Too Much Design Collaboration

One thing has really bothered me to this point in the season. Design collaboration. For example, I’ve seen a thousand 8059A robots to this point. It might be a good design, but how is this going to play out at worlds? With such similar robots, how are you guys planning on getting an edge over a team with the same robot as you? Design collaboration has certainly been an issue this year.

Not to mention the fact that it sort of takes out the whole step of designing and prototyping different designs for inexperienced/younger teams. I mean, you don’t really need to troubleshoot your design if you just copy exactly the same robot as a team you see on the forum.

I have actually been thinking the same thing and I completely agree. Teams that are showing their robots too much, are basically giving them the answers during a test. And about worlds. The only thing that will probably get you on edge with a team is by having a lift.

coding and pid things, problem is there basically three types of popular launchers. flywheel, puncher, and catapult

But what Mr.Vex and I are trying to say is that there is too much replicas of a robot, not just designs.

What I’m saying is their all going to look alike in some way. Not everyone wants to put a piece of metal in a different place to make it unique

Yes but it’s not just about robots looking alike, we’re also concerned about worlds.

then it comes down to programming and efficiency of the build, aswell as some luck.

Now that part, I agree on ^_^.

@Mr. VEX I agree. I got into the competitions to teach the engineering design process. What I’ve run into is a lot reverse engineering of YouTube videos which is comparable to stealing Intellectual Property in the adult business world.

That being said the first thing I have my teams do is research the problem, and that includes seeing previous solutions which are readily available on YouTube. I encourage my students to use it as a starting point if they are struggling, but focus on coming up with an innovation or unique design.

Blank readily gave out their design details too, which is kind of like Tesla making their battery technology open source. So of course people will use it.

Each scenario is a great learning experience for team members. It is disappointing though.

This. There’s a difference to seeing an amazing robot and building an amazing robot.

Wha…

I really like this analogy because it carries a lot more truth than you might initially think. When Tesla makes all of its secrets public it does a few things

  1. It says “hey this is a difficult problem and we would like to see a lot more participation in solving it so join me. We admit it was difficult to get to this point and have no interest in watching everyone needlessly suffer through it”
  2. It says in terms of competition “at least be this good, by the time you are I will have far outstripped this technology”

Now in terms of design convergence this year, there really has been very little. I even anticipate that this is one of the better years at worlds for multiple designs to flourish. None of those will look exactly like 8059(as good as that robot was). This will even be an interesting question for me to ask Tom, the head builder of 8059, his opinion on it in tonight’s VEXCast.
https://vexforum.com/t/vexcast/32273/1

I know that next year I want to completely take all of my parts from my robot from a successful robot in the Asian countries that have already started their season and in the preseason learn all I can about what I might need. This isn’t nearly as cool or fun I guess but you get a better robot in less time because of the lack of research and development. I don’t see why a team that starts their season late wouldn’t copy a robot system that has already been developed. Unfortunately I don’t see how vex could stop this from happening. Each robot will still be unique but the ideas in each category are already almost identical.

But who really wants luck to decide the winner of worlds. We want different robots competing, each believing that their robot will have an edge over the other robot.

I’m not saying that it’s just the 8059A robots repeating themselves. I’ve seen many different iterations of the slip gear catapult launcher, and many of them are similar, and because the are so similar, they are producing extremely similar results.

Everyone is trying to do as well as they can. Of course is if someone does well, other teams will try to copy their success. I don’t really see any way for this to not happen.

A thread like this pops up every year. Design convergence will happen no matter what, yet at worlds, there are always robots that are better/more efficient than others(along with strategies) and those are the ones that will come out on top.

Oh what would vex be like if everyone patented their robot… Probably a very bad one…

Because of the limited possibilities for robot designs, design convergence has become very common.
Calling it work copying undermines all of their hard work.
We designed our robot to be very original, but last month, we went to a competition, and discovered that 2A,R&Z’s had robots that looked exactly the way we wanted to build ours. Now people think that we were just copying their success.
Anyway, why should we care if one team that has released their design to the pubic, and answers questions about their gets some of their ideas taken? Wasn’t that part of the reason that they made a reveal?

I think one major concern is the game itself this year. Since the task (aside from lifting and auton) is so repetitive, there are only a few ways to do it.

Look at last years game. Robots had to lift to 5-6 feet, process posts at different heights, and intake a large, hollow cube (or even 2 or 3 or 4.) Basically, there are a lot of ways to do that. Even in division elimination at worlds last year, cube intakes differed. There were the side tread intakes that looked a lot like tossup intakes (8095A), needle intakes (2915A), and bottom tread intakes (62). And then there were the teams with skyrise claws (2Z) and arms (2915A). And as far as arms go, some teams had vertical arms (62) while others had horizontal arms (everyone else). As for lifts, there were plenty of double reverse four bars (2915A), but also scissor lifts (7682). And I saw a pretty decent number of 8 bars (Hawaiian Kids) and elevator lifts (6007) in middle school. In other words, while some designs may have trumped others, even the best teams had very, very different designs.

This year, the only real tasks are shooting and intaking. Thanks to the magnitude of balls on the field top roller have become the overwhelming majority. And since most teams have 12 motors rather than pneumatics, even innovative intakes like the one on Elevated from 26 have become a small fraction of teams. As for shooters, aside from (brilliant) teams like 974X, most every flywheel iteration is a single one. And while catapults and pinball launchers are theoretically more efficient, they are just not able to keep up with 0.1 seconds per shot (LEER’s single flywheel) because the balls have to settle in the launcher. To be honest, had 62 not pioneered the cam gear mechanism, it may never had caught on. I’m sure slip gears and catapults would have, but even as it is, “flywheel alternatives” make up a very small minority of teams at regional events. And they will continue to get worse as the season progresses since they cannot shoot from up close without a worm gear and a ball adjuster (again, purely the ideas of 62; I haven’t seen anyone else implement both of these together.) The only real alternative I can see is a pinball for full-court and a flywheel for up close (2R and 2Z), but this design is complicated, which means most teams won’t implement it. Personally, I have had trouble getting 1 launcher to work, so I can’t imagine building, tweaking, and maintaining 2!

Then there’s lifting. This aspect of the game has inspired some incredibly brilliant dual use of motors, clever mechanisms, and true innovation. 8900 has used their 5 catapult motors on their lift as well, while some team (I can’t find them) has an unfolding scissor lift. 974X, building off of previous designs, found a purely pneumatic lift that relies on rubber bands rather than motors for torque. 1344B and 2Z have both found 2 motor self-lifters, and of course there are ramps, with varying degrees of cleverness and compatibility. Unfortunately, many teams, even halfway through the season, cannot lift at all! This means that if a team can’t find any good ideas, they are perfectly free to build no lift and still win competitions. For instance, cyber hawks G, with a really good flywheel, drive base, and intake, were able to take the robot skills, programming skills, first seeded team, and tournament champion at the Streets of Bakersfield Extravaganza, all without a lift. Their noted lack of a lift is not a criticism of them in any way; the team that builds the robot that can score the most points wins. And I suppose the extra motors gave them more than a 50 point boost, to such an extent that they decided they could do without a lift.

In other words, I sincerely believe that the convergence we are seeing this early and to this extent is the fault of poor game designing, not poor building or “laziness.” Sorry this post was so long, but I felt I needed a lot of examples to make my point clear.

Only if that IP is actually protected. Plenty of things companies do are not protected by IP regulations. And there are entire markets (China) where IP is largely ignored and the result is less friction in the economy and faster innovation. (Ironically, those are the things the patent system is supposed to achieve - whoever decided to get lawyers involved in engineering was, to put it politely, making a poor decision).

Design convergence is much less prevalent this year than it has been previously.

There are also only a few ways to make each design. It doesn’t make sense to get grumpy at people for “copying 8059”, because any single roller flywheel is going to look more or less the same as their robot. And it’s inevitable that many teams will build single roller flywheels. When you simplify each design to its fundamentals there are only about ten different ways you can throw a ball with a VEX robot, and there are several thousand teams. For any design that works well, even without information sharing that design would be built by hundreds of teams.