Robot Vacuum Suction Claims, Decoded

A robot vacuum suction claim is comparable only when it is given in pascals, and most of the biggest claims are not. iRobot publishes multiples against an older Roomba, so a 70X and a 180x figure rank against each other but against nothing else. Roborock publishes 18,500 Pa, a real pressure measurement that no iRobot page read states. The two numbers cannot share a table. This guide reads five product pages, three from iRobot and two from Roborock, checked October 10, 2026, to show which rows can be compared, which cannot, and which figures none of the five publishes at all.

A suction multiple compares a product to its ancestor, not to a rival

iRobot's headline figures are ratios with a footnote. The Roomba Max 705 Vac page claims "180x more extreme power-lifting suction", measured against the Roomba 600 series in spot clean mode with a full battery.

The cheaper Roomba 105X Vac page claims "70X more power-lifting suction" against the same baseline. Because the baseline matches, those two figures can be ranked against each other.

The baseline then changes. The Roomba s9+ page claims "100% more suction" compared to Roomba i Series robots, a different reference product. That figure cannot be placed on the same scale as the other two.

The reading rule: find the footnote before the number. A multiple without a named baseline, a named mode and a named battery state is an advertising claim, and two multiples with different baselines cannot be subtracted.

Pascals compare across brands, but they are not a cleaning score

Roborock states an absolute figure. Its Qrevo Curv page lists "Industry-leading 18,500 Pa HyperForce", a pressure measurement that can in principle be set beside any other maker's pascal figure.

The figure then vanishes within the same brand. Roborock's Saros 10R page, read on the same day, publishes no pascal rating at all, which makes the two models in one catalogue incomparable to each other.

Pascals also measure the wrong thing on their own. Suction pressure at the intake does not describe pickup from carpet, where brush design, seal and airflow matter. None of the five pages read publishes airflow in cubic feet per minute or litres per second.

Treat a pascal figure the way you would treat a decibel rating with no test distance: real, comparable, and only one input among several. The same caution applies to the airflow numbers in how to read an air purifier spec sheet.

Navigation names are brands, with one testable claim between them

Four trademarks cover three systems at iRobot: "ClearView LiDAR" on the 105X, "ClearView Pro LiDAR" plus "PrecisionVision AI Technology" on the Max 705, and "Imprint Smart Mapping" on the s9+. None of those names is a specification.

Roborock adds two more: "PreciSense LiDAR Navigation" with "structured light and an RGB camera" on the Qrevo Curv, and a "StarSight Autonomous System 2.0" with "Dual Transmitter Solid-State LiDAR" on the Saros 10R.

One claim in the set can be checked. Roborock states its system "detects and avoids 62 object types across 20 categories". iRobot describes PrecisionVision as navigating around "cords, socks, and pet waste" and publishes no count.

iRobot publishes the other testable figure: a mapping time of "8 minutes" for 1000 square feet, or 93 square metres. A speed with an area attached is a specification; a trademark is not.

The dock claim is measured in days, and days depend on your floors

All three makers quantify the dock in time. iRobot says the Max 705 dock traps debris "as small as 0.7 microns in an enclosed bag" and keeps a household clear for "75 days", while the s9+ Clean Base is listed at "up to 60 days".

Roborock states "up to 7 weeks" between emptyings for the Qrevo Curv. By editorial arithmetic that is 49 days, so the three claims rank 75, 60 and 49 under whatever conditions each maker assumed.

Those conditions are unpublished. No page read states the floor area, the number of occupants, or whether pets are present behind its days figure, and a two-bedroom flat and a four-bedroom house with a dog are not the same test.

The checkable version is missing too. None of the five pages publishes a dustbin volume in millilitres or a dock bag capacity in litres, which is the figure that would let a reader verify the days claim instead of taking it.

Run time, noise and height are published unevenly

Run time is absent from all five pages. iRobot lists the battery as "Lithium Ion" without a capacity in milliamp hours or watt hours, and Roborock's two pages list no battery figure at all, so cleaning area per charge cannot be estimated.

Only Roborock publishes noise, and only for one mode: "as low as 55dB" in mop mode. An "as low as" figure for a single quiet mode is the weakest form of a decibel claim, and the vacuum mode figure is unpublished.

Height is the row that decides whether the robot reaches under furniture, and here all three are close. The Max 705 is listed at "13.8 x 14.0 x 4.1" inches, the 105X at "13.23 (L) x 13.19 (W) x 4.09 (H)", and the Qrevo Curv at "352*347*103 mm".

By editorial conversion, 103 mm is about 4.06 inches, so all three sit within five hundredths of an inch. Measure your lowest sofa clearance first.

Warranty and price are the rows nobody fills in

Not one of the five pages read publishes a warranty term. Not the Max 705, not the 105X, not the s9+, not the Qrevo Curv, not the Saros 10R, as of October 10, 2026.

That is a category outlier. The power station makers covered in portable power station numbers that matter print a five-year term on the product page, and the appliance warranty documents in what appliance warranty pages leave out run to several pages.

Price is nearly as thin. Only the 105X states one, at "$249.99". The Max 705 page says it is "Available at select retailers", the s9+ page states the product "is not available", and both Roborock pages show no price.

A complete robot vacuum spec sheet would state all of the following:

• Suction in pascals, and any multiple claim with its baseline, mode and battery state
• Airflow in cubic feet per minute or litres per second
• Dustbin volume in millilitres, and dock bag capacity in litres
• Dock interval in days, with the floor area and household it assumes
• Battery capacity and run time in minutes, per cleaning mode
• Noise in decibels for the vacuum mode, not only the quiet mode
• Robot height, and the brush type and filter rating
• A warranty term in years, and a price

Frequently asked questions

Does a higher Pa rating mean better cleaning?

Not on its own. Pascals measure suction pressure at the intake, not pickup from carpet, which also depends on brush design, seal and airflow. None of the five pages read for this guide publishes an airflow figure in CFM or litres per second.

What does 180x more suction actually mean?

It is a ratio against an older product. iRobot measures the Roomba Max 705 claim against the Roomba 600 series in spot clean mode with a full battery. The Roomba s9+ instead claims 100% more suction than i Series robots, a different baseline entirely.

How long does a self-emptying dock last before it needs a new bag?

The makers claim 75 days for the Roomba Max 705 dock, up to 60 days for the s9+ Clean Base and up to 7 weeks for the Roborock Qrevo Curv, which is 49 days. None states the floor area or household size behind the figure.

Do robot vacuum pages list a run time?

The five pages read on October 10, 2026 do not. iRobot lists the battery only as Lithium Ion with no capacity, and Roborock's pages list no battery figure, so area covered per charge cannot be worked out from the published specifications.

What warranty comes with a robot vacuum?

The product pages read here do not say. No warranty term appears on the Roomba Max 705, 105X or s9+ pages, or on the Roborock Qrevo Curv or Saros 10R pages. Ask the seller for the term in writing before purchase.

Bottom line

Suction is the loudest row on a robot vacuum page and the least comparable. Rank multiples only against the same baseline, use pascals only when both makers publish them, and ignore navigation trademarks in favour of the two figures that carry units: object counts and mapping time.

Then look for what is missing. Run time, dustbin volume, airflow, vacuum-mode noise and the warranty term are absent from all five pages read, which is a wider gap than the ones found in monitor specs or wireless headphone spec sheets.

Ask the seller for those rows in writing, and price the electricity separately using what energy labels say about running cost. A listing that publishes a five-digit pascal figure and no warranty has chosen which number it wants you to read.