The cheesecake runaway

[Paul Copioli]

First off, I tend to agree with DR. Joe on this one, but don't expect everyone to agree with me.

However, I think something needs to be changed with respect to "Cheesecaking" (thanks Corsetto for such a horrible word for this activity), "fix-it crews", etc.

For those of you not paying attention, this has been going on for years. It has manifested itself from fixing broken partners, helping a team pass inspection, and modifying teams to be more compatible for eliminations.

In many cases cheesecaking was stripping off an arm and putting additional weights (anvils, steel plate, bags of bolts, etc.) on a third partner in order to play defense.

This year, increasing a teams functionality with can grabbers was the most popular method of cheescaking. This is directly related to the game design.

Game design can definitely influence the type of modifications teams will make at an event.

To be clear, I am not an advocate of making rules to specifically disallow adding functionality to teams at events, but I am personally against selecting less functionally capable teams because they are better cooking pans for my cheesecake.

This is is a tough subject to address, unless the game just doesn't encourage it. This year's game definitely encouraged it.

[marshall]

Originally Posted by Paul Copioli

This is is a tough subject to address, unless the game just doesn't encourage it. This year's game definitely encouraged it.

I think I agree. I think it has to be addressed through game design. The game this year massively encouraged it. I think the games have to be designed such that they discourage it. To be clear, I don't think that is an easy thing to do but I think it has to be thought about.

[Lil' Lavery]

Originally Posted by Paul Copioli

For those of you not paying attention, this has been going on for years. It has manifested itself from fixing broken partners, helping a team pass inspection, and modifying teams to be more compatible for eliminations.

Or in some cases, fixing your opponents after they eliminate you.
http://www.chiefdelphi.com/forums/sh...d.php?p=620506

[Spartan710]

Originally Posted by Andrew Lawrence

It's not for the sake of winning, it's for the sake of the alliance. I always say that when you get to the elimination rounds, it's no longer just about your team - it's about the success of the entire alliance. If you're not putting in your all to achieve your alliance's goals (which should be inspiring your students through accomplishment and success [aka winning]), then you're doing a disservice to the other teams on your alliance.

Each team should ask not what their alliance can do for them, but what they can do for their alliance.

I would like to say I totally agree with this. Last year we we were the third pick of the poofs. My team so excited that a team like that would pick us out of the 60 teams that remained. Though we did not get cheesecaked we did what was best for the alliance. We only played one match because Allsparks had issues. It made sense for us to sit out because the kiss pass between 469 and 2848 was much smoother and 2848 was a taller robot, so defense was easier. We were just happy to help when needed. So we took a lot for this experience and would do it again in a heart beat.

[JaneYoung]

Originally Posted by Paul Copioli

First off, I tend to agree with DR. Joe on this one, but don't expect everyone to agree with me.

However, I think something needs to be changed with respect to "Cheesecaking" (thanks Corsetto for such a horrible word for this activity), "fix-it crews", etc.

For those of you not paying attention, this has been going on for years. It has manifested itself from fixing broken partners, helping a team pass inspection, and modifying teams to be more compatible for eliminations.

In many cases cheesecaking was stripping off an arm and putting additional weights (anvils, steel plate, bags of bolts, etc.) on a third partner in order to play defense.

This year, increasing a teams functionality with can grabbers was the most popular method of cheescaking. This is directly related to the game design.

Game design can definitely influence the type of modifications teams will make at an event.

To be clear, I am not an advocate of making rules to specifically disallow adding functionality to teams at events, but I am personally against selecting less functionally capable teams because they are better cooking pans for my cheesecake.

This is is a tough subject to address, unless the game just doesn't encourage it. This year's game definitely encouraged it.

I'm going to digress a bit. Forgive me.

Mentors like Paul carry the knowledge and the history of this ever-evolving program. Important and relevant knowledge and history. This would make a wonderful conference topic at the Championship Event, presented by a distinguished and knowledgeable panel.

Jane

[rick.oliver]

This has been an interesting read.

First, I will point out that minibots were specifically encouraged to be shared and teams earned Coopertition Points for sharing their minibots. So, I don't consider those as having been "cheesecake" ... more like "ice cream" for my "apple pie".

I believe that it is appropriate for an alliance captain to assemble the strongest alliance possible ... this is after all a competition. And I agree with Carolyn_Grace that each team may choose what measures they may take to fulfill their own objectives and goals.

As long as everything is done within the rules, I have no complaints. I agree with Paul Copioli that rules should not be implemented to discourage or inhibit teams from helping other teams increase their functionality.

I do think that "cheesecake" should be a gift which remains with the receiver, but that, too is between the teams involved.

[marshall]

Originally Posted by rick.oliver

This has been an interesting read.

First, I will point out that minibots were specifically encouraged to be shared and teams earned Coopertition Points for sharing their minibots. So, I don't consider those as having been "cheesecake" ... more like "ice cream" for my "apple pie".

I believe that it is appropriate for an alliance captain to assemble the strongest alliance possible ... this is after all a competition. And I agree with Carolyn_Grace that each team may choose what measures they may take to fulfill their own objectives and goals.

As long as everything is done within the rules, I have no complaints. I agree with Paul Copioli that rules should not be implemented to discourage or inhibit teams from helping other teams increase their functionality.

I do think that "cheesecake" should be a gift which remains with the receiver, but that, too is between the teams involved.

I feel like this is a really enlightened stance on the subject.

[mprikril]

Let me state that what was accomplished to 1923's and 900's robot in that amount of time was nothing short of amazing. Please don't think this was simply bolting a new mechanism on a robot, I was able to witness three teams come together to achieve something I would have thought impossible had I not witnessed it. The work required to design something that would function and (mostly) fit, then to get it mechanically integrated, then to get the controls in place in time for the matches on Curie is just nuts. They pulled it off.

That being said, I am curious on how robots are inspected mid-competition against rule compliance for cost of materials used if a part from another team is on their robot.

In the past, if I design a part for my team’s robot, my time is not charged to the cost of that part since my company is a partner of this team. If I needed a complicated part made by wire EDM and had to go to an outside machine shop, the cost of that part (which includes its labor) would have to be captured on our BOM. Even if that vendor machine shop did us a favor and did the work for free, we would have to capture the fair-market value of the part if they aren’t a team partner.

If one robotics team designs a part and has the on-site machine shop build it, wouldn’t the receiving team for that part be required to account for the engineering design time on the cost of that part in their BOM? How is donated time at events captured? How about design time that was “spent” prior to the event but for parts given to other teams at the event?

[BrendanB]

I too would agree with some of the opposing arguments against Cheesecaking but that is a discussion for another day.

Regardless of those opinions it was hard not to walk through the Curie pits by where 148, 900, 1114, and 1923 were setup and not be blown away by the dedication and drive these teams had. As it has been mentioned before the 148 & 1114 pits were a non-stop Nascar pit crew doing everything they could to prepare for Saturday. I'm sure their work in the pits in St. Louis barely touches the surface of what they have been doing in the weeks leading up to St. Louis and reflects on how much their entire team gets it with mentors and students working side by side with each other and collaboratively among teams.

While this specific Curie alliance has been talked about a lot with their Cheesecake this weekend there were countless teams doing the same on Saturday morning and again it was an amazing sight to behold with teams working together to do what they could to have the highest average.

Its all in how the game was created that those four green buckets created a huge demand among alliances looking to make it far in the tournament and many alliances including the 148, 900, 1114, and 1923 alliance did so to the best of their ability while following the rules.

[George Nishimura]

I am struggling to see how game design determines anything but the type of modification. Based on the incredible feat performed by team 900 and co, how long would it take them to build a great d-bot for any of the previous games? Why can't teams give out custom kitbot-ready gearboxes?

With increased availability and ability of COTS parts, it's become a harder problem to outlaw.

I always thought the rules were bound to have loopholes, but it was up to the teams themselves not to try and exploit them.

Unless FIRST tries to explicitly ban it again, there's a culture change, or the role of the third robot becomes trivial, I think "cheesecaking" could be here to stay.

[BigJ]

Originally Posted by George Nishimura

I am struggling to see how game design determines anything but the type of modification. Based on the incredible feat performed by team 900 and co, how long would it take them to build a great d-bot for any of the previous games? Why can't teams give out custom kitbot-ready gearboxes?

With increased availability and ability of COTS parts, it's become a harder problem to outlaw.

I always thought the rules were bound to have loopholes, but it was up to the teams themselves not to try and exploit them.

Unless FIRST tries to explicitly ban it again, there's a culture change, or the role of the third robot becomes trivial, I think "cheesecaking" could be here to stay.

The cost to benefit ratio of any "cheesecake" for a game task that isn't "Do this or lose by default" (read: can race) is probably not good enough.

[efoote868]

Originally Posted by George Nishimura

I am struggling to see how game design determines anything but the type of modification.
...
Unless FIRST tries to explicitly ban it again, there's a culture change, or the role of the third robot becomes trivial, I think "cheesecaking" could be here to stay.

This game might have been a perfect storm of conditions; cheesecaking doesn't have to be specifically outlawed to achieve the desired interaction (or lack there of).

If the game pieces, floor space and opportunity to score isn't as limited, a 3rd bot can contribute more than just a platter to hold cheesecake (decrease reward).

If the interaction between robot and game piece is non-trivial and requires more fabrication than a bolt on mechanism, cheesecaking becomes much more difficult (increase risk).

[George Nishimura]

Originally Posted by BigJ

The cost to benefit ratio of any "cheesecake" for a game task that isn't "Do this or lose by default" (read: can race) is probably not good enough.

Cheesecaking occurs when there is a gap between the necessary contribution of a third robot and the actual ability of the pool of possible third robots. The difference in this year's game is that each alliance needed a significant contribution from the third robot beyond having a basic drivetrain. I don't know if anyone sees that as "bad game design".

It's interesting to me that the World Champions in general never needed to rely on a cheesecaked robot (I believe that's true). They just had an incredible steal of a third robot (1671).

How I interpret comments such as "it's the games fault" is that the game is at fault for making a game where you need three good robots to succeed, and that making a "good" robot is an actual engineering challenge. Or that one such robot could be engineered within the witholding allowance.

[BigJ]

Originally Posted by George Nishimura

...The difference in this year's game is that each alliance needed a significant contribution from the third robot beyond having a basic drivetrain. ...

That "significant contribution", at least for some number of elimination alliances was:

• Absolutely required to not just lose in a match where all center cans would be taken by an opponent otherwise
• A small enough portion of the match that a vast majority of teams had better things to be working on to attempt to seed highly and contribute points in qualification rounds
• difficult enough to do on a competitive level that very few capable robots would be left for deep 3rd/4th picks
• Easily capable of being small enough, light enough, and autonomously operated (by necessity) to bolt onto an alliance partner's robot

Only the 4th point there is not a fault of the game design of Recycle Rush, in my opinion.

[ASmith1675]

Originally Posted by George Nishimura

It's interesting to me that the World Champions in general never needed to rely on a cheesecaked robot (I believe that's true). They just had an incredible steal of a third robot (1671).

How I interpret comments such as "it's the games fault" is that the game is at fault for making a game where you need three good robots to succeed, and that making a "good" robot is an actual engineering challenge. Or that one such robot could be engineered within the witholding allowance.

It's the game's fault that it was even a viable strategy to play with two robots. Its also the game's fault that they needed a third robot with a very specific attribute (sub-0.5s can grabber) which could be relatively easily attached to another robot (as we saw).

You make a good point about the world champs... so its definitely not an instant win button that all teams are going to use from now on. I really think this game encouraged this type of strategy by its nature. The parts we've seen added in the past have been largely passive and defense oriented. I never saw many complaints in that area.