900's Championship Cheesecaking Chronicles

I’m really interested to hear more about this… and how it worked.


We have our students working on a white paper. We’ll be sure to have them post it when they’re done!

We are going to get a white paper out as soon as we can. Here is a link to the PDF version of the presentation we had up on our screens in the pits for Curie: https://drive.google.com/open?id=0B7SC9BMzdll7RVdra2pTMjBoT1k&authuser=0

You can get the missing video over here:

And one of it in action here:

We were fortunate that Nvidia took an interest and we’re hoping to follow up with them as well. We’re tired of the same old vision targets and dead reckoning autonomous routines. We’ve been striving for the last few years to bring computer vision into our designs and this year we saw a huge payoff. I’m hoping it continues to improve and our new contacts over at Nvidia are going to help improve even further.

Sweet because this may be something I have our programmers do just as an off season challenge.

To add to Jon’s comments, the inspection took a while but was fun and intense. All teams were involved and were very appreciative of the inspectors even when we pointed out issues. The teams clearly worked closely together and the final robot was quite impressive. This will be one of those inspections I will never forget not only because of the robot and the situation, but because of the students we got to interface with and how gracious they were in such a tense situation.

I saw these at the Championships just off to the right of the Einstein fields, I was HOPING to see these in action and was completely deflated when they didn’t get on the field.

But perhaps not for the reason most of you are thinking.

I wanted this to happen because I think that this would have be the moment that 1114 and Karthik had finally put a stake into the heart of Cheesecaking - The moment those harpoons had won a World Championship, I would have sung like the Coroner in the Wizard of Oz:
As Coroner I must aver, I thoroughly examined her, and she’s not only merely dead, she’s really most sincerely dead.

Team 900, no disrespect, but I believe your story is a horror show, start to finish.

If the Harpoons had taken the field and carried the day on Einstein, Pandora’s box would have been opened so far and wide that everyone would see the problems that Cheesecaking pose: From that day forth, for the lower 90% of teams, the World Championships would become an exercise in whether and when to sell their souls to the devil.

Our team built a robot, we love our robot, but as soon as it becomes clear that we have no chance of making it to Einstein, it’s time to figure out how we can toss that robot in the dumpster and build a platter to hold cheesecake (largely) conceived of, designed & built by a team that still has a chance to get to the Big Show.

It is my sincere hope that FIRST can figure out a way to prevent this in the future.

Joe J.

Thank you for the nod to the inspection staff. We are happy to be of help, you really challenged us. This device/assembly/(add your own words) was a challenge to evaluate for legality under the robot rules and to believe it to be safe for the teams and any human close by. I am happy to report that everyone we were involved with, acted with the utmost GP in assisting us with the process. It was obvious that the design team had thought and planned for everything I thought might be an issue. While scary in looks and operation, it contained significant safety features and system backups should anything fail during deployment or operation. I would like to restate for the team involved that you acted with all the decorum expected of a Hall of Fame team and I am proud to have been involved in your endeavor.
Attached is a picture of the device in the transport configuration behind the Einstein fields. Yes it fit!

I’m going to take your comment with the spirit in which I think you meant it and not as an insult. Saying “no disrespect” and then “horror show” in the same sentence does not come off as respectful. I don’t think you meant it that way and I think I know what you meant.

I personally feel like the fix for this is better game design. I don’t feel like you can stop teams doing what we did without seriously hurting something truly wonderful about collaborate at-event designs and re-designs. We used COTS items and KOP components to do what we did. The few custom components were well under the 30 pound limit. I would not want to remove the ability of any team to redesign their robot at an event.

I don’t know that you can make it illegal but you can remove the incentive for anyone to do it. I think that is the best method forward.

I’m glad we are all talking about this though. It should be talked about.

I was a bit disappointed when I saw the harpoons field-side without them being ever deployed, but I imagine it was great fun and a real learning process to create an entirely new robot from scratch in only a matter of hours. I’ll bet it was quite inspirational for all who were involved, both on the “donor” team (1114), the “recipient” team (900), and the other teams on the alliance (148 and 1923). You could even say the fact that some think it was a horror show is merely a form of recognition of their achievements in such a short time schedule.

While I understand that teams and individuals (me included) get attached to their specific robots, I fail to see the downsides of allowing teams to help others improve their robots. My team sends out members of our electrical, mechanical, and software sub-teams to help out anyone who asks on Thursday and Friday at regional events, and has even gone so far as to assist in building a robot on Thursday with a team that ended up on the winning alliance come Saturday! I personally would have loved to have experienced what the members of all four of those teams went through to design, build, inspect, and tweak a design like that in a matter of hours. I find the idea of collaborating with some of the greatest well-known minds in FRC (and even those who aren’t well known!) to create an elegant solution to a problem immensely inspiring. Cheesecaking also helps people who enjoy the competition aspect more than the inspiration by creating a more competitive environment. If teams weren’t allowed to make changes after the six weeks, except to replace broken parts, the whole year would go on looking like week one events.

Cheesecaking is something that, to me at least, is a great example of people working together to accomplish a difficult task in as little time as possible. That’s what you will be expected to do in the workforce, even if it means taking cues from your competitors. Rebuilding is a part of the iterative design process and should not be feared to the extent that it is for some.

On the topic of the harpoons themselves, I can’t wait to read the paper on them! I stopped by 1114’s pits on Friday during lunch to see if I could take a look at their robot and was surprised instead by the four harpoons in the process of being assembled. I’m looking forward to hopefully seeing them successfully deployed at IRI this July, so I might be able to see them in action!

The amount of work put into the in-event modifications was nothing short of amazing. I felt I really needed to say how AMAZED I am by the speed at which teams were able to produce new pieces and modifications. An unbelievable feat to witness. Truly remarkable!

My opinion: I support “cheesecaking” (still not the biggest fan of the term). BUT there is a difference between improving or adding a feature to a robot, versus builing a whole new robot.

At multiple events, can grabbers or ramps were added that made a world of a difference for teams. I support this, especially if the team learned from the experience.

in 900’s case, (from what I can tell) a whole entire new AM14U2 chassis and superstructure was created. The only main thing from the original robot was electronics, from what i have read. This in my eyes is a whole new robot (especially since it was even said that 900’s original robot was mechanically intact). I appreciate that all rules were followed to the dot, and especially love how each team dealt with it proffesionally, and I can tell everyone (even 1114) learned from the experieince, but my hopes are that in the future a rule is created that prevents “cheesecaking” to this degree.

Overall im happy (and not upset) that 900 did what they did, and learned so much in such a short time (probably a seasons worth of knowledge), but hope that a rule is created that prevents alterations/improvements to the point of creating a 2nd/new robot.

Side Note: Cudos to 900, for the crazy idea of approaching 1114 about being there “cheesecake platter”. The amount of effort you guys put forth such as CADing a new drivetrain overnight, is beyond inspirational. 1114 could of tried to find a team themselves but 900 put fate into their own hands and made sure they were the chosen team. INSPIRATIONAL, BRILLIANT, AND SMART! 900 has shown that they arent afraid to go with a crazy, out of the box idea (another example: 2014 - shooting the ball almost full court)

My team’s pit was directly next to 900’s and across the aisle from 1114, so we watched this process the entire weekend. First, simple observations - I’ll attempt to avoid opinions:

  • It was clear that 1114 had designed the harpoons before arriving. They had a crew that began working on them almost immediately (Wed.?) in St. Louis. It’s hard for me to know the precise start time as they spent time working on both their robot and the (at the time) unidentifiable cheesecake.

  • It was also clear that the harpoons, by design, were more than a little cheesecake. They would have to replace most, if not all of a robot.

  • Team 900 did start formally working with 1114 Friday afternoon/evening.

  • Though, until Friday, it was clearly the work of 1114, once the Zebracorns were involved, their students were actively involved in the construction process - there were typically more of them around the bot working than 1114.

  • Before eliminations started, team 900’s entire bot (sans electronics?) were put into their crate. They did, in fact, build an entirely new robot.

  • The new robot was clearly very small and only meant to make it legally defined as a robot. It was not going to be picking up and moving the harpoons anywhere. It also had no other ability than to perform the autonomous routine.

  • The students on 900 seemed fine with the arrangement.

At what point does improving a robot become a new one all together? I have no issues with improving a robot, but when you pour your heart and soul into a robot that you call your own, then 95% of it gets replaced by someone else is it really you improving your bot?

I have great love and respect for the teams involved, it just makes me sad to see whole robots thrown aside.

This is a great question, and one I’ve mused over with many people. So far, I’ve found it impossible to come up with a rule wording that can draw the line clearly at a certain amount of cheesecaking- how do you allow an assembled gearbox without allowing an entire arm? How do you allow and arm without allowing an entire robot? A vague rule could be created that would require on-site interpretation for each presented solution, but I don’t see how we can make it a plain black and white difference.

I also think it’s important in this thread to be clear on one thing: any dislike of the rule that allows cheesecaking should not extend to these teams or this situation. They played the game according to the rules given and to the best of their abilities - if this rule hadn’t been present, then I highly doubt we would have seen this take place :). One of the great things about FRC is seeing teams push the limits and boundaries in a gracious manor, and forcing the rest of us to rethink our basic conceptions of how to approach the game challenge. They definitely accomplished that here!

It’s just some metal and electronics. Yeah, work went into it but the point is to inspire and build students, not just robots. Does a robot make the team or does the team make the robot? The determination and drive of a team is what gets you to Einstein, not just a robot.

Our team looks at engineering challenges as just that, engineering challenges. The goal is to win the challenge and come out on top. Sometimes that involves going back to the drawing board. Our strategists predicted a machine like the one we built with 1114 would be required to help win the event and I’m glad we got to build it. It wasn’t going to win a regional but it very well could have decided Einstein.

Now, my opinions…

  • I do not believe the teams did anything against the rules. (Okay, I was wondering whether or not the harpoons would be considered “safe,” but figured that was something the inspectors and/or referees could decide.)

  • I got the idea that the harpoons may not have been fully ready for competition… The last time I saw them fired (during the Curie finals), they grabbed two bins, but knocked the other two over to the other side.

  • I am not a fan of this level of cheese-caking. The harpoons were clearly a device that very few teams were able to construct and, were we to compete against them, we would have felt like we were competing against an alliance of 148-1114-1114. That just does not seem right to me. Moreover, if a team is going to be competing on the championship fields, should it not have to use its own robot?

  • Are we really at a point that the key to being asked onto the #1 alliance is to toss the competition bot you’ve played with all seasons, bring a KoP kit and ask an elite team to cheesecake you? This just strikes me as bad - and quite un-inspirational - especially to teams who were far stronger than 900 this weekend and had to watch the elimination rounds.

  • It also begs the question, with all the emphasis and prestige of winning, does 900 now get to claim they reached Einstein? I suppose. It just seems strange to say so as the team finished in 75th place out of 76 teams and didn’t enter into a single playoff match.

  • I don’t really blame the game-design for this. Designing a game that has never before existed or been played is difficult. It is nearly impossible for a group to foresee all the possible ways teams will interpret and/or play the game before it is actually played. It’s not like they can look at last year’s games and see what this year would be like. No, this is largely a question of culture. What do we as a community consider “acceptable?” How do we define “sportsmanship” and “gracious professionalism”? These are questions FIRST will have to tackle and we’ll likely get some sort of rule clarification from it. Of course, none of that will do any good unless the FIRST community can come to some sort of consensus about whether this sort of activity is acceptable.

I didn’t check (was a little busy Saturday) but I’m willing to guess there are other alliances who had not yet played their 4th bot when reaching Einstein.

Of course they ‘get’ to claim they reached Einstein. They put in more work in one day than I think I’ve ever seen in my now 15+ years as a FIRST observer.

They were on our alliance. Even if just for that one day, we’re a family. We win and lose as a unit.


First of all, 900 is one of my favorite team in FRC. We had a qualification match with them in Archimedes 2014, and they were fantastic to work with. We tried some crazy ideas in our practice match with them before it that most teams wouldn’t have been willing to try (full-court catching strategy. It would have been sick if it worked.
900 consistently thinks outside the box and has an overwhelming desire to win at the highest levels, which I admire about them so much.

From what I’ve heard from everyone involved, it was an incredibly inspiring experience for students on that whole Alliance, and I don’t think there’s anything wrong with it from that perspective.

I can understand why an opponent might be upset at this, but I don’t blame any of the teams involved, I do blame the game design.

Everyone knew from then beginning of the season that finals on Einstein was going to be decided by can races- that was an obvious game flaw, and we knew how an arms race would end- we saw it happen in 2011, where teams with more resources were able to create faster minibots and won a world championship because of it.

Congrats to Team 900 on doing what it takes to reach the Einstein semifinals, as well as their alliance partners and supporters.

I’ll be using 900 as an example of doing what it takes to win for anyone who will listen for the next year or so, until they reach Einstein again.

Bold prediction: 900 reaches Einstein again in 2016. Calling it right now.

Good luck, and I hope to compete with you next year.

You are a brave man. I hope it comes true but only time will tell. I can tell you that next year will be just like this year. My students, fellow mentors, and myself will be working towards the same goal we had this year.