Chief Delphi

Chief Delphi (http://www.chiefdelphi.com/forums/index.php)
-   Extra Discussion (http://www.chiefdelphi.com/forums/forumdisplay.php?f=68)
-   -   pic: Here is our final primary design. (http://www.chiefdelphi.com/forums/showthread.php?t=133331)

staplemonx 22-01-2015 09:46

pic: Here is our final primary design.
 

JohnSchneider 22-01-2015 09:48

Re: pic: Here is our final primary design.
 
Why did you limit your stacks to 4 high?

x86_4819 22-01-2015 09:56

Re: pic: Here is our final primary design.
 
I really like the idea of having totes slide through the robot. I will have to present this idea to my team.

staplemonx 22-01-2015 10:29

Re: pic: Here is our final primary design.
 
Quote:

Originally Posted by JohnSchneider (Post 1432077)
Why did you limit your stacks to 4 high?

We could do more, but after watching the double helix prototype where it takes them nearly 20 seconds to build a stack of 4, and 30 seconds to build a stack of 6 in nearly perfect conditions. We felt it makes more sense to use the 10 seconds to go find container to put on top then to wait the extra 10 seconds to get 4 points.

https://www.youtube.com/watch?v=9QmJ...nCuJv3aKb8Os0G

Chris is me 22-01-2015 10:37

Re: pic: Here is our final primary design.
 
Quote:

Originally Posted by JohnSchneider (Post 1432077)
Why did you limit your stacks to 4 high?

Why stack to five or six? The majority of robots won't be able to cap recycle bins on stacks five tall, and almost none will be able to cap on an existing six stack. The rare robots that can do this can probably also handle putting a single bin on an existing four stack. Finally, at most if not all regionals, the goals will not run out of room for new bins when stacked four high. It's a completely reasonable choice, especially if it makes the robot simpler to build or more effective (less tippy, etc).

cadandcookies 22-01-2015 11:08

Re: pic: Here is our final primary design.
 
Quote:

Originally Posted by JohnSchneider (Post 1432077)
Why did you limit your stacks to 4 high?

I think a lot of teams are thinking tall stacks "the thing" to get success in this game. Fundamentally most games come down to scoring a many points as fast as possible-- if you can build more four stacks faster you can potentially score more points than stacking only a few six stacks-- and you make it easier for an ally to find a stack to cap.

FRC teams are fantastic at overestimating what they can do in a given game-- something which hurts them come competition.

Excellent design is driven by excellent constraints-- striking a balance between defining the right project constraints and leaving enough room for creativity can be huge for keeping team directed to building an effective robot.

dellagd 22-01-2015 11:26

Re: pic: Here is our final primary design.
 
Quote:

Originally Posted by cadandcookies (Post 1432141)
I think a lot of teams are thinking tall stacks "the thing" to get success in this game. Fundamentally most games come down to scoring a many points as fast as possible-- if you can build more four stacks faster you can potentially score more points than stacking only a few six stacks-- and you make it easier for an ally to find a stack to cap.

Well said! In your teams initial strategic analysis of the game, you shouldnt be looking at what single actions give the most points, but rather what actions give the most points per unit time, your point scoring efficiently. How those points are obtained is irrelevant. Of course there are other considerations, but that's pretty much the jist of it. Through this process is possible that the conclusion may be to not attempt the action worth the most amount of points, like many teams in 2013 with climbing the pyramid. Once resources are factored in, I think not doing stacks of 6 may be the best course of action for many teams.

JohnSchneider 22-01-2015 11:29

Re: pic: Here is our final primary design.
 
Quote:

Originally Posted by cadandcookies (Post 1432141)
I think a lot of teams are thinking tall stacks "the thing" to get success in this game. Fundamentally most games come down to scoring a many points as fast as possible-- if you can build more four stacks faster you can potentially score more points than stacking only a few six stacks-- and you make it easier for an ally to find a stack to cap.

FRC teams are fantastic at overestimating what they can do in a given game-- something which hurts them come competition.

Excellent design is driven by excellent constraints-- striking a balance between defining the right project constraints and leaving enough room for creativity can be huge for keeping team directed to building an effective robot.

But scoring is linear. 5 stacks of 3 and 3 stacks of 5 are the same. Building taller stacks keeps you from being dependent on the center bins to reach your maximum scoring potential.

Their design seems capable of stacking higher so it was a question of why they chose 4 over 3 or 5 or 6 or whatever.

dellagd 22-01-2015 11:46

Re: pic: Here is our final primary design.
 
Quote:

Originally Posted by JohnSchneider (Post 1432164)
But scoring is linear. 5 stacks of 3 and 3 stacks of 5 are the same. Building taller stacks keeps you from being dependent on the center bins to reach your maximum scoring potential.

Their design seems capable of stacking higher so it was a question of why they chose 4 over 3 or 5 or 6 or whatever.

While I do think their design is probably capable of doing highest stacks as well, depending of how you look at it the 'scoring' may not be linear.

You are totally correct in that those stack combos would yield the same points, however, we're looking at scoring efficiency here. If you look at the game from that perspective, as you're really just battling the clock in this game, and if you didn't use any containers, the stack height wouldn't matter at all, apart from the increase in trip time. Once you do factor in the containers, a non-linear increase in points-per-second is found if you look at creating and topping off stacks of different heights. Factor in the can scarcity with the difficulty of topping off a taller stack with a container and the challenge quickly deviates from simply 'stack as high as you can'

JesseK 22-01-2015 11:52

Re: pic: Here is our final primary design.
 
I really like this design, from an 'overall' perspective. It fixes several issues with a single-orientation human-player-fed lift. It also looks to keep c.g. in check when lifting 3 totes high enough for a 4th tote to to slide down the ramp underneath it. Those totes don't stand a chance!

I applaud your choice to do a 4-stack. A 5-stack is definitely cool, but I agree with others who've stated that capping a 5-stack is tough - probably not doable by the average team on a consistent basis, when compared to what else the average team could be doing to create points.

You may also have a nice setup to add a passive guide system for a noodle into a RC. You're already at the right place to get a noodle, and I can't imagine a noodle guide would be terribly complex or interfere with the lift. That may be another avenue to investigate once this bot is built.

JohnSchneider 22-01-2015 12:05

Re: pic: Here is our final primary design.
 
Quote:

Originally Posted by dellagd (Post 1432173)
While I do think their design is probably capable of doing highest stacks as well, depending of how you look at it the 'scoring' may not be linear.

You are totally correct in that those stack combos would yield the same points, however, we're looking at scoring efficiency here. If you look at the game from that perspective, as you're really just battling the clock in this game, and if you didn't use any containers, the stack height wouldn't matter at all, apart from the increase in trip time. Once you do factor in the containers, a non-linear increase in points-per-second is found if you look at creating and topping off stacks of different heights. Factor in the can scarcity with the difficulty of topping off a taller stack with a container and the challenge quickly deviates from simply 'stack as high as you can'

Assuming they can't pick up bins and are only doing totes then you'd want stacks of 6. Since you cannot reduce stacking time you want to reduce travel so you build them as high as you can. In this situation 1 stack of 30 is optimum but not realistic.

Assuming they have a partner who can score bins the optimization objective becomes to build the tallest possible stack that can be capped by that partner since you reduce travel time for yourself, but fewer taller stacks reduces time for your partner.

So my original question was "why 4 " because it's a weird number to optimize for. If their plan was to just score on totes then a larger number, maybe even > 6 might be optimum. If their plan is to have a partner cap then 6 is optimum.

Just curious as to what led them to that decision.

excel2474 22-01-2015 13:44

Re: pic: Here is our final primary design.
 
Quote:

Originally Posted by JohnSchneider (Post 1432183)
Since you cannot reduce stacking time you want to reduce travel so you build them as high as you can.

You also need to consider that you may have to drive slower with a stack of 6 than a stack of 4, so carrying more does not equal less travel time in all cases. Maybe there is an optimum number like 5.

staplemonx 22-01-2015 15:20

Re: pic: Here is our final primary design.
 
Quote:

Originally Posted by JohnSchneider (Post 1432183)
Assuming they can't pick up bins and are only doing totes then you'd want stacks of 6. Since you cannot reduce stacking time you want to reduce travel so you build them as high as you can. In this situation 1 stack of 30 is optimum but not realistic.


We can pick up upright containers using the same ram them into the wall method as the 3 day Indiana build.

The idea would be that we, pick one up, go to feeder station, build stack of 4 under it, drive to scoring platform drop off, go get next container...repeat.

staplemonx 22-01-2015 15:23

Re: pic: Here is our final primary design.
 
Quote:

Originally Posted by JohnSchneider (Post 1432164)
Their design seems capable of stacking higher so it was a question of why they chose 4 over 3 or 5 or 6 or whatever.

Good observation, we wanted to start tall so that if we get a faster stacker partner we might just switch to topping a 4 foot stack only for that match. If tipping becomes an issue we can start cutting the legs down.

asid61 22-01-2015 15:46

Re: pic: Here is our final primary design.
 
I'm certain that the game will come down to the cans at the high level, maybe even at elims at regionals. If you "waste" cans by using them on 4-5 tote stacks, you are dependent on either the clock running out too fast for everybody or on the inefficiency of other teams to get cans. If you have good can grabbers, then it makes sense to make shorter stacks as time becomes the limiting factor.
JMO though. The real test will be when competitions actually begin.


All times are GMT -5. The time now is 17:42.

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