Roomba / iRobot is years behind the Chinese companies. We just got an Ecovacs X1 Omni - a mopping vacuum bot with actual rotating mop scrubbers, AI vision obstacle avoidance, LIDAR mapping, and most importantly fully automated water fill & mop pad cleaning at the base station. We fill a clean water tank & empty the dirty one about once a week. Floors are spotless - huge difference from static cloth mopping bots.
Meanwhile, iRobot still hadn’t figured out how to make an effective stand-alone mop bot, much less an auto fill station or combined bot that actually works. Roborock and Ecovacs products launch in China 6-12 months before their US launch. By the time Roomba has a decent mop bot, the X2 or X3 will probably be out offering integration with home plumbing to avoid any manual intervention for months at a time.
EDIT: commenters are asking what works offline from Roborock / Ecovacs. For the self-cleaning-mop generation like the X1 Omni:
- Roborock obstacle avoidance is local, but works better with cloud.
- Roborock can be controlled by a local home automation server without internet access.
- Ecovacs obstacle avoidance requires cloud.
- Ecovacs control requires cloud.
I remembered these stats from a Youtube video that I can't find; probably in my partner's history since we watched reviews on the TV together.
I picked the X1 Omni over the Roborock because it has superior mopping performance on tile floors, but I think both would be good options. It seems like Roborock is more privacy compatible.
Yeah I went from a Roomba 960 (2018) to a Xiaomi Roborock S7 and in my research it was eye opening to see the level of tech difference between the two, even the modern Roomba is using a vision only model, and poorly based on reviews. For the same price, Roborock has Lidar scanning, AI obstacle avoidance and mopping. The thing is, it actually works in the Roborock, it can even spot cords and other misc items around my house and avoid them entirely.
Additionally since it maps using Lidar it completes a complete cleaning of my living room + kitchen area in 22 minutes, whole house in under 40 minutes. That is with it covering literally every square inch of carpet & tile instead of going around in random patterns...
I have an S7 and it’s been ridiculously stupid. My old Roomba was less aggravating. The S7 takes 50 minutes to clean 800sqft, always gets stuck in a high rug or chair support. Frequently gets tangled on cables and toys. I can clean tht floor in 10 min with a stick vacuum. Anyone wanna buy an old S7? Barely used, frequently kicked.
The only annoying Xiaomi moments are when it recognizes a sock as dog poo then refuses to do anything in that area (or I couldn't figure out how to reset it). Ah, and it constantly gets stuck on bar stools pedestals.
I've owned Neato, Roomba and Roborock. Hands down Roborock is what I recommend people to buy. Neato had such a great leg up when it first came to market by having the first, and best, Lidar implementation - but their product was marred by horrible customer support, batteries that are garbage and just general failures of hardware that are impossible / too expensive to fix. By the time I gave up on Neato I had 4 of the same robot that I was piece parting out as components died (again and again). Talk about a company that banks on eWaste to keep the sales cycle going.
I'm thankful Amazon bought Roomba - one less reason to even consider them as an option considering all of the current gen vacs are sporting onboard cameras + CV (poop avoidance, is that really a huge concern?).
I've had great experiences with the Xiaomi products I've bought (34" monitor, earbuds, security cam, air purifier, soundbar). They are popular here in Thailand and at a really, really great price point
The Roomba j7+ cord and object detection is working well for me, although it's (probably) more expensive than the Chinese offerings. Regardless, it was a huge step up from my precis Neato Botvac connected (which was probably 5-6 years old though).
Wow, those both look incredibly slick. I've got a cheap like $150 chinese vacuum one and a mopping one but I might have to upgrade if they ever die. So do they actually handle floor vs carpet sweeping/mopping, don't get stuck on stuff on the floor, refill/empty, and you don't have to change the mop pads each time?
That is wild as that basically handles all the prep I have to do for mine. Albeit, 1000-1500 is quite a bit more.
Absolutely agree. The lack of LiDAR (or equiv) seems to really hamper iRobot's effectiveness - I had an iRobot i7 and its mapping/pathfinding was ramming in to things at near top speed and ricocheting off the walls.
They also still sell vacuum and mopping* robots as seperate units which just drives up the cost. My old setup of the two robots took somewhere on the order of 5-6 hours to do a ~100m2 apartment.
I moved to a Roborock and then Dreame bot (someone really needs to talk to Xiaomi about all the sub-brands :P) comparitively, they both took around 1.5 to just under 2 hours to do a much better job.
I actually felt huge guilt selling my two iRobot's to someone, wanting to tell them they were complete junk and they were wasting their money.
*There is always mixed opinions on mopping effectiveness, but the iRobot roll-and-spit model of mopping is truly horrendous.
Based on the comments here I did a quick look at Amazon and was surprised to not even see a single vacuum only robot from Evovacs. At least not prominently showcased.
Why are there only mixed models and what are the benefits?
If I look at my place the ground floor is tiled and the top floor is all carpet.
That means I would need two vacuums anyway as they haven’t learned yet to climb stairs.
Why wouldn’t I want a vacuum only model for upstairs?
For downstairs the universal model is great obviously. But upstairs?
I don't understand this, I had a Neato before I switched to iRobot and the lidar just seemed to be the worst idea for a home environment.
On an almost weekly basis it would fail to start because it would think it was moved if something just moved a little bit nearby or a cat just happened to be sleeping nearby. This generally forced me to remap
Then it would constantly get lost and be confused about what room it was in. Seemingly for the same reasons of things moving around.
A home isn't a static environment and a camera just seems much more efficient. Not that a camera is perfect, but at least a camera can account for things moving around were lidar cannot.
Or was this a case of Neato just not being good? The lidar issues was why I went with iRobot. It was a worse vacuum but if it actually works consistently than I can just have it run more often.
Once my i7 has generated a proper map, it will will slow down when it approaches known walls, or even places where obstacles have been several times in the past. It will need to bump into the obsticle or wall to confirm its position though.
It does use a mostly predictable back and forth pattern, taking note of areas with walls/obstacles, and after the main sweeping it will combe back to go around the obstacles, and finally (optional) do a sweep of the walls. Then it moves onto the next room.
The system does have limitations. For example, if it needs assistance in the middle of a clean job, after you fix it, there is a reasonable chance that it will do dumb stuff afterwards, like missing a large section of floor. (It seems to even know this, but have problematic coding that prevents it from doing the right thing).
But yeah, I'd not call the mapping good, just acceptable.
There are other annoyances like the i7 battery being so small, especially since they have a giant battery compartment. Like seriously big enough like ni-cad batteries battery of equivalent capacity. But they only offer a slightly larger capacity battery as part of their Costco exclusive model, and try to lock out third party batteries. The net result is that even when brand new, it could not quite do a single floor of my house in one go, needing to go recharge for like 40 minutes before it could do the last 10 minutes of cleaning.
The app is not very good. It does not even do some basic things like prompt the user for regular maintenance periodically based on number of hours used.
The floor maps in the app are weird. For example, on one of my floors, the online map decided that the entrance to one of my rooms was on a completely different wall from where the entrance really was. It thought the actual door was just a wall. However the actually mapping data on the robot was just fine. It knew where the door really was, and never tried to drive though the wall where the online map though the door was.
You are absolutely right, but I’m not sure I want to have a free roaming camera and LiDAR sensor from a Chinese company in my home. Absolute piss poor track record of data collection, I’ll happily stick to my roomba. It’s definitely semi dumb but it’s predictable in its quirks. It’s also repairable with copious supply of original parts. Note also that they still innovate, including the self empty bin that actually works, which others have now copied. Just sayin’.
This is exactly my concern. Imagine a DoD contractor with a remote-controllable, lidar+camera (+mic?) equipped robot based in the USA's main rival country. Does that seem like a good idea? Should we put a fleet of Chinese robots in SpaceX's facility? Lockheed's? Why not?
It's not just "Murica first". If these were European companies, I'd jump ship very quickly to an iRobot competitor with a superior product, thanks to EU's data collection laws and USA's OK relationship with EU. But they aren't, they operate in a regime where questionable data collection is the norm and theft of IP is called "R&D".
Yes, Amazon will use iRobot to vacuum more data than dirt, but at least they are part of the US and EU laws. They provide gov services for a ton of US DoD/NASA infra! China could and would just thumb their nose at us without much concern.
> You are absolutely right, but I’m not sure I want to have a free roaming camera and LiDAR sensor from a Chinese company in my home. Absolute piss poor track record of data collection
I am not sure if Amazon is more trustworthy than that.
I don't really want continuous surveillance of the interior of my home by any company from any nation. There should be some places where you get to just exist without being monitored.
But what is the difference, is the track record of a democracy less grim with your privacy? " Amazon admitted its Ring security cameras have sent recordings to police without the knowledge or consent of the people who own the cameras.
Responding to an inquiry from Sen. Ed Markey (D-Mass.), Amazon said in a letter dated July 1 that it has handed over private recordings to police 11 times in 2022. The company said it was complying with an "emergency request."
I guess is better to have a device that will share your data with domestic entities that actually spy on you and harm your privacy. If someone
"has to" own my data, I prefer that someone to be in another jurisdiction.
Is there a tangible difference between chinese and american companies when it comes to privacy? It seems either will sell your data to the highest bidder.
How easy are the Chinese models to repair? I can't speak for iRobot's mop bots, but the older model vacuums were insanely easy to repair. I had a motor die on one, and it was about ten phillips head screws to get to it, remove it, and pop in a $10 replacement. I was amazed to see a modern consumer item that was so easy to repair.
Of course, this is a model I bought used almost ten years ago. Not sure how similar current models are.
I can only speak about the Roborock S5, but I assume the newer ones are fairly comparable.
I once had it give me an error that the main fan (for suction) was broken. In this case, there was a small particle lunged into it, so it wouldn't spin. All it took was a little wiggle, and it spun again.
It's definitely more than 10 phillips head screws, but less than 30. I guess it depends which part broke, but I had to disassemble pretty much the whole thing. I can promise you that my repair skills or experiences are very novice, but I managed just fine. Had two screws left over after putting it all back together as is tradition, but I suppose they weren't mission critical, as this was like 3 years back and Dobby is still cleaning like a champ.
Like half a year ago, the LIDAR broke. Ordered a new one on amazon for 20 bucks, unscrewed now about 28 screws and exchanged the broken part. Put it all together, again tossed away a screw that left me wondering where it belonged, and it still works to this day.
This was a very long way of saying "Roborocks repair fairly easily, and there are numerous tutorials online."
Edit: BTW, Dobby has been cleaning for almost 500 hours now. At that price point, there is no cheaper hourly cleaning staff :)
I recently went from a Roomba S9+ to a RoboRock S7 MaxV Ultra and it was like night and day. The RoboRock has not gotten stuck on a single cat toy, cable, or sock yet. It doesn't bump into furniture repeatedly. It is considerably quieter. I feel like it is the first time I can actually safely start running a robot vacuum while not at home, and not worry that I'll come home to a robot tangled in some random object.
How does it avoid cables? I have an earlier Roborock version and while I love it and will defend it until the day it dies, I've found that cables are the absolute nemesis. Does your version have computer vision?
I sincerely desire to avoid having a mobile, internet connected device that equips many sensors and cameras in my home that is designed, manufactured and supported by a Chinese company.
What I don't understand is how iRobot, once an iconic innovator, could move so slow and innovate so little that Roborock, a company founded in 2014, could beat the shit out of iRobot.
We went from a "dumb" Eufy RoboVac 11 to a Roomba i3+ EVO 3550 and I honestly wish we stayed with the Eufy or done more research. I just assumed Roomba was like the gold standard of robot vacuums. Its had so many stupid issues including randomly lighting up and making sounds at night which wakes me up since its in our bedroom. I haven't talked to their support because honestly I'm sure it'll be like all other consumer electronics and they'll just waste my time running through BS troubleshooting steps and not actually fix anything
You can turn off the lights in the app. This was annoying me too. They added it as a feature recently. I think the idea is that you know your Roomba is docked correctly and charging if the light is flashing
I can second this. Roborock was already years ahead when we bought ours (3 years ago?) and when I look at what their latest models can do it's pretty clear that Roomba forgot to invest into new technologies.
Maybe they are buying the brand, more than the tech. I bought an amazon basics vac for something like 60 USD which just stole the design of the latest cheap vac, so presumably can plunder the more advanced models too, once the market is ready for them.
What about effectiveness of the vacuuming? That's where Roomba is superior afaik, and the feature is pretty essential since you buy them to clean floors and carpet, not to delight you with the smoothest navigation.
Does it still double as a poop spreader or did they finally add a sensor for that?
As a former roomba owner, the issue I always had was that it was pretty terrible about coverage. The random walk it did often left large areas vacuumed as it had no sense of where it had and had not vacuumed.
It’s crazy tho, even the cheapest roborock w/ LiDAR smokes every roomba out there. It’s hard to justify the more expensive models unless but the advanced feature sets are tempting.
That's exactly what makes this a good acquisition, right? Roomba has the brand/trademark recognition in the US and solid marketshare but a stagnant product, which iRobot apparently does not have the expertise needed to keep improving. For Amazon that should be a pretty straightforward technical/ML problem to solve to get back to the forefront of where Xiaomi, Ecovacs, etc. are.
> will probably be out offering integration with home plumbing to avoid any manual intervention for months at a time.
I can't even get a chicken waterer to work for months at a time, no way I'd let a robot access. Even so, is it so bad to let China take the lead, western design can go on vacay and rip off the results when they get back!
> can go on vacay and rip off the results when they get back
You can clone the design if there's capital left after consumers have been buying the competitor for years, but can you clone the manufacturing process or price as easily when you're years behind?
We see with Qualcomm CPUs and Android phones - the internals are still a few years behind iPhone after about a decade, and Apple still takes the lions share of profits in the phone market.
What’s the repair story? I have one of the earlier vacuum Roombas (620, bought 8 years ago, before most advancements happened), and I can still find both 1st and 3rd party replacement parts and repair almost everything myself.
I have a Roborock S4 and I love the thing (though I want to upgrade to a mopping one at some point)
But on the S4 I find sometimes it underperforms with navigation: it ends up relying on the bumpers and it smashes into table and desk legs even though the lidar should give it a heads-up (the legs are visible on the map it creates).
I'm wondering if it updates the map as it cleans but only _uses_ the updated map once the current clean is finished. Hmmmm
> Meanwhile, iRobot still hadn’t figured out how to make an effective stand-alone mop bot, much less an auto fill station or combined bot that actually works
But then maybe that's where the synergy is. Amazon is probably ahead of everyone on the planet with machine vision. They have the Amazon Go shops, and nobody else has anything that comes close. If they take iRobot's robotics knowledge and couple it with their own machine learning expertise, who will be able to compete with them?
It had better be truly amazing, it's 4x the price I paid for my old Roomba 780 many a year ago that did a decent job of it until it died after 5-6 years of service.
Do you know how they work on tile? I see everyone recommend for wood flooring, but my house is mainly a textured tile. Things like a regular vacuum make loud clanks rolling over and between them. Wondering if a robovac would even work, or get stuck, or just be obnoxiously loud rolling around. Anyone know?
> a mopping vacuum bot with actual rotating mop scrubbers, AI vision obstacle avoidance, LIDAR mapping, and most importantly fully automated water fill & mop pad cleaning at the base station.
And here I am using a broomstick 5 min every day... I feel like we're going to hit the great filter pretty soon
> We just got an Ecovacs X1 Omni - a mopping vacuum bot with actual rotating mop scrubbers, AI vision obstacle avoidance, LIDAR mapping, and most importantly fully automated water fill & mop pad cleaning at the base station
Well for 1500 Eur it better does all of it and even more
Does mop pad really work? I only have roborock with no fancy mop pad cleaning base station but ignoring that all it does is dragging a wet cloth around. I don't see that helping much
My previous bot was a "wet mop pad" Ecovacs model without any of the bells and whistles, and it was pretty worthless. I got the X1 Omni specifically because it has dual rotating scrubbers, self-clean/dry, and self-fill. Night and day difference. I feel like I could eat off the floor every morning.
Does the list of obstacles these things avoid include pet accidents like poop, cat puke, etc? Cause we have a pukey cat and a robo vac might make things much much worse.
It will steer around the rug automatically I think, and you can mark the rug locations as no-go zones or no-mop zones. The Roborock is better for mixed floor types because it will actually raise the mop up to vacuum in no-mop zones. The X1 needs to have the mop removed manually to do a vacuum-only pass.
I mean not trying to start a flame war, but everyone is
praising how great roborock is but the reality is, most of Xiaomi/Roborock technology is stolen/reversed engineered from Neato. Chinese robot vacuum would not even be close to their current capabilities if it wasn’t for the LiDAR on the XV11. If you look at the tear down from the first couple gen, the PCB and components were exactly the same compared to Neatos. (I owned both Neato and Roborock vacuums).
I was at iRobot for holiday season 2003; 1st gen Roomba was out, MANY of the returns piled up along the hallways of HQ.
One day the "director of sales"? walked into the standup meeting like Vince McMahon into the arena and slammed down this robot onto the coffee table that was a pitch-perfect copy of the Roomba, except it was tea green and had more colorful LEDs and sounds. "F this and F that" like we had any ability to stop that from happening. I did the teardown and it was a near-perfect mechanical copy like somebody walked the molds across the street. The PCB was different where they chose cheaper connectors. It was way prettier looking IMO. What they couldn't copy was the logic; the SW/uC was handled through a different chain so unless you literally stole a pile of programmed parts, you'd be on your own. The Roomba development team consoled themselves that the performance of floor coverage of this knock-off was worse than the original...but to be frank, both were only useful under limited circumstances that I found nobody I knew had. I'd have better luck with an old sweep-broom.
As an intern, those returns piling up were so they wouldn't get resold elsewhere or scavenged. And there were a LOT of returns; its a complicated design with a LOT of cables and connectors and some marginal components going out of the expected tolerances. One of my jobs was to strip out the batteries, lay these robots out in the back parking lot, and using a pointed steel rod, 'spear-fish' them to render them unusable. To make my work quick I'd aim for the 'brain' - the middle where the case was thin and I'd bust through the microcontroller part of the main PCB. Aim, 'crunch', make sure I 'killed' it, and then fling it into the dumpster. Nearly 20 years on, I fondly remember thinking 'I can't believe I get paid to do this!'
From what I can tell, Chinese companies will freely steal technology from other countries and companies, which is obviously frustrating and not great, but then they relatively quickly start exceeding the technology they stole, quickly adding features not present in the original products. So it's a bit dangerous for American companies to sit on their heels saying Chinese companies are stealing. Yes, they are, and it's frustrating, but then those Chinese companies add features at a rapid pace that the American companies can seemingly not keep up with.
I'm not sure how many Chinese HN users we have...but I'd personally love to see them chime in occasionally on this topic.
My personal experience/conclusions drawn from business, experiences in grad school etc. (so obviously not worth much and poorly sourced from a data perspective) are that certain seemingly obvious concepts like "cheating" or "stealing" are just fundamentally understood differently and not ingrained in a Western compatible way culturally in China. There seems to be deep differences there that manifest in all kinds of ways.
In Chinese culture, copying (I would not use the word "stealing") symbolizes respect for a product and has roots in Confucianism (however, let's avoid potential exploitation of this). But don't forget that Americans behaved similarly in the 19th century [1], and Brits did the same to Dutchs. So I guess this is a common pattern for raising powers.
Sadly I think this is just encouraging US companies to make products that phone home, and require a service contract. It's much harder to copy that part, and more lucrative.
Just look at the tear down videos on the XV11 compared the any roborock tear down. I would link you the XV11hacking site but seems like it’s down. It had everything from firmware to schematics that hobbyist did in the early 2010s since it was the cheapest way to get a 2D LiDAR (pulling it from a the vacuum itself).
And it's quite disingenuous for IT people to make a fuss about it. EVERYTHING in tech started with some sort of piracy, of music, of web content, of videos, of books, of hotels, of taxis, it's persistently the right strategy. And that's good, the world can't hang onto gatekeepers forever.
It matters because people who take risk and innovate should be rewarded. When that fails, theres no point for anyone to take risks and release new products.
That type of engineering that Compaq performed was very specific clean room reverse engineering or clean room design, which is specifically designed as a defense against copyright infringement because it relies on independent creation by going off the specification (in Compaq case the Bios specification) and not the original design.
https://en.wikipedia.org/wiki/Clean_room_design
Why does large-scale corporate IP theft matter? Well, for one thing, I can't imagine the next Neato being willing to invest in all that R&D if a chinese company is going to rip it off and sell it for peanuts, just so the internet can say "wow, this chinese product is amazing, and I don't know why corporate IP theft matters!"
Amazon Robotics has been actively poaching talent from iRobot already, but I think it's probably a little of everything: talent, data, and the bump they can give to iRobot sales.
The schema of your property is already publicly available in home schematics records, and they already have your address to look up.
I think Occam's Razor applies here. iRobot is a profitable smart home product maker, their revenue/profit has continuously grown, and their IP + talent will allow Amazon to expand into other smart home verticals.
My robot vacuum sits docked away from me, on a separate floor, during the day. It runs at night time and I never see it. It's connected to its own IOT network. How would it know how many people live in my home and when I'm not there?
Not to mention the location of valuables, security cameras (other than Ring, as they can just turn those off), any security dogs or cats. How do you think Jeffrey Beesoz got so rich?
It's also possible that iRobot owns patents related to technology Amazon is planning to use. Besides brush/cleaning-related patents, iRobot owns the rights to several techniques used to map and zone a space.
Looking at the reviews for spare parts for my model i7, i hope amazon gets rid of the crap replacements, but equally lowers the extortionate prices on the official, various, eu website(s).
Apparently iRobot is going to Amazon Devices and not Amazon Robotics, but who knows if they'll silently transfer a lot of that knowledge/talent after the acquisition
Warehouse automation is a completely different ballgame to consumer, though.
If an Amazon warehouse can replace a $20,000/year human with a Kiva robot, they're happy to pay $10,000 for it. Consumers, on the other hand, would find $500 on the steep side for a robot vacuum, considering they'll also need a regular vacuum to cover the stairs and so on.
Consumers' price sensitivity is why Roomba spent years with a plastic product navigating at random by bumping into things and relying more on a brush than a vacuum.
I once attended a Foxconn seminar presented by Jay Lee where the topic was "dark warehouses"; certainly wouldn't be surprised if Amazon has a strong desire to reach that level of machine autonomy in their own warehouses.
I've always wondered how iRobot hasn't been more successful. They have a cool name, brand recognition, first-mover advantage. It's been nearly 20 years since their IPO. I would have expected by now that robots would have taken over more household cleaning tasks, but it doesn't seem like the technology has improved much in the past two decades.
I gave up on my roomba because it would always get stuck in something before it could finish cleaning. I now use a Dyson stick vac.
Leaving specifics of Roomba vs. competitors aside, the basic concept just doesn't work well in a lot of situations. My house ends up having a bunch of cords and clutter. And even on the "single level" ground floor which is fairly open and fairly heavily trafficked, there are a number of small level changes between rooms because of additions over time.
I'm with you. I considered a Roomba a few years back but came to the conclusion that a Dyson cordless vac essentially solved my problem by making it trivial to spend a few minutes pulling out and running a vac as opposed to hauling out a canister vac between housecleanings.
My brother, on the other hand, has dogs and has a newly constructed single level house that they keep pretty free of clutter. A robo-vacuum works pretty well for them.
Newer robots from companies iterating on the technology, like Roborock, can detect cables/shoes/stuff left on the floor and just drive around them. They're a lot better than what Roomba is selling, and much better at dealing with the average home. I've been really impressed with ours after having a similar, frustrating experience with one from iRobot.
I've always felt like these products were mainly designed for people who live in single-level small-ish apartments, but then it's confusing because manually vacuuming those isn't very hard either. I just don't understand the market.
Yeah I had a Roomba 10 years ago. It was starting to age a bit (still working! but would always get stuck). So I was thinking.. it's been a decade.. time to buy a new model, surely they improved!
Nope, nothing is better, absolutely nothing. The "AI" (if we can call it that) to vacuum hasn't improved one bit. It somehow seems so suck a lot less on the carpet than the previous one. It's somewhat better to manage my cat's hair, but that's about it. Sure now I have wifi on it, to re-start it because it often stops too early before it's done. But nothing, no improvement in 10 years.. in ROBOT TECHNOLOGY!
Same experience. Bought a Roomba in 2008 and didn’t like it much. Bought another in 2021 and was shocked at how little it had improved in that time. It’s better, but not by much. And clearly behind the technology in some of the Chinese vacs.
Their robots started to suck after a while ! They are more interested in extracting more from their customers than creating better robots. Best example is their mopping robot that they try to market using a kind of "subscription" model with proprietary pads. The chinese roborock is so much better and they get better after each release.
Their mopping robot was an extreme letdown. We bought the nicer one and it has such a tiny tank that it can't clean my downstairs without needing refilling halfway through.
Plus any cleaning solution besides theirs breaks the robot.
Plus the cleaning pad it uses to mop needs to be replaced after every run
Plus it's a lot more aggressive than the regular roomba and slams itself into things, knocking all kinds of things over.
The cleaning solution is very expensive. The pads have DRM and are very expensive. The robot is very expensive.
Thanks to all it's drawbacks it's not really able to be automated. So you wind up with a mop that you still have to manage, direct to run, and feed expensive soap and cleaning pads. In reality, I have the robot, and it's been plugged in for 4+ years and probably not been ran for 3+ years.
I wonder how much of it is also people like me. I work in tech, but I despise single-purpose household "gadgets." I have a vacuum cleaner and a set of working eyes. I can clean the floor in 5 or 10 minutes, rather than having a "robot" roaming the room randomly, bumping into things, and making noise for an hour. When I'm done, I store it away in a closet and don't have to look at an unsightly lump of plastic parked in a charging station. Bottom line, these cleaning robots don't solve any problem that I have, and they make my home less enjoyable to be in.
The technology has come a long way in these years and the modern models are pretty good devices. I use a Miele canister vac for the deep cleans but I also have a Eufy and a Roborock robot vaccum, one does the random bouncing and the other uses LIDAR to map and clean intentionally, and they're both great. I run them several times a week and the amount of stuff they pull out of the carpet, rugs and off the hard floors is incredible. I would never bust out the Miele several times a week.
Also faster iteration. Back in my mechanical engineering days, we'd design prototypes and then have to wait months for them to come in (of course they were built off-shore).
Doing it onshore was difficult (few could handle our requests, or they were exorbitantly expensive and couldn't scale, and terrible quality control.)
Thus we'd take a long, long time to work out design issues. A Chinese company can iterate much faster because of this.
It wasn't just China. Roomba was really slow to move mapping technology into their lower end products. There was a lot of reputational damage done by making mediocre products their primary volume sellers.
The idea that they’ll have ring outside, a camera inside monitoring every movement is kinda nuts, and Alexa monitoring all audio is insane. We already know Amazon has a service to sell data to police. I don’t know what people will expect here.
It's plain not true, Amazon has shared information with the police, but they don't make money from it. It's also not large-scale with only 11 requests in 2022.
There are plenty of reasons to dislike Amazon, no need to invent one:
> Ring says it will only "respond immediately to urgent law enforcement requests for information in cases involving imminent danger of death or serious physical injury to any person." Its policy is to review any requests for assistance from police, then make "a good-faith determination whether the request meets the well-known standard, grounded in federal law, that there is imminent danger of death or serious physical injury to any person requiring disclosure of information without delay."
It's interesting because I would totally go for that if I could run all of the services locally with some digital assistant AI to tie it together and manage them. It's tragic, really. The cool future I imagined is here but there is a catch. There's always a catch.
How long do folks think until this will lead to actual change within what's now iRobot? Am I naive to assume nothing really will change in the next year or two? I just bought this thing a year ago. It's frustrating to think it's now part of the Amazon data collection system.
I wonder if there'll be a way to sever the device from the cloud and use it through HomeAssistant exclusively. As far as I know it is running an MQTT server but I have no idea how much it relies on the cloud to function.
Absolutely every American company that get a legal request for data from law enforcement will comply. Small companies don't make the news the way Amazon does, but if robot vacuums have ever collected evidence of a crime, then I guarantee iRobot has been issued and complied with requests for data too. Police routinely collect security camera footage from small businesses. Nobody is immune to due process.
I have heard from a former irobot employee that they were besieged by the same issues I had while working in consumer robotics: endless priority changes and lack of long term product vision. This is why their lawn mower robot apparently failed to launch
I feel like if a company just focused on the basics (ie reliable cleaning) instead of a bunch of different marketing gimmicks, it will work out long term.
> focus on the basics instead of marketing gimmicks
This reminds me of what Steve Jobs said about Xerox[0]. After achieving market leadership in their industry, engineering at iRobot seems to have been pushed aside in the company's decision making process.
Is this a consequence of being a publicly traded company? After all, pushing consumables and subscription services requires less capital and makes more money compared to developing a better robot. The $1.7B exit seems like a good outcome for a stagnant (relative to its peers) company.
I ran into this issue even in a private company. Hardware takes a long time, and I think leadership doesn't realize how hard it actually is do get the basic functionality done RELIABLY. That is - having a demo of a working lawn mower / vacuum cleaner / whatever is fairly easy (I've seen Arduino based projects online), but having it work in variety of settings and conditions without user intervention for years takes a very long time to work out the bugs.
I think over time leadership gets frustrated with seemingly lack of progress on the foundation, while efforts to change the high level features (app, smart home integration, GPS, etc) distract from the core product. Just my 2c
Now we wait for the cameras and microphones. Gotta hand it to Bezos, they're building a solid fleet of home espionage devices. The modern trend of replacing everything with smart devices reminds me, ironically, of the movie "I, Robot." I wonder how long it'll take for a corporation to go rogue and lock the country down. Maybe not now, or even in 20 years. But they have a lot of power.
Imagine that. A corporate-run national coup, all thanks to people's laziness.
I just bought a roomba and was happy that they weren’t owned by Google or Amazon. This is unfortunate.
I owned nests before Google bought them and it was sad to both see their innovation decline (basically the same now as 8 years ago) and usability decline (forcing Google logins).
I’m normally skewering how Google manages their physical products, but Tbf Nest declined due to mismanagement by Tony Fadel, one of the founders. Google gave Nest tons of money which didn’t lead to anything major materializing instead of delay after delay.
Fadell left Nest in 2016. Maybe the early stagnation can be blamed on him. But Google has had the reigns for 6 years and nest has really not done anything substantial.
I have two Nest-E devices in the home and have not enabled Google login. If you can find the older versions of the thermostats, you can factory reset them, stick them on a guest network without internal access, and never use the Google login, just the old Nest login.
Not sure how long that'll stay like that, but it's doable at present.
Time to do what I did to every hardware product I owned which Google acquired: Rip it out, delete my account, and try to make my money back selling it to someone else.
I actually made a profit on both my Nest and my Fitbit so far! Though the Nest was from energy rebates before I resold it, and the Fitbit was from it's recall for burn risk... Recall paid MSRP, not the price I purchased it for.
Readit News
Meanwhile, iRobot still hadn’t figured out how to make an effective stand-alone mop bot, much less an auto fill station or combined bot that actually works. Roborock and Ecovacs products launch in China 6-12 months before their US launch. By the time Roomba has a decent mop bot, the X2 or X3 will probably be out offering integration with home plumbing to avoid any manual intervention for months at a time.
EDIT: commenters are asking what works offline from Roborock / Ecovacs. For the self-cleaning-mop generation like the X1 Omni:
- Roborock obstacle avoidance is local, but works better with cloud.
- Roborock can be controlled by a local home automation server without internet access.
- Ecovacs obstacle avoidance requires cloud.
- Ecovacs control requires cloud.
I remembered these stats from a Youtube video that I can't find; probably in my partner's history since we watched reviews on the TV together.
I picked the X1 Omni over the Roborock because it has superior mopping performance on tile floors, but I think both would be good options. It seems like Roborock is more privacy compatible.
Additionally since it maps using Lidar it completes a complete cleaning of my living room + kitchen area in 22 minutes, whole house in under 40 minutes. That is with it covering literally every square inch of carpet & tile instead of going around in random patterns...
For example can you use them completely offline or do they 'need' to phone home to a remote server in order to work?
I'm thankful Amazon bought Roomba - one less reason to even consider them as an option considering all of the current gen vacs are sporting onboard cameras + CV (poop avoidance, is that really a huge concern?).
That is wild as that basically handles all the prep I have to do for mine. Albeit, 1000-1500 is quite a bit more.
They also still sell vacuum and mopping* robots as seperate units which just drives up the cost. My old setup of the two robots took somewhere on the order of 5-6 hours to do a ~100m2 apartment.
I moved to a Roborock and then Dreame bot (someone really needs to talk to Xiaomi about all the sub-brands :P) comparitively, they both took around 1.5 to just under 2 hours to do a much better job.
I actually felt huge guilt selling my two iRobot's to someone, wanting to tell them they were complete junk and they were wasting their money.
*There is always mixed opinions on mopping effectiveness, but the iRobot roll-and-spit model of mopping is truly horrendous.
Why are there only mixed models and what are the benefits?
If I look at my place the ground floor is tiled and the top floor is all carpet.
That means I would need two vacuums anyway as they haven’t learned yet to climb stairs.
Why wouldn’t I want a vacuum only model for upstairs? For downstairs the universal model is great obviously. But upstairs?
On an almost weekly basis it would fail to start because it would think it was moved if something just moved a little bit nearby or a cat just happened to be sleeping nearby. This generally forced me to remap
Then it would constantly get lost and be confused about what room it was in. Seemingly for the same reasons of things moving around.
A home isn't a static environment and a camera just seems much more efficient. Not that a camera is perfect, but at least a camera can account for things moving around were lidar cannot.
Or was this a case of Neato just not being good? The lidar issues was why I went with iRobot. It was a worse vacuum but if it actually works consistently than I can just have it run more often.
Once my i7 has generated a proper map, it will will slow down when it approaches known walls, or even places where obstacles have been several times in the past. It will need to bump into the obsticle or wall to confirm its position though.
It does use a mostly predictable back and forth pattern, taking note of areas with walls/obstacles, and after the main sweeping it will combe back to go around the obstacles, and finally (optional) do a sweep of the walls. Then it moves onto the next room.
The system does have limitations. For example, if it needs assistance in the middle of a clean job, after you fix it, there is a reasonable chance that it will do dumb stuff afterwards, like missing a large section of floor. (It seems to even know this, but have problematic coding that prevents it from doing the right thing).
But yeah, I'd not call the mapping good, just acceptable.
There are other annoyances like the i7 battery being so small, especially since they have a giant battery compartment. Like seriously big enough like ni-cad batteries battery of equivalent capacity. But they only offer a slightly larger capacity battery as part of their Costco exclusive model, and try to lock out third party batteries. The net result is that even when brand new, it could not quite do a single floor of my house in one go, needing to go recharge for like 40 minutes before it could do the last 10 minutes of cleaning.
The app is not very good. It does not even do some basic things like prompt the user for regular maintenance periodically based on number of hours used.
The floor maps in the app are weird. For example, on one of my floors, the online map decided that the entrance to one of my rooms was on a completely different wall from where the entrance really was. It thought the actual door was just a wall. However the actually mapping data on the robot was just fine. It knew where the door really was, and never tried to drive though the wall where the online map though the door was.
I mean, yes, the first time, but my i7 takes like 2 hours to do my ~140sqm apartment?
It's not just "Murica first". If these were European companies, I'd jump ship very quickly to an iRobot competitor with a superior product, thanks to EU's data collection laws and USA's OK relationship with EU. But they aren't, they operate in a regime where questionable data collection is the norm and theft of IP is called "R&D".
Yes, Amazon will use iRobot to vacuum more data than dirt, but at least they are part of the US and EU laws. They provide gov services for a ton of US DoD/NASA infra! China could and would just thumb their nose at us without much concern.
I am not sure if Amazon is more trustworthy than that.
https://www.wired.com/story/amazon-ring-police-videos-securi...
Responding to an inquiry from Sen. Ed Markey (D-Mass.), Amazon said in a letter dated July 1 that it has handed over private recordings to police 11 times in 2022. The company said it was complying with an "emergency request."
Don't you already have that in your pocket?
It's also user-serviceable.
Dead Comment
Of course, this is a model I bought used almost ten years ago. Not sure how similar current models are.
I once had it give me an error that the main fan (for suction) was broken. In this case, there was a small particle lunged into it, so it wouldn't spin. All it took was a little wiggle, and it spun again.
It's definitely more than 10 phillips head screws, but less than 30. I guess it depends which part broke, but I had to disassemble pretty much the whole thing. I can promise you that my repair skills or experiences are very novice, but I managed just fine. Had two screws left over after putting it all back together as is tradition, but I suppose they weren't mission critical, as this was like 3 years back and Dobby is still cleaning like a champ.
Like half a year ago, the LIDAR broke. Ordered a new one on amazon for 20 bucks, unscrewed now about 28 screws and exchanged the broken part. Put it all together, again tossed away a screw that left me wondering where it belonged, and it still works to this day.
This was a very long way of saying "Roborocks repair fairly easily, and there are numerous tutorials online."
Edit: BTW, Dobby has been cleaning for almost 500 hours now. At that price point, there is no cheaper hourly cleaning staff :)
Deleted Comment
Gosh, we're approaching the marketing singularity
Deleted Comment
As a former roomba owner, the issue I always had was that it was pretty terrible about coverage. The random walk it did often left large areas vacuumed as it had no sense of where it had and had not vacuumed.
It costs $1400 here, literally $1000 more than the basic Roomba models, and still $400-600 more than buying a Roomba mop and vacuum separately.
I can't even get a chicken waterer to work for months at a time, no way I'd let a robot access. Even so, is it so bad to let China take the lead, western design can go on vacay and rip off the results when they get back!
You can clone the design if there's capital left after consumers have been buying the competitor for years, but can you clone the manufacturing process or price as easily when you're years behind?
We see with Qualcomm CPUs and Android phones - the internals are still a few years behind iPhone after about a decade, and Apple still takes the lions share of profits in the phone market.
But on the S4 I find sometimes it underperforms with navigation: it ends up relying on the bumpers and it smashes into table and desk legs even though the lidar should give it a heads-up (the legs are visible on the map it creates).
I'm wondering if it updates the map as it cleans but only _uses_ the updated map once the current clean is finished. Hmmmm
But then maybe that's where the synergy is. Amazon is probably ahead of everyone on the planet with machine vision. They have the Amazon Go shops, and nobody else has anything that comes close. If they take iRobot's robotics knowledge and couple it with their own machine learning expertise, who will be able to compete with them?
It had better be truly amazing, it's 4x the price I paid for my old Roomba 780 many a year ago that did a decent job of it until it died after 5-6 years of service.
And here I am using a broomstick 5 min every day... I feel like we're going to hit the great filter pretty soon
Well for 1500 Eur it better does all of it and even more
10 million customers = data about 10 million homes. Plus some revenues, why not.
- Ecovacs control requires cloud.
No thanks. I'm not giving my home to China, America or whatever.
Roborock seems to be better. Anyway it should be blocked by the home firewall, just to be safe.
Dead Comment
One day the "director of sales"? walked into the standup meeting like Vince McMahon into the arena and slammed down this robot onto the coffee table that was a pitch-perfect copy of the Roomba, except it was tea green and had more colorful LEDs and sounds. "F this and F that" like we had any ability to stop that from happening. I did the teardown and it was a near-perfect mechanical copy like somebody walked the molds across the street. The PCB was different where they chose cheaper connectors. It was way prettier looking IMO. What they couldn't copy was the logic; the SW/uC was handled through a different chain so unless you literally stole a pile of programmed parts, you'd be on your own. The Roomba development team consoled themselves that the performance of floor coverage of this knock-off was worse than the original...but to be frank, both were only useful under limited circumstances that I found nobody I knew had. I'd have better luck with an old sweep-broom.
As an intern, those returns piling up were so they wouldn't get resold elsewhere or scavenged. And there were a LOT of returns; its a complicated design with a LOT of cables and connectors and some marginal components going out of the expected tolerances. One of my jobs was to strip out the batteries, lay these robots out in the back parking lot, and using a pointed steel rod, 'spear-fish' them to render them unusable. To make my work quick I'd aim for the 'brain' - the middle where the case was thin and I'd bust through the microcontroller part of the main PCB. Aim, 'crunch', make sure I 'killed' it, and then fling it into the dumpster. Nearly 20 years on, I fondly remember thinking 'I can't believe I get paid to do this!'
My personal experience/conclusions drawn from business, experiences in grad school etc. (so obviously not worth much and poorly sourced from a data perspective) are that certain seemingly obvious concepts like "cheating" or "stealing" are just fundamentally understood differently and not ingrained in a Western compatible way culturally in China. There seems to be deep differences there that manifest in all kinds of ways.
[1] https://foreignpolicy.com/2012/12/06/we-were-pirates-too/
Why can't American companies study the Chinese products and do the same in turn?
I'm interested, please link to that.
Does it really matter?
Dead Comment
All companies are reverse engineering the Universe anyway. It's not like you build these technologies out of thin air.
I think Occam's Razor applies here. iRobot is a profitable smart home product maker, their revenue/profit has continuously grown, and their IP + talent will allow Amazon to expand into other smart home verticals.
Create an all-in-one Alexa robot.
Deleted Comment
If an Amazon warehouse can replace a $20,000/year human with a Kiva robot, they're happy to pay $10,000 for it. Consumers, on the other hand, would find $500 on the steep side for a robot vacuum, considering they'll also need a regular vacuum to cover the stairs and so on.
Consumers' price sensitivity is why Roomba spent years with a plastic product navigating at random by bumping into things and relying more on a brush than a vacuum.
This is almost equivalent to saying that Boeing cares about paper planes because they make airliners.
I gave up on my roomba because it would always get stuck in something before it could finish cleaning. I now use a Dyson stick vac.
I'm with you. I considered a Roomba a few years back but came to the conclusion that a Dyson cordless vac essentially solved my problem by making it trivial to spend a few minutes pulling out and running a vac as opposed to hauling out a canister vac between housecleanings.
My brother, on the other hand, has dogs and has a newly constructed single level house that they keep pretty free of clutter. A robo-vacuum works pretty well for them.
Nope, nothing is better, absolutely nothing. The "AI" (if we can call it that) to vacuum hasn't improved one bit. It somehow seems so suck a lot less on the carpet than the previous one. It's somewhat better to manage my cat's hair, but that's about it. Sure now I have wifi on it, to re-start it because it often stops too early before it's done. But nothing, no improvement in 10 years.. in ROBOT TECHNOLOGY!
https://youtube.com/c/VacuumWars/videos
Plus any cleaning solution besides theirs breaks the robot. Plus the cleaning pad it uses to mop needs to be replaced after every run Plus it's a lot more aggressive than the regular roomba and slams itself into things, knocking all kinds of things over.
The cleaning solution is very expensive. The pads have DRM and are very expensive. The robot is very expensive.
Thanks to all it's drawbacks it's not really able to be automated. So you wind up with a mop that you still have to manage, direct to run, and feed expensive soap and cleaning pads. In reality, I have the robot, and it's been plugged in for 4+ years and probably not been ran for 3+ years.
Are you buying multiple robots over time?
Christ, and here I am thinking $1000 CAD is a lot of money.
I dunno, I would hope they started to suck right out of the box.
(Old MS joke...)
Is this a euphemism for outright IP theft?
Doing it onshore was difficult (few could handle our requests, or they were exorbitantly expensive and couldn't scale, and terrible quality control.)
Thus we'd take a long, long time to work out design issues. A Chinese company can iterate much faster because of this.
Deleted Comment
The idea that they’ll have ring outside, a camera inside monitoring every movement is kinda nuts, and Alexa monitoring all audio is insane. We already know Amazon has a service to sell data to police. I don’t know what people will expect here.
There are plenty of reasons to dislike Amazon, no need to invent one:
> Ring says it will only "respond immediately to urgent law enforcement requests for information in cases involving imminent danger of death or serious physical injury to any person." Its policy is to review any requests for assistance from police, then make "a good-faith determination whether the request meets the well-known standard, grounded in federal law, that there is imminent danger of death or serious physical injury to any person requiring disclosure of information without delay."
https://arstechnica.com/tech-policy/2022/07/amazon-finally-a...
I wonder if there'll be a way to sever the device from the cloud and use it through HomeAssistant exclusively. As far as I know it is running an MQTT server but I have no idea how much it relies on the cloud to function.
Deleted Comment
I feel like if a company just focused on the basics (ie reliable cleaning) instead of a bunch of different marketing gimmicks, it will work out long term.
This reminds me of what Steve Jobs said about Xerox[0]. After achieving market leadership in their industry, engineering at iRobot seems to have been pushed aside in the company's decision making process.
Is this a consequence of being a publicly traded company? After all, pushing consumables and subscription services requires less capital and makes more money compared to developing a better robot. The $1.7B exit seems like a good outcome for a stagnant (relative to its peers) company.
[0] https://youtu.be/NlBjNmXvqIM?t=56
I think over time leadership gets frustrated with seemingly lack of progress on the foundation, while efforts to change the high level features (app, smart home integration, GPS, etc) distract from the core product. Just my 2c
Imagine that. A corporate-run national coup, all thanks to people's laziness.
I owned nests before Google bought them and it was sad to both see their innovation decline (basically the same now as 8 years ago) and usability decline (forcing Google logins).
Not sure how long that'll stay like that, but it's doable at present.
I really like my nest and hope I can keep them for a long time.
I actually made a profit on both my Nest and my Fitbit so far! Though the Nest was from energy rebates before I resold it, and the Fitbit was from it's recall for burn risk... Recall paid MSRP, not the price I purchased it for.