Chief Delphi

Chief Delphi (http://www.chiefdelphi.com/forums/index.php)
-   Rules/Strategy (http://www.chiefdelphi.com/forums/forumdisplay.php?f=6)
-   -   Why the low Gear love (http://www.chiefdelphi.com/forums/showthread.php?t=153800)

Rangel 17-01-2017 17:21

Re: Why the low Gear love
 
Quote:

Originally Posted by Cory (Post 1632513)
Averaging 6 gears per match would make you one of the top 5-10 robots in the world. Top robots in 2011 were hanging ~6 tubes on average in teleop. topping out at probably 8. And that was with a more open field, a protected scoring zone, and the ability to sit in the protected zone and grab tubes that were thrown cross field to you.

Others brought up good points but one huge factor is that making an effective gear scorer is arguably a lot easier to make than making something that scores tubes at tall various heights(and sometimes over another already placed tube). I think the shear reduction in the barrier of entry will make 6 gear averages a lot more common. Maybe I'm wrong and most regionals will be decided on gears alone but I hope that I'm not.

pmattin5459 18-01-2017 00:57

Re: Why the low Gear love
 
Quote:

Originally Posted by mrnoble (Post 1632464)
One way to attempt to overcome the point values that Kevin rightly pointed out is to design a very efficient gears-only bot. A team that can consistently cycle rapidly and get three rotors turning on their own would do alright. This would require a ton of automation and driver practice, but I don't think that the only top-level bots we see will be ball shooters. Someone, somewhere is going to break gears.

I think that there should be more focus on designing these sorts of bots. I also feel that vision tracking is a must on a top gearbot, as otherwise lining up will probably be challenging (I know from experience it's a lot harder to do these sorts of things than it looks, especially with a drivetrain that can't strafe). Two such bots would be nearly unstoppable, especially with the potential addition of defense, climbs, or a fuel bot.

For example (in elims):

2 bots, 6 gears each: 180 + another 20 pts if another auton bot + additional 100 pts for 4 rotor bonus.

Add another 90 pts (40 pts + 50 bonus for 40kpa) for an ideal fuel bot.

We end up with 390 pts. Up to 540 with three climbs. Good gearbots that can score 6 gears each + climb make up the majority of the points with this strategy, with the fuel bot bringing the cherry on top.

This is an ideal scenario, but it illustrates what a championship alliance may be able to accomplish with good gearing.

Michael Hill 18-01-2017 06:16

Re: Why the low Gear love
 
I'm predicting a lot of teams are putting too much faith in the little hook on the end of the spring. In our testing with the "official" spring (with team stand), if the gear was placed less than half way on the spring, the pilot would have to pull up rather slowly because the spring can deflect a large enough amount to make the gear want to slip off.

Gdeaver 18-01-2017 07:43

Re: Why the low Gear love
 
I have to agree with Hill. We received our McMaster spring last nigh and it is not as stiff as I thought it would be. An active mech that pushes the gear in should have no problem. A passive mech that only stets the gear at frame perimeter is very touchy. Pilots, pull smoothly. Placing any further back is risky. If a gear is dropped will doing this makes that station unusable. A team can take the penalty and move it. If they move a second dropped gear is that a yellow card? A third a red card? Gears are not easy. Balls are not easy. This is going to be a wild dynamic game.

Raysaran 18-01-2017 07:53

Re: Why the low Gear love
 
I think one of the things that people overlook is the defense capabilities on the gear. It's only 5 points for a typical foul and 140 points for the last gear. There's going to be incredibly aggressive defense against that last gear, and if you don't manage to place that last gear you have just wasted time on the other 5.

mman1506 18-01-2017 08:28

As teams start the integration process I think many will realize the true difficulty of fuel. These balls aren't small relative to the robot and holding a large amount while still being able to properly index and feed them consistently to a shooter at a high rate of fire is going to be difficult.

I'd bet many teams at this point have prototyped shooters capable of firing 5+ balls per second but haven't spent nearly as much time feeding the balls at that rate. With limited space this year stacking just 10 balls for auton is difficult to package in a "do everything robot"

A case study we often looked at on my team was 254's 2013 robot. It was by all accounts a great "do everything" robot, good shooter, good intake and a sweet climber but it's indexing mechanism held it back severely in the upper levels of competition due to its relatively low rate of fire. This was probably a result of limited options in the packaging/integration stage of design. In the end much simpler robots were able to be more competitive.

My prediction is a lot of upper middle tier robots will shoot at a high rate of fire with low consistency due to poor feeding which may be fine for teleop but will struggle in auton or shoot 3-8 indexed balls at a high rate of fire and then significantly slow down as un indexed balls feed into the indexing system. Although there will definitely be teams who are all able to overcome this.

Kevin Sevcik 18-01-2017 09:24

Re: Why the low Gear love
 
Quote:

Originally Posted by Raysaran (Post 1632814)
I think one of the things that people overlook is the defense capabilities on the gear. It's only 5 points for a typical foul and 140 points for the last gear. There's going to be incredibly aggressive defense against that last gear, and if you don't manage to place that last gear you have just wasted time on the other 5.

You're waiting till the last gear? You must be very confident in your defender(s). If I'm otherwise ahead (because I gave up on gears and fueled at 3 rotors), I'd throw 2-3 robots into defense with 3 gears left. If you can lock them down to 3 gears left with 15-20 on the clock, you can pull off and climb with reasonable confidence that they're doomed.

Unless 4 rotors becomes utterly mundane and easily achievable, gear heavy alliances are going to have to get creative to keep opponents in the dark about just how many gears they have remaining.

Ginger Power 18-01-2017 09:38

Re: Why the low Gear love
 
Quote:

Originally Posted by mman1506 (Post 1632826)
My prediction is a lot of upper middle tier robots will shoot at a high rate fire with low consistency which may be fine for teleop but will struggle in auton or shoot 3-8 indexed balls at a high rate of fire and then significantly slow down as un indexed balls feed into the indexing system. There will definitely be teams who are all able to overcome all this though.

Completely agree with your analysis. Teams that haven't been using CAD to integrate systems from the beginning are going to find themselves with small hoppers, poor indexing systems, and most will be better off just avoiding Fuel when it comes time for their events.

On the other hand, if you've been planning your systems from the beginning, and have utilized CAD effectively, then it isn't terribly difficult to maintain a large hopper, and set up an effective indexing system. My question becomes, how big of an indexing system will separate great Fuel robots from good and poor Fuel robots. Will the top teams be able to empty their 120 ball hoppers in 12 seconds? Or will they choose to go with shorter bursts of rapid fire shooting (say 20 Fuel in 2 seconds) and just mix Fuel scoring into their typical Gear cycles.

pmattin5459 18-01-2017 09:41

Re: Why the low Gear love
 
Not sure how possible a 120 fuel hopper is. I'd say around 40-50 would be about the max you could store. Inevitably someone will prove me wrong, however.

JesseK 18-01-2017 10:32

Re: Why the low Gear love
 
Quote:

Originally Posted by pmattin5459 (Post 1632786)
For example (in elims):

2 bots, 6 gears each: 180 + another 20 pts if another auton bot + additional 100 pts for 4 rotor bonus.

Add another 90 pts (40 pts + 50 bonus for 40kpa) for an ideal fuel bot.

We end up with 390 pts. Up to 540 with three climbs. Good gearbots that can score 6 gears each + climb make up the majority of the points with this strategy, with the fuel bot bringing the cherry on top.

This is an ideal scenario, but it illustrates what a championship alliance may be able to accomplish with good gearing.

Going to throw this out there just in case you mis-read point values (like I've done in the past). The 40kpa threshold gives a 20 point bonus, not 50, according to table 4-1 in the manual. As far as I can tell, it is the only place in the manual which states the point value of the threshold.

Ginger Power 18-01-2017 10:36

Re: Why the low Gear love
 
Quote:

Originally Posted by pmattin5459 (Post 1632867)
Not sure how possible a 120 fuel hopper is. I'd say around 40-50 would be about the max you could store. Inevitably someone will prove me wrong, however.

CAD + math says it's very possible. I don't think we'll keep our 120 Fuel hopper though in favor of a slightly smaller hopper with a faster indexing system. I will say that in order to have a large hopper size, you're giving yourself less room for your gear mechanism, and your climber. Teams have to determine if these tradeoffs are worth it.

Raysaran 18-01-2017 10:41

Re: Why the low Gear love
 
Quote:

Originally Posted by pmattin5459 (Post 1632867)
Not sure how possible a 120 fuel hopper is. I'd say around 40-50 would be about the max you could store. Inevitably someone will prove me wrong, however.

No you're pretty right, I don't think you can get more than 50-60 unless you are only a shooter bot. We've tried many ways to improve storage capacity as well.

Raysaran 18-01-2017 10:53

Re: Why the low Gear love
 
Quote:

Originally Posted by Kevin Sevcik (Post 1632853)
You're waiting till the last gear? You must be very confident in your defender(s). If I'm otherwise ahead (because I gave up on gears and fueled at 3 rotors), I'd throw 2-3 robots into defense with 3 gears left. If you can lock them down to 3 gears left with 15-20 on the clock, you can pull off and climb with reasonable confidence that they're doomed.

Unless 4 rotors becomes utterly mundane and easily achievable, gear heavy alliances are going to have to get creative to keep opponents in the dark about just how many gears they have remaining.

My intention wasn't on waiting till the last gear, but rather making sure that last gear does not get scored as a primary game objective against a gear focused alliance. I was trying to emphasis the problem with having a gear only alliance, is that it is incredibly difficult to get 12 gears even with mediocre defense. Assuming your drivetrain operates at a 10ft/s when traversing the field, not accounting for defense, each robot will take around 20ish seconds to place a gear. Capping the best gear robots at 5-6 gears. Playing defense around the feeder station, where you have significant vision advantage, is incredibly easy, you could easily knock out 2-3 gears from the opposing alliance even if you have mediocre unpracticed drivers just by getting in their way. Obvious this speculation is alliance dependent and is incredibly situational based on the pick/strategy situation. As a fundamental rule having a good flywheel shooter is essential in high level matches based on how the gear points scale. It will be the difference maker, assuming climbing capabilities on both alliances are similar. It's an interesting prospect because for once, diversifying your interests in FIRST is actually a good thing, something that was not the case for previous years. There is a high-risk high reward element coming from hitting that 4 rotor threshold, and a very intense game of cat and mouse is going to come from either the defending alliance or the offensive alliance. Regardless, my money is on the cat.

KJaget 18-01-2017 17:24

Re: Why the low Gear love
 
Quote:

Originally Posted by JesseK (Post 1632545)
You just need 900's game piece vision tracking. Then 2-gear and 3-gear autos are totally doable :rolleyes: .

Or just street number, uh, robot number/bumper detection https://static.googleusercontent.com...hive/42241.pdf

Yep, totally practical :)


All times are GMT -5. The time now is 20:36.

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