Ultimate Ascent - Most Difficult Ever? Too difficult?

[rick.oliver]

I agree that this challenge is the most difficult to solve completely as compared to what I've seen in my 11 seasons with FRC. I will not be surprised to see teams which find a complete solution. We applied the strategy development process which Karthik trains; it drove us to a decision early.

We evaluated the scoring options and have opted to focus on scoring discs in the high goal and hang for 10 points as our highest priority and then work in the ability to pick from the floor and score high as weight and space allow.

The additional 20 or 30 points (if the GDC decides to increase the value) did not seem worth the effort and risks.

I am really looking forward to seeing all of the solutions. I think this is going to become my favorite game of all time.

[Grim Tuesday]

Not only is this game difficult to design for, it's difficult to analyze. Every time I look at the score distribution I think one of two thoughts:

"Wow the pyramid is really overpowered"

and

"Wow the pyramid is really underpowered"

I think this arises from an extreme difficulty of estimating how other teams will be able to play this game. Disc shooters we've seen on CD are all pretty good, reasonably accurate. I haven't seen any good way of getting the discs into the shooters prototyped nor have I seen any climbers beyond level one. But if the shooters are as good as is to be believed from what we've seen so far, the effort spent to climb seems like a big waste: If your climber takes 30 seconds and you can score one trip of white discs and climb for 10 in that time, then you've just scored 22 points making that very complex and fancy climber worth 8 points.

But on the other hand, I wonder if it will turn out like Rebound Rumble where some factors arise (ball squishiness for instance) -- maybe this year it will be getting the disc into your shooter, or your drive train not getting caught on them as you drive around the field -- that will cause the average disc score to go down precipitously. Maybe it will be the rampant defense that is conceivable. In this case, climbing is worth it no doubt.

But that's the whole issue: I don't know which will end up happening. We chose one of these and moved on. I think this game is an excellently designed one, at least for the difficulty of analyzing it. Every time you start thinking one way is definitively better than another, something (cough cough colored discs) comes out and proves you wrong and brings both ideas back to balance.

[Anupam Goli]

This year's game most certainly is interesting, and the end game is something I don't think we've ever seen, but I'm not sure to say it's the most difficult or even too difficult, but rather say that last year was likely one of the (relatively) easier games that FIRST has come up with. This year's game piece is entirely new, and I will admit, I was skeptical at first at being able to fire frisbees, but now I'm more confident in our ability to do so accurately. Compared to last year, with a very similar game piece to 2006, and somewhat similar end games to 2001 and 2006, I think this year's game is certainly more difficult than 2012, but I don't want to say it's too difficult, just this past year we were given an easy game and now we're thrown into a harder one.

[Jay O'Donnell]

Originally Posted by Grim Tuesday

"Wow the pyramid is really overpowered"

and

"Wow the pyramid is really underpowered"

I would say it's neither. Don't be surprised if it is a good balance this year between the three portions of the game.

[pfreivald]

Originally Posted by Grim Tuesday

Not only is this game difficult to design for, it's difficult to analyze.

Interestingly enough, I think it was very easy to analyze the "killer robot" application -- a great shooter with floor pickup as well as slot loading, that can climb and score disks in the top of the pyramid. The hard part is translating that down into a doable robot -- which functions are critical, which are not?

[Grim Tuesday]

Originally Posted by pfreivald

Interestingly enough, I think it was very easy to analyze the "killer robot" application -- a great shooter with floor pickup as well as slot loading, that can climb and score disks in the top of the pyramid. The hard part is translating that down into a doable robot -- which functions are critical, which are not?

Well, arguably in every game the 'killer' robot would be the one that can do literally everything. It's just that some years those functionalities are compatible (tube hanging and minibot scoring, ball scoring and bridge balancing, etc...) whereas this year requires a true tradeoff for all teams but the very elite. Like I said, what makes this so difficult is the difficulty of estimating what your opponents will be able to do.

[dellagd]

Originally Posted by Grim Tuesday

Like I said, what makes this so difficult is the difficulty of estimating what your opponents will be able to do.

So true. I really have no idea how well we're doing with our progress and how well others will do too.

I don't like it

[MrForbes]

I don't think it matters too much what your opponents will do. You do the best you can do, and that depends on what your team is capable of doing. As I see it, the "easy points" come from flinging Frisbees into the higher point goals, and from a low hang. The difficult points come from climbing, and dumping Frisbees into the pyramid goal.

But we don't all see things the same.

[jhaugen]

I believe the key to this competition is to choose your strategies and be consistent.

[Andrew Schreiber]

Originally Posted by Grim Tuesday

Well, arguably in every game the 'killer' robot would be the one that can do literally everything.

I give you as a counter-example FRC71 in 2002. Arguably one of the most dominant robot to ever exist in FRC. It didn't touch the balls in that game at all. (video)

I've said it time and time again, a ROBOT does not win championships. An ALLIANCE does. Build a robot that does it's job the best and build out an alliance that plays the game the best.

I'm glad this game is "hard" I think that it forces us to make choices for the first time in many years.

[Billfred]

Originally Posted by Andrew Schreiber

I'm glad this game is "hard" I think that it forces us to make choices for the first time in many years.

Agreed. In my eyes, the last game that really made you make a conscious "We're-gonna-do-that!" decision was FIRST Overdrive. Freshmen that played that game are now sophomores in college.

[Kusha]

I think one of the difficult things about this game is that it's fairly unique, and that uniqueness makes it hard to predict how other teams are going to play this game. That the time allocated vs everything you can do makes this game more difficult than the others I've participated in.

It'll be nice not to see robots that look really generic, unlike last year.

[karomata]

Originally Posted by Koko Ed

Has FIRST ever made a game that was easy to do?

Overdrive looks the simplest, but FIRST added the aspect of the track balls to complicate it, so I don't think you can support an argument that FIRST has made games that are too easy.

[Kevin Ray]

Originally Posted by Andrew Schreiber

I give you as a counter-example FRC71 in 2002. Arguably one of the most dominant robot to ever exist in FRC. It didn't touch the balls in that game at all. (video)

I've said it time and time again, a ROBOT does not win championships. An ALLIANCE does. Build a robot that does it's job the best and build out an alliance that plays the game the best.

I'm glad this game is "hard" I think that it forces us to make choices for the first time in many years.

I agree with Andrew. In fact, a better video to prove your point would have been the Einstein finals for that year when Team Hammond (FIRST Team 71) was finally stopped in the finals when the opponents were able to race across the field and knock them off target. 71's ally, FIRST Team 173 Team Rage vacuumed almost all of the balls on the field and scored enough points to win the nationals--AS ALLIES.

It seems as though the GDC hit on the perfect formula for this game. There are aspects that understaffed and underfunded rookie teams can still do well (shoot frisbees) while giving elite teams a run for their money if they wish to continue being the "Do everything very well" type of robots. The middle of the road robots can excel in one or the other and still make themselves very competitive by doing so.

[Pault]

I feel like the challenge in this game isn't that every task is too difficult to perform, it's a combination of the game being hard to predict and parts of the game being too enticing.

The full court shot is a great example of this. At first glance, it seems as though it is your team's pathway to having an absolutely amazing robot. But it is hard to tell if it will work in competition, mainly because of defense. A big question for anybody considering a full court shot is "will there be robots who can block us?". Depending on the answer, you could either have the next 469 bot from 2010 or a strategy that only works if you have a lucky match schedule.

Another thing that is hard to predict is how many disks will be on the floor, which will affect decisions on whether to have a pick up system or not. The 30 point climb is an enticing thing, but many teams probably do not have the resources to achieve it*. Even 341 said in their vlog that they were struggling with it.

So is this a difficult game? Depends on your definition on difficult. I think that by far the most difficult part of this game is deciding what your robot will do.

*Of course, unless there is some super easy climbing method which I am just oblivious to.