Chief Delphi

Chief Delphi (http://www.chiefdelphi.com/forums/index.php)
-   Rules/Strategy (http://www.chiefdelphi.com/forums/forumdisplay.php?f=6)
-   -   Fuel Seems Undercosted (http://www.chiefdelphi.com/forums/showthread.php?t=153119)

Kpchem 07-01-2017 15:30

Re: Fuel Seems Undercosted
 
Quote:

Originally Posted by Trey178 (Post 1626496)
Since the boiler takes about 2 seconds to process fuel...

Where are you pulling this from? The boiler goals process 4-5 Fuel per second as stated during the webcast.

darklink2458 07-01-2017 18:05

Re: Fuel Seems Undercosted
 
Quote:

Originally Posted by Joe G. (Post 1626469)
Just because a task closely resembles the "main" task of old games, doesn't mean it's the "main" task of this one.

Also, fuel has an effectively infinite scoring potential, whereas gears are finite. True, the gears limit is relatively high and will take many alliances the full match or more to reach, but others will be able to finish them off with plenty of time to spare. In these matches, where both alliances do this, fuel will be the difference maker.

Additionally, there's two very key differences between fuel and most recent shooting tasks: the ball is tiny compared to the robot, and storage is unlimited. Unlike recent games, and especially unlike last year, where you had to make every shot count, pure rate of fire is arguably more important than actual accuracy, and the size difference opens up some interesting possible ways to make this happen.

Yes but you can calculate the top amount of fuel that can be processed and even then in perfect conditions accuracy and speed it is marginally better than gears.

Hitchhiker 42 07-01-2017 18:10

Re: Fuel Seems Undercosted
 
Quote:

Originally Posted by mman1506 (Post 1626503)
The fuel is recycled onto the field so capacity is essentially infinite.

I'd assume so since they are scored differently.

No, but how much overflow can the boiler handle (say if you shoot at >5 fuel/sec)?

ollien 07-01-2017 18:19

Re: Fuel Seems Undercosted
 
Quote:

Originally Posted by Hitchhiker 42 (Post 1626655)
No, but how much overflow can the boiler handle (say if you shoot at >5 fuel/sec)?

I think this depends on the geometry of the funnel. I can't find it in the field drawings, but it would be nice to see someone calculate the maximum possible number of fuel.

wheelerka 07-01-2017 18:38

Re: Fuel Seems Undercosted
 
As my team discussed this topic, it was brought up that if scoring gears has the most point potential, why wouldn't everyone do it? The answer seems to be that gearing is an intimidating task that may prove to be much more difficult than previously thought.

On the other hand, dumping fuel to achieve the boiler ranking point is comparatively less difficult. However, the big thing we seemed to realize was that while scoring ranking points via fuel may get you to playoffs, it just isn't enough to win playoffs against teams scoring ranking points via gears. A gearing ranking point is worth 100 pts, while the fuel point translates to 20 points; leaving a gap of 80 points at the very least, which would be extremely difficult to overcome by only scoring fuel.

While it may seem undercosted point wise, strategy will be crucial as fuel provides a comparatively easier way to rack up ranking points, which is what really matters in terms of your shot at playoffs if you may not be able to always win matches point wise (similar to breaching the defenses last year)

Rangel 07-01-2017 18:43

Re: Fuel Seems Undercosted
 
The thing about fuel is that it's fairly important in order to rank high and very important if both alliances can max out on gear points. 120 balls in the high boiler seems to be a very achievable task to get the ranking point while 12 gears is going to be very difficult for any single robot to do. Climbing is also not to be overlooked. One climb is worth 150 high boiler shots! Elimination matches are certainly going to be interesting and i expect a lot of upsets from specialized alliances such as a gear only alliance. Lots of different strategies and counters to different types of robots.

engunneer 07-01-2017 19:08

Re: Fuel Seems Undercosted
 
I see the fuel RP as equivalent to the breach - 60-80% of the alliances should get it. the Gear/Rotor RP is like capturing, where things have to go well for the whole alliance.

Also when working out the cost/benefit of gears vs fuel, remember that the points scored per gear goes down through the match (fist gear is 40 or 60 points, while the last 6 gears are needed to get the last 40 points). An alliance that only scores 11 gears gets the same points as an alliance that stopped after 6. Fuel scoring is more linear.

Jarren Harkema 07-01-2017 19:14

Re: Fuel Seems Undercosted
 
Quote:

Originally Posted by ollien (Post 1626660)
I think this depends on the geometry of the funnel. I can't find it in the field drawings, but it would be nice to see someone calculate the maximum possible number of fuel.

We tried to do this and came up with some rough numbers.

The Cylinder itself appears to be ~ 1.78 cubic feet, and the funnel is ~ 1.63. Assuming that the trapezoid below the cylinder is used for processing fuel, thats ~ 3.41 cubic feet total.

The large clear bins used in the videos look like they hold approximately 50 balls, and are 4.24 cubic feet.

Using this, the number of balls that could be held in the boiler stack is approximately 40. If you were shooting at a rate of 6 balls per second non-stop, the boiler stack would fill up in 40 seconds. So you need 240 balls going in non-stop.

Obviously the voids in a rectangular container will be different than the voids in a cylindrical container, but it's close enough for me tonight. There are bigger fish to fry, like figuring out how to flood the boiler in the first place.

faraz5557 07-01-2017 20:22

Re: Fuel Seems Undercosted
 
Quote:

Originally Posted by Jarren Harkema (Post 1626704)
3.41 cubic feet total.

Using this, the number of balls that could be held in the boiler stack is approximately 40.

I'm going to dispute your 3.41 cubic feet figure (I'm getting like 4 cubic feet, 26 in tall cylinder with an approximately 9.1875 in radius which is less than the given number of 10.5 in bc the tube is lower radius than the frustrum), but more importantly how did you get 40? 3.41 cft is 5892 cubic inches, each ball is 65.5 cubic inches for 90 balls, but since sphere packing is only about 74% efficient this is 66 balls in only the cylindrical and frustrum part of the boiler room. How did you only get 40 balls?

engunneer 07-01-2017 20:35

Re: Fuel Seems Undercosted
 
Quote:

Originally Posted by faraz5557 (Post 1626773)
I'm going to dispute your 3.41 cubic feet figure (I'm getting like 4 cubic feet, 26 in tall cylinder with an approximately 9.1875 in radius which is less than the given number of 10.5 in bc the tube is lower radius than the frustrum), but more importantly how did you get 40? 3.41 cft is 5892 cubic inches, each ball is 65.5 cubic inches for 90 balls, but since sphere packing is only about 74% efficient this is 66 balls in only the cylindrical and frustrum part of the boiler room. How did you only get 40 balls?

packing is 74% best case. see wikipedia. I'd estimate we get loose or poured for most cases in this game, so 60% packing seems reasonable.

Jamesa47 07-01-2017 20:35

Re: Fuel Seems Undercosted
 
Quote:

Originally Posted by faraz5557 (Post 1626773)
I'm going to dispute your 3.41 cubic feet figure (since sphere packing is only about 74% efficient this is 66 balls in only the cylindrical and frustrum part of the boiler room. How did you only get 40 balls?

74% packing efficiency is maximum possible if you pack perfectly, in reality I think it should be more around 60-62%.

I addressed it in a little more detail here:

https://www.chiefdelphi.com/forums/s...d.php?t=153169

It should be more around 54 or so? I may have calculated that inaccurately but it's a bit less than 66 balls.

Jarren Harkema 07-01-2017 21:49

Re: Fuel Seems Undercosted
 
Quote:

Originally Posted by faraz5557 (Post 1626773)
I'm going to dispute your 3.41 cubic feet figure (I'm getting like 4 cubic feet, 26 in tall cylinder with an approximately 9.1875 in radius which is less than the given number of 10.5 in bc the tube is lower radius than the frustrum), but more importantly how did you get 40? 3.41 cft is 5892 cubic inches, each ball is 65.5 cubic inches for 90 balls, but since sphere packing is only about 74% efficient this is 66 balls in only the cylindrical and frustrum part of the boiler room. How did you only get 40 balls?

Quote:

Originally Posted by Jamesa47 (Post 1626786)
74% packing efficiency is maximum possible if you pack perfectly, in reality I think it should be more around 60-62%.

I addressed it in a little more detail here:

https://www.chiefdelphi.com/forums/s...d.php?t=153169

It should be more around 54 or so? I may have calculated that inaccurately but it's a bit less than 66 balls.

Again, as explained by the numerous "about" and "approximately" disclaimers, my numbers were a quick 15 minute back-of-a-napkin estimate. With either number, the possibility that enough fuel can be constantly supplied to overflow the high goal is very slim in my opinion.

I suppose the best way to know for sure is have someone dump a bunch of balls in the high goal of an official field boiler. I think that one is on a shipping container as we speak though...

gigaboggie 08-01-2017 10:36

Re: Fuel Seems Undercosted
 
I agree with the roughly 60-65% packing. That would be for both the robot capacity and for the capacity in the boiler. Operating on the assumption that the boiler is approximately 3.41 ft^3, we multiply (3.41 ft^3)*(.60 usable ft^3/ft^3) giving us roughly 2.05 usable ft^3. Since balls take up a volume of roughly 4/3*pi*2.5^3 that would give us a volume of 65.45 in^3. The boiler volume converts from 2.05 usable ft^3 to 3535.5 usable in^3. Dividing those gives you roughly 54 balls. So it is certainly safe to assume that at least 50 balls will fit within the boiler while waiting to be processed.

I would like to argue that balls are actually being incredibly under-valued in many discussions so far. I am going to operate under the assumption of 60% packing due to packing error and that a team can dedicate roughly 25% of their robot to ball storage which does not seem to be unreasonable. Assuming tall configuration (excluding bumpers is 24in x 26in x 36in tall), this gives us a total robot volume of 22,464 in^3. This gives us roughly 5,616 in^3 of ball storage. Multiplied by .6 to find usable space yields 3,370 usable in^3. Divided by the volume of a ball (65.45 in^3), gives a conservative robot capacity of 50 balls.

Now to calculate the number of cycles, I will assume a bad accuracy of 30% into the high goal. High achieving teams may have an accuracy of upwards of 90%, but 30% is a conservative estimate for your average team. Since 120 balls will be necessary to achieve the goal of a ranking point, that means that it would require that the team shoot 120/.3 balls which is equal to 400 balls. This would require 8 cycles for an inaccurate team.

The time for a cycle in this game I believe would be rather short allowing for that number of cycles without many problems. This is primarily due to the fact that many teams could find a way to increase their ball capacity. Additionally, with a fly wheel shooter, the amount of balls that could be fired off per second is large. This would mean that the only time-consuming portion of scoring balls is collection. Balls are very readily available both on the field and from feeder station-like inputs. All of these estimates are likely underestimates of both the capacity of a robot and the accuracy of a capable team. I, therefore, believe that it would appear to be an almost guaranteed ranking point for teams with high goal scoring capabilities.

Grim Tuesday 08-01-2017 11:14

Re: Fuel Seems Undercosted
 
I don't think you will have time for 8 cycles and shooting 400 balls.

Jarren Harkema 08-01-2017 12:04

Re: Fuel Seems Undercosted
 
Quote:

Originally Posted by Grim Tuesday (Post 1627145)
I don't think you will have time for 8 cycles and shooting 400 balls.

Plenty of teams last year performed way more than 8 cycles, all while avoiding defense. This year, shooting is mostly protected, or at least risky enough for most drivers to avoid. Something else I think is being underestimated is how much of the field is obscured by the airships. Looking at the top view, possibly two of the three driver stations won't be able to clearly see their opponent's key.

400 balls does seem like a long shot, but if you improved that accuracy to 50%, 240 seems more reasonable to me.


All times are GMT -5. The time now is 20:34.

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