Floor pickup Vs. feeder station

[xSAWxBLADEx]

You don't need a floor pick-up nor a shooter
http://www.youtube.com/watch?v=jqXTZyHvM4A

[JosephC]

Originally Posted by xSAWxBLADEx

You don't need a floor pick-up nor a shooter
http://www.youtube.com/watch?v=jqXTZyHvM4A

That makes 0 sense when paired with that video.

[xSAWxBLADEx]

Originally Posted by JosephC

That makes 0 sense when paired with that video.

You could feed another bot to score the full auto. points

[Cory]

Originally Posted by xSAWxBLADEx

You could feed another bot to score the full auto. points

Someone still has to pick the other 6 off the ground.

[NelsonMichael]

To answer the question, I believe that we will see more teams use the feeder station than to pick up Frisbee from the floor, Now whether or not one is better than the other is totally depends on the alliance. Like mentions before, 148 & 118 worked well together because 1 worked the feeder well and 1 could pick up well.

So what would happen if there was an alliance of 3 that picked up from the floor. And another alliance that only fed from the human players?

Point is, there is no definite answer. This is FIRST, Whatever works well with your team is the best.

[xSAWxBLADEx]

Originally Posted by Cory

Someone still has to pick the other 6 off the ground.

That is what you guys are for.

[Zebra_Fact_Man]

Originally Posted by Tom Line

Ground pickups have an added advantage. They are complimentary to full-field shooters. However, full-field shooters have already been largely negated, so that point is arguable.

The only reason (I have observed) that cross-court shooters have been negated is that alliances have not drafted/strategized to ensure their cross-court shooter is unaffected by the opposing alliance's defense.

In other words, alliance a cross-court shooter, a ground pickup bot, and an offensive lineman to protect the cc shooter and we'll see if the cc shooter has been negated.

That's like saying a pocket passing QB has been negated because he can't avoid the rush without his O-line.

The best example I've seen so far of this was 51 blocking 4779 from deflecting 3604's cc shots. But even that was only in passing to/from attempting their 5pt disc shots.

I really, really, REALLY would like to see a "complete" cross-court shooting alliance assembled to legitimize this strategy's effectivity.

[Kevin Leonard]

We found that our ground pick-up paired with 2791's great feeder station bot made a great alliance at WPI.
However, technical issues with bots certainly didn't help our alliance much, that's for sure.

[BJC]

This is an interesting discussion; Let's add some math.

Assumptions:

-the “floor pickup” robot is a robot that can score 42 points in auto (under pyramid) or 30 points (from midfield) and can hang in 5 sec for 10 points.

-the “30 point hanger” is a robot that scores 18 points in auto and can hang in 15sec for 30 points.

-All robots are 100% accurate in auto and 90% accurate in teleop. All robots always hang.

-Teleop cycles from the human player station and back are 20 seconds. (seems reasonable with high level play including Defense.) Floor loading and shooting from around your side of the pyramid takes 10 sec.

-Assume that the robots are not getting in each other’s way.

-A third pick should not be able to meet the above (or they are the most ridiculous sleeper in FRC history.) I will be running through examples with only two robots (presumably alliance captain and first pick.)
--------------------------------------------------------------------------------
Example 1: “Floor pickup” + “30 point climber”

Auto points: 42 “floor pickup” + 18 “30 point climber” = 60 auto points

Teleop for “30 point climber” = 54+30 (climbing) = 84
120sec-15sec = 105sec in match to work with
105sec/20sec per trip = 5.25 trips round down ~5 trips
5trips * 4 disks = 20 disks
20 disks *90% accuracy = 18 disks
18 disks * 3 points = 54 points

Teleop for “ground pickup” =66 +10(climb) = 76
120sec - 5sec = 115sec in match to work with
20sec per trip * 5 trips = 100 sec of feeder station trips
5 trips * 4 volleys = 20disks
20 disks *90% accuracy = 18disks
18disks * 3 points = 54 points
At this point in the match there are 4 disks on the ground and 15sec before hanging.
4disks on ground * 3points= 12 points
54 points + 12 points = 66 points

Total teleop points: 84 points +76 points= 160 points

Total match points: 60auto + 160 teleop = 220 points

Example 2: “30 point climber” + “30 point climber”

Auto points: 18 + 18 = 36 points

Teleop for “30 point climber”: 54 + 30(climb) = 84 points

Total teleop points: 84 points * 2 robots = 168

Total match points: 36auto + 168teleop = 204 points

Example 3: “ground pickup” + “ground pickup”

Auto points: 42+30= 72 points

Teleop for “ground pickup”: 76 points

Teleop for “ground pickup” two: 64 points
See “ground pickup” in example 1 excluding the 12 extra points from picking up off the ground (because the first “ground pickup” robot got those.)

Total teleop points: 76 + 64 = 140 points

Total match points: 76auto + 140teleop = 216 points

Total match points without middle field auto: 60auto + 140teleop = 200 points

------------------------------------------------------------------------------------
In summary:

-The “ground pickup” + “30 point climber” is the most effective at 220 points.
-The “ground pickup” + “ground pickup” was the second most effective with 216 points.
-The “30 point climber” + “30 point climber” was the third most effective with 204 points.
-The “ground pickup” + “ground pickup” was the least effective if the second “ground pickup” could not get the disks in the middle of the field.

The scores are obviously very close. I think that both strategies are very evenly matched. If we had real teams competing in any of these scenarios it would come down to what the third robot did during the match.

Huh, longest post ever.
Regards, Bryan

[z_beeblebrox]

Originally Posted by BJC

This is an interesting discussion; Let's add some math.

Assumptions:

[i]-the “floor pickup” robot is a robot that can score 42 points in auto (under pyramid) or 30 points (from midfield) and can hang in 5 sec for 10 points.

-the “30 point hanger” is a robot that scores 18 points in auto and can hang in 15sec for 30 points
...

Sounds like 1114 and 2056...

The scores that you estimated were just a little above the maximum that 1114 and 2056 scored. Perhaps robots getting in each other's way or defense made up for the difference.

[z_beeblebrox]

Originally Posted by z_beeblebrox

Sounds like 1114 and 2056...

The scores that you estimated were just a little above the maximum that 1114 and 2056 scored. Perhaps robots getting in each other's way or defense made up for the difference.

Edit: or autonomous misses

[NXTGeek]

Originally Posted by BJC

Assumptions:

-All robots are 100% accurate in auto and 90% accurate in teleop. All robots always hang.

-Assume that the robots are not getting in each other’s way.

Thanks for putting that analysis together!
I quoted these assumptions because defense crushes in teleop, and I have seen wk2 7pt autos to be less accurate in percentage than teleop, which is rarely a 90% in accuracy already.
I think that as the game progresses, autonomous will become more accurate, and matches will be more viciously defensive.
Staying around the safe pyramid as much as possible is a strategy made possible with a floor pickup as well. It also makes getting to the pyramid for climbing more reliable for a 30pt or even a 10pt climber as well.

[Kevin Leonard]

Originally Posted by BJC

This is an interesting discussion; Let's add some math.

Assumptions:

[i]-the “floor pickup” robot is a robot that can score 42 points in auto (under pyramid) or 30 points (from midfield) and can hang in 5 sec for 10 points.

-the “30 point hanger” is a robot that scores 18 points in auto and can hang in 15sec for 30 points.

That's assuming those are the only types of robots.
Obviously those are going to be common and effective, but you also have teams like 1986 with Pickup+Climb 30
I also think that your ground pickup bot could score even more if they got fed a couple discs fed across the field to them by either the third bot or the 30 Point Hanger.
It'll be interesting to see what other combinations teams come up with as we go into week 3 and beyond.
Some teams are looking at a full court shooter- usually a feeder station bot- with a bot playing counter-defense which could be a powerful strategy as well.
I think 2168 King-Tec did that, but I only heard that through word-of-mouth as I wasn't watching their webcast.
Anyway- cheers to strategy!

[BJC]

Originally Posted by NXTGeek

Thanks for putting that analysis together!
I quoted these assumptions because defense crushes in teleop, and I have seen wk2 7pt autos to be less accurate in percentage than teleop, which is rarely a 90% in accuracy already.
I think that as the game progresses, autonomous will become more accurate, and matches will be more viciously defensive.
Staying around the safe pyramid as much as possible is a strategy made possible with a floor pickup as well. It also makes getting to the pyramid for climbing more reliable for a 30pt or even a 10pt climber as well.

I wasn't trying to simulate an actual match, just compair the two "types" of robots people are generally arguing about. I believe that 90% accuracy is definatly achieveable in this game. As to cycle times I just picked a reasonable number. Your welcome to run any numbers using different cycle times. It would be interesting to see what percentage of points comes from auto, teleop, climbing based on different cycle times. It's also not that hard to calculate aproximatly how many disks a "floor loader" could shoot if being fed by, say, the 3rd pick robot.

Originally Posted by Thunder910

That's assuming those are the only types of robots.
Obviously those are going to be common and effective, but you also have teams like 1986 with Pickup+Climb 30
I also think that your ground pickup bot could score even more if they got fed a couple discs fed across the field to them by either the third bot or the 30 Point Hanger.
It'll be interesting to see what other combinations teams come up with as we go into week 3 and beyond.
Some teams are looking at a full court shooter- usually a feeder station bot- with a bot playing counter-defense which could be a powerful strategy as well.
I think 2168 King-Tec did that, but I only heard that through word-of-mouth as I wasn't watching their webcast.
Anyway- cheers to strategy!

Everything you say is true, except one thing. The robots in my example are not common. The floor loader scores 118 points per match. The 30 point climber scores 102. 1986 using the standards of my example is at 126 points/match, only 8 points higher. Their also appears to be great benifits to feeding (shooting downfield from the feeder station without the intent of scoring) because it would dirastically increase the speed with with the floor loader could score in teleop. All in all, we'll see a great variety of successful stratigy within the next couple weeks; I'm looking forward to see how it all pans out.

Regards, Bryan

[Kevin Leonard]

Originally Posted by BJC

Everything you say is true, except one thing. The robots in my example are not common.
Regards, Bryan

Very true- even 1114 and 2056 weren't doing that- they were close- but not quite that much. Their third bot was also doing quite a lot- especially in auto- to help them get their massive score.