Chief Delphi

Chief Delphi (http://www.chiefdelphi.com/forums/index.php)
-   General Forum (http://www.chiefdelphi.com/forums/forumdisplay.php?f=16)
-   -   Percentage of EFFECTIVE Limbo Robots (http://www.chiefdelphi.com/forums/showthread.php?t=142944)

BenGuy 01-02-2016 09:51

Re: Percentage of EFFECTIVE Limbo Robots
 
I think that most teams that try to get under the low bar will be able to. If you're going to design your entire robot to fit under that bar, most teams are going to make sure they can.

Andrew Schreiber 01-02-2016 09:56

Re: Percentage of EFFECTIVE Limbo Robots
 
Quote:

Originally Posted by BenGuy (Post 1533327)
I think that most teams that try to get under the low bar will be able to. If you're going to design your entire robot to fit under that bar, most teams are going to make sure they can.

In 2014 most teams designed their robots to hold the ball. Most teams couldn't.

(I have more examples)

wjd13 01-02-2016 09:57

Re: Percentage of EFFECTIVE Limbo Robots
 
I definitely think that the number of effective Limbo Robots will be greater after week 4 than before, in the early weeks. After teams try using their robots and strategies in a real match, they'll find the flaws, and be able to fix them as the season goes on, whether on their second robot, in the programming, or whatever other method.

I also think that the best limbo bots need to shoot goals of some sort, they can't just breach the defenses and be done. This is because, as mentioned earlier, the biggest benefit of being able to go under the low bar is the faster cycle time for getting boulders. But if you don't design a robot that can take advantage of this, you'll only be mediocre at best.

dirtbikerxz 01-02-2016 10:01

Re: Percentage of EFFECTIVE Limbo Robots
 
Its honestly not too hard to design a effective robot that can go under the low bar, but yes I agree with the fact that most rookie teams will not be on chief delphi, so all these percentages that we are seeing in the polls represent a good number of teams that already have experience with or have an idea on how to design how to do this.

Our robot will be able to cross every defence, and shoot in low and high goals. The only function im not a hundred percent sure will function is scaling the wall, but even that has a 90 percent chance of working (will find out later this week).

IndySam 01-02-2016 10:25

Re: Percentage of EFFECTIVE Limbo Robots
 
Quote:

Originally Posted by Andrew Schreiber (Post 1533314)
Maybe I'm just a grumpy old man already.

Welcome to the club.

dirtbikerxz 01-02-2016 10:41

Re: Percentage of EFFECTIVE Limbo Robots
 
Quote:

Originally Posted by Andrew Schreiber (Post 1533314)
0-19%... why? Because that's about the percentage of robots that seem to reliably do the game challenge anyway.

How many will continue to tell you they do it and are 110% accurate into the high goal? If the number of teams in 2014 who told me they always scored a ball in auto is any indicator it'll be something like the remaining 81%. And they'll all scream about me being "un GP" when I point out data to the contrary of their delusions.

I totally agree with you, the number of teams last year that told me (during scouting) that they could do multiple x stacks (like 3 stacks of 4 or etc), and would have them ready for me to cap off were huge. But when we got to the actuall match, they crapped out, did nothing, or could do maximum stacks of two.... at maximum. Some teams even pointed me towards their reveal video that showed them doing stacks of 5 and 6, but that was in a perfect world, on the field, they couldn't even do stacks of 3.

And one thing that really annoyed me, is when some teams had bots that were nothing but a drive train, and a simple grabber arm that could lift one tote at a time (don't get me wrong there is nothing wrong with having a not as good bot, everyone has to start learning somewhere), but what annoyed me was when they started claiming they could stack a 4 or 5 tote high stack AND a recycle bin on top, but they could barely push one tote onto the scoring zone.

JesseK 01-02-2016 10:42

Re: Percentage of EFFECTIVE Limbo Robots
 
If a team demonstrates the ability to go under the low bar, and then in the next match demonstrates the ability to bolt on a scaling mechanism that precludes low bar but enables the scaling, is that still considered effective for the low bar?

Billfred 01-02-2016 10:56

Re: Percentage of EFFECTIVE Limbo Robots
 
Quote:

Originally Posted by JesseK (Post 1533373)
If a team demonstrates the ability to go under the low bar, and then in the next match demonstrates the ability to bolt on a scaling mechanism that precludes low bar but enables the scaling, is that still considered effective for the low bar?

If it's bolt-on-bolt-off, I'd give it to them. Operational flexibility never hurt anybody!

IronicDeadBird 01-02-2016 13:06

Re: Percentage of EFFECTIVE Limbo Robots
 
I've been eyeing this thread a lot and I didn't know how to approach it but I think a robot will be effective at going under the low bar is a weird thing to say (like everyone else has pointed out). Otherwise "effective limbo" is just teams that can utilize the low bar so it wouldn't actually boil down to an if you can, but an if you do does it help more statement.

MrJohnston 01-02-2016 13:17

Re: Percentage of EFFECTIVE Limbo Robots
 
When I try to define an "effective" low-bar robot, I consider this:

* Should be able to cross the low bar during autonomous play.
* Must be able to easily go under the low bar with minimal slowing.
* Must have some primary role based on its ability to go under the low bar (cycling through secret passage - and scoring most of the time, damaging other defenses, very quickly ferrying boulders through for an opponent, carry boulders across several different defenses)
* Must either be so good at its primary purpose that there is generally no reason to do anything else OR must have a solid secondary purpose.
* Climbing and Autonomous points not involving the low bar are bonuses.

Examples:
Robot A: Can go under the low bar, pick up a boulder from teh secret passage and score it in the low goal. During tele-op, it generally can manage four full cycles, scoring twice.... It can cross Category B and D defenses, but is labored in doing so and will often lose a boulder in the process. Auto: Can cross the low bar - but nothing else. I view this robot as "ineffective" - but would make a decent third robot in the right alliance in district eliminations. (24 pts., including the act of rolling up to the tower)

Robot B: Goes under the low bar, picks up boulders from the secret passage and scores in the low goal - nearly 100% of the time - and can complete about five cycles per match. Additionally, it can cross all the Category B, C and D defenses, carrying a boulder across. Though, admittedly not as effectively as the low bar. During Autonomous, it can cross the low bar and score in the low bar, or simply cross a category B defense. I view this robot as "effective" and would be a good second robot or, possibly, a low level captain in district events. (40-50 pts., including threatening a tower.).

I figure that less than 20% of all robots will be as strong as Robot B.

Monochron 01-02-2016 14:46

Re: Percentage of EFFECTIVE Limbo Robots
 
Quote:

Originally Posted by ToddF (Post 1533324)
In my view, the most damaging action a robot can take is to shoot for the high goal and miss, robbing your alliance of the second bonus. The possible benefit of a high goal over a low goal (3 points) is not worth the risk of taking that shot (losing 25 points) unless your shooter is greater than 80% accurate (not bloody likely) AND you have time to make all 8 shots into the high goal. (Which, at 80% accuracy, means attempting 10 undefended shots.)

I'm surprised this hasn't been brought up more. Shooting percentage needs to be factored in when considering shooting for the high goal. Top level teams are going to have accurate shooters regardless, and this poll ask about being an "effective" robot, two entirely different things. I think a mid level team is going to get a similar amount of points if they choose the high goal or the low goal. The percent of missed high shots is going to bring their final score down close to what they would have made if they did nothing but low goals.

MrForbes 01-02-2016 14:53

Re: Percentage of EFFECTIVE Limbo Robots
 
Quote:

Originally Posted by Monochron (Post 1533519)
I'm surprised this hasn't been brought up more. Shooting percentage needs to be factored in when considering shooting for the high goal.

Yup, our (low) robot design will be able to attempt both high and low goals....if it's not so good at high shots, we'll do low shots. As usual, we won't know until we play the game.

JesseK 01-02-2016 15:15

Re: Percentage of EFFECTIVE Limbo Robots
 
Quote:

Originally Posted by Monochron (Post 1533519)
I'm surprised this hasn't been brought up more. Shooting percentage needs to be factored in when considering shooting for the high goal. Top level teams are going to have accurate shooters regardless, and this poll ask about being an "effective" robot, two entirely different things. I think a mid level team is going to get a similar amount of points if they choose the high goal or the low goal. The percent of missed high shots is going to bring their final score down close to what they would have made if they did nothing but low goals.

I think this grossly underestimates the effect of defense on low goal. The low goal-only bot has fewer options for scoring. Far shots into the low goal have the same margins of error as far shots into the high goal, thus the only guaranteed low goal shot is the one made from really up close. Since these are two static spots on the field, and (unlike 2014) the spots are relatively close to each other, I bet a good defender can easily and completely shut down a good low-goal-only fast-cycler bot.

Having 1 offensive robot completely shut down by a defender means the other offensive robot needs to perform the tower weakening by itself, all else equal. If the low goal-only bot is also distracted from defense weakening, it's gravy on top for the defender.

Though I do agree that sacrificing the ability to do low goal in order to go after high goal is an error. That was a tough lesson from 2014.

IKE 01-02-2016 15:23

Re: Percentage of EFFECTIVE Limbo Robots
 
Quote:

Originally Posted by Joe Johnson (Post 1533116)
According to the poll Low Bar 90% of teams are planning on being able to go under the low bar.

I believe that 90% of teams are going to TRY to go under the low bar, I just don't think that there will be that many that are effective at it. By which I mean that their robot can not only go under the low bar but can do something else that adds significant value to their alliance (e.g. cross many/most other defenses, score boulders into the high goal, scale the tower, block opponent shots/play defense effectively, ...).
...snip....

Is scoring in the low goal not considered useful? do they need to do all those things, one of them, or some combination?

Dezion 01-02-2016 15:35

Re: Percentage of EFFECTIVE Limbo Robots
 
Quote:

Originally Posted by IKE (Post 1533532)
Is scoring in the low goal not considered useful? do they need to do all those things, one of them, or some combination?

It can be useful because it will still weaken the tower, which will give you a chance to capture the tower (for the RP point). As well, it definitely would be very useful to be able to do some of the other goals.


All times are GMT -5. The time now is 01:02.

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