Looking Back: 3 Day Robots

[DampRobot]

Originally Posted by Katie_UPS

It is a competition aimed at high school students. The challenges are hard, but they should not be mind boggling for a seasoned engineer with experience in the sport. There is no "correct" solution for basketball... but there is a difference between the pros and the high school leagues.

How is this different from how some teams function normally? Almost all the perineal powerhouses (as well as some teams that don't always do so well) are driven by dedicated, seasoned engineers with experience in the sport. They do so well because the challenges are easier for professional engineers who have been in FRC for years than for inexperienced high school kids who graduate every four years.

Personally, I'd rather have those professional engineers share their experience with everyone than keep it to their own teams. But pick your poison, I guess.

[Katie_UPS]

Originally Posted by DampRobot

How is this different from how some teams function normally? Almost all the perineal powerhouses (as well as some teams that don't always do so well) are driven by dedicated, seasoned engineers with experience in the sport. They do so well because the challenges are easier for professional engineers who have been in FRC for years than for inexperienced high school kids who graduate every four years.

Personally, I'd rather have those professional engineers share their experience with everyone than keep it to their own teams. But pick your poison, I guess.

Last post and I'm out.

I'm maintaining my analogy:

An NBA star coaching a basketball team is wonderful and fantastic. I love that there are awesome mentors out there helping students and working with teams.

Mentors playing robots with students is great, fantastic, and I love it. I like seeing mentors being awesome with students. I've worked with teams that have seen the whole spectrum of mentor involvement and have no qualms.

I think robots in three days is weird because it cuts out the students and becomes, like I've said, watching an professional athlete play in high school sports.

Mentors playing robots in a high school competition (without students) is weird to me, and I find it off-putting. I'm not trying to imply anything further than the situation of the Ri3D.

[Nemo]

At a minimum, it was really interesting to see what the teams could come up with in three days. I'd probably agree that it feels a little odd for the "pros" to take over the game for a few days without any students. Still, it's only three days of work, so even really brilliant people aren't going to come up with a world champion caliber robot that fast. It's more of a baseline that other teams can start from.

Our team certainly benefited from those designs, using a bunch of ideas from the Ri3D and Build Blitz teams. I don't feel too bad about that, because creating a completely original idea is tough in a competition where so many solutions to similar problems have already been created in the past. Our team had a lot of fun and learned a lot from testing and refining the ideas we used. I think engineering is more tweaking and iterating and refining than outright invention. We made a roller collector, a catapult, and a winch with a ratchet wrench + pneumatic release. All of those were different than the originals in some pretty significant ways, and there was plenty of sweat and TLC put into those modified designs.

I wonder what we would have built without Ri3D? I suspect that we would have looked more closely at a Simbot SS design. We would have been researching previous designs in any case.

I'm particularly glad that roller collectors came out in the 3 day robots. That was a really doable mechanism that any team can pull off, and the more robots that can collect a ball, the better. The 3 day teams didn't invent the first ever roller collectors for FRC, but they showed everybody that they are effective for this game.

[GBK]

I am still not a fan of the 3 day builds. Our team did watch the videos, but not until the Thursday after kickoff, after we made decisions on what direction we wanted to go. We did it as a team and discussed how these robots would work with our design. I guess if a team does not have the creative design resources, then they are a good thing. It seems like so many teams got stuck in the design influence of the 3 day builds.
I think if they are going to continue with this, the teams doing the three day build should be required to do it with a kit chassis, that way they are truly helping teams that do not have the creative design resources.

[pfreivald]

Originally Posted by GBK

I think if they are going to continue with this, the teams doing the three day build should be required to do it with a kit chassis, that way they are truly helping teams that do not have the creative design resources.

This is not a top-down activity, where anyone has the authority to dictate to anyone what they can or can't do. Nobody has any ability to tell a non-FRC team group of people what to do or how to do it, so there's no such thing as "requiring" any of these groups to do anything.

[AllenGregoryIV]

Originally Posted by GBK

I think if they are going to continue with this, the teams doing the three day build should be required to do it with a kit chassis, that way they are truly helping teams that do not have the creative design resources.

Only one, Boom Done, of the 6 72 hour builds used a completely custom chassis and they did it show that it was possible in such a short time. All the other chassis were basically COTS. Part of the reason for the 72 hours builds is promote products which isn't a bad thing. A more diverse understanding and use of COTS parts helps teams build better robots.

[Caleb Sykes]

I absolutely love BB/Ri3d, we used some of their ideas very effectively on my team. The only thing that I have against the Ri3D/BB is that certain individuals seem to think that anyone that builds anything worse than one of those robots has "failed". I will touch on a few reasons why this is ridiculous:

time-These guys build their robots in 72 hours, and from what I saw, most of them literally worked for all 72 hours. Making an estimate of my team's total hours: 15hrs/week*6weeks - 10 hours in early build season + 10 hours in late build season = 90 hours, with a good chunk of that time going into set-up/take-down. Since we only have maybe 5 students working at any given time, our total man-hours will fall well short of what the 3-day builds do.

experience-Some individuals working on these robots have more years of FRC experience than all of the mentors and students on my team have, put together. Even a few years out, when we have more experienced students, they will start to leave. The groups building these robots will always have more combined experience than the average team out there.

resources- Most teams have to work on a budget, and have to use some of that precious build time waiting for parts to arrive. Also, making quick changes (different motor, different gearing, etc...) for them is easy, but quick changes like this are impossible for many teams.

Again, I love these 3 day build robots, but I don't enjoy people looking down on us when our robot is worse than one of them.

[Dunngeon]

For 955, Ri3D is used to validate concepts that have gone to prototyping. Our inital design phase is completed in 3 days, so we are usually independent of Ri3D. We really liked Ri3D, because it presented weaknesses with packaging and also early validation data. As an example, our team chose to package our catapult and intake on opposite ends of the robot to eliminate the rock that Team 1.0 had. Initally, we didn't follow BuildBlitz at all, but we found the cam gear through that channel which is a brilliant mechanism to use with our catapult.

Overall, Ri3D provides good validation data and helps show early problems with specific designs, saving us time and money.

I do have a gripe with Buildblitz though, both robots on there could win regionals without any additional iterations. They are refined bots, and I think they possesed qualities that the Ri3D teams lacked, and qualities that many teams lack, which is polish. The Buildblitz robots are pressed out, they can do everything well and if a team possesses the knowhow or the money, they could re-manufacture the robot with ease. My point here is that the Buildblitz robots are too refined, and they could have too much influence on the sport.

Obviously recreating these robots is another challenge

Sorry if I rambled a bit, it's late

[evanatch]

I haven't been involved with FRC very long, but I see plenty of design variation this year (maybe even more than last year, at the top level especially).

I think something worth mentioning is that there are certain mechanisms and certain teams that are copied more than others. I've seen far more JVN and Boom Done style builds than say, Team Copioli copies. To me, that suggests that Ri3D serves as a kind of prototyping for many teams - they are exposed to a variety of mechanisms and can determine for themselves what type of mechanisms will suit their strategy the best. I think this is very helpful to teams that do not have the resources to prototype a variety of mechanisms for themselves.

Our team's robot looks very little like any of the 3-day robots, but we were inspired to use a choo-choo gear to wind our shooter from Team JVN. I think that presenting a variety of working robots to view before the build season gives teams a base to build from, without ruining the game. All of the Ri3D robots are competitive but not dominant designs so teams still need to push them further to be winning - especially if other teams have 3-day style robots to build from as well.

Overall, I think the 3-day robots raise the achievable bar for struggling teams, which is always a good thing - this year especially.

[Dancin103]

Originally Posted by evanatch

I haven't been involved with FRC very long, but I see plenty of design variation this year (maybe even more than last year, at the top level especially).

I think something worth mentioning is that there are certain mechanisms and certain teams that are copied more than others. I've seen far more JVN and Boom Done style builds than say, Team Copioli copies. To me, that suggests that Ri3D serves as a kind of prototyping for many teams - they are exposed to a variety of mechanisms and can determine for themselves what type of mechanisms will suit their strategy the best. I think this is very helpful to teams that do not have the resources to prototype a variety of mechanisms for themselves.

Our team's robot looks very little like any of the 3-day robots, but we were inspired to use a choo-choo gear to wind our shooter from Team JVN. I think that presenting a variety of working robots to view before the build season gives teams a base to build from, without ruining the game. All of the Ri3D robots are competitive but not dominant designs so teams still need to push them further to be winning - especially if other teams have 3-day style robots to build from as well.

Overall, I think the 3-day robots raise the achievable bar for struggling teams, which is always a good thing - this year especially.

So there's a few things here with Ri3D: lots of pros, lots of cons. I think Evan here has nailed a handful of the pros - it helps struggling teams, it helps rookies, and it gives ideas for mechanisms not whole machines. And this is why one half of me likes it. The reason the other half doesn't like it - it limits thinking and innovation, a common theme we are seeing not just here in FIRST, but in our society as a whole.

Just my 2 cents.

[Tungrus]

It's sad to see how the well intended philosophy of mentors guiding/sharing experience with students is sliding down hill. Not going to turn this into ugly thread of who/what is right or wrong, I am just moving out FLL!

[cadandcookies]

I still have yet to see any hard evidence that "thinking" or "innovation" were any lower this year than in previous years. I'd really appreciate the next person to bring up this assertion providing non-anecdotal evidence.

[AllenGregoryIV]

Originally Posted by cadandcookies

I still have yet to see any hard evidence that "thinking" or "innovation" were any lower this year than in previous years. I'd really appreciate the next person to bring up this assertion providing non-anecdotal evidence.

I completely agree, most great ideas come from combining other ideas. That requires first that you gain a ton of experience with other ideas. The 72 hour builds do a great job of exposing students to a large set of ideas that they can build on.

Even the teams that build robots that are very similar (I don't know of any team that exactly copied a robot) to the 72 hour robots they still had a to learn a lot of the details of the design and I'm sure they will be better at designing robots next year. The problems they faced this build season were just as real as other teams and they had to learn to solve them. No one expects the next great innovation in the auto industry to be designed by someone that has never seen a car before why do we expect the same from our students. Take what other people have done, learn from it, and make it better.

[ASmith1675]

I personally am a fan of the 3 day robot builds. I think over the last 2 seasons in which they have been taking place, I have seen the floor of the competition increase significantly. (I also think that the availability of high quality COTS parts specifically designed for use with FIRST robots from the likes of VEX and AndyMark, among others is also a large reason)

I have mentored FRC teams for a total of 8 years (2005-2008 & 2011-2014). Several other of the engineering mentors on our current team have also been involved for similar periods of time. This gives us a large experience base of designs of both robots and mechanisms that have both worked and failed in past years. Over the past several years we have had games that have closely mirrored other games in the not too distant past (see: 2006 & 2012, 2007 & 2011, 2008 & 2014). Because we had mentors that knew the games in the past, we could easily say to our students, "Hey, this game is pretty close to 2008, lets take a look at some robots that played that game well for inspiration"

The robot in 3 days concept gives this same advantage to all teams. Not every team can have a mentor that even would have known about the game in 2008 this year. Luckily, now we have the resource of the 3 day robots to help teams which do not have the luxury of mentors that have been involved in the program for nearly a decade. The team that I was first involved with, 677, was mentored entirely by college students from Ohio State. So not only were we turning over our students every 4 years, we were also turning over all of our mentors as well. These type of projects would have helped both our students and our young mentors get up to speed a little quicker, had they been available at the time.

We all draw inspiration from somewhere. I don't see how drawing inspiration from a 3-day robot is any different than drawing inspiration from 1114's 2008 robot, or seeing roller claw mechanisms in 2007 and incorporating them in 2011. Yes, you can argue that the 3-day robots are specifically designed for this game, but as others have said, nothing is a direct copy. Even if a design were to be a direct copy, there is still likely to be a lot of testing and troubleshooting of the systems involved to get them to be as effective as possible.

I don't tend to get too preachy, but in the end its about inspiration. Some are inspired by the 3-day robots to say, "Hey, that doesn't look too hard, we could do that". Others say, "Yeah, that's a nice design, but I think it would be better with X, Y, and Z changes. Still others will challenge themselves to say "I don't want to copy the 3-day robots, I want to try something completely different". Personally, I think teams can learn from any of these ways of thinking.

Anyway, I've rambled for far too long now.

[Siri]

Originally Posted by cadandcookies

I still have yet to see any hard evidence that "thinking" or "innovation" were any lower this year than in previous years. I'd really appreciate the next person to bring up this assertion providing non-anecdotal evidence.

Personally, I agree with you--I'm a huge fan of Ri3D. But what non-anecdotal evidence could possibly be offered for either case? It's not like we're taking a census here. Would you be looking for something in particular?