2023 Ultimate Robot Vacuum and Mop ComparisonJune 1, 2023
In my opinion, robotic vacuums are the single most useful item you can add to a smart home and are one of the biggest advances in home tech in the last 10 years. But the big question is whether it’s worth the significant increase in cost to look for premium mopping features along with your robotic vacuum. Today I’ve got 9 top-of-the-line robotic mop and vacuum combos and we’re going to test their vacuuming performance on both carpet and hard flooring, auto-empty bin performance, ability to resist pet hair clogs, long hair tangle avoidance, real world mopping performance, unattended mopping capacity, automated mop pad cleaning and drying, AI obstacle recognition and avoidance, baseboard and furniture damaging behaviors, app features, and smart home integrations.
As a reminder there are no sponsored reviews on this channel, and all the data you see in this video is from my personal testing instead of relying on manufacturer claims. Feel free to use the timestamps in the video to jump around, and links and coupon deals are down in the description.
Starting with the least expensive flagship robot, the Ecovacs T10 OMNI was just released with an MSRP of $1200, but a street price of around $950. The T10 uses Ecovacs combination rubber and soft bristle roller for vacuuming, and dual spinning pads for mopping. The T10 uses a spinning LIDAR unit on top of the robot for mapping, and a front facing camera for AI object detection and avoidance. This is the T10 OMNI so it includes the OMNI base, which does it all including automatic dustbin emptying, automatic mop washing, and hot air mop drying to avoid mold and mildew growing on the pads.
Next, with a street price of around $999, is the Dreametech L10s Ultra. The L10s uses a single rubber roller for vacuuming and dual spinning pads for mopping, but unlike the Ecovacs T10, the Dreametech can lift its mopping pads when going over carpet or returning to the base station. The L10s uses a top mounted spinning LIDAR unit for mapping, and front facing camera for AI object detection and avoidance. The L10s Ultra comes with the Dreametech Ultra base which includes automatic dustbin emptying, automatic mop washing, an automatic mopping solution dispenser, and hot air drying.
Then, also with a street price around $999 is the Narwal Freo. The Freo has a combination rubber and soft bristle roller for vacuuming, and dual spinning pads for mopping, and like the Dreametech the Narwal can raise its mopping pads when vacuuming carpet or returning to the base station. The Freo uses a top mounted LIDAR unit for mapping but lacks any advanced object avoidance. The Narwal Freo base station doesn’t have automatic emptying for the dustbin but does have large water tanks for automatic mop washing, an automatic mopping solution dispenser, and hot air drying.
The next $999 robot is the Ecovacs X1 OMNI, which is very similar to the T10 using a combination rubber and soft bristle roller for vacuuming and dual spinning pads for mopping. Like the T10, the X1 has a top mounted spinning LIDAR module for mapping, but the X1 has a slightly more advanced front mounted camera and sensor set for AI Object detection and avoidance. The X1 OMNI also includes the OMNI base which has the same features as the T10 OMNI including automatic dustbin emptying, automatic mop washing, and hot air mop drying.
Next for $1099 is the iRobot Roomba J7+ Combo. The J7+ Combo is the first combination robotic vacuum and mop from Roomba, so in addition to Roomba’s usual dual roller design, the J7+ combo also has a unique folding mop that can completely stow away when not in use. The J7+ Combo lacks any LIDAR sensors and instead uses a front facing camera to map out your house and enable cleaning of specific rooms. Unlike the rest of the bases we’ve seen so far, the Roomba base only includes automatic dustbin emptying, and doesn’t have any mop cleaning, refilling, or drying capabilities.
After that for a street price around $1399 is the flagship Roborock S7 MaxV Ultra. The S7 MaxV uses a single rubber roller bar for vacuuming and Roborock’s patented Vibrarise mopping system. The Vibra portion is this rapidly vibrating center section of the mopping pad, and the rise part refers to the ability to raise the entire mopping pad when traveling over carpet or returning to the base station. The S7 MaxV uses a top mounted spinning LIDAR unit for mapping, and a front facing camera for AI object detection and avoidance. This Ultra base station has automatic dustbin emptying, and automatic mop washing, but lacks a hot air drying system, though an add-on unit is available for $100.
For $200 more with an MSRP of $1599 is the new Roborock S8 Pro Ultra. For vacuuming, the S8 Pro uses a brand-new dual rubber roller bar system that was previously patented by Roomba, and for mopping the S8 Pro has the upgraded Vibrarise 2.0 mop that has an extra vibration zone for more efficient mopping. The S8 Pro uses a top mounted LIDAR unit for mapping but swaps out the front facing camera on the S7 MaxV for a more privacy focused laser object detection array. The S8 Pro Ultra also includes the new Ultra base which is almost exactly the same as the previous base with automatic dustbin emptying and automatic mop washing, but with the addition of hot air drying, and some slightly different styling.
And last, the most expensive robot I’ll be testing is the Lefant T1, which is a mopping only robot with an MSRP of $1699. The T1 is significantly smaller than the other robots and uses dual spinning pads for mopping but has no vacuuming capabilities at all. The T1 uses a top mounted LIDAR unit for mapping but lacks any other advanced object avoidance system. The T1 base has automatic detergent dispensing, automatic mop washing, and hot air drying. I reached out to Lefant multiple times asking why the T1 was more expensive than much more feature rich robots, but they assured me that the T1’s performance would speak for itself, so I’m expecting to be wowed by its mopping capabilities.
It should be noted that all these vacuums are brand new except for the Ecovacs X1 OMNI and Roborock S7 MaxV Ultra that I’ve been using daily in my house for the last year. However, I did replace their roller bars and filters before testing.
Starting with the carpet vacuuming performance test, I sprinkled a mixture of 10g each of flour, salt, flax seed, and rice grains onto my high pile carpet and weighed the dust bins of each vacuum before and after an area cleaning and between testing I used an upright corded vacuum to remove any leftover debris from the carpet before the next round.
The top performer in the carpet pickup test was the Roborock S7 MaxV Ultra, which picked up 36 grams or 90% of the flour, salt, flax seed, and rice mixture. After that the Ecovacs T10 OMNI and Roborock S8 Pro Ultra both picked up 30 grams or 75% of the mixture, then the Roomba J7+ Combo picked up 72.5% and the Dreametech L10s picked up 70%. The Narwal Freo struggled, only picking up 57.5% from the carpet, and the most surprising results were from the Ecovacs X1 OMNI that only picked up 47.5% of the mixture, or 19 grams total, and really struggled to pick up the heavier particles like the flax seed and rice grains.
I repeated the same test on hard flooring where the Roborock S7 MaxV Ultra also finished on top, picking up all 40g for a score of 100%. On hard flooring the Narwal Freo also did very well with a 97.5% pickup score, after that the Roomba J7+ Combo had 90% and the Roborock S8 Pro and Dreametech L10s picked up 87.5% of the mixture. The bottom two performers were the Ecovacs T10 OMNI with 77.5% and again, surprisingly, the Ecovacs X1 OMNI finished last with a 67.5% pickup score. Also, just out of curiosity I ran the Lefant T1 to see if it could mop up this debris, and it went about as poorly as you’d expect.
So that means that for vacuuming performance the Roborock S7 MaxV Ultra was first with an average pickup score of 95% followed by the Roomba J7+ Combo and Roborock S8 Pro Ultra that both finished with an average of 81.3%.
To measure the performance of the Auto Empty Bins, I weighed the dustbins after the auto empty process and compared it to the starting weight before the vacuuming pickup tests.
In both the hard flooring and carpet testing the Roborock S7 MaxV Ultra cleared 100% of its dustbin, and the Roborock S8 Pro Ultra performed similarly with 0g left after the carpet pickup test, and only 4g left after the hard flooring test. The Dreametech L10s also performed very well with 2g left after the carpet test, and 4g after the hard flooring test. The Ecovacs T10 OMNI and X1 OMNI both performed relatively poorly with a combined 28 and 29 grams left in their bins after the two tests and the Roomba J7+ Combo also did very poorly with a combined 31 grams left in the bin after the auto empty process.
However, having emptied quite a few robotic vacuum bins, I know that larger debris is much less common than carpet fluff, and if you own pets then matted dog and cat hair can cause real problems for these auto empty systems.
To evaluate their resistance to pet hair clogging I ran a series of tests with stretched and combined cotton balls to determine the maximum amount of matted fibers each auto empty system could clear. I started with 3 large cotton balls and increased the number of cotton balls by 3 after each successful auto empty process.
In this test the Dreametech L10s was by far the best performer and has a great system for emptying its bin which uses a forced air inlet and a separate waste outlet on the side of the robot. The L10s was able to clear 12 cotton balls on its first try, and 15 cotton balls after 2 tries, which barely fit into the bin in the first place.
The Roomba J7+ Combo successfully cleared 12 cotton balls after 2 attempts, which I thought was pretty impressive based on the small outlet of the auto empty system. It’s also worth noting that the J7+ is unlikely to ever need to empty a clog as large as 12 cotton balls since Roomba also has a patent on a dustbin sensor that will cause the Roomba to return to the base and empty its bin mid-clean if it starts to get too full.
In my house though, pet hair isn’t the big problem, human hair is. Specifically, my wife and daughter’s long hair can completely disable a brush roller in a matter of weeks. So, to test how these vacuums deal with long hair I had them pick up ten 18” strands of plastic easter grass and noted where the strands ended up, and I repeated that test twice per robot.
The Roborock S7 MaxV Ultra performed the best here and managed to avoid tangles using its roller that’s designed to pull hair inside of the hollow recesses on either side. Both ends of the roller are completely removable, which then allows you to pull out the tangled hair without the use of tools. These results weren’t surprising to me since, as I mentioned, I’ve been using the S7 MaxV every day for the last year, and this is completely consistent with the real-world long hair performance that I’ve experienced.
The Dreametech L10s also uses a similar strategy and has a similar style roller but didn’t manage to completely avoid tangles and had strands wrapped around both the side brush, and one wheel.
The Narwal Freo avoided getting hair wrapped around its wheels and side brushes but didn’t seem to have any real strategy for avoiding brush roll tangles, and disappointingly the S8 Pro Ultra’s dual brush roller system performed significantly worse than the S7 MaxV Ultra’s single roller.
Both Ecovacs units performed identically and were able to complete their vacuuming runs but had multiple tangles that would need to be dealt with to avoid performance issues, and the Roomba J7+ Combo performed the worst, and in both trials got tangled enough to cause an error that stopped the vacuuming task.
So, combining vacuuming performance, auto empty bin performance and tangle resistance the Roborock S7 MaxV Ultra is the undisputed champion of vacuuming, followed closely by the Dreametech L10s, but these aren’t just vacuums, so next let’s test out their mopping abilities.
At first, I designed a torture style test using dried coffee, maple syrup, and ketchup on my vinyl flooring, but after running the mops through that test I honestly couldn’t quantify that one mop had performed significantly better than another. Also, even though I’m definitely guilty of using these extreme style tests in the past, I don’t think they accurately portray the purpose of these robotic mops, which aren’t an excuse to drop a tablespoon of ketchup on the ground and let it dry.
Instead, I tried to replicate the most common mopping scenario in my house, which is when someone with dirty feet or dirty shoes steps on a wet floor. To set up this test I made a solution of Florida mud and spread 3mL of it on the floor. Then I used a hairdryer for 5 minutes to completely dry the stains to the flooring and let the flooring cool down to room temperature. I set all the mops to medium water flow and 2 passes and started a room clean for just the kitchen.
After cleaning I did 2 even sprays of water over the testing area and wiped it clean with a paper towel, and judged mopping performance based on the amount of leftover dirt on the paper towel.
For reference I also included a paper towel wipe that was done without any mopping, and one that was done after manual mopping with floor cleaning solution, which I did in between each trial.
Both visually and by feel the Narwal Freo was the top performing mop, and it left a very similar amount of dirt to my manual mopping results. After the Freo the next best mopping performance was from the Dreametech L10s, and surprisingly the Ecovacs T10 OMNI was close behind that, despite the fact that the Narwal Freo and Dreametech L10s were using mopping solution from their automatic dispensers while the Ecovacs T10 OMNI was using just plain water. A surprisingly poor performer was the Ecovacs X1 OMNI, which I suspect is due to the fact it has mopped my floor approximately 100 times over the last year and even though I wash the mopping pads regularly, it might be time for a full replacement of both sets.
However, the worst performer by a significant margin was the Lefant T1, which is both the most expensive, and most specialized of all the robotic mops I’m testing, and you can see that despite that, it was designed with a gap between the mopping pads meaning it misses a whole 1” section on each pass, leaving a striped floor behind.
Aside from performance, the other important factor for mopping is how often the clean and dirty water tanks will need to be refilled and emptied. In the case of the Narwal Freo and Ecovacs T10 OMNI, the only time the mopping pads get wet is when they are cleaned at the base station, while the Roborock S8 Pro Ultra, Roborock S7 MaxV Ultra, Dreametech L10s, and Ecovacs X1 OMNI can fill their onboard water tanks using the clean water reservoir in the base station, and the Roomba J7+ Combo, and Lefant T1 have on board water tanks that need to be filled manually. Still, the majority of the water will be used by the station itself to wash the mopping pads, and that metric can be used to estimate the total mopping capacity of each system.
I measured the amount of water in the clean and dirty water tanks before and after a mop washing cycle, and I also measured the total amount of water that the clean water tank could hold and using those values I was able to calculate the total number of mop-washing cycles for each robot and base station.
Each mopping session always begins and ends with a mop wash, and using default settings the mops will generally be cleaned about every 100 to 150 square feet. So, if you mop 2 normal sized rooms per session that would be 3 total mop washes and doing that 3 times a week would represent 9 washes total. Using that metric, the Ecovacs X1 OMNI and Dreametech L10s would need to have their mop water tended to every week, while the Narwal Freo and Roborock S8 Pro would need attention every 2 weeks. The Roborock S7 MaxV Ultra could go 3 weeks, and the Ecovacs T10 OMNI could last 4. The Lefant T1 had both the largest capacity and the smallest amount of water per wash resulting in only needing to empty the dirty water tank and refill the clean water once every 5 weeks.
Quality of mop washing is also a consideration and I found that the Narwal Freo, DreameTech L10s, Ecovacs X1 OMNI and T10 OMNI did the best job washing their mops. The Lefant T1 didn’t use enough water per cleaning and sometimes left the mop pads feeling slimy depending on how much of a mess it was cleaning up and the Roborock S7 MaxV Ultra and S8 Pro Ultra had trouble getting to all the sections of the mop using the scrubbing area of the base, and as I mentioned the S7 MaxV Ultra station lacks drying functionality. The Narwal Freo has by far the best base design for keeping the base station clean with an entire removable tray, but the DreameTech L10s also has a cool feature that lets you fill the mop washing area with water by holding the home button, then you can clean out any debris, and when you hold the home button again it will suck all that gross water into the dirty water tank.
The Roomba J7+ Combo doesn’t have mop washing or drying abilities and is actually even worse than that because it keeps it’s mopping pad stowed away which doesn’t even allow it to air dry, and after my 2 weeks of testing there were already signs of mildew growth on the mop pad.
The last consideration for mopping is the ability to raise the mopping pads to travel over carpet. Even if you don’t want to vacuum the carpet in mopping mode, the robot needs to be able to pick up its pads to travel over carpeted areas. The DreameTech L10s, Narwal Freo, Roborock S8 Pro Ultra, and Roborock S7 MaxV Ultra can all raise their mopping pads between 5 and 7 millimeters to avoid getting your carpets wet and dirty, while the Roomba J7+ Combo can completely stow its mop, reducing any chances of cross contamination to zero.
The Ecovacs T10 OMNI, X1 OMNI and Lafant T1 have no ability to raise their mops and will need an uninterrupted stretch of hard flooring to travel from one room to another. The T10 OMNI and X1 OMNI also need to have their mopping pads manually removed in order to switch to vacuum only mode, which definitely reduces their ability to clean autonomously.
Combining all the mopping scores, the top performers specifically for mopping capabilities were the Narwal Freo, DreameTech L10s, and Ecovacs T10 OMNI by a pretty significant margin, with the Freo being the best at mopping, the Dreametech L10s being the most full featured robot that is also good at mopping, and the Ecovacs T10 OMNI providing the most hands free mopping experience as long as your house has a carpet free path between rooms that need to be mopped since the T10 will detect carpet and refuse to pass over it with mops attached.
Speaking of which, map creation and map customization are absolutely critical to being able to use your vacuum the way you want to. LIDAR based robots are significantly better at generating accurate maps than camera-based robots like the Roomba J7+ Combo, but not all LIDAR maps are created equal.
The Roborock and Dreametech apps specifically are a cut above the rest and allow for easy room creation and modification, easy to place no go zones, and lots of options for naming and customization. The Roborock S8 Pro Ultra even allows you to designate the orientation of your flooring planks so that the robot mops along the grain of the wood.
The Ecovacs and Narwal apps work fine, but can occasionally fail to load maps, or inexplicably throw errors when modifying rooms in the floorplan.
The iRobot app performs as well as it possibly could for a robot without LIDAR sensors, but it is absolutely beyond me why iRobot keeps resisting the use of LIDAR when it makes the process of map creation 100x easier and more accurate than camera-based navigation.
What cameras are good for is detecting potential trouble areas and navigating around them. That includes things like shoes left on the floor, cords, and the dreaded pet waste poo-pocolypse.
As I mentioned, the Roomba J7+ Combo, Roborock S7 MaxV Ultra, Ecovacs X1 OMNI, Ecovacs T10 OMNI and DreameTech L10s all have front facing cameras that should enable them to correctly identify and avoid problem areas using computer vision.
To test this, I set up a relatively simple obstacle course with two different colored piles of pet waste, an untied shoe, a charging cable, and some of that plastic easter grass from before. In order to score a perfect 5 points, the robots would need to vacuum the carpet, avoiding the shoe, cords, and pet waste, while still cleaning up the strands of easter grass.
In practice this is an area where these robots need a little more common-sense coding, and I think Roomba is the only one that got it right. Scoring 4 out of a possible 5 points the Roomba was extremely cautious navigating around each obstacle, so much so that it refused to vacuum up the pink easter grass. The Ecovacs T10 OMNI also did a good job identifying and avoiding objects, but in my opinion, it got WAY too close to each one. I personally wouldn’t be upset if the vacuum decided to just stop its job completely if it identified pet waste, but both the Roborock and Ecovacs AI decided they wanted to clean as close as possible to it. The T10 managed to avoid the poop but got caught on the shoelaces which resulted in an eventual poop event, while the Roborock S7 MaxV Ultra, Dreametech L10s, and Ecovacs X1 OMNI all made eventual navigation errors that resulted in complete failure and an eventual poo-pocolypse.
As expected, the robots without camera-based navigation had no ability to see those obstacles and made no attempt to avoid the poop, shoes, or cords.
They do, however, have front facing object avoidance arrays that allow them to avoid smashing into your furniture and baseboards and getting stuck in corners.
One thing I noticed after a year of having two robotic vacuums cleaning every night was that my baseboards were getting pretty beat up. Specifically, there was a common swirling pattern that I could see all around the house, but I couldn’t figure out which robot was responsible for them.
To test this, I repainted the baseboards in my bathroom and had each robot clean a total of 10 times in that confined space where it was likely to get stuck and run into the walls. After each robot completed those 10 runs, I inspected the area and repainted as needed.
Using this method, I found the culprit which appears to be the side brush of the Roborock robots. Both the Roborock S7 MaxV and Roborock S8 left small swirl patterns on my quarter round, which wiped off easily, but if not dealt with could compound with frequent cleaning leading to dirty looking baseboards that need to be repainted.
The Roomba also frequently rammed into walls and furniture, but it must have a specifically designed front section to be able to handle those frequent crashes because I never saw any scuff marks or damage as a result of the impacts.
So, for object avoidance, map generation, and general navigation the Ecovacs T10 OMNI came out on top, but I’d love to see the same cautious behavior of the Roomba J7+ Combo implemented by the rest of the manufacturers, or even an option that says, “If pet feces are detected, stop cleaning immediately”, because it’s just not worth the risk.
Okay, last, and I think this one is really important. Let’s talk about smart home integrations because it is surprisingly useful and convenient to be able to ask your vacuum to clean a specific room when there is an unusual mess. I added each vacuum to Amazon Echo, Google Home, Siri Shortcuts, and Home Assistant to see which functionalities were available on each platform.
The most disappointing thing to me was that the only fully featured integration with Amazon Echo was the Roomba J7+ Combo, probably because Amazon owns iRobot. With the Echo integrations, you can ask your Roomba to vacuum or mop a specific room, but the other vacuums are limited to on/off only meaning there’s no way to clean a specific area or designate mopping or vacuuming mode, and the Narwal Freo unfortunately didn’t have any Echo integration at all.
Thankfully Google Home is a different story and room specific cleaning is available for all the Ecovacs, Roborock, Roomba and Dreametech vacuums that I tested but unfortunately the Lefant T1 was still on/off only, and the Narwal also didn’t have any integration with Google Home.
As for Siri Shortcuts, they were pretty full featured for Roomba, Ecovacs and Roborock vacuums including specific room and area cleaning, but the Narwal Freo and Dreametech L10s had only limited Siri functionality, and the Lefant T1 had no ability to work with Siri or HomeKit at all.
For more advanced automators, the Home Assistant integration for Roborock vacuums works extremely well when using the Roborock app and functions locally, but at this time you can’t manually block your vacuum from the internet without putting it in an endless reconnection loop.
Ecovacs vacuums also work very well via the Deebot 4 Home Assistant HACS integration but rely on a cloud connection and can’t be blocked from the internet.
To my knowledge Roomba is the only advanced robotic vacuum that can function locally in Home Assistant, though the J7+ Combo requires a little more effort than most. But when you integrate your Roomba via MQTT it actually loses connection with the Roomba cloud and only communicates via your local network, which will break all your integrations with Google Home or Amazon Echo, but if you’re looking for strictly local control, you’re probably okay with that.
Okay, conclusion time: That was a lot of information and there isn’t just one clear winner, so let me make some recommendations:
If you’re looking for an all-around great robot that can vacuum every night, do a great job mopping one to two times a week and only needs to be tended to once a month then the DreameTech L10s is probably the right robot for you and my personal choice. I’ll personally be setting up the L10s with three separate schedules: 4 nights a week I’ll have it vacuum the entire house with the mopping pads dry and lifted, then on the other 3 nights I’ll set up two schedules, first it will vacuum the high pile carpet areas, and since the mopping pads should still be completely dry and lifted, there’s no chance of dragging dirty water over the carpet. On those same nights I can set up a second schedule after the carpets are vacuumed and use the cleaning sequence mode to mop all the rooms that don’t have carpet and using this schedule with the capacity of the Dreamebot, I should only need to tend to the clean and dirty water tanks twice a month.
If you have zero carpet, like actually ZERO and you want larger capacity for more frequent mopping, better object detection, and better smart home integrations, then the Ecovacs T10 OMNI is the right pick, and I’m not sure Ecovacs intended it this way, but the more budget friendly T10 OMNI seems to be slightly better than the flagship X1 OMNI from last year.
If you just want the best possible mop, you don’t have pet hair or pet feces issues, and you’re willing to empty a dustbin and refill the water tanks every two weeks, the Narwal Freo is by far the most competent robotic mop and was consistently my go-to for cleaning up my floors after a full day of testing. It’s hard to fully quantify, but something about the feel of the floors after a Narwal cleaning was just better than the rest of the vacuums.
If you’re choosing between the S7 MaxV Ultra and S8 Pro Ultra, the S7 MaxV is better in just about every way, but if mopping is important to you then you should also consider buying the add-on mop drying attachment for the S7 MaxV Ultra base.
As for the Roomba, the J7+ Combo is pretty worthless. Mopping performance was unexciting, even after adding mopping solution manually, the pad has no way to dry after being lifted, and the water reservoir needs to be manually refilled each time you want to mop. The J7+ Combo is $1100 while the J7+ without the mop is only $800 and still does all the things that the J7+ combo does well, but better.
As far as the Lefant T1, it’s not good and it’s really expensive. It’s got a buggy app, it can’t vacuum, and it doesn’t mop very well. I guess if I had to say something nice about it, I could say that it is small, and I kinda like the color… but that’s about it.
I’ve got links for all these vacuums below, and if you appreciate the time, effort, and money that it takes me to make a video like this I’d appreciate if you’d consider using those links since as an Amazon affiliate, I do earn a small commission on the sale at no cost to you.
I’d also like to thank all of my patrons over at Patreon for their continued support of my channel and if you’re interested in supporting my channel, please check out the links below. If you enjoyed this video go ahead and hit that thumbs up button and consider subscribing to my YouTube channel, and as always, thanks for watching The Hook Up.
Top Pick: Dreametech L10s Ultra
Best Mop: Narwal Freo
Best for 100% Hard Flooring: Ecovacs T10 OMNI