Game Design and the Perils of the Vacuum

Below is a series of the eight potential field profiles that can be generated through the randomization of the plates as described in the manual, WPILib documentation, and further addressed in this Q&A question. (chart by Matt Lythgoe)

You will quickly notice how 4 of the 8 profiles are symmetrical, and 4 of the 8 are not symmetrical. Is this good? Is this bad? I don’t know. Is this providing an advantage to one alliance over the other? It depends on a number of factors and assumptions we can’t confirm without a real match with real robots in the books yet, so it’s hard to say. It’s somewhat agreeable that different plate assignments will result in different sides of the field to target. I would like to centrally argue that the middle four profiles will, without question, force opposing alliances to play the game in 2 different ways at the same time. This effect will

a) end up causing an unnecessary controversy in a high level match and
b) be like a series of other GDC decisions made in the past, somewhat avoidable, if not at least recognizable.

I am taking a giant leap onto a thinly padded “jumping to conclusions” mat here, but it feels like we have seen design challenges be tacked onto the game ever since most every robot in Aerial Assist did the exact same thing, if not look similar.

The 2015 season featured a game that was pretty much a design challenge top-to-bottom. No real matches were played excluding finals elimination matches. No wins, losses, or ties were assigned until the second, or third to last match at an event. This game, to its credit, brought about some of the most impressive and dominant robots I have ever seen, but I would wager that most veterans of the community have been somewhat denigrating of its memory, and the GDC itself has yet to really incorporate any newly introduced elements from 2015 into later games.

The 2016 season featured two attempts at a design challenge: one low bar that a lot of average teams could actually use to their advantage, and a pair of largely underused and occasionally derided defenses. I remember the chorus of groans after the audience selected the Drawbridge at the CHS DCMP in 2016. Very few teams incorporated the completion of that task into their design, and fewer still spent their whole season using that ability to their match-deciding advantage.

2017 brought us a pair of very frustrating design challenges. A spring that performed as expected (it broke a lot), and a hopper switch that never wanted to trigger in the way described by the GDC in the manual. The hopper switches could have swayed some Einstein matches if not for luck of the draw or larger factors coming in to play. These were unfortunate mechanical design inconsistencies that could not be fixed within the season, on top of what I can only assume were hundreds of thousands of dollars spent on keeping the field functioning last year.

Now, circle back to the plates. The randomization factor has resulted in a unique design challenge. I think this is definitely great! It gets decidedly less great after thinking about possible answers to these 2 questions:

a) How many auto routines does the GDC expect teams to complete with 4 possible field configurations for the part of the field they can access in auto when they also make you put the robot in a bag 45 days after they tell you this? (Keep this in mind for future threads or interactions where you hear someone spin the web of lies that we are building these machines for $5000 in 6 weeks)

b) Does the GDC not expect there to be a significant difference in the inside patterns, or do they not care?

I feel as though we are in an unnecessary pattern of the GDC creating design challenges for the robots in the vacuum of the game design, without considering the potential factors to their fullest intent (Will this result in a fair game? Will this result in a robust field that can be replicated to an acceptable standard of quality over 2 dozen times? Will this actually be a design challenge teams care about? Will this game actually be a game?

The window is still open to further marinate on and take action on a potential change. I think considering the removal of the profiles 2, 3, 6, and 7 from the chart above (when reading left to right, top to bottom) should be something looked in to by the GDC now.

Does this alone sound crazy?