Chief Delphi

Chief Delphi (http://www.chiefdelphi.com/forums/index.php)
-   Rules/Strategy (http://www.chiefdelphi.com/forums/forumdisplay.php?f=6)
-   -   Efficient Points (http://www.chiefdelphi.com/forums/showthread.php?t=154174)

Lil' Lavery 25-01-2017 18:46

Re: Efficient Points
 
Quote:

Originally Posted by Ty Tremblay (Post 1636294)
So does the true bet become "how many teams will claim 12 gears per match in pit scouting?"

We're going to insist on using our pilot so we can claim 13 gears per match.

Joe G. 25-01-2017 19:01

Re: Efficient Points
 
Quote:

Originally Posted by Monochron (Post 1636219)
Assuming a 3 Gear auto, that is only scoring 9 Gears in Teleop. World class teams can probably pull off a 3 Gear auto, and powerhouse teams could probably do 9-10 Gears in Teleop.

I could believe that a specialized, purpose built robot centered around the task, to the moderate or extreme detriment of every other aspect of the game potentially including teleop gear ability, could be built to place three gears in auto, picking the second two off of nonmobile partners' bumpers. It would brush up against the limits of the pilot's ability to place them, as well as a generous interpretation of the edges of the "throwing gears" rule, but I believe it's at least technically feasible.

I expect a team that underestimates the difficulty to attempt a robot centered around this, and I also believe that there are teams with the resources and ability to make this happen. I don't expect them to overlap, and I don't think we'll see it actually achieved. A team with the resources to pull this off is much better off building a "do everything" robot than an auto gear specialist. The point advantage is big, but not big enough to be game-breaking. If a 3 gear auto represented a chokehold or similar condition, I do think we'd see it.

I do think two gear autos are more likely, though they'll be pretty rare due to most powerhouses focusing on getting the fuel bonus in auto. Additionally, by high levels of play, reliable one gear autos will be common enough that spending time on this will become less worthwhile. I think a lot of high level alliances will end up scoring three gears in auto, but it will almost always come from three robots.

PayneTrain 25-01-2017 19:30

Re: Efficient Points
 
Quote:

Originally Posted by Lil' Lavery (Post 1636306)
We're going to insist on using our pilot so we can claim 13 gears per match.

at that point you might as well claim the preplaced gears as well, really throw pit scouting for a loop

Monochron 25-01-2017 22:49

Re: Efficient Points
 
Quote:

Originally Posted by Paul Copioli (Post 1636292)
A 3 gear auto is possible IF the two alliance partners agree to just set their gears on their bumpers and drive away in Auto.

Every historical game I compare the gear cycle shows me there is no way a team will AVERAGE 8 or 9 gears a match by themselves. Some may do it once or twice the entire season.

So that auto would be similar to 2014? We only saw it happen at a handful of events last year, but there it was. I certainly don't imagine that any teams will be averaging 9 gears a match (especially while there is 40kPa to get) but, like you said, we may see it. If I had to guess I would say that a powerhouse team could pull it off, but I may be wrong there. Specifically, the point was about "never" seeing it happen.

Monochron 25-01-2017 22:54

Re: Efficient Points
 
Quote:

Originally Posted by Joe G. (Post 1636313)
A team with the resources to pull this off is much better off building a "do everything" robot than an auto gear specialist. The point advantage is big, but not big enough to be game-breaking. If a 3 gear auto represented a chokehold or similar condition, I do think we'd see it.

Do you think a world class team would need focus their whole robot on gears to do a 3 gear auto?

JesseK 25-01-2017 23:21

Re: Efficient Points
 
3 gears by a single bot in auto ain't happenin'. It isn't about a world class robot - doing so requires a ballet of all 3 robots at warp speed so the pilots have enough time to pull the gears up and spin the rotors.

9 gears is a great 'bucket list' target for a top gear bot. If the bot hits its auto, partners drop their gears, and the opponents are careless with even one gear, that's "only" 5 full-field cycles and 3 close ground pickups. If one partner doesn't do much of anything, and the other isn't confident its auto will hit from the side (and is therefore willing to drop its gear to go down field in auton), this could be a common quals scenario. Seems like a reasonable prediction to say we'll see 1 bot do 9 gears in a match least once a week this season.

Monochron 25-01-2017 23:27

Re: Efficient Points
 
Quote:

Originally Posted by JesseK (Post 1636447)
3 gears by a single bot in auto ain't happenin'. It isn't about a world class robot - doing so requires a ballet of all 3 robots at warp speed so the pilots have enough time to pull the gears up and spin the rotors.

3 from a single bot requires a ballet from all 3?
I think it won't happen because teams will prioritize other scoring options in Auto.

Lil' Lavery 25-01-2017 23:46

Re: Efficient Points
 
Quote:

Originally Posted by Monochron (Post 1636433)
Do you think a world class team would need focus their whole robot on gears to do a 3 gear auto?

Yes. Multi-gear autos are harder than multi-tube autos in 2011, and the only team that approached a 3-tube auto in 2011 never pulled it off in real play. When trying to score multiple game pieces in such a restricted amount of time, it involves some serious optimization to complete that task. That means sacrificing some other features to optimize for that one. There's one very specific commonality you can find with almost every robot that could score additional game pieces in 2012 and 2016, as well as with 233 in 2011 and 254 in 2014.

Joe G. 26-01-2017 00:20

Re: Efficient Points
 
Quote:

Originally Posted by Monochron (Post 1636433)
Do you think a world class team would need focus their whole robot on gears to do a 3 gear auto?

Yes. The only real point of reference we have for this type of auto is 233's 3 tube auto from 2011. We'll say that the slight decrease in distance travelled is roughly cancelled out by the increased pickup difficulty and the need to complete the task in time for the pilots to act on it, so a robot doing a 3 gear auto needs to be about as "optimal" for the task as 233's machine was.

In 2011, 233 posted this video, never pulled off the feat in competition in large part because competent single-tube autos were common enough that it was rarely even attempted, and no other team came even close to achieving this, with most two tube autos being buzzer beaters for the 2nd tube. 233 was so much faster than everyone else because of a feature somewhat unique to their robot that year, the ability to pick up from the reverse side of the bot, and score on the other. In fact, they actually gain significant reach due to their arm length on the pickup, greatly reducing the distance the robot has to travel. It follows that a three gear auto likely has to follow a similar pattern, never turning around, and reducing travel distance of the drivebase by moving the gear forwards within the robot as it moves. Certainly, at minimum, the gear has to be picked up on the back of the robot, and scored on the front, so that the robot doesn't have to execute four precision 180 degree turns. Designing for the gear to pass through the middle of the robot like this provides little or no benefit during teleop, and would seem to make the having even a bare-minimum size fuel hopper nearly impossible.

This is not an equivalent case to the other three gamepiece auto of note, 254 in 2014. Putting a single ball through a high goal was a harder task than placing a single gear this year, and even the best teams in the world missed a non-trivial amount of the time. 254's 2014 robot was probably the most accurate and effective finishing robot in the world that year. Even teams with excellent autonomous modes gladly gave up their balls to put them in 254's robot, and the build style that enabled the three ball auto also provided extremely tangible benefits to 254 throughout the match. Here, every sketch I've come up with that seems like it could possibly achieve this makes me ask "but why give up so much," especially when I consider the relatively low difficulty of a single gear auto, and the number of teams I expect to be achieving this with competency at high levels of play.

Donut 26-01-2017 02:35

Re: Efficient Points
 
Quote:

Originally Posted by JesseK (Post 1636159)
Point 2 makes an assertion that in matches where a gear threshold was met (3rd or 4th rotor) that the win margin will be much greater than for the losing matches when the gear threshold was not met. It means that this game is about hitting the marks for the rotors, and win/loss will likely be determined by who can do that more reliably.

Point 3 does contradict Point one, but Point 1 was a supposition rather than an assertion. Your second paragraph is correct - I presume that more points = more wins, and since starting the next rotor means many more points than an average cycle of balls into the boiler, it is likely a gear bot will win more. Sorry this wasn't clear.

I agree, rotor count is pivotal in this game. I don't agree a gear bot will win more often, and I'll outline how I think fuel is effective below, but we really just have to wait and see how this game plays out to know.

Quote:

Originally Posted by Ginger Power (Post 1636184)
...In the case of the Super Cycle robot, continuing the assumption that you're with 2 average robots, you'll be able to deliver 3+(2 or 3)+(2 or 3) = 7 - 9 Gears. I personally like this situation much better. You have a decent buffer against defense/mistakes and you are also shooting 3 hoppers worth of Fuel into the Boiler. The value of this Fuel will be completely dependent on the effectiveness of the shooter, but some Fuel points are basically guaranteed. These Fuel points will more often than not be the difference in matches where both alliances turn 3 Rotors and have the same number of climbs.

TL;DR: Designing your robot to handle solely Gears is risky because you're dependent upon alliance partners to achieve a Rotor advantage vs. the opposing alliance.

This is similar logic to what I was thinking. If alliances in general tend to end up with 3 rotors, then fuel will be the difference, assuming climbs work out the same.


The important thing to remember in comparing gears versus fuel is that gears have a non-linear points for effort curve. The first two gears are worth a lot of points for the effort to score them, especially since your alliance can start with those gears in robot already. The next 4 take quite a bit more work though, and the last 6 are going to be beyond the average alliance capability. Scoring 2 gears is easier than scoring substantial quantities of fuel, but 4+ is a different story.

My team is pursuing fuel as our top priority, with gears as a second. The key to be effective with fuel, is to count on staying within 1 rotor of a gear focused alliance.
  • Our alliance has to ensure we get at least the 2nd rotor since those are easy points and we can't overcome an 80+ point gear deficit. If our alliance can't manage 2 gears in a match there's no point in further strategic analysis, we lose.
  • Next priority after the 2nd rotor is to go after 40kPa in the boiler. This will be worth 40 points in the match, equivalent to the next rotor, while guaranteeing 1 RP.
  • Priorities from here depend on alliance composition on both sides. If our opponents have no chance of getting to 12 gears and can't score fuel or out score us in autonomous/climbing, we get to 41kPa and guarantee the win, since they can't score more than 40 gear points above our total.
  • If our opponents do have an edge in autonomous or can score a small amount of fuel, but can't reach 12 gears, we want to ensure we match their rotor count. This means scoring 2 and playing heavy defense to limit them to 5 or less gears if our alliance can't manage 6, or getting to 6 ourselves. At that point we are relying on outscoring them in fuel since we have a more dedicated fuel scorer than they do. If the opponents are very skilled with gears and get to 6 very early in the match, the fuel bot may have to abandon the boiler at some point to run counter D to ensure 6 gears get scored and rely on the fuel up to that point as the tiebreaker.
  • If our opponents can get 12 gears, you either play enough defense to hold them to 11, or you get to 6 gears and score 41kPa.

We're trying to ensure we can do 2 gears a match while "super cycling" as Ginger coined it to ensure we get the 2nd rotor and pull our weight for the 3rd one. From there we are all fuel focused to give a 1 rotor cushion and get the RP. We're not doing climbing because we don't want to overextend ourselves and our team is not great at building mechanisms for high loads. We feel we will be more reliable handling balls and gears than trying to lift our robot's weight (and potentially falling). I also am in the minority camp that thinks climbing will be difficult to do reliably in match, and if it isn't then someone can cheesecake a climber onto us for elims. Based on this assumption I think most matches will have 1 successful climb per side and will balance out. Since we can't climb we can sell out on defense at match end to prevent an opposing climb if needed.

Joseph Smith 26-01-2017 06:52

Re: Efficient Points
 
Quote:

Originally Posted by Monochron (Post 1636429)
So that auto would be similar to 2014? We only saw it happen at a handful of events last year, but there it was. I certainly don't imagine that any teams will be averaging 9 gears a match (especially while there is 40kPa to get) but, like you said, we may see it. If I had to guess I would say that a powerhouse team could pull it off, but I may be wrong there. Specifically, the point was about "never" seeing it happen.

Three ball auto in 2014 was significantly easier than a 3 gear auto would be. My team's robot was designed from the ground up that year to control 3 balls at the same time specifically for auto. (Though we mostly stopped running our 3 ball auto after districts, because most teams had a reliable one ball by state champs. We did run it successfully in all but one of our matches at Troy district though.) This year, with only being able to control one gear at a time, there's a lot more precision and speed required.

SweetSalty 26-01-2017 09:10

Re: Efficient Points
 
Honestly I believe that Fuel bots will be more meta this year. They can efficiently if done correctly apply points to the team and can hold their own for most of the match

SweetSalty 26-01-2017 09:12

Re: Efficient Points
 
Quote:

Originally Posted by SweetSalty (Post 1636537)
Honestly I believe that Fuel bots will be more meta this year. They can efficiently if done correctly apply points to the team and can hold their own for most of the match

I wish I could use Dean's computer to go back in time and change the ball design for this years game.

pmattin5459 26-01-2017 09:14

Re: Efficient Points
 
Even if your robot was specifically designed for gears and picked up on one side, dropped off on the other, there's still the problem of trying to pick up gears that have fallen in a random position from an alliance partner's bumper. Plus, you would most likely only use one gear peg, as that is much easier and faster than using different pegs each time. Could a 3-gear auto be done? In theory, yes. But your robot has to be very specifically designed to do so, and the advantages offered are somewhat minimal compared to the challenge of doing so (I gather that this year, the ability to handle gears, however minimally, is extremely valuable and should be on every robot).

JesseK 26-01-2017 09:22

Re: Efficient Points
 
Quote:

Originally Posted by Donut (Post 1636496)
I agree, rotor count is pivotal in this game. I don't agree a gear bot will win more often, and I'll outline how I think fuel is effective below, but we really just have to wait and see how this game plays out to know.

We will have to see how this plays. Right now the only real comparison we can do is 1-bot vs 1-bot, and then start running rock-paper-scissors-lizard-spock scenarios with permutations of types and capability-levels of partners.


All times are GMT -5. The time now is 08:45.

Powered by vBulletin® Version 3.6.4
Copyright ©2000 - 2017, Jelsoft Enterprises Ltd.
Copyright © Chief Delphi