded to 66.3.

The loser is the Philips i9000 Prestige Ultra at 15.3, and it is the most expensive electric here at Philips’ own $329.96. It advertises a cut below skin level on a page selling skin comfort, prints a 50 percent figure with no comparator, and its own support desk tells new owners to expect two to three weeks of irritation. Move my weights around all you like. That combination does not climb far.

If You Want an Annual Blade Bill, Only Merkur Publishes Both Numbers

A running cost needs two published figures: what the consumable costs, and how long it lasts. Across five systems, exactly one pairing gives you both.

System	Consumable price, from the maker	Replacement interval, published	Annual cost you can actually compute
Merkur 34C	$0.35 a blade in the 100-pack, $1.00 a blade in the 10-pack, on Merkur’s own US store	Not from Merkur. The AAD puts a single-blade razor at “5 to 7 shaves.”	$9.10 to $36.40
Panasonic ARC6	WES9600P foil and blade set, $125.99	“once every 1-2 years”	$63.00 to $125.99
Braun Series 9 PRO+	Head 94M — Braun’s US site publishes no price I could retrieve	“every 18 months”	Interval known. Cost unknown.
Philips i9000	SH91/52 — no price on Philips’ US page	SH91 says one year. SH98 says within two years. Same 4.5 million hairs.	Neither figure is firm.
Gillette SkinGuard	$3.75 a cartridge, from Gillette’s $15 four-pack	Gillette publishes none	Price known. Interval unknown.

The Merkur arithmetic, shown. The NHS advises against shaving every day, so take every other day: 182 shaves a year. At the AAD’s 5 to 7 shaves per single-blade, that is 26 to 36.4 blades. At Merkur’s own $0.35 blade, $9.10 to $12.74 a year. At its own $1.00 blade, $26.00 to $36.40.

Note what I did not do. I did not apply the AAD’s 5-to-7 figure to the Gillette. That number is stated for a single-blade razor, and a four-edge cartridge is a different object. Borrowing it would have produced a tidy annual number built on nothing.

One more finding from the same shelf: Merkur’s own store sells blades at $1.00 each in tens and $0.35 each in hundreds. A 2.9× spread on one maker’s site. Anyone telling you DE blades cost pennies is quoting the hundred-pack.

Prices are approximate, current as of publication, and change frequently — confirm the latest price with the retailer before buying.

If You Are Down to One of Five, Here Is Who Each One Is Wrong For

Merkur 34C. Solingen-made, listed on Merkur’s German store at €52.00 including VAT; I found no US dollar list price on a Merkur-operated page for the razor itself, only for the blades. It wins this ledger on evidence and on cost, and it demands technique you will not have in week one. Skip this if you shave in a hurry before a commute, or if the phrase “learning curve” applied to a razor sounds absurd to you. It is not an unreasonable reaction.

Gillette SkinGuard, $20 for handle and four cartridges. The only product here with a peer-reviewed paper behind its skin claim, and its guard geometry is aimed at exactly the mechanism dermatology keeps naming. Then it charges $3.75 a cartridge and tells you nothing about how long one lasts. Skip this if you want to know your annual spend before you commit to a system.

Panasonic ARC6 ES-LS8A-K, $399.99, or $499.99 with the cleaning station. The most disclosed of the electrics. Panasonic names its comparator, states that its 10 percent is about pressure, publishes the part number and the price of the replacement set, and publishes an interval. Everything I asked of a maker, Panasonic did. It just did it for a mechanical claim rather than a skin one. Skip this if $126 every year or two for a foil set is not in the budget.

Braun Series 9 PRO+ 9677cc. Braun publishes the honest bit — “The average Braun shaver cuts 6 million hairs every 18 months” — and then does not publish what the replacement head costs. A retailer will sell you one. Braun will not quote you one. Skip this if you want the full ownership cost from the manufacturer before you buy.

Philips i9000 Prestige Ultra XP9405/84, $329.96. Philips has also quietly split its own branding: the older SP9840/90 page is still live under the SkinIQ name while the newer models are SenseIQ, so both names sit on Philips’ US site at once. Skip this if your reason for buying an electric was your skin. That is the one job this ledger says it is worst at.

A note on availability and safety, since both get skipped. All five are currently sold on their makers’ own stores as of August 31, 2026. A search of CPSC’s published recalls found no open recall on any of them; the Norelco entry that surfaces is a 1986 charging-stand notice, and the Gillette entry is a 2019 Venus cartridge recall, a different product line. Neither touches what is on this page.

If You Will Not Learn a Safety Razor, the Tie-Breaker Is 5 Points Wide

Most people reading this are not going to buy a double-edge razor, and I am not going to pretend otherwise. That makes the real contest second place: Gillette SkinGuard at 49.3 against the Panasonic ARC6 at 44.3. Five points. Inside the noise of any weighting you or I choose.

My own money goes on the Gillette, and for one reason that has nothing to do with the 60 percent. The guard is built to cut the hair at the surface rather than beneath it, and surface-level cutting is the single mechanism the AAD, the NHS, DermNet and the Cutis military review all independently point at. The Panasonic is the better-documented company. The Gillette is the better-aimed tool.

Change one weight and that flips. Drop “support behind the skin claim” to 15 percent and raise “what the maker discloses” to 30, and Panasonic takes second. That is the point of printing the axes.

Three other posts here go at the same aisle from different angles: the foil-or-rotary question, where brands come second, three $200-plus electric shavers graded on three-year cost, and safety razor against cartridge against disposable on cost per shave. The rest of it sits in shaving and oral care.

If you have a paper comparing electric to blade on skin outcomes that I missed, send it. I would rather rewrite this page than keep the finding.

July 16, 2026

Closet Systems: Rent-Friendly to Renovation-Grade

Some links on our site may earn us a commission at no extra cost to you. Learn more.

Updated August 2026: Restructured as a fork-by-fork decision guide with a new capacity-per-dollar breakdown across wire, modular, and freestanding systems, and refreshed every price this month.

Most closet-system advice starts with the products. That’s backwards — a wire kit, a modular wood system, and a freestanding rack aren’t really competing for the same buyer, because two of the three questions that actually decide which one you need have nothing to do with the closet itself. Work through the decisions in order and the “best” system stops being a matter of opinion.

The short answer: renters and no-drill situations land on freestanding, budget-first owners land on wire, and anyone who wants a furniture-grade result and can absorb roughly triple the cost per linear foot lands on modular. The forks below show exactly where that split happens.

Fork one: do you own the space, or does a landlord?

This eliminates two of the three options before price ever comes up. Wire and modular systems both mount to the wall — track-and-bracket hardware for wire, wall anchors and ledger strips for modular — which means both need either ownership or explicit landlord permission to install. Freestanding racks and towers touch no walls and need no permission, which makes them the only real option for a rental where drilling isn’t allowed. If you own the space or have sign-off, keep reading; if you don’t, freestanding is your answer and the rest of this comparison is about which freestanding piece to buy.

Fork two: will you actually do the installation?

Wire’s advantage here is real: a level, a stud finder, and a weekend get a track-and-shelf system installed and adjustable forever after. Modular wood systems ask for more — precise leveling, solid wall anchoring, and reviewers routinely describe the install as a full weekend with instructions that test patience. A half-finished modular install is genuinely worse than either finished alternative, so this fork is worth answering honestly before buying: if a multi-step weekend project won’t happen, wire (or paying for professional install) is the safer bet regardless of budget.

Fork three: what does your budget actually buy, linear foot by linear foot?

This is where sticker price stops telling the real story. We priced one representative kit from each format and divided its total linear feet of hanging rod plus shelf space by its cost.

System	Example kit (current price)	Linear feet included	Cost per linear foot
Wire (ClosetMaid SuperSlide, 5–8 ft kit)	$71.04	24 ft (15 ft shelf + 9 ft hanging rod, per the manufacturer spec)	~$2.96/ft
Modular/wood (ClosetMaid Selectives Basic Standard, up to 121 in)	$239.68	~22 ft (estimated: three 48-in top shelves = 12 ft, plus an estimated 10 ft across the tower’s three expandable hanging rods — ClosetMaid doesn’t publish a linear-footage spec, so this is calculated from the kit’s stated components)	~$10.89/ft
Freestanding (Amazon Basics Adjustable Double Rod Rolling Garment Rack, 36 in wide)	~$37.99	6 ft (two 3-ft rods, double-hung)	~$6.33/ft

Prices are approximate, current as of publication, and change frequently — confirm the latest price with the retailer before buying.

Wire wins capacity per dollar by a wide margin — less than a third of modular’s cost per linear foot. Freestanding sits in the middle: pricier per foot than wire, but with zero installation cost or wall commitment. Modular’s premium buys drawers, closed storage, and a furniture-grade look wire and freestanding can’t match — the question is whether that’s worth roughly 3.5x the per-foot price.

Fork four: hanging-heavy wardrobe, or folded-heavy?

Most people over-buy hanging space and under-buy drawer and shelf space relative to how they actually dress, which is where modular’s configurability pays for itself if you have a lot of folded items, shoes, or accessories to organize — wire handles hanging and open shelving well but needs bins for anything small, and freestanding racks are hanging-only by design. If your wardrobe is genuinely hang-dominant (coats, dresses, suits), wire or freestanding cover it at a fraction of modular’s cost; if it’s folded- and accessory-heavy, modular’s drawers start to justify the premium.

Neatly organized modular closet drawers and shelves with folded clothes — Closed drawers and shoe shelves are where modular’s per-foot premium actually shows up — useless for a hang-dominant wardrobe, decisive for a folded-heavy one.

Where you should have landed

Renting, or no wall-anchor permission: freestanding — the only format that needs zero installation and zero landlord sign-off, at a reasonable middle price per foot.
Own the space, comfortable with a weekend install, budget matters most: wire — the clear capacity-per-dollar winner, and $20 of shelf liners solves its one real flaw (thin items slipping through the wires).
Own the space, want a furniture-grade result, and can absorb roughly 3.5x wire’s per-foot cost: modular — sized honestly around your actual hang-to-fold ratio, not over-built on hanging space you don’t use.

All prices and linear-footage figures above come from each manufacturer’s or retailer’s current listing, checked this month, with the source named in the sentence. Where a manufacturer doesn’t publish a linear-footage spec, we show the arithmetic used to estimate it rather than presenting a guess as a verified number. See our editorial standards for how we research and verify pricing. For a related no-drill storage option, our storage tote comparison covers bins that pair well with any of the three systems above.

Closet System Questions, Answered

Can I install a closet system without professional help?

Wire systems are the most DIY-friendly — a level and basic tools handle a weekend install; modular wood systems are doable but demand more precision, and a professional install is worth considering if the weekend project genuinely won’t happen.

How much hanging space does a typical closet actually need?

It depends entirely on wardrobe size, but most people overestimate their hanging needs and underestimate shelf and drawer needs relative to how their clothes actually break down — worth an honest inventory before buying any system.

Do wire shelves damage clothes?

They can leave shelf-line marks on soft knits over time; a few dollars of shelf liner eliminates the issue entirely and is the standard fix reviewers recommend.

Is a modular closet system worth the extra cost over wire?

Only if you need what the extra cost buys — closed drawer storage and a furniture-grade look — since on a pure capacity-per-dollar basis, wire outperforms modular by a wide margin.

July 16, 2026

A $1,129 Corded Pool Cleaner Beats a $2,299 Cordless One

Some links on our site may earn us a commission at no extra cost to you. Learn more.

Updated September 2026: Maytronics’s, Polaris’s and Beatbot’s current prices, warranty terms and pool-size ratings re-verified against each maker’s own product pages, cross-checked against an authorized Dolphin dealer’s listing and against independent reviewer and owner-report data from The Pool Nerd and Birdeye.

Prices are approximate, current as of publication, and change frequently — confirm the latest price with the retailer before buying.

Best for most pools: the Maytronics Dolphin Cayman, at $1,129.00 direct from Maytronics’s own store — $564.50 for every year of its 2-year warranty, the cheapest of the four here. It runs floor and walls on a built-in weekly timer with no battery to manage. Skip the cord only if your pool deck has nowhere safe to plug one in; that’s the one problem the Polaris Freedom and Beatbot AquaSense 2 Ultra solve that a Dolphin doesn’t.

The Beatbot is the one to skip outright right now. At $2,299.00 on sale — $3,150.00 at list — it’s the most expensive robot in this comparison and the worst-reviewed: The Pool Nerd’s independent review calls it “Pool Nerd Disapproved” and reports weak suction that gets weaker as the battery drains. Its own coverage spec isn’t even measured the same way as the other three, which is its own finding below.

What Does Each Robot Cost per Year You’re Covered?

A pool robot’s warranty is the maker’s own bet on how long it lasts. Divide today’s price by that bet and you get a real number: what a single covered year actually costs. Every price below is what each maker (or, for Polaris, an authorized retailer) charges today, pulled September 2, 2026 — not carried over from an old listing.

Model	Current price	Warranty	Cost per covered year
Dolphin Cayman	$1,129.00, per Maytronics’s own store	2 years	$564.50
Polaris Freedom	$1,199.00, per Leslie’s Pool Supplies (Polaris’s own product page lists no price)	2 years	$599.50
Dolphin Premier	$2,499.00, per Maytronics’s own store (Poolbots currently lists it on sale at $1,999.00)	3 years	$833.00
Beatbot AquaSense 2 Ultra	$2,299.00 on sale, $3,150.00 list, per Beatbot’s own site	3 years	$766.33 at the sale price, $1,050.00 at list

The Cayman wins this even before you get to pool size: cheapest sticker, shortest warranty, and still the cheapest year-by-year math of the group. The Premier costs $833.00 a covered year — the most of any Dolphin — but it’s also carrying a full extra year of coverage the Cayman doesn’t offer, on a bigger pool.

Rated pool length is the second real number, and it’s where the four stop being comparable on one column. Maytronics’s own Cayman page and an authorized Dolphin dealer’s listing don’t even agree with each other on the Cayman — that disagreement gets its own section below. Here’s what each currently states, plus what that works out to per foot:

Model	Rated pool length	Cost per foot
Dolphin Cayman	Up to 40 ft, per Maytronics and per The Pool Nerd’s independent review; a Dolphin dealer’s own listing says 33 ft	$28.23 at 40 ft / $34.21 at 33 ft
Polaris Freedom	Up to 50 ft, per Leslie’s Pool Supplies listing (Polaris’s own page states “in-ground pools” with no foot figure)	$23.98
Dolphin Premier	Up to 50 ft, per Maytronics	$49.98
Beatbot AquaSense 2 Ultra	Not rated by length at all — see below	Not comparable

The Freedom actually posts the cheapest cost per foot of the four, even though it isn’t the cheapest cost per covered year — the Cayman’s shorter warranty term drags its per-year number back down below the Freedom’s. Neither number is “the” answer; a buyer sizing a large pool on a budget should weight the per-foot column harder, and a buyer who just wants the cheapest total commitment should weight cost-per-year harder. Both are real math from the same two prices.

Beatbot doesn’t publish a maximum pool length for the AquaSense 2 Ultra at all. Its own product page rates the robot by area instead — 3,875 square feet of coverage per cleaning cycle — which isn’t the same kind of measurement as a linear foot rating and doesn’t convert into one without guessing. Rather than force a number that isn’t there, it sits out of the per-foot column above. That gap is worth knowing on its own: every other maker in this comparison tells you how long a pool the robot is built for, and Beatbot doesn’t.

Which Pool-Length Number Is Real — Maytronics’ 40 Feet or the Dealer’s 33?

Two sources say 40 feet. One says 33. Here’s what’s actually printed where, plus two more places a maker’s page and an independent reviewer land on different footing.

What’s claimed	Source	Verdict
Dolphin Cayman covers pools up to 40 ft	Maytronics’s own Cayman product page; The Pool Nerd’s independent review separately notes the Cayman’s 50-ft cord is “adequate for pools up to 40 feet”	Two sources agree on 40 ft. An authorized Dolphin dealer’s own listing page states 33 ft for the same model. If your pool sits between 33 and 40 feet, size to the smaller number.
Beatbot AquaSense 2 Ultra runs “up to 10 hours”	Beatbot’s own spec sheet	True only for surface-only skimming. The same page lists 5 hours for floor cleaning and 4.5 hours for wall-and-waterline cleaning — the modes that actually clean the pool run less than half the headline number.
Beatbot’s own retail reviews	Fakespot authenticity rating, reported by The Pool Nerd	F, versus an A for the Cayman on the same rating system. Treat Beatbot’s own product-page star rating with real skepticism before treating it as owner sentiment.
Polaris Freedom cleans fine debris like sand	A Birdeye-verified owner review of the Freedom	The reviewer reports the unit doesn’t ship with an ultra-fine filter and paid roughly $100 extra for one to handle sand. Budget for the accessory if your pool sees fine sediment.

Does Cordless Actually Mean You Do Less Work?

The pitch for both cordless robots here is that you stop thinking about pool cleaning entirely. Neither one’s own numbers back that up. Polaris’s own Freedom page states a single run takes “2.5 hours or less” and a full recharge takes “less than 5 hours” — so between cleanings, someone is pulling the robot out and putting it on a charger. Beatbot’s numbers run the same way: 4.5 to 10 hours of runtime depending on mode, against a 4.5-hour recharge, per Beatbot’s own spec sheet. A corded Dolphin, by contrast, sits on a built-in weekly timer and runs itself; nobody retrieves it between cycles unless they’re taking it out for a swim day.

Pool maintenance worker cleaning a swimming pool with a long-handled pole — Someone still has to retrieve and recharge a cordless robot between runs — the manual work doesn’t disappear, it just moves to a different task.

There’s a real safety reason to take that extra handling seriously, too. The CPSC has recalled two different cordless robotic pool cleaners from Aiper over the past few years — the Elite Pro in 2023 and the Seagull Pro in 2025 — both over charging setups that could overheat and cause burns or fire. Neither Aiper model appears in this comparison, and none of the four robots here carry an open recall as of this writing. But it’s the reason to buy an established brand, register the product, and follow the maker’s charging instructions exactly with any battery device that lives near water.

How Do the Four Score on Power, Coverage and Reviews?

Four axes, weighted by what actually determines whether a pool robot does its job: cleaning power and coverage, cost per covered year (from the table above), what owners and independent reviewers actually report, and how much hands-on work it demands between cycles. These are our own weighted scores, built from the sourced facts in this piece — not a lab result, and not anyone else’s rating system.

Axis (weight)	Dolphin Cayman	Dolphin Premier	Polaris Freedom	Beatbot AquaSense 2 Ultra
Cleaning power & coverage (30)	24	30	16	10
Cost per covered year (25)	25	15	19	4
Owner & reviewer sentiment (25)	23	24	20	5
Hands-off convenience (20)	20	20	6	4
Total (100)	92	89	61	23

Coverage is the one axis where the Premier legitimately beats the Cayman — 4,500 GPH and full floor-wall-waterline cleaning against the Cayman’s 4,000-plus GPH and floor-and-walls-only. It’s also the reason The Pool Nerd’s reviewer, who has reviewed more than 30 robotic cleaners for that site, names the Premier a personal favorite. The Cayman closes almost the entire gap on price and convenience instead. The Freedom’s biggest score loss is convenience, not cleaning — 784 Birdeye reviewers rate it 4.0 out of 5, but every one of those reviews happened around a retrieve-and-recharge routine the corded models don’t require. The Beatbot loses on every axis at once, which is a different problem than losing on one.

Who Should Buy the Dolphin Cayman?

The Cayman is dual-motor, wall-climbing, and rated for over 4,000 gallons an hour of flow, per Maytronics’s own spec sheet — enough to clean the floor and walls of a pool up to 40 feet on its own timer. The Pool Nerd’s independent review backs the reliability case specifically: it reports “zero downtime between cleaning sessions” and no measurable loss of power over time, the kind of claim a battery robot structurally can’t make. The one drawback either source names is minor — the power cord lacks the anti-tangle swivel found on pricier Dolphins.

Buy it if your pool is 40 feet or under and you want the cheapest real ownership math here. Skip it if you’re under trees — it doesn’t clean the waterline, and it has no leaf bag for heavy debris. That’s the Premier’s job.

Who Should Buy the Dolphin Premier?

The Premier is Maytronics’s flagship for a reason: 4,500 GPH of flow, dual scrubbing brushes, full floor-wall-waterline coverage, and a multi-media filtration system built around an oversized leaf bag that the Cayman doesn’t have. It’s rated for pools up to 50 feet and backed by a 3-year warranty, the longest of the four here. The Pool Nerd’s reviewer, who has reviewed 30-plus robotic cleaners for that site, calls it a personal favorite and reports a specific unit “running strong for over three years without a single hiccup” — a first-hand claim from that reviewer, not from us, and worth reading as one data point rather than a guarantee.

Buy it if your pool sits under trees, tops 40 feet, or both — the leaf bag and waterline cleaning genuinely earn the extra cost there. Skip it if your pool is small and mostly clear of debris; you’re paying $833.00 a covered year for capability the Cayman’s $564.50 already covers.

Who Should Buy the Polaris Freedom?

Polaris has built pool equipment for decades, and the Freedom is the sober cordless choice on that reputation alone — an established dealer network matters if something needs warranty service in year two. Birdeye’s 784 verified reviews average 4.0 out of 5, with reviewers specifically praising waterline cleaning and how it handles fine particles. The catch sits in the fine print of those same reviews: the standard filter isn’t the ultra-fine one, and at least one reviewer paid roughly $100 more to add it for sand. A single reviewer also reported it doesn’t always stop reliably at the deep end.

Buy it if your pool deck genuinely has no safe outlet and you’ve priced in the ultra-fine filter as a real add-on cost. Skip it if you have an outlet — a corded Dolphin outperforms it on this Ledger at a lower cost per covered year, with none of the retrieve-and-recharge routine.

Who Should Buy the Beatbot AquaSense 2 Ultra?

The AquaSense 2 Ultra is the only robot here that skims the surface as well as cleaning the floor, walls and waterline, with AI-driven mapping across 27 sensors, per Beatbot’s own spec sheet. On paper, that’s the most capable machine in this group. The Pool Nerd’s independent review says the paper doesn’t hold up in the water: weak suction that gets noticeably weaker as the 13,400 mAh battery drains, debris missed in corners, and a 26.5-pound unit that’s genuinely heavy to lift out for its daily recharge. The same review flags an F rating on Fakespot for the product’s own retail reviews — a signal to discount Beatbot’s star rating on its own listings rather than take it at face value.

Buy it if the surface-skimming feature specifically matters to you — floating leaves and pollen are a real nuisance a corded Dolphin can’t touch — and you’re prepared to pay $766.33 a covered year, even at the current sale price, for that one capability. Skip it for straightforward floor-and-wall cleaning; three other robots here do that job better, more reliably, and for less money, on both our numbers and an independent reviewer’s.

So Which One Should You Actually Buy?

Buy the Dolphin Cayman unless a specific reason points you elsewhere. It’s the cheapest robot here per covered year, the independent review backs its reliability claim specifically, and a built-in timer means you never touch it once it’s set. Go to the Premier only if your pool is over 40 feet or sits under real tree cover — the leaf bag and extra flow are worth the jump from $564.50 to $833.00 a year. Take the Freedom only if a cord genuinely can’t reach your pool, and budget for the ultra-fine filter separately. Leave the AquaSense 2 Ultra to buyers who want surface skimming specifically and have already accepted the price and the weaker independent review that come with it.

Two more stops on the backyard: our above-ground pool comparison covers what’s left to buy after a 2025 recall pulled two popular models, our quiet window AC shortlist handles the heat once you’re out of the water, and our portable grill comparison covers the rest of a pool day.

July 15, 2026

Nanit vs Eufy Baby Monitor: Which Expands Better?

Some links on our site may earn us a commission at no extra cost to you. Learn more.

Updated August 2026: rebuilt this comparison around adding a second (or third) camera to a system you already own. We focus on cross-generation compatibility, per-camera subscription cost, and multi-camera app support, rather than ranking these four monitors as a first purchase. We also re-verified every price and compatibility claim directly against each brand’s own site this week.

A baby monitor, with one camera or five, is a convenience tool, not a safety device. Nothing in this comparison should be read as a claim that any product substitutes for direct caregiving or your pediatrician’s safe-sleep guidance. Follow the manufacturer’s instructions for mounting any additional camera and its cord well out of a crib’s reach. Don’t treat more cameras as more safety — it’s more coverage, which is a different thing.

The nanit vs eufy baby monitor decision comes down to three criteria — hardware compatibility, subscription cost, and multi-crib app support, not raw video quality. Buying a second camera is a different purchase than buying your first one. The question isn’t which brand has the sharpest video anymore. It’s whether the camera you already own will even talk to the new one, what a second camera does to your subscription bill, and whether the app was actually built to show more than one crib at once. Those three questions rarely show up in a first-time buying guide. So this one is built around them specifically, for the parent who already owns a monitor and is now adding a sibling’s room, a second angle, or a hand-me-down camera to the mix.

Nanit vs Eufy Baby Monitor at a Glance

Criterion	What to check
Hardware compatibility	Does a new camera pair with your existing parent unit?
Subscription cost	Does adding a camera add a monthly bill?
Multi-crib app support	Was the app built to show more than one crib?

Nanit vs Eufy Baby Monitor: What Matters When You Add a Camera?

The three criteria that matter once you already own a monitor are whether a new camera pairs with your existing hardware, whether adding one adds a subscription, and whether the app actually supports more than one crib. These questions only matter once “which monitor” has already been answered. “How do I add to it” is the actual question at that point.

Criterion one: does a new camera pair with your existing hardware?

Not always. Pairing depends on the camera generation, since even Infant Optics won't let a DXR-8 parent unit pair with a DXR-8 PRO camera. Baby monitor brands update their camera hardware over the years. A newer add-on camera doesn’t always speak to an older parent unit or app, or vice versa. Buying the wrong add-on camera can mean it simply won’t pair, full stop. That’s true no matter how similar the box looks on a store shelf.

Criterion two: does adding a camera add a subscription?

A second camera on a cloud-connected system can mean a second monthly bill, not just a second one-time purchase. Or it might not, depending on how the brand structures its plans. This is worth confirming before buying, because the sticker price of the add-on camera is sometimes the smaller number.

Criterion three: was the app built for more than one crib?

Some apps treat a second camera as a first-class feature — real split-screen, per-camera settings, quick switching. Others clearly weren’t designed with it in mind. There, a second camera becomes something you tolerate rather than something that works well.

How Do Nanit, Owlet, Infant Optics, and Eufy Baby Monitors Compare?

Only Infant Optics publishes a clear limit. That limit is a documented multi-camera cap with zero subscription cost at any camera count. Nanit’s own pages leave its maximum camera count and subscription impact unclear.

How Do Nanit and Eufy Baby Monitors Stack Up Here?

	Nanit Pro	Owlet Dream Duo / Cam 2	Infant Optics DXR-8	Eufy E21
Add-on camera price (verified)	~$289.99 (no bundle discount vs. buying separately — a twin bundle runs $579, essentially 2× single price, per Nanit’s own store)	$69.99, per Owlet’s own site	$73.49, per an authorized Infant Optics reseller; requires the original DXR-8, explicitly not compatible with the DXR-8 PRO	$129.99, per Eufy’s own site — listed as sold out at the time of this update
Documented max cameras per system	Not published in a way we could independently verify; Nanit markets bundles of up to three cameras but doesn’t state a hard cap	Owlet’s own support confirms multiple Cam 2 units can connect to one account, but doesn’t publish an exact maximum	Up to 4 Camera Units per parent unit, explicitly stated, per the add-on camera’s own listing	Up to 2 cameras on split-screen, per independent testing from a 2026 E21 review; not a figure Eufy’s own site states outright
Cross-generation compatible?	N/A — one current camera generation	Not documented either way for older Owlet Cam vs. Cam 2	No — original DXR-8 and DXR-8 PRO add-on cameras are explicitly not interchangeable	Model-specific — Eufy sells separate add-on cameras for different monitor models rather than one universal add-on
Extra subscription triggered by a second camera?	Unclear from Nanit’s own membership page; some owners report on Nanit’s community forum that a second camera needs its own Insights plan to get full analytics	Unclear — Owlet360 ($99.99/yr) is described as covering “compatible devices,” not explicitly priced per camera	No subscription exists for this system at any camera count	No subscription exists for this system at any camera count

What Should You Budget for the Add-On Camera?

Prices are approximate, current as of publication, and change frequently. Confirm the latest price and current stock with the retailer before buying.

Nursery room interior with a crib, the kind of second room a growing family adds a camera to — A second camera usually means a second room, not a second angle on the same crib — which is exactly where cross-generation compatibility questions show up.

Which System Costs Least Per Expansion Feature?

To make the table above comparable in a single number, we counted four specific, independently verifiable expansion features. Those are an add-on camera priced under $100, a publicly documented maximum camera count, confirmed or not-applicable cross-generation compatibility, and zero confirmed added subscription cost. We then divided the add-on camera’s price by how many of those four each system actually has. This only credits features we could confirm; anything marked “unclear” above did not count in a brand’s favor.

System	Add-on price	Features confirmed (of 4)	Price ÷ features
Infant Optics DXR-8	$73.49	3 of 4 (under $100, documented max, no subscription)	~$24.50 per feature
Infant Optics DXR-8 PRO	$66.99	3 of 4 (under $100, documented capacity via “4 channels,” no subscription)	~$22.33 per feature
Eufy E21	$129.99	2 of 4 (independently documented 2-camera max, no subscription)	~$65.00 per feature
Owlet Dream Duo / Cam 2	$69.99	1 of 4 (under $100 only)	~$69.99 per feature
Nanit Pro	$289.99	1 of 4 (single-generation compatibility only)	~$289.99 per feature

Infant Optics wins this specific math decisively, on both the original DXR-8 and the DXR-8 PRO. It’s the only system here with a plainly published multi-camera limit and zero subscription of any kind at any camera count. But its own cross-generation gap is the sharpest in this comparison. The two Infant Optics product lines don’t talk to each other’s add-on cameras at all. So the “which one do I already own” question matters more with this brand than any other here.

What we couldn’t verify, and why that matters

Two gaps are worth naming directly rather than glossing over. First, neither Nanit nor Owlet publishes a clear, findable answer to whether a second or third camera requires its own subscription tier for full features. Nanit’s own community forum contains an owner reporting they needed a second Insights plan for a second camera. That’s a user report, not an official policy statement, and we’re flagging it as exactly that. Second, Eufy’s E21 add-on camera showed as sold out on Eufy’s own site at the time of this update. That’s itself a real expansion obstacle no comparison table captures well. Confirm current stock before assuming you can expand a Eufy system today.

Which Nanit or Eufy Baby Monitor Makes Sense to Expand?

Already own an Infant Optics DXR-8 or DXR-8 PRO? Adding a camera is the cleanest expansion path in this comparison — cheap, subscription-free, and pairs up to four cameras. Just buy the add-on that matches your specific model, since the two don’t cross-pair. Already own a Eufy E21? Expansion works and stays subscription-free, but check current stock and budget for the priciest add-on camera of the four. Already own an Owlet or Nanit system? Confirm the subscription math directly with the brand before assuming a second camera costs only the hardware price. Our research couldn’t fully resolve that question from either company’s own published pages, and it’s the kind of gap worth a support-chat before you buy.

This is a deliberately narrower question than “which baby monitor is best.” For that broader comparison, including video quality, vitals tracking, and privacy trade-offs on the newer camera generations, see our Nanit Pro vs. Owlet Dream Duo vs. Infant Optics DXR-8 Pro comparison. If the real question behind your purchase is whether you want a camera on the internet at all, see our separate breakdown of WiFi vs. closed-RF baby monitor privacy. This article assumes that decision is already made. It focuses purely on what happens when you add a second camera to whichever system you picked.

How We Verified Nanit, Owlet, Infant Optics, and Eufy’s Camera Specs

Here’s how we built this comparison. We focused on three criteria for expanding an existing system: cross-generation camera compatibility, per-camera subscription impact, and documented multi-camera capacity. We verified them directly against each brand’s own product pages and support content this week, rather than re-ranking these four monitors as a first purchase. Where a brand’s own documentation didn’t clearly answer a question, we said so explicitly instead of guessing. Our full research and sourcing standards are on our editorial standards page. Prices are approximate, current as of publication, and change frequently — confirm the latest price and stock with the retailer before buying.

Nanit Eufy Baby Monitor FAQ

Can I mix an Infant Optics DXR-8 parent unit with a DXR-8 PRO camera?

No. The two lines use separate add-on cameras, and each is explicitly listed as not compatible with the other model. Buy the add-on that matches your specific parent unit, not just the brand name on the box.

Does a second Nanit or Owlet camera cost extra in subscription fees?

We couldn’t confirm this cleanly from either brand’s own published pages. A Nanit community forum post from an owner suggests a second camera may need its own Insights plan for full analytics, and Owlet’s Owlet360 page doesn’t spell out per-camera pricing. Confirm directly with either company’s support before assuming the add-on camera’s sticker price is the whole cost.

Which system supports the most cameras on one account?

Infant Optics publishes the clearest number: up to 4 Camera Units per parent unit, for both the DXR-8 and DXR-8 PRO lines. The two lines’ cameras aren’t interchangeable with each other, though. Eufy’s E21 is independently documented at up to 2 cameras on split-screen. Nanit and Owlet don’t publish a clear maximum in the pages we could verify.

Is a cheaper add-on camera always the better expansion choice?

Not automatically — check compatibility with your specific existing hardware first. A cheap add-on camera that doesn’t pair with your parent unit or account isn’t a deal. This is exactly the mistake cross-generation incompatibility can cause, as documented with the Infant Optics DXR-8 and DXR-8 PRO lines.

Do any of these systems make it unsafe to monitor two rooms at once?

No — adding a camera is a coverage question, not a safety upgrade or downgrade. None of these systems, with one camera or several, replaces following your pediatrician’s safe-sleep guidance. That guidance still applies in every room a baby sleeps in.

July 15, 2026

Hair Dryers Decoded: Basic vs. Ionic/Ceramic vs. Premium High-Speed
Some links on our site may earn us a commission at no extra cost to you. Learn more.

Updated September 1, 2026: the site-standard affiliate and price disclosures were added to the top of this page. The tier breakdown and buying advice below are unchanged since the July 25, 2026 review and were not re-checked today — treat the price bands as rough guides, not current quotes.

Hair dryer marketing is a physics word salad — ionic, ceramic, tourmaline, infrared — wrapped around a price range that runs from $20 to $500. Strip the jargon and three real tiers emerge. The face-off.

Prices are approximate, current as of publication, and change frequently — confirm the latest price with the retailer before buying.

The contenders

Basic dryers — the $20–40 drugstore workhorses: a heating coil, a fan, hot air.

Ionic/ceramic mid-range — the $40–150 tier: ceramic or tourmaline components for evener heat, negative-ion features marketed for frizz, better motors and attachments.

Premium high-speed dryers — the $200+ tier built around brushless digital motors: dramatically faster airflow, intelligent heat control, salon-grade engineering.

Round 1: Drying speed

Premium high-speed dryers win decisively — the modern brushless-motor machines cut drying time by a third to half versus basics, and for thick or long hair, reviewers describe the difference as the entire reason to pay. Mid-range models beat basics modestly. Basics get there eventually, which is the polite way reviews put it.

High-speed brushless motors cut drying time by a third to half, especially on thick or long hair.

Winner: Premium — and speed is also gentleness, since less time under heat is the real hair-health lever.

Round 2: Hair health and finish

Here’s the honest decoding: “ceramic” means more even heat (real, useful, prevents scorch spots), and “ionic” means negative ions marketed to cut frizz — a modest, real-for-many effect that reviewers with frizz-prone hair value and reviewers with fine hair sometimes skip (ions can flatten volume; several mid-range models include an ion on/off switch for exactly this reason). Premium models’ actual advantage is intelligent heat regulation — measuring and capping air temperature — plus that shorter exposure time. Basics run hot, uneven, and unmonitored; the damage tier.

Winner: Premium on engineering; mid-range ceramic is the value sweet spot.

Round 3: Features that matter (and don’t)

The ones reviews validate: a true cool-shot button (sets the style), a concentrator nozzle (smooth blowouts), a diffuser (non-negotiable for curls), multiple genuine heat/speed settings, and weight — arm fatigue is the top complaint about cheap-but-heavy and old premium models alike. The ones that are mostly stickers: tourmaline vs. ceramic distinctions, “infrared,” and ion-count numbers nobody can verify.

Winner: Draw — check the feature list, not the mineral vocabulary.

Round 4: Value

Fine, short, or wash-and-go hair genuinely doesn’t need more than a decent basic or entry mid-range — reviews from that cohort call premium dryers lovely and unnecessary. Thick, long, curly, or daily-blowout hair extracts full value from the premium tier: minutes saved daily, visibly less heat stress, multi-year build quality. The mid-range is the smart default for everyone between.

Winner: Mid-range for most; premium earns it for heavy users.

The verdict
- Best overall: A quality mid-range ceramic/ionic dryer with cool shot, real attachments, and light weight — 80% of the premium experience at a quarter of the price.
- Best for thick/long hair and daily stylers: A premium high-speed dryer — the time and heat savings are real and daily.
- Best budget: A basic with a concentrator — for occasional dryers and gym bags; just keep it moving and finish cool.
Whatever the tier: heat protectant first, keep the dryer moving, finish with the cool shot — technique reviews agree those three habits outperform any $300 upgrade.
July 15, 2026

Espresso at Home: Effort, Mess, and Cost per Shot

Some links on our site may earn us a commission at no extra cost to you. Learn more.

Updated August 2026: Rebuilt as a table-first cost-per-shot comparison with current 2026 hardware, bean, and capsule prices, replacing the older matchup format.

What does a single shot of home espresso actually cost, once the machine itself is paid for? Sticker price answers a different question than that one — a $180 pod machine and a $700 manual setup aren’t really being compared on the same axis until you count beans or capsules, cleaning supplies, and how many shots the hardware needs to pull before it pays for itself.

The short answer: a manual espresso setup costs roughly $0.54 per shot once amortized, a super-automatic runs about $0.61, and pod machines land near $1.32 — pods are cheapest to buy and most expensive to drink from, while manual is the reverse. The table below shows the full math.

Format	Hardware (current price)	Amortized per shot*	Beans/pod per shot	Cleaning per shot	Total per shot
Manual (Breville Bambino Plus + Baratza Encore ESP grinder)	$499.95 + $199.95 = $699.90	$0.19	$0.32 (18g dose, Lavazza Espresso Italiano whole bean at ~$7.95/lb)	$0.03	~$0.54
Super-automatic (De’Longhi Magnifica Evo)	$899.95	$0.25	$0.32 (same bean, built-in grinder)	$0.05 (descaler + grinder-cleaning tablets)	~$0.61
Pod (Nespresso Vertuo Next)	$179.99	$0.05	$1.25 (average current capsule price)	$0.02	~$1.32

*Amortized over an assumed 5-year hardware life at roughly two shots a day (~3,650 shots) — a household pulling more shots per day lowers this per-shot hardware cost further, and a lighter user raises it.

Prices are approximate, current as of publication, and change frequently — confirm the latest price with the retailer before buying.

Hardware cost: the number on the sticker

Pods win the sticker-price column outright — a Nespresso Vertuo Next runs $179.99, versus $699.90 for a manual machine plus a real grinder, or $899.95 for a super-automatic that grinds, tamps, and pulls the shot itself. This column is exactly where most buying decisions stop, and it’s also the least useful column for predicting what espresso actually costs over time.

Amortized cost: what the machine adds to every shot

Spread over roughly 3,650 shots across five years, even the priciest hardware here adds well under a quarter per cup — a super-automatic’s higher price tag only costs about six cents more per shot than manual once amortized, and a pod machine’s low sticker price barely registers per shot at five cents. The hardware column that looked decisive at checkout nearly disappears once it’s divided by every cup it will ever pull.

Beans and pods: the column that actually decides the ledger

This is where the totals really separate. Manual and super-automatic machines both run on the same whole beans — Lavazza’s mainstream Espresso Italiano blend prices at roughly $7.95 per pound, working out to about 32 cents for an 18-gram double shot. Nespresso Vertuo capsules range from $1.00 to $2.25 each depending on the blend, averaging around $1.25 — nearly four times the per-shot bean cost, and the single biggest reason pods lose the long-run math despite winning on hardware.

Cleaning and consumables: the cost nobody budgets for

Manual machines need occasional descaling and portafilter-cleaner backflushing, running only a few cents per shot in bulk. Super-automatics carry more internal parts — brew group, milk system, integrated grinder — so their descaling tablets and cleaning-cycle supplies add slightly more per shot. Pod machines need the least maintenance, just occasional descaling, though that’s already the smallest line item on the ledger next to $1.25 capsules.

Total cost per shot: what actually wins

Manual comes out cheapest per shot once everything is counted, followed closely by super-automatic, with pods costing more than double either — even though pods were the cheapest to buy. The crossover point is straightforward: at roughly two shots a day, a pod drinker spends the difference between a Vertuo Next and a full manual setup within about a year, then keeps paying the premium indefinitely.

Beyond the cost math: cup quality and morning effort

Cost isn’t the only variable, and it shouldn’t be the only one that decides this. Manual machines win the quality ceiling by a wide margin — with a decent grinder and some practice, home shots can rival café work, though the learning curve is real and early shots often go down the drain. Super-automatics produce consistently good espresso that plateaus below a manual machine’s best, in exchange for one-button simplicity. Pods produce small, identical, decent coffee every time; the ceiling is the capsule, and it doesn’t move no matter how much practice goes into it.

Reverse the podium for effort: pods take about 30 seconds with zero skill, super-automatics add a button press plus tank refills and periodic cleaning cycles, and manual espresso is a genuine hobby wearing an appliance costume — five to ten minutes per drink while learning.

Rich espresso with golden crema pouring from a portafilter — With a quality grinder and practice, a manual machine pulls shots that rival café work — the cost math above shows that ceiling costs less per cup, not more.

Who should skip each format

Skip manual if the morning window is measured in seconds or a stretch of imperfect practice shots sounds like a dealbreaker. Skip a super-automatic if repair anxiety and countertop space are real concerns — more internal parts means more that can eventually need service. Skip pods if you’re drinking three shots a day; at that volume the subscription math buys a full manual setup within a year or two. Still deciding between a machine and simpler brewing entirely? Our pod-versus-ground-coffee cost breakdown covers the broader question.

Every hardware price, bean price, and capsule price above was checked against the manufacturer’s or retailer’s current listing this month, with the source named in the sentence. The per-shot amortization assumes a 5-year hardware life at roughly two shots daily — a reasonable household baseline, not a guarantee for every machine or user. See our editorial standards for how we source and verify pricing.

Home Espresso Questions, Answered

Is a manual espresso machine worth it for a beginner?

Only if the learning curve itself is appealing — expect several weeks of inconsistent shots before results are reliable, in exchange for the lowest long-run cost per shot and the highest quality ceiling of the three formats.

How often does a super-automatic espresso machine need repairs?

It varies by brand and water hardness, but the added internal complexity (built-in grinder, brew group, milk system) is the tradeoff for one-button convenience — buying from a brand with an established service network reduces the risk.

Are Nespresso pods actually more expensive than ground coffee?

Yes, on a per-shot basis — averaging around $1.25 per capsule against roughly $0.32 for the same size shot pulled from whole beans, pods cost close to four times as much per cup.

Do you need a separate grinder for a manual espresso machine?

Effectively yes — pre-ground coffee stales too fast and can’t be dialed to espresso’s fine, consistent grind, so skipping the grinder caps how good a manual machine’s shots can get regardless of the machine itself.

July 15, 2026

Adjustable Dumbbells: Six Recalls, One Failure Mode

Some links on our site may earn us a commission at no extra cost to you. Learn more.

Updated August 2026: every CPSC adjustable dumbbell recall since 2021 is pulled and tabulated from CPSC’s own pages, and the cast has changed: PowerBlock’s Elite EXP 404s on PowerBlock’s US store and is absent from its own US collection page, so the Elite USA 90 replaces it here. BowFlex’s recall page states that JRNY membership is only available if the claim is submitted by June 4, 2028 — narrower than the general claim deadline widely quoted for it.

The verdict. The REP Fitness QuickDraw takes the tally at 78.3 out of 100, on a clean CPSC dumbbell record, published dimensions at every weight, and a lifetime home-use warranty with its exclusions written down. The PowerBlock Elite USA 90 is 4.5 points back at 73.8 and holds 90 pounds per hand in less floor space than a BowFlex holds 52.5. The BowFlex Results Series 552 is the named loser at 29.5: its predecessor line is the largest recall this category has ever had, and BowFlex backs the current one for two years.

Prices are approximate, current as of publication, and change frequently — confirm the latest price with the retailer before buying.

Nearly Four Million Units Were Recalled, but Only One Thing Broke

Six adjustable dumbbell recalls sit on CPSC’s site from November 2021 forward. Read the hazard sentences side by side and they are the same sentence.

Recall	Date	US units	Reports	Injuries	CPSC’s hazard wording
BowFlex SelectTech 552 and 1090 (25-311)	5 Jun 2025	3,844,200	337 + 12	111	“The weight plates can dislodge from the handle during use”
iFIT ProForm Rapid Strike 50 (24-310)	25 Jul 2024	54,400	8	3	“The weight plates can dislodge from the handle during use”
Tzumi FitRx SmartBell Quick-Select (26-431)	23 Apr 2026	50,000	115+	6+	“Weight plates can dislodge from the handle during use”
iFIT NordicTrack 50 LB iSelect (24-102)	1 Feb 2024	15,400	40	8	“The weight plates can dislodge from the handle during use”
Amazon Basics 55 Lbs. (26-486)	14 May 2026	2,900	8	0	“can fail to engage completely during weight adjustment and dislodge”
PowerBlock Commercial Pro 100 (25-020)	24 Oct 2024	1,845	4	0	“Weight plates above 55 lbs… can dislodge from the 5 lb grip handles”
Totals		3,968,745	524+	128+	One mechanism, six times

Every figure is CPSC’s own, from the six recall pages linked in this section. Totals are my addition of those figures. “+” marks the two CPSC states as “more than” or “at least.”

The arithmetic: 3,844,200 + 54,400 + 50,000 + 15,400 + 2,900 + 1,845 = 3,968,745 US units. Reports 337 + 12 + 8 + 115 + 40 + 8 + 4 = at least 524. Injuries 111 + 3 + 6 + 8 = at least 128.

One correction worth making, because the number is widely misquoted. The BowFlex figure is often repeated as “349 reports.” CPSC never prints that number. It splits the incidents by firm: “Nautilus received 337 reports of the plates dislodging during use for units it sold, including 111 resulting in injuries such as concussions, abrasions, broken toes or contusions,” and separately “Johnson Health Tech Trading has received 12 reports of the plates dislodging during use with no injuries for units it sold.” 349 is a sum you compute, not a figure CPSC published.

BowFlex is 96.9 percent of the recalled units on that table, across a sales window CPSC dates from “as early as 2004 through April 2024.” Twenty years of the same failure. The unit on shelves now is the redesigned Results Series, and the recalled ones are the ones behind it.

Two things this table does not prove, and I want them said plainly. A clean recall record is not a safety test — it is an absence of one particular kind of bad news. And there is no publicly readable, methodologically documented independent test of plate retention on any of these products. Consumer Reports runs a dumbbell programme and says it evaluates ease of use, speed of adjustment and cost; its ratings sit behind a paywall. The category’s defining failure mode has never been independently tested in public.

PowerBlock Was Recalled Too, but for 1,845 Units and No Injuries

Worth being fair about, because a recall headline flattens everything. Recall 25-020, announced 24 October 2024, covers the Commercial Pro 100 weight set sold with 5 lb grip handles — a roughly $1,500 commercial product. CPSC’s wording: “PowerBlock has received four reports of weight plates slipping beyond the handle during use. No injuries have been reported.” The Elite line is not named anywhere on that page.

1,845 units against 3,844,200 is a factor of 2,084. Both are recalls. They are not the same event.

iFIT is on the table twice, which surprised me. The NordicTrack 50 LB iSelect, model NTSAW23020.0, was recalled in February 2024 with 40 reports and eight injuries, remedy a full refund. The ProForm Rapid Strike 50 LB, model PAMSDB20, followed in July 2024 with eight reports and three injuries, remedy a replacement moulded tray.

Here is a loose end I could not tie off and will not guess at. Retailers list the NordicTrack Select-A-Weight 55 lb under model code PAMSDB20 — the same code as the recalled ProForm 50 lb. The weights differ, the brands differ, and iFIT publishes nothing on its own site that resolves whether the two share a tray. If you own a Select-A-Weight, that is a question for iFIT, not for me.

REP Has a Recall on File, but Not on a Dumbbell

REP Fitness has one CPSC recall: strap safety brackets for its 4000 and 5000 series squat stands, February 2021, about 1,940 units, three reports of improperly welded brackets breaking, no injuries. That is a rack part. Search CPSC for a REP adjustable dumbbell recall and there is nothing.

So “REP has never been recalled” is false and “REP has never had a dumbbell recalled” is true. The distinction matters if you are the person being told the first one.

The Elite EXP Is Still Sold in Europe, but It Is Gone From the US Store

PowerBlock’s Elite EXP is no longer sold on PowerBlock’s own US store. As of 31 August 2026 the product URL returns a 404. PowerBlock’s own US adjustable dumbbells collection page lists the Pro 100 EXP, Elite USA 90, Pro 50, Pro 32 and Sport 24, plus the commercial range, and the Elite EXP is not on it. The same product still resolves on PowerBlock’s European store, and PowerBlock still sells Elite EXP expansion parts.

PowerBlock does not state anywhere that the Elite EXP is discontinued, so I will not say it is. What I can say is that it is no longer listed for sale in the US, and that the US model in its place is the Elite USA 90 at $469 for the 50 lb stage, $668 for 70 lb and $867 for 90 lb — prices PowerBlock’s product page publishes and its own collection page does not render at all.

Dumbbells resting on a rack in a home gym — A rack of fixed dumbbells. The floor area a rack like this occupies is the whole argument for buying one adjustable pair instead.

Three Makers Promise Seconds, but Not One Publishes a Method

Adjustment speed is the thing every one of these is sold on. Go looking for a measured figure and it evaporates.

The claim	Where it appears	What is behind it	Verdict
“Go from 5lbs to 60lbs in seconds”	REP’s QuickDraw product page	No number, no stated method, no definition of what the clock covers	Unmethodized
“Change weights quickly and easily with the secure magnetic pin”	PowerBlock’s Elite USA 90 page	No number at all	No claim to check
“15-in-1” compactness framing	BowFlex’s Results Series 552 page	No seconds claim published anywhere on the page	No claim to check
5 to 52 lb range	BowFlex’s collection page	BowFlex’s own product page for the same dumbbell says 5 to 52.5 lb, and its settings list ends at 52.5	BowFlex disagrees with BowFlex
Product weight	BowFlex’s 552 and 1090 product pages	Both return “454” with no unit. A pair of 552s is roughly 105 lb and a pair of 1090s roughly 180 lb, so one figure cannot describe both	Broken data field, not a spec
Claim deadline “June 4, 2028”	BowFlex’s recall page, and widely repeated	The sentence is “JRNY membership is only available if your claim is submitted by June 4th, 2028” — which is narrower than a claim deadline	Repeated more broadly than written

Claims check, run against each maker’s own current pages on 31 August 2026.

Consumer Reports explicitly tests speed of adjustment. It is the one metric no maker will put a stopwatch to in public, and the only body that measures it charges for the answer.

PowerBlock Holds 90 Pounds in the Least Floor Space, but You Lift Inside a Cage

You buy adjustables to get a rack’s worth of weight out of a corner. So price the corner.

Take each maker’s own published length and width for one dumbbell, multiply for footprint, divide by the maximum weight that dumbbell reaches. Square inches per pound. Lower is better.

Product	Published footprint (L x W)	Area	Max per hand	Sq in per pound
PowerBlock Elite USA 90	16.25″ x 6″	97.5 sq in	90 lb	1.08
REP QuickDraw, dumbbell only	18.5″ x 7.3″	135.1 sq in	60 lb	2.25
BowFlex Results Series 552	16.9″ x 8.3″	140.3 sq in	52.5 lb	2.67
REP QuickDraw, counting its cradle	20.6″ x 8.45″	174.1 sq in	60 lb	2.90

Footprints are each maker’s own published figures. The division is mine. REP is the only one of the three that publishes cradle dimensions, so it is the only one that can be compared honestly both ways.

PowerBlock is 2.5 times more space-efficient per pound than the BowFlex and 2.1 times more than the REP. That is not marketing. It falls out of the cage design having no cradle at all — the block sits on the floor.

Note what is missing from the table. BowFlex publishes no tray dimensions and neither does iFIT, so a tray-inclusive comparison across all three cannot be built from maker sources. I left the cells out rather than fill them with estimates.

Now resolution, which is quieter and matters more than the settings count suggests. PowerBlock advertises 27 selectable weights against BowFlex’s 15, which sounds decisive until you divide by the range each covers. PowerBlock: 27 settings across 85 pounds, one every 3.27 lb. BowFlex 552: 15 settings across 47.5 pounds, one every 3.39 lb. Effectively the same density. PowerBlock’s ladder also skips 12.5, 22.5 and 32.5 pounds, and BowFlex’s step widens from 2.5 lb to 5 lb the moment you pass 25 lb — right where a new lifter is actually progressing.

REP ships two 2.5 lb micro-adjustment plates per dumbbell as standard, which turns 12 settings into 23 across 55 pounds: one every 2.50 lb, the finest of the three, with a 60 lb ceiling and nowhere to go after it.

The Tally Puts REP First, but PowerBlock Owns the Floor Space

Five axes, weighted for this particular matchup, each scored out of 100. Recall record carries the most because the category has one dominant failure mode and a twenty-year history of it. Move that weight and the answer moves.

Axis (weight, and the reason for it)	REP QuickDraw	PowerBlock Elite USA 90	BowFlex Results 552
CPSC record on this product line (30%) — one failure mode, six recalls, twenty years	90	85	20
Floor area per pound of max load (20%) — the reason to buy adjustables at all	45	95	30
Warranty as written, exclusions included (20%) — the maker’s own bet on its mechanism	95	60	20
Weight resolution per pound of range (15%) — what progressive overload actually needs	70	55	45
Numbers the maker publishes for checking (15%) — you cannot verify a spec that is not printed	85	60	45
Weighted total	78.3	73.8	29.5

Weights are mine and are printed so you can move them. Weight the floor-space axis at 40% and PowerBlock takes it.

REP takes it at 78.3. Lifetime warranty with the exclusions spelled out — “wear and tear due to normal usage” and “any damage to any part resulting from abnormal/abusive use (consistent drops)” — no dumbbell recall, and the only published dimension set detailed enough to check. PowerBlock is 4.5 points back, which is a real gap rather than a tie, so nothing gets flipped for it.

BowFlex Results Series 552 is the named loser at 29.5, and the score is not about the redesign. It is about two years of parts coverage on a mechanism whose predecessor produced 111 injuries, a warranty page that 404s on BowFlex’s own site so the exclusions cannot be read, and extended-coverage plans sold at $69 to $99 against a $379 dumbbell.

What the Ledger does not settle: how a dumbbell feels in your hand. No maker and no independent body publishes a figure for that, and it is the thing people change their minds about.

All Three Will Get You Strong, but Each Has a Buyer It Is Wrong For

Skip the REP QuickDraw if you are already pressing 50s. It stops at 60 pounds per hand with no expansion path, and the cradle adds a footprint the others do not have to declare. Buying it at your current strength and outgrowing it in two years is buying twice.
Skip the PowerBlock Elite USA 90 if you dislike lifting inside a cage. You put your hand through the block, and the ladder skips 12.5, 22.5 and 32.5 pounds. If you want small jumps at the bottom of the range, that gap is where you live.
Skip the BowFlex Results Series 552 if you want the maker’s terms in writing. Two years of parts coverage, exclusions unavailable on BowFlex’s own site, and a collection page that disagrees with its own product page about the weight range. If you own a pre-recall 552 or 1090, stop using it and file — and check BowFlex’s recall page for what the remedy actually is, because it is a voucher rather than cash.

One thing none of this covers, deliberately: how each mechanism wears out and what a broken one costs. That belongs to a different page — our longevity comparison of Bowflex, PowerBlock and NordicTrack maps where each mechanism fails and runs the cost-per-warranted-year table. If you are still deciding whether adjustables are the right purchase, bands against dumbbells against kettlebells covers the budget-and-space case, and the bench you will want next is in the bench and power tower comparison. The rest is in Home Gym.

July 14, 2026

Smart Lock Grades: Schlage Cites One, Ultraloq Doesn’t

Some links on our site may earn us a commission at no extra cost to you. Learn more.

The verdict. The Schlage Encode Plus at $329 takes this smart lock ledger at 78.0 of 100, and the gap is not close: the Yale Assure Lock 2 lands at 46.3, the August Wi-Fi Smart Lock at 45.3, and the Ultraloq U-Bolt Pro WiFi at $179.99 finishes last at 34.5. Schlage wins on paperwork rather than hardware. It is the only one of the four that names the standard its grade was awarded under, the only one that publishes the conditions behind its battery claim, and the only one with three years on the electronics.

Updated August 31, 2026: every price, grade claim, battery figure and warranty term re-verified against each maker’s own pages. Added the certification audit, the battery-cycle arithmetic against BHMA’s own durability test, and the CVE check. August Wi-Fi is still sold at $199.99 on August’s own store and has not been folded into Yale.

Published July 14, 2026 · Updated August 31, 2026 · By Nate Colburn

Schlage Is the Only One That Names a Standard Edition

Five axes. The weights are mine, the reason for each is printed beside it, and every entry names its source. Move a weight and the tally moves; that is what the table is for.

Axis and weight	Why it carries that weight	Schlage Encode Plus	Yale Assure Lock 2	August Wi-Fi	Ultraloq U-Bolt Pro WiFi
Certification you can name — 30%	The grade is the only security number on the box. A grade without a standard number is a marketing word.	70 — cites ANSI/BHMA A156.40-2015 Grade AAA and A156.36 Grade 1; the 2015 edition has since been superseded twice	30 — three of Yale’s own pages say three different things	40 — claims no grade, and has a structural reason: it reuses your deadbolt	25 — claims “ANSI Commercial Grade 1 Certification” with no standard number and no lab named
Battery claim, and its conditions — 20%	These locks run on AA cells. A figure with no stated duty cycle is not a figure.	75 — six months, and Schlage publishes the cycle ceiling and the Apple Home penalty	15 — publishes no duration on any of its three product pages or in the manual	40 — “up to 3 months” on 2 CR123, unqualified and the shortest here	15 — 8 months on the headline, 3–6 months in its own FAQ
Electronics warranty — 20%	The mechanism outlives the circuit board. Lifetime mechanical coverage does not cover the part that fails.	90 — 3 years electronics, lifetime mechanical and finish	55 — 2 years electronics; lifetime cover ends if you move, and does not transfer	25 — one year flat, no split published	45 — 18 months electronics; batteries named as an excluded consumable
Ongoing cost — 15%	A lock that meters your own door back to you is a different product.	90 — Schlage’s app page lists history, codes and sharing with no tier	85 — free Yale Access app; the paid Yale subscriptions are alarms and other regions	85 — no subscription found for lock features	60 — ULTRALOQ + Air is called “lifetime free service” with no pricing table published
Published vulnerability record — 15%	A lock is now a radio. The vendor’s disclosure history is part of the product.	70 — no CVE for the BE499	70 — no CVE for the Assure Lock 2	50 — a chipset SDK flaw names “August Smart Lock” generically, availability only	40 — no CVE for this model, but a CVSS 8.1 unlock-code flaw on its own sibling
Weighted total		78.0	46.3	45.3	34.5

Schlage’s arithmetic, shown: 0.30 × 70 = 21.0, plus 0.20 × 75 = 15.0, plus 0.20 × 90 = 18.0, plus 0.15 × 90 = 13.5, plus 0.15 × 70 = 10.5. Total 78.0.

The loser is the Ultraloq U-Bolt Pro WiFi at 34.5, and it is also the cheapest lock on this page at $179.99. That is the whole trade. Six ways to open a door, a fingerprint reader the others do not have at the price, and a maker whose own website cannot keep its battery figure or its certification straight from one page to the next.

Prices are approximate, current as of publication, and change frequently — confirm the latest price with the retailer before buying.

BHMA’s Certified Products Directory Will Not Open Without a Login

Start with what the grades actually mean, because the marketing uses two different scales interchangeably.

The commercial scale is ANSI/BHMA A156.36, Auxiliary Locks. BHMA’s own summary puts real numbers on it: “Grade 1 dead locks, for example, must pass a rigorous test through 250,000 cycles of opening and closing,” plus “a 1350 pound force on the door against the dead bolt for Grade 1,” and key-operated locks that “must open with a maximum torque of 9 lbf-in.”

The residential scale is different. ANSI/BHMA A156.40, Residential Deadbolts, grades three categories separately — Security, Durability, Finish — and awards letters, which is where AAA comes from. The current 2025 edition specifies “Seventeen aggressive tests,” one of which involves “loading the door against the deadbolt with 1,350 pounds of force,” and puts Grade A residential dead latches at “400,000 cycles of opening and closing.”

Now the audit. Schlage’s support article states the Encode Plus holds ANSI/BHMA A156.40-2015 Grade AAA and A156.36 Grade 1. That is the most specific claim of the four, and it carries an edition year — 2015, which A156.40 has since replaced twice, in 2020 and again in 2025. Ultraloq claims “ANSI Commercial Grade 1 Certification” and names no standard number at all. Yale’s main Assure Lock 2 page publishes no grade, its Touch page claims “Grade 2 BHMA/ANSI rated,” and its Plus page settles for “tested to industry-leading BHMA standards.” August publishes nothing, which is defensible: it replaces your interior thumb-turn and leaves the graded deadbolt in your door exactly as it was.

I tried to check all four against BHMA’s certified products directory. The directory exists, it lists Schlage, Yale and U-tec among its certified manufacturers, and its live search subdomain redirects to a sign-in wall. I could not corroborate a single one of these four grade claims against a certification record. Treat every grade on this page as manufacturer-stated. BHMA itself draws the distinction between products it has certified through third-party testing and products whose makers merely say they “meet standards” — and Yale’s “tested to” phrasing and Ultraloq’s numberless “certification” both sit on the wrong side of that line.

Yale Does Not Publish a Battery Life for the Assure Lock 2

Four locks, four battery claims, and the fine print is where the differences live.

Lock	Cells	The claim	Conditions the maker attaches
Schlage Encode Plus	4 AA	“up to six months of battery life with typical use”	“Less than 10-15 lock unlock cycles a day is acceptable.” Alkaline recommended, lithium discouraged. Paired to both Schlage Home and Apple Home, “This dual communication significantly increases battery consumption.”
Yale Assure Lock 2	4 AA	No figure published	Not on the standard page, the Touch page, the Plus page, or in the installation manual
August Wi-Fi	2 CR123	“up to 3 months”	None stated. No cycles-per-day, no Wi-Fi-off alternative
Ultraloq U-Bolt Pro WiFi	8 AA	“Up to 8 months” on alkaline, “Up to 1 year” on lithium	See below. Ultraloq’s own pages give three incompatible answers.

Schlage comes out of this well by being candid rather than by being generous. Six months is not the longest number on the page. It is the only one attached to a duty cycle you can check yourself against your own front door, and Schlage volunteers that running Apple Home alongside its own cloud costs you battery. That is a maker arguing against its own feature list.

Yale’s silence is the finding. Three product pages and an installation manual, all naming four AA alkaline cells, none naming a duration.

Ultraloq’s Own Support Page Contradicts Its Own Headline

On the U-Bolt Series page, the headline reads “Up to 8 months with Wi-Fi connection and Alkaline batteries.” Further down that same page, in the FAQ, the answer reads “U-Bolt WiFi: 3-6 months.” One page, two answers, a factor of about two apart.

Ultraloq’s support site offers a third framing, in accesses rather than months: “up to 6000 times access,” and separately “the battery life is up to 1 year on average use of 20 times access per day.” Do that division. 6,000 divided by 20 is 300 days. Ten months, not twelve. The support page’s own cycle count does not reach the support page’s own headline.

Your Batteries Will Not Reach BHMA’s 250,000-Cycle Test

Here is a piece of arithmetic worth the paragraph. Take Schlage’s own published condition — six months at 10 to 15 lock-unlock cycles a day — and count it out. Six months is about 182 days. At 10 cycles a day that is 1,820 cycles per set of batteries; at 15, it is 2,730.

BHMA’s Grade 1 deadlock durability test runs 250,000 cycles. Divide: 250,000 divided by 2,730 is 92 sets of batteries, and 250,000 divided by 1,820 is 138 sets. At four AA cells apiece, reaching the mechanical durability figure printed on the box takes somewhere between 368 and 552 batteries.

That is not a criticism of Schlage. It is the shape of the whole category. The bolt is certified to a standard that assumes a mechanical lifetime measured in hundreds of thousands of throws, and the electronics driving it are specified in units of a few thousand. Whichever of these four you buy, the deadbolt is the part that will still be working.

The Cyber Trust Mark Has Not Labelled a Single Smart Lock

If you have read that a US government cybersecurity label is coming to connected devices, it is real and it is stalled.

The programme sits in FCC PS Docket No. 23-239. A January 2026 FCC public notice records that “UL Solutions withdrew as Lead Administrator effective December 19, 2025,” and opens a fresh filing window for replacements between January 7 and January 28, 2026. Eleven Cybersecurity Label Administrators are conditionally approved, and a companion notice states that conditional approval does not let them certify products to use the label yet.

So: no lead administrator as of the most recent FCC notices I could find, no labels being issued, and no product of any kind carrying the mark. I found no statement from Schlage, Yale, August or U-tec about participating. Do not let a retailer sell you a lock on this label. It does not exist yet.

Every CVE I Found Names a Sibling, Not These Four

I searched the National Vulnerability Database for all four brands. No published CVE names the Schlage BE499 Encode Plus, the Yale Assure Lock 2, the current August Wi-Fi Smart Lock, or the Ultraloq U-Bolt Pro.

Their relatives are another matter. CVE-2022-46480, scored 8.1 High, describes “Incorrect Session Management and Credential Re-use in the Bluetooth LE stack of the Ultraloq UL3 2nd Gen Smart Lock Firmware 02.27.0012” that “allows an attacker to sniff the unlock code and unlock the device whilst within Bluetooth range.” That is a different Ultraloq model, and it is a genuine open-the-door flaw. CVE-2019-17518 is a Dialog Semiconductor Bluetooth SDK bug that NVD notes “affects, for example, August Smart Lock,” with availability-only impact — a crash, not an entry. Schlage’s entry, CVE-2020-9059, is a battery-drain issue on the Z-Wave BE468, a different radio in a different lock. Yale’s two are a Bluetooth key app and a keyless RFID tag, neither of them the Assure Lock 2.

A clean record is weak evidence. It may mean the model is sound; it may mean no researcher has looked. I scored it as vendor history rather than model safety, which is why Ultraloq loses points here for a lock it does not sell you.

August Does Not Split Its Warranty, and Neither Does Its Price

Four locks, four different answers to what happens when the electronics quit. Here is who each one is wrong for.

Close-up of a deadbolt lock on a residential door — The bolt and the strike plate behind it are the parts every grade in this comparison is actually measuring.

Schlage Encode Plus, $329. Limited lifetime mechanical and finish, limited three years on the electronics. It is the most expensive of the four and no longer Schlage’s flagship — the Sense Pro now sits above it from $399. Worth knowing before you pay top of range for a lock the maker has already positioned as second tier. Skip this if a fingerprint reader is what you actually wanted; the Encode Plus does not have one at any price.

Yale Assure Lock 2, $189.99 to $309.99 depending on configuration. The warranty is where this one gets interesting, and not in a good way. Yale’s lifetime mechanical and finish cover lasts only “as long as the original end user purchaser continuously resides at the original installation location,” electronics run two years, and the terms state in capitals that the warranty “SHALL NOT APPLY TO SUBSEQUENT OWNERS.” Non-consumer installations drop to three years on the mechanical side. Skip this if the lock is going on a short-term rental or a property you plan to sell, because both of those cases fall outside the cover you thought you were buying.

August Wi-Fi Smart Lock, $199.99, or $229.99 bundled with the keypad. Still sold on August’s own store, which now carries “a Yale Company” in its footer and sells Yale hardware alongside its own. One year of warranty, flat, with no split between electronics and mechanical — though with no exterior hardware to warrant, that is less of a gap than it reads. It is also the only one here that leaves your existing deadbolt and your existing keys alone. Skip this if your existing deadbolt is the cheap builder-grade one that came with the house, because August inherits whatever security that bolt has and adds none.

Ultraloq U-Bolt Pro WiFi, $179.99. Eighteen months on the electronics, lifetime on the mechanism, and an exclusion list that names “consumable parts, including batteries.” A plain Bluetooth-only U-Bolt Pro is no longer listed on U-tec’s own collection page; the Wi-Fi version is what is sold now. Skip this if you want to be able to look up the grade on the box, because right now that trail ends at a marketing sentence.

A note on recalls, since it should be checked and rarely is: a search of CPSC’s published recalls turned up nothing for Schlage, Yale, August or U-tec. The lock recalls that do surface belong to Best Lock, Stanley Security Solutions and dormakaba, all commercial hardware, all entrapment hazards, none related to these four.

Yale Does Not Beat August by Enough to Settle It

Yale 46.3, August 45.3. One point. That is not a result, it is a rounding difference, so here is the tie-breaker.

My own money goes to August, and the reason is that the two locks fail differently. Yale’s weakness is that it will not commit — no battery figure anywhere, three grades across three pages, and a lifetime warranty that quietly ends when you move house. August’s weakness is that it is honest and modest: three months of battery, one year of cover, and no pretence about hardware it does not supply. Given a one-point gap, I would rather own the limitation I can see.

The caveat that outranks all of it comes from Consumer Reports, which swings a 100-pound steel battering ram at locked doors eight times and reports that the cause of failure “is more often than not the short screws used to attach the strike plate to the doorframe.” Replacing the strike plate with a reinforced one, CR says, “costs as little as $10” and “improves the lock’s resistance to kicking by magnitudes.” Four locks separated by 43 points on my ledger, and a $10 steel plate moves the number that matters more than the choice between them does. Buy the plate whichever lock you pick.

If you are wiring up the rest of the door, we have gone at the neighbouring purchases the same way: video doorbells that store footage locally instead of charging monthly, camera subscriptions priced as the product they really are, and the smart plug that actually holds up. The rest sits in smart home.

If you can pull a BHMA certification record for any of these four models, send it. I will print it and adjust the top axis.

July 14, 2026