Questionalble Robotic Printer Kickstarter Using Parallax Inkjet Kit
Duane Degn
Posts: 10,588
We all see bad Kickstarters but this one really bugs me.
They put some wheels on a Parallax Inkjet Kit and are calling it a portable printer.
The demo video is honest and shows the robot printing a few words of text on a single line. What isn't honest are the pictures on the page showing the "printer" on sheets of paper with graphics and technical drawings.
Anyone with any experience trying to get a robot to travel in a predictable route knows how impossible what they are proposing is.
I think this goes beyond "biting off more than one can chew" to down right fraud.
You can find out more information about one of the "developers" by viewing his questions on the Let's Make Robot's forum.
This Kickstarter bugs me enough that I paid $1 so I could post a comment calling it a scam.
I thought Kickstarter projects had to have a working prototype? Why would this one be allowed?
Edit: After reading Gordon's post below, I realize that a robot capable of using a an inkjet cartridge would be a use device. Others have also commented about the apparent misleading aspects of the project so backers who take the time to read the comments can now make an educated decision about backing the project. While I have reason to believe the creators will not be able to deliver a robot capable of making the prints shown on the main page, I don't have reason to believe they won't deliver some sort of printing robot. While I still think what I said was true, I regret using such harsh language.
They put some wheels on a Parallax Inkjet Kit and are calling it a portable printer.
The demo video is honest and shows the robot printing a few words of text on a single line. What isn't honest are the pictures on the page showing the "printer" on sheets of paper with graphics and technical drawings.
Anyone with any experience trying to get a robot to travel in a predictable route knows how impossible what they are proposing is.
I think this goes beyond "biting off more than one can chew" to down right fraud.
You can find out more information about one of the "developers" by viewing his questions on the Let's Make Robot's forum.
This Kickstarter bugs me enough that I paid $1 so I could post a comment calling it a scam.
I thought Kickstarter projects had to have a working prototype? Why would this one be allowed?
Edit: After reading Gordon's post below, I realize that a robot capable of using a an inkjet cartridge would be a use device. Others have also commented about the apparent misleading aspects of the project so backers who take the time to read the comments can now make an educated decision about backing the project. While I have reason to believe the creators will not be able to deliver a robot capable of making the prints shown on the main page, I don't have reason to believe they won't deliver some sort of printing robot. While I still think what I said was true, I regret using such harsh language.
Comments
In any event, maybe I should do a Kickstarter with my drawbot?
I'm kidding of course, I'd love to make a kit out of it, but I don't have the time.
I don't think they need to be posted to ScamWarners.com, At least not yet anyway..
Ordinarily, I would suggest not using your real life information for this kind of hunting, but I think you have this well in hand, and should be safe enough.
Kind of bothered by the fact one would have to spend a dollar just to post any opinion on a Kickstarter project, Just doesn't seem right to me.
Seems like most folks aren't like Duane, and would pass up the opportunity to post a warning like that...
Thanks Duane.
Maybe I'm naive, but didn't want to come across as an anonymous troll. I also wanted people to be able to search my name to see what sort of experience I have with hobby robotics.
While I wasn't thrilled to pay a dollar to cause I didn't like, I think the $1 price keeps away a lot of possible trolls.
The funding seems to have slowed down but it still looks very likely they will reach their goal and a lot of people will have lost $200.
Hopefully the Kickstarter stalls where it is at, but there are many uneducated people with money out there. If they get close to the total, I could see them initiate another trick that occurs with Kickstarter. The Kickstarter author will fund the last bit to push it over the edge. Gotta figure if it is a scam and it gets to say $390k, it wouldn't be too difficult to have someone back you with $10k to make it succeed, if you promise to hand them $15k a few weeks later from the funds you scammed.
Duaune,
Is it an actually Parallax Inkjet Kit? or something they put together.
I couldn't find a reference to Parallax, but I might have missed it.
That kit has not be available for eons if I recall. Ask Matt G.
Jim\
BTW, I've been accused of having "ulterior motives" and the creator has be misquoting me pretty badly. Fortunately they include links to my nay saying so people can see what I really said.
Here's the reference.
I see it's possible it's not a Parallax kit but "like" a Parallax kit.
The questions asked by the developer sure makes it clear they don't know a lot about robotics.
I don't see anything wrong with using of the shelf parts to show the idea and then use kickstarter money to make a custom (and smaller) final product
My problem is they have a working 1D printer and promising a 2D printer. IMO, the added dimension makes their task next to impossible.
Now that's I've stated my concerns in the comments, the project doesn't bug me as much as it had. I'm afraid most people have no idea how hard it is to get a robot to move precisely and don't realize the huge risk they're taking by backing the project.
So it will only function the same way as the proto type or they fold the cards out of frustrations and you got ripped off.
I think 50/50 is way off. Just because there are two outcomes does mean each outcome has the same odds of happening. If you buy a lottery ticket you can win or not but your odds are not 50/50. I think the odds of getting a printer working as well as they suggest is about the same as winning the lottery (though the lottery is probably a safer bet).
It is up to you, but I won't contribute at any funding level for the right to complain even if the product does offend me in some way.
Getting out of alignment is a problem, I would try to use optical sensors to scan the above pass lowest pixels (keep a list of the lowest pixels from precious line pass)
Though hexbot movement would not be good to fine-tune movement as you go.
P.S lottery on average is 50/50 payout, spend 1mill and you are pretty much guaranteed to win 500k
Though I can see that their design addresses some of the issues raised -- the X rollers are driven by a single motor, and with an accurate enough translational sensor they *might* be able to correct for some drift -- I have to ask, "why bother?" Are mobile users REALLY that desperate to print things? I never am. You'd think if printing were all that important people at Starbucks would actually use the printers they supposedly have connected to their WiFi. I've never seen anyone use them. Maybe it costs a dollar or two, but it's color, and it takes just a few seconds. A printing robot, though cute, has got to take several minutes for a single B/W page.
I'm thinking the backers are just looking for a small $180 robot to hack. That's cool, and perfectly acceptable. But why anyone would want to print in monochrome at crude resolution (Epsons print at 360 px/inch or higher) and very low speeds is beyond me. I have no comment on the technical reliability of the robot because they don't provide enough detail, but I'd question its rationale. It seems like a solution to a problem that doesn't exist.
While it's no proof that the robot actually printed it (you don't see the blank page prior to the print), I doubt it's faked.
I am glad you took my reply in the spirit it was intended, I wrote that just after i left a different place, one that embraces and rewards paranoia..
So, My apologies to you and the forum in general...
Your response to the developers is accurate. The pictures are misleading, and the technology seems to be out of reach... :nerd:
Keep up the good fight.
-Tommy
Take a look at this video. That is what I call a working prototype (if it's real). There are several videos on the channel that look very convincing. Much more impressive than the video I linked to in post #16.
If these guys can do it maybe the Kickstarter robot can? (I still think not.)
As I said, it IS an intriguing application, but the thing's gotta be slow. I don't see the point other than as a toy, in which case the quality as-is is probably just fine. In fact, as a toy, it would probably be an enormous seller.
I'm impressed with how well the robot appears to be working. I think the wide margins limit the printer's usefulness but it could be a fun gadget.
I think Zuta Labs should make a remote control app for the robot so when the robot runs out of ink people could still drive it around on their desk.
I'm not surprised the robot remains on the paper the entire print. I think the transitions from paper to desktop back to paper would cause alignment problems for the printer.
In an attempt of maximize the very limited printable area Zuta Labs reduced the size of the robot from the original prototype. Apparently normal printer cartridges were too large to fit in this smaller robot so they had to modify the printer cartridge.
Many of the backers are upset by the switch to a proprietary cartridge and many have said they'd rather the robot be larger than having to use a proprietary cartridge. I don't think these backers realize what a larger robot would mean with regards to the printable area of the paper.
I still think the photos on the "Story" page were misleading. I think it's safe to say the current printer could not print any of the five documents shown on the KickStarter site.
One of my main complaints about this Kickstarter is amount of warning of the risk to the backers.
Under the section "Risks and Challenges" they state:
I think they gave backers the impression they were further along with the development of the robot than they really were.
If they really had gotten the printer "to work well" printing in straight parallel lines, they would have surely known they needed to keep the robot on the page throughout the print. If they knew this then they knew the example documents they were displaying couldn't be printed with their printer.
I've tried to figure out why this particular KickStarter bugs me so much. I think a lot of it has to do with my own experiments with omni wheels. I also think their asking for advice on Let's Make Robots about how to get their omni wheels to roll correctly made it clear to me they were misrepresenting themselves as know more than they did.
Apparently Zuta Labs is offering refunds to backers who are upset about the proprietary cartridge. If they're willing to offer refunds and they're being honest about what the robot can presently do, then I don't have anything to complain about (with regards to this KickStarter campaign).
As I've previously mentioned, I regret using such harsh language when I originally commented about this project.
I was curious about their 2 DOF technique and I wondered what sort of precision one could achieve with such a design. I let my curiosity get the better of me and I started building a 2 DOF robot myself.
In order to limit the robot to 2 degrees of freedom, I needed a long axle to join each pair of wheels. The long axles needed to be able to pass by each other so the two sets of omni wheels needed to be different sizes.
Vex's 4" and 2.75" omni wheels were good sizes to use in this sort of robot.
I got busy with other work and I haven't done much with this robot. I doubt I'll keep this robot intact for very long. The 2 DOF drive mechanism doesn't allow the robot to rotate (at least not on purpose) and makes for a pretty boring robot.
Since virtually no one prints without a computer -- likely a laptop when out of the office -- it makes far more sense to have a little "strip" printer that is based on age-old printer technology. It can mount on the top lip of the laptop cover, and print pages more accurately while you're typing. Standard inkjet paper, standard inks, faster printing, good to great resolution. Boring technology-wise, but that's not bad for a product that's actually useful and cost effective.
What if you wanted one line of text at the top of a page, and one line of text at the bottom of the blank page with nothing in between? If it uses the side of the paper to align then scan all the way down to find the bottom I guess that would work but edge detection doesn't seem like something they're using. If it is it would be clunky because of the whole contrast issue. Ultrasonic paper detectors could use for edge detection I guess. Cheap desktop scanners can detect a double feed really well these days. It looks like it is using its own text as the optical scanner reference. An inkjet printer knows where the edges of the page are, that is the only way a printhead can be accurate. I don't see how it can fix accumulated errors.
Does it record the paper fibers into a map like you'd use a sensor to map a room? I guess that could work with some really good optics. That would remove the edge detection issues too.
My biggest objection is the rest of the world is moving away from paper. The rest of the world already hates inkjet. Inkjet sucks. But, hey check out my robot it can print on this piece of paper! That would be AWESOME! Until... the ink cartridge nozzles dried up because you never use it. Besides sucky ink cartridges, the battery isn't user replaceable, another thing to think about.
A couple people on this forum had made drawing robots, and of course there is the Scribbler. People without that ability would love to share in the same feeling you got the first time you saw it print. I don't think you need to build a highly technical device for kickstarter. You need to build something simple that people will enjoy and that you can deliver faster than promised. I'm surprised more people here aren't running Kick Starters, even if you don't like Kick Starter it's a super high conversion rate for their percentage. My point was there are a lot of people making stuff that could be just as successful when it comes to cool stuff with wow factor.
I don't think it's that advanced. I'm pretty sure it's not using any feedback mechanism. It's just driving the steppers a certain number of steps and hoping the robot ends up where it was intended. If you look at 0:32 of the video, you can see how the top three lines aren't very well aligned. At 0:40, they've switched the paper to a page which has better aligned text. I'm starting to wonder if the robot really printed the whole page. There is very little new text shown being printed in any one shot. Most of the video showing the robot in action show the robot passing over text which was already on the page. The only new text I saw were the few words shown at 0:26. I found it odd how many cuts occurred just as the robot was about to reveal the freshly printed line.
I personally think this is a cool (though probably nearly useless) gadget. But I think the campaign has a very slimy feel/smell to it. I get the very strong impression the creator is not being open about the limitations of the robot.
They're letting the print head half spray the same line twice? I'd turn that off because those lines look the worst.
One more thing, the product name is wrong in my book, it's too big for pockets. Calling it something it is not, then the pics of stuff it didn't do shouldn't be tolerated. Crowdsourcing will turn into a scammer paradise.
Ouch! You're right.
It's pretty clear from later in the video (0:40), those top three lines are supposed to be evenly spaced (and parallel).
If that gets left in the demo video, I wonder how a typical page looks.
Assuming you mean the video in post #23, the shot at 0:40 is clearly edited in. It's hard to know if they're just being lazy, and didn't bother to reshoot the partial page at 0:32, or if there's trickery involved. The two printouts are clearly not the same piece of paper, as the video suggests. Video editing like this doesn't (and shouldn't) engender trust. Reshoot the video if the machine has gotten better, for crying out lout. They've gotten half a million dollars from would-be purchases. (And no, don't say it's an old video so it doesn't count. It was posted to YouTube just five days ago.)