0:02 It has been 6 months since my last
0:03 robotic vacuum review, and it seems like
0:05 these companies are releasing new models
0:08 with flagship features every single
0:10 month. So, in this video, I've got six
0:12 new top-of-the-line models from Robbo
0:17 Rock, Narwhal, Dream, MUA, Eovax, and
0:19 3i, and we're going to see if any of
0:20 them are better than the overall winners
0:22 from February, which were the Robbo Rock
0:25 Sorrow 10R and the Dream X50 Ultra. And
0:27 as always, there are no sponsored
0:28 reviews on this channel. First, we'll
0:30 test vacuuming performance on carpet and
0:32 hard flooring. Then, we'll see how well
0:34 their mops work in different situations.
0:36 After that, we'll see how well they deal
0:38 with pet hair and long human hair. We'll
0:39 test their object recognition and
0:42 avoidance, mapping, and ability to clean
0:43 in ways that previous generations of
0:45 robotic vacuums couldn't. And last,
0:47 we'll evaluate their specific apps,
0:48 smart home integrations, and privacy
0:50 features. To test their vacuuming
0:52 performance, I prepared a mixture of 10
0:54 grams each of rice, flax seed husks,
0:57 salt, and flour to simulate differentiz
0:59 dirt and debris. And I ran each vacuum
1:01 on their maximum suction power using a
1:03 vacuum only two pass cleaning. I weighed
1:05 their dust bins before and after each
1:07 cleaning and then again after the auto
1:08 emptying process. And I thoroughly
1:10 vacuumed with a corded upright in
1:12 between tests. And we'll start out with
1:13 the least expensive vacuum in this
1:17 video, the $649 3i P10 Ultra. 3i
1:19 released their super unique S10 flagship
1:22 in 2024 that could pull water out of the
1:24 air to refill its mopping tanks. But the
1:26 P10 doesn't have any of that and it's
1:28 just a normal vacuum and mop with a
1:29 pretty standard set of features. And in
1:32 its 45minute vacuuming run, the 3IIP10
1:35 used 26% of its battery and it collected
1:39 25.6 g or 64% of the flour, salt, flax,
1:41 and rice mixture. And after the auto
1:43 empty process, there was 1.1 g left in
1:45 the bin. but its small air filter was
1:47 almost completely clean. After that is
1:51 the $899 Dream L40S Ultra, which is
1:53 considerably less expensive than their
1:56 flagship X50 and has almost all the same
1:58 features except for the retractable LAR
2:00 turret. And it only has 19,000
2:01 kilopascals of suction instead of
2:04 20,000. And in this test, the L40S Ultra
2:06 cleaned for 30 minutes and used 23% of
2:11 its battery to pick up 25.4 4 g or 63.5%
2:13 of the debris mixture, which is half a%
2:15 less than the 3II, putting it in second
2:17 place. And after the auto empty process,
2:20 the Dream had 4.8 g left in its dust
2:22 bin, and that was mostly caked into the
2:24 air filter. Next, for 1299 is the
2:26 Narwhal Flow, which is their highest
2:28 suction vacuum ever at 22,000
2:30 kilopascals. But in its 44-minute
2:32 vacuuming run on my high pile carpet,
2:34 the Narwhal was only able to pick up
2:38 16.2 2 g or 40.5% of the flour, salt,
2:41 flax, and rice, but it did only use 20%
2:43 of its battery. And after the auto empty
2:45 process, the Narwhal had 0.6 g left in
2:47 its bin with a surprisingly clean air
2:49 filter, indicating that its auto empty
2:51 system is very effective and this vacuum
2:53 should be less likely to lose suction
2:55 over time. After that, for $13.99 is the
2:58 MUA V50 Ultra Complete. And MUVA is
3:00 another newer company that debuted at
3:02 CES 2025, but it's definitely tied
3:04 directly to Dream. and you'll see that
3:06 they use a lot of the same technology
3:07 and have a lot of shared features. The
3:10 V50 advertises a massive 24,000
3:12 kilopascals of suction, which helped it
3:16 pick up 26.2 g or 65.5% of the debris
3:18 mixture in its 33minute vacuuming run,
3:20 putting it in first place. But all that
3:23 suction took a lot of power, using 44%
3:25 of its battery. And after the auto empty
3:27 process, the MUA still had a concerning
3:30 8.9 g of debris left in its bin, which
3:32 similar to the Dream was heavily caked
3:34 into the air filter. Next for $14.99 is
3:38 the brand new Eavax X11 Omnicyclone. And
3:40 I reviewed the Eavax X8 when it came out
3:42 just 6 months ago. And since then,
3:44 they've released the X9, skipped over
3:46 the X10, and now we're at the X11. And
3:49 the X11 Omnicyclone has 19,500
3:52 kilopascals of suction, which is 1,500
3:55 more than the X8. But the X11's carpet
3:57 test was kind of a disaster. first
3:59 getting stuck by wedging itself under my
4:01 couch, then telling me that it completed
4:02 its entire vacuuming task after only
4:04 doing a perimeter clean. And then when I
4:06 finally got it to do a full cleaning
4:08 run, it got stuck under the couch two
4:10 more times. And to top it off, in 36
4:12 minutes of cleaning, the X11 Omnicyclone
4:16 picked up just 20.5 g or 51.25% of the
4:18 debris mixture, putting it in second to
4:20 last place in this group, which is
4:22 especially surprising because the X8 had
4:23 the second highest carpet score that I
4:25 have ever recorded. However, the new
4:27 bagless auto empty system on the
4:29 Omnicyclyone was very effective, and the
4:31 bin and filter were relatively clean
4:33 with only one gram of leftover flower
4:37 dust. And last, for a staggering $2,599,
4:38 we've got the most technologically
4:41 advanced robot vacuum on the market that
4:43 vacuums, mops, and picks up after you,
4:47 the Robot Soros Z70. The Z70's 22,000
4:49 kilopascals of suction matches the
4:52 lowerpriced Soros 10R, but the Z70 has a
4:54 completely different brush roller system
4:56 that uses recessed blades to avoid hair
4:57 tangles instead of the popular and
4:59 effective Duo Divide brush that we find
5:02 on the Soros 10R. And in its 31minute
5:05 vacuuming run, the Z70 picked up 23.7 g
5:07 or 59.25%
5:09 of the mixture, putting it in fourth
5:11 place. And after the auto empty process,
5:13 the Z70s small dust bins still had 2.7
5:15 gram of debris in it, which was almost
5:17 all caked into the air filter. And that
5:19 means that the Mua V50 Complete had the
5:21 best score for this set of vacuums at 65.5%.
5:22 65.5%.
5:24 And the 3IP10 Ultra was close behind
5:26 that. But they were all significantly
5:29 worse than the X8 Pro Omni, Soros 10R,
5:31 and X50 Ultra from my February review.
5:33 Next, repeating that same vacuuming test
5:36 on my LVP hard flooring again on maximum
5:38 suction two pass vacuum only runs. All
5:40 the vacuums performed significantly
5:42 better with the Robbo Rock Saros Z70
5:45 recording a perfect 100% score. The MUA
5:48 came in second at 98.5%. The Dream had 97.5%,
5:50 97.5%,
5:53 the 3i had 96% and the Narwhal and Eavax
5:55 filled out the bottom two positions at
5:58 95.5 and 94.5% respectively. So that
6:00 means for combined vacuuming
6:01 performance, the overall top performer
6:05 was the MUA V50 Ultra Complete with 65.6
6:07 g picked up and the Dream was in second
6:09 with 64.4. 4. But as I said, those
6:10 numbers still don't really come close to
6:14 the Soros 10R, Dream X50 Ultra, or Eavax
6:16 X8 Pro Omni from earlier this year. In
6:17 this video, I also measured their corner
6:19 brush performance by putting a teaspoon
6:21 of dried breadcrumbs in four separate
6:23 locations in my kitchen and sending the
6:25 robots out on a single pass vacuuming
6:26 run. And to score them, I gave the
6:28 vacuums two points for cleaning up a
6:30 pile completely and one point for a
6:32 partial clean. And these results were
6:33 really strange because the top two
6:36 scores were the Robbo Rock and the 3II,
6:38 scoring seven points each, meaning they
6:39 completely cleaned all the piles except
6:41 for under the refrigerator. And the odd
6:43 part about that is that the Dream, MUA,
6:46 and Robok all have fancy articulating
6:48 side brushes that are specifically for
6:50 reaching into corners. And the Narwhal
6:52 has a new hinged side brush that extends
6:53 when it's spun in reverse to be able to
6:56 reach into corners. But the 3i just has
6:58 a fairly traditional side brush design,
7:00 and it somehow still ended up tied for
7:02 first place. Similarly, odd, Eovax has
7:04 switched to a front-facing lidar system
7:06 to avoid that top-mounted lidar bump.
7:07 And as a result, it should be able to go
7:09 under more furniture, but it decided for
7:11 some reason that it couldn't fit under
7:13 my stools, and it didn't even attempt to
7:14 clean that pile, even though it would
7:16 have had plenty of clearance. And last,
7:17 for vacuuming, I measured the noise
7:19 level at their highest suction power.
7:21 And here they are from the quietest to
7:41 Moving on to mopping performance. I like
7:42 to simulate a realistic test in my
7:44 house, which is when wet mud gets
7:46 tracked in on shoes and feet and then
7:48 dries on the floor. So to do that, I
7:50 spread 10 milliliters of Florida mud
7:52 over a twoft x two foot square. And then
7:54 I dried it with a haird dryer. Each
7:56 vacuum then gets set to maximum water
7:58 flow in two passes for a combined
8:00 mopping and vacuuming run. And after
8:02 each run, I spray the area with two
8:03 sprays of water and wipe it clean with a
8:05 paper towel. And then I set the paper
8:07 towel aside to air dry. And in between
8:09 each test, I mop the floor with a
8:11 traditional wet mop. And I hand dry it
8:13 with a towel. And this year I had a few
8:15 anomalies in this test. First, the MUA
8:17 V50 Ultra Complete is the first robot
8:18 that I've ever seen to decide that the
8:21 mud was too messy and chose to avoid a
8:22 large portion of it on its first run.
8:24 But it later decided that it was going
8:25 to clean the stain and it didn't do much
8:27 better in its second attempt. And the
8:29 Eavax also had some similarly strange
8:30 behavior, cleaning erratically on its
8:32 first pass, but then doing a completely
8:34 normal second pass. And in this group of
8:36 robots, there was really only one
8:38 standout mopping performance, which was
8:40 from the Narwhal Flow that not only has
8:42 the new roller style mopping system with
8:44 an onboard clean and dirty water tank,
8:46 but it also has a double wide roller.
8:48 And this is the first time that I have
8:50 ever had a completely clean paper towel
8:52 test. And I was so impressed with it
8:54 that I reran the test a second time just
8:55 to see if it could replicate those
8:57 results. And they were both equally
8:58 perfect. And maybe the most surprising
9:00 part about that is that all the other
9:02 vacuums have automatic detergent
9:03 dispensers that are supposed to help
9:05 with tough stains. But the Narwhal
9:06 requires you to add detergent directly
9:08 to the clean water tank if you want to
9:10 use it. And for these tests, I was just
9:11 using plain water with the Narwhal,
9:12 while the rest of the mops were using
9:14 their mopping solutions. A lot of these
9:16 models also make vague statements about
9:18 mopping with warm or hot water. So, I
9:20 recorded each mopping session with a
9:21 thermal camera. And while some of the
9:23 robots themselves were clearly warmer
9:25 than others, there wasn't any evidence
9:26 that I could see that the floors were
9:27 being mopped with significantly warmer
9:29 water than room temperature. However,
9:31 the vacuums do use heated water to clean
9:33 their mopping pads at the base, and they
9:34 dry their mops with hot air. And all of
9:36 that uses power. So, I measured the
9:38 energy required for each robot to mop a
9:41 100 ft space, wash and completely dry
9:43 their mop. And the Narwhal used the most
9:46 energy at 0.412 kwatt hours. The Eavax
9:48 and Robot Rock were around 0.3 kwatt
9:50 hours and the rest of the robots were
9:53 under 0.2 kwatt hours per mopping run.
9:54 All these robots also have the ability
9:56 to extend their mops to clean under
9:58 cabinets and close to the wall. So, I
9:59 tested those abilities by putting a
10:01 spear of hot sauce in five separate
10:03 locations in my kitchen and scoring each
10:05 robot's mopping performance by awarding
10:07 one point for a partially clean spot and
10:09 two points for a completely clean one.
10:11 And coming out on top again, the Narwhal
10:13 Flow used its extra- wide extending
10:15 roller brush to carefully hug the
10:17 corners and reach under the stool,
10:18 completely cleaning three of the spots
10:20 and mostly cleaning the other two. The
10:22 Dream also did very well, and it had a
10:24 new corner cleaning algorithm that
10:26 stops, rotates, and extends the mop for
10:28 maximum coverage. The Robbo Rock had
10:29 decent coverage, but struggled to absorb
10:31 the hot sauce on its mopping pads, and
10:33 it left a lot of streaks and smears. The
10:35 3II started out pretty well and then it
10:37 failed to clean under the corner cabinet
10:39 and it also missed the outside corner by
10:41 the stove. The Eavax kind of just looked
10:42 like it was cleaning randomly and it
10:44 didn't really make any effort to hug the
10:46 walls or corners. And the MUA had the
10:48 worst corner mopping score with a
10:50 combination of poor edge detection like
10:52 the 3II and poor mop pad absorption like
10:54 the Robbo Rock. And for the last mopping
10:56 test, I looked at whether the vacuums
10:58 could mop and vacuum a room with both
11:00 carpet and hard floors without getting
11:01 the carpet wet. And I found that despite
11:03 them being able to lift their mopping
11:06 pads, the MUA, Narwhal, and ThreeI all
11:08 got my high power carpet wet when doing
11:10 a combined vacuum and mopping run. The
11:12 Dream and Robarok were able to keep the
11:14 carpet completely dry by recognizing
11:15 that the room had both carpet and hard
11:17 flooring and then detaching their
11:19 mopping pads at the station to vacuum
11:20 the carpet first. And after that was
11:22 done, they returned to their base to
11:23 grab their mopping pads before cleaning
11:25 the hard floors. And oddly, the MUA also
11:27 has the ability to leave its mopping
11:28 pads at the station, but it isn't
11:30 programmed the same way that the Dream
11:32 and Robot Rock are. And it only detaches
11:34 its mopping pads when it's doing a
11:35 vacuum only run. And the biggest
11:37 surprise during this test was the Eavax,
11:39 which doesn't have detachable mopping
11:41 pads, but it somehow managed to mop the
11:43 floor and then travel directly onto the
11:44 carpet without getting the carpet wet.
11:46 So, I'm not sure if it can just lift its
11:48 mopping pad that high or if it's just
11:49 scraping all the moisture out of the mop
11:51 before vacuuming the carpet. But either
11:53 way, it was pretty cool to see. And that
11:54 means that overall for mopping, the
11:56 Narwhal was in a league of its own with
11:58 by far the best hard floor mopping
12:00 performance and the best corner mopping
12:01 with its main weakness being the
12:03 inability to cross high pile carpet
12:04 without getting it wet. However, while
12:06 roller style mops like the Narwhal Flow
12:08 and the Eavax X11 are significantly
12:10 better at cleaning up big messes, they
12:12 also require a bit more maintenance.
12:13 Specifically, you'll need to clean the
12:15 onboard reservoir where it constantly
12:16 scrapes dirty water off the mopping
12:18 roller with every revolution. And while
12:20 you can get away with just throwing
12:21 spinning mopping pads in the washing
12:23 machine, you will need to actually get
12:24 in and brush out the reservoir on the
12:26 roller mops about once a month to keep
12:28 them draining well. Another big problem
12:30 that these vacuuming brands have been
12:32 trying to solve is hair pickup and
12:34 tangle avoidance. And in this group,
12:35 there are four different potential
12:38 solutions. First, the Dream, MUA, and
12:40 Narwhal are all using tapered brush
12:42 rollers that are only connected to the
12:44 vacuum on one side. So any hair that
12:45 wraps around the roller will eventually
12:47 travel down the brush and into the
12:49 suction port. The Robot Rock C70 is
12:51 using recessed blades on two sides of
12:53 the rollers that have the function of
12:55 pulling in any tangled hair and cutting
12:56 it in half so that it can be more easily
12:59 collected in the bin. The Eavax uses a
13:00 detangling combined with their
13:03 anti-tangle brush roller. And the 3IIP10
13:05 Ultra actually has a blade on its base
13:07 that runs across the brush roller before
13:09 the auto empty process to hopefully cut
13:11 any tangled hair free. But the question
13:13 is, which of these solutions actually
13:15 works? And for the first test, I started
13:17 with a very realistic task of cleaning
13:19 up 30 individual strands of real 12-in
13:21 human hair from my bathroom floor. And
13:23 then after the auto empty process, I
13:25 carefully inspected each robot for any
13:27 hair tangles. And in this test, most of
13:29 the vacuums performed very well, with
13:31 the Narwhal coming in first with not a
13:33 single piece of hair tangled anywhere on
13:35 the robot. The Dream and Robbo Rock each
13:37 had one stray piece of hair tangled up
13:39 in their wheel axle. The Eavax had two
13:41 strands caught up in its wheels. The 3II
13:42 had a bit more than that all tangled up
13:44 in the brush roller itself. And the Mua
13:46 was mostly tangle-free, but it had an
13:48 issue with its auto empty system where
13:49 clumps of hair would get caught in the
13:51 back flap and then dragged around until
13:53 they fell off at random places in my
13:55 house. I also ran a second less
13:57 reasonable test, and I spread out 1 g of
14:00 synthetic 16-in blue hair and two cotton
14:02 balls separated into toughs to simulate
14:04 pet fluff on my high pile carpet. And
14:06 then I ran each robot on a one pass
14:08 vacuum only run. Afterwards, I collected
14:10 any unclean pet hair toughs and any
14:11 leftover synthetic hairs on the carpet.
14:13 And I inspected each robot for tangles
14:16 and clogs. And as expected, this test
14:18 was much more extreme. And starting with
14:20 leftover hair and fluff on the carpet,
14:21 the robot rock basically didn't clean
14:23 any of the pet fluff, and it just matted
14:25 it into the carpet. And there was also a
14:26 significant amount of synthetic hair
14:28 left over. The Dream also struggled to
14:30 pick up the fluff on the carpet, but I
14:31 think that it was mostly because it saw
14:33 them as obstacles that it needed to
14:35 avoid. And the 3II also had some erratic
14:36 behavior around those fluff balls. The
14:38 Eavax cleaned in a normal pattern, but
14:40 it still left quite a bit of pet fluff
14:42 and even more synthetic hair. The MUA
14:44 did a very respectable job with just a
14:46 couple leftover pieces, but the Narwhal
14:48 did a truly awesome job, and it cleaned
14:50 up 100% of the hair and fluff on the
14:52 carpet. And even more incredibly, the
14:54 Narwhal also managed to almost
14:56 completely avoid any tangles with just a
14:58 single strand of blue synthetic hair
15:00 stuck in between the rubber and the
15:01 plastic of the wheel, which as I said is
15:03 a ridiculously amazing performance on
15:05 its own and the best that any robot has
15:07 ever performed in this test. But it's
15:09 even more amazing considering how much
15:11 hair and fluff was on the other vacuums.
15:13 And overall, the tapered brushes seem to
15:14 perform significantly better than any of
15:16 the other methods of preventing hair
15:17 tangles. And the Robbo Rock cutting
15:20 brush was completely matted. The Eavax
15:21 brush had a large tangle in the center
15:23 of the roller. And the three-eyed hair
15:25 cutting system wasn't effective in this
15:26 test. And the brush roller was still
15:28 completely tangled. And here's an
15:30 overview of the amount of hair and fluff
15:32 each robot left on the ground. And these
15:33 piles represent what was tangled on the
15:35 robot itself. And that means that twice
15:37 now in this video, object avoidance has
15:39 stopped these robots from cleaning my
15:41 testing area completely. But now it's
15:42 time to see how well it works when it's
15:44 supposed to. So I set up my living room
15:46 with a pair of untied shoes, some rubber
15:48 dog poop, and a loose extension cord.
15:50 And I sent each vacuum on a one pass
15:52 vacuum-only run. And to score their
15:54 object avoidance, I gave them two points
15:56 for avoiding an object and one point for
15:58 moving it with no catastrophic outcomes
16:00 like getting stuck or running over dog
16:02 poop. And the only perfect score of six
16:04 came from the Movea V50, which kept
16:06 appropriate distances from each object,
16:08 meaning cleaning right up against the
16:10 extension cord and shoes, but keeping a
16:12 safe distance from the dog poop. The 3i
16:14 P10 Ultra earned five points, avoiding
16:16 the dog poop and shoes, but tangling
16:18 itself up on the cord, pulling it
16:19 several feet away before breaking itself
16:22 free. The Robbo Rock Saros Z70 also
16:23 scored five points, again, avoiding the
16:25 dog poop and shoes, but moving the
16:27 extension cord several times. The Eavax
16:29 scored four points, successfully
16:30 avoiding everything on the first part of
16:32 his cleaning task and then running
16:34 directly over the dog poop. And the
16:35 Dream knew that it shouldn't run over
16:37 the dog poop, but it got a little too
16:38 close with its side brush, ultimately
16:40 still scoring four points, but it would
16:42 have been a total disaster. And coming
16:44 in last with just three points total,
16:46 the Narwhal first got itself snagged on
16:47 the cord and then it completely
16:49 disregarded the dog poop, running it
16:51 over multiple times, and it really made
16:52 no effort to avoid it. Also, in this
16:54 test, we can talk about the main selling
16:56 point of the Robot Rock Z70, which is
16:58 the fact that it is the first robotic
17:00 vacuum to also have a five-axis
17:02 mechanical arm that can recognize and
17:03 move objects out of the way to clean
17:05 under them, and it can even identify and
17:07 sort objects into different locations.
17:10 And my impressions of the Z70s arm are
17:12 mixed. On the one hand, I love to see
17:14 new innovation enabling these vacuums to
17:16 become even more useful than they
17:17 already are. But on the other hand, the
17:19 advertised use cases like picking up
17:21 used tissues from the floor and throwing
17:23 them away and finding socks on the
17:25 ground and sorting them into a separate
17:27 bin really seem like personal hygiene
17:28 problems that shouldn't need a
17:30 state-of-the-art robotic vacuum to
17:32 solve. And the SRO Z70 seems like a huge
17:34 amount of technology to counteract what
17:35 could be a pretty simple lifestyle
17:37 change. And about the only real
17:39 application that I can think of is if
17:40 you had a dog that leaves stuffed
17:42 animals around, the SRO Z70 could put
17:44 them all away in some central location
17:46 before cleaning. But if it's your kids
17:48 leaving their toys around, the Soros Z70
17:49 is less likely to be able to pick up
17:51 non- stuffed animal toys. And as a
17:53 father myself, my personal opinion is
17:54 that teaching your kids to put away
17:56 their own toys is probably better than
17:58 relying on your vacuum to do it. That
17:59 said, these vacuums do have a couple
18:01 flagship features that I do find very
18:03 useful. And the first is being able to
18:05 climb higher than normal thresholds
18:07 between rooms. And to test this, I built
18:09 three different size thresholds at 13
18:13 mm, 30 mm, and 50 mm. And each of the
18:15 thresholds has both a ramp side and a
18:16 flat side. And in this test, all the
18:18 vacuums were ultimately able to cross
18:21 the 13mm threshold, but the Narwhal and
18:22 especially the Dream struggled quite a
18:24 bit more than I was expecting on the
18:26 flat side. At 30 mm, the Narwhal Flow
18:28 and Move V50 crossed both the ramp side
18:30 and the flat side with no issue. And the
18:33 Robbo Rock Z70 eventually crossed the
18:34 flat side after a little bit of
18:37 struggle. But the 3II, Dream, and Eavax
18:39 all failed to pass back over the 30mm
18:42 flat threshold. At 50 mm, the Narwhal,
18:43 Dream, and Eavax were able to cross the
18:45 ramp side, but not the flat side. The
18:47 Robbo Rock and ThreeI weren't able to
18:49 cross in either direction, and the Move
18:51 V50 was the only robot that was able to
18:53 cross in both directions using the same
18:55 technique and hardware as the Dream X50
18:57 Ultra from my last video. So, if you
18:58 have an abnormally large threshold in
19:00 your house, the Move V50 or Dream X50
19:02 Ultra are the most likely to be able to
19:03 handle it. The way that the robots
19:05 handle mapping and navigation is also
19:07 evolving quite a bit. And the Dream and
19:09 ThreeI are the only ones that still have
19:11 a traditional top-mounted LAR unit,
19:13 which in theory should give the other
19:15 robots a clearance advantage. But in
19:17 reality, the Dream L40S Ultra has a
19:20 total height of 95 mm, including its LAR
19:22 bump, while the Narwhal Flow is 95 mm
19:25 without a LAR bump. And the Eavax X11 is
19:28 3 mm taller than that at 98 mm. And
19:29 unfortunately, in my testing, switching
19:33 from 360° LAR to front or rear only
19:35 ultimately decreases mapping and
19:36 navigation efficiency, and it also seems
19:38 to affect the robot's ability to tell
19:40 which furniture items it can and cannot
19:42 fit under. And neither the MUA nor the
19:44 Eavax decided they would be able to fit
19:45 under my kitchen stools that have plenty
19:47 of clearance, but at the same time, the
19:49 Eavax continually wedged itself under my
19:51 couch until I set up a no-go zone. And
19:53 when it comes to navigation, clearance,
19:54 and correctly determining furniture
19:57 height, Robbo Rock's Starite navigation
19:59 is in a league of its own. And it seems
20:01 to know exactly where it can and can't
20:03 clean. And the fact that it's only 79 mm
20:05 tall means that it has an additional 16
20:07 mm of clearance over the Dream and
20:09 Narwhal, making it by far the best
20:11 option for cleaning under furniture. And
20:12 the last things we need to cover in this
20:14 video are their apps, smart home
20:16 integrations, privacy, and local
20:18 control. And while the apps from 3II and
20:20 Narwhal are definitely usable, MUA,
20:21 Dream, and Robbo Rock have nearly
20:23 perfected their apps with a delicate
20:25 balance of ease of use and customizable
20:27 features. And for me, the Eavax app
20:29 falls somewhere in the middle of that
20:30 with basically all the same features as
20:32 Dream, MUA, and Robbo Rock, but it's
20:34 just not quite as easy or intuitive to
20:36 use. I also really like to be able to
20:37 control my robotic vacuum with voice
20:39 assistance to tell it to clean up a
20:40 specific room without getting out my
20:42 phone. And while all the vacuums support
20:45 Amazon Echo integrations, the 3i doesn't
20:47 support room specific cleaning. And for
20:48 the rest, you just have to know the
20:50 right thing to say. I also tested Google
20:51 Home integrations and the only one that
20:53 I wasn't able to get working at all was
20:55 the Narwhal. And the robot rock, Dream,
20:57 MUA, and Eavax all support room specific
20:59 cleaning commands, while the 3i only
21:01 supports on andoff commands. If you are
21:03 a home assistant user, the robot rock
21:05 Z70 is the only one that gives what I
21:07 would call a fullfeatured integration,
21:09 and the built-in EVAX integration
21:11 doesn't support the X11. Dream's custom
21:13 integration doesn't support the L40S
21:15 Ultra, and the 3i and Narwhal don't have
21:17 integrations at all, but for the first
21:19 time in a very long time, I'm happy to
21:21 report that we do have some form of
21:23 local control by way of a matter
21:25 integration for both EVAX and Robbo
21:27 Rock. And I did check and this is a
21:29 completely local connection that allows
21:30 for some minimal amount of customization
21:32 and cleaning even when the vacuum is
21:34 completely blocked from the internet.
21:36 And impressively, the robot rock even
21:37 maintained all of its robotic arm
21:39 functionality without an active internet
21:41 connection, which means that all the
21:42 image processing and recognition is
21:44 being handled on the robot itself. But
21:46 for both the Eavax and Robbo Rock, you
21:48 will need to connect at least once
21:49 through the app to set up your map and
21:51 your preferences. So, looking at the
21:53 total scores for each vacuum, none of
21:55 them stand out as clear winners like in
21:56 my previous video, with four vacuums
21:58 coming in with an average score around
22:00 three. The MUA V50 Ultra Complete
22:03 finished in first with an average of 2.9
22:04 thanks to its object avoidance and
22:06 threshold crossing abilities. But at
22:09 $1399, that's $200 more than the Dream
22:11 X50 Ultra, which has the same threshold
22:14 crossing, the same object avoidance, and
22:16 the same mopping, but better overall
22:17 vacuuming scores. And the same is true
22:20 for the Robox C70, which had an average
22:22 score of 3.2. But if you don't need the
22:24 robotic arm, which is a cool concept,
22:26 but ultimately not ready for prime time,
22:28 then the robot rock 10R has all the
22:31 great parts of the Sorrow Z70, like low
22:33 furniture clearance, best-in-class
22:34 navigation, and excellent smart home
22:36 integrations, including matter support
22:38 for less money. And as a bonus, it also
22:40 had better vacuuming scores and much
22:42 better hair tangle avoidance thanks to
22:44 Robok's Duo Divide brush system. And 6
22:45 months later, I still think that the
22:48 Robok SORS 10R is the best all-around
22:50 vacuum and mop on the market right now.
22:52 However, as I mentioned earlier, the
22:54 Narwhal Flow is the best mopping robot
22:56 that I have ever tested by a fairly
22:57 significant margin. And even though its
23:00 overall score was just average at 3.0,
23:02 if your main goal is mopping and you
23:03 don't have high pile carpet that needs
23:05 to be vacuumed, then the Narwhal Flow is
23:07 an easy choice. And the fact that it has
23:09 great hair pickup and tangle avoidance
23:11 is just an added bonus. I also think
23:13 that the Dream L40S Ultra is worth
23:15 considering since it's basically just a
23:16 slightly more budget friendly version of
23:18 the X50 Ultra. So, if you don't need
23:20 things like the ultra high threshold
23:22 crossing ability or the first person
23:24 remote control, the L40S Ultra gets you
23:26 a very similar performance to the X50
23:28 for quite a bit less money. As always,
23:30 there are no sponsored reviews on this
23:32 channel, but I do have links to all the
23:33 vacuums down in the description. And as
23:35 always, I appreciate when you use those
23:37 links since as an Amazon affiliate, I do
23:38 earn a small commission on the sale at
23:40 no cost to you. I also want to thank all
23:42 of my awesome patrons over at Patreon
23:43 for their continued support of my
23:44 channel. And if you're interested in
23:45 supporting the channel, please check out
23:47 the links down in the description. If
23:48 you enjoy this video, please consider
23:50 subscribing. And as always, thanks for
Chrome Extension