Go to Post Honesty and Integrity is the key to success. - team4384 [more]
Home
Go Back   Chief Delphi > FIRST > General Forum
CD-Media   CD-Spy  
portal register members calendar search Today's Posts Mark Forums Read FAQ rules

 
Reply
Thread Tools Rating: Thread Rating: 3 votes, 5.00 average. Display Modes
  #16   Spotlight this post!  
Unread 01-09-2018, 01:42 PM
John Pugsley's Avatar
John Pugsley John Pugsley is offline
Registered User
FRC #1720 (PhyXTGears)
Team Role: Driver
 
Join Date: Apr 2017
Rookie Year: 2013
Location: Upland Indiana
Posts: 4
John Pugsley is an unknown quantity at this point
Re: Unfairness in plate assignment

Truly random means red and blue have equal probabilities of getting any given field configuration; some configurations may be more likely, but there is no bias towards red or blue. Yes some matches will seem to be easier for one alliance, but good teams will be able to score regardless.
__________________





2015 - EE, C, PhyXT-Red-Card (1741,1592)
2016 - J, W (868,1024), J, IC
2017 - EE, IC, F (71,1529), IC, F (1741,4926)
Reply With Quote
  #17   Spotlight this post!  
Unread 01-09-2018, 01:55 PM
Mr. Tatorscout's Avatar
Mr. Tatorscout Mr. Tatorscout is offline
Registered User
AKA: Stuart King
FRC #2122 (Team Tators)
Team Role: Mentor
 
Join Date: Dec 2013
Rookie Year: 2010
Location: Idaho
Posts: 123
Mr. Tatorscout has a brilliant futureMr. Tatorscout has a brilliant futureMr. Tatorscout has a brilliant futureMr. Tatorscout has a brilliant futureMr. Tatorscout has a brilliant futureMr. Tatorscout has a brilliant futureMr. Tatorscout has a brilliant futureMr. Tatorscout has a brilliant futureMr. Tatorscout has a brilliant futureMr. Tatorscout has a brilliant futureMr. Tatorscout has a brilliant future
Re: Unfairness in plate assignment

Quote:
Originally Posted by Bennett548 View Post
What is strange to me is that the GDC left this bit of random unfairness that could have easily been eliminated.

They (the GDC) generally try to eliminate random unfairness where possible. The qualification match scheduler does a ton of work to make the match schedule as fair as possible. In 2016 the audience selection was the same for both alliances. In 2008 the track ball positions were random, but the same for each alliance (your color on the right side of your overpass I think). In 2005 the position of vision tetras was random, but reflectively mirrored to make it fair.
Have you forgotten Aerial Assist auto? You had a 50/50 shot of being the robot with the bonus light. And if you have a 2ball auto and got the bonus, woohoo! The randomness forces the coding sub team to design autos that compensate for it. It's not unfair, just more challenging.
Reply With Quote
  #18   Spotlight this post!  
Unread 01-09-2018, 02:02 PM
RoboChair's Avatar
RoboChair RoboChair is offline
He who fixes with hammers #tsimfd
AKA: Devin Castellucci
FRC #1678 (Citrus Circuits and 5458 Digital Minds)
Team Role: Mentor
 
Join Date: Jan 2007
Rookie Year: 2005
Location: Davis, CA
Posts: 1,097
RoboChair has a reputation beyond reputeRoboChair has a reputation beyond reputeRoboChair has a reputation beyond reputeRoboChair has a reputation beyond reputeRoboChair has a reputation beyond reputeRoboChair has a reputation beyond reputeRoboChair has a reputation beyond reputeRoboChair has a reputation beyond reputeRoboChair has a reputation beyond reputeRoboChair has a reputation beyond reputeRoboChair has a reputation beyond repute
Re: Unfairness in plate assignment

Code for all scenarios you need to handle and pray to RNGesus.
__________________

12 Years and counting! Over a third of my life has been spent with FRC.
Reply With Quote
  #19   Spotlight this post!  
Unread 01-09-2018, 02:08 PM
Rangel's Avatar
Rangel Rangel is online now
John Rangel
FRC #0842 (Falcon Robotics)
Team Role: Mentor
 
Join Date: Jan 2010
Rookie Year: 2009
Location: Phoenix, AZ
Posts: 897
Rangel has a reputation beyond reputeRangel has a reputation beyond reputeRangel has a reputation beyond reputeRangel has a reputation beyond reputeRangel has a reputation beyond reputeRangel has a reputation beyond reputeRangel has a reputation beyond reputeRangel has a reputation beyond reputeRangel has a reputation beyond reputeRangel has a reputation beyond reputeRangel has a reputation beyond repute
Re: Unfairness in plate assignment

Quote:
Originally Posted by Mr. Tatorscout View Post
Have you forgotten Aerial Assist auto? You had a 50/50 shot of being the robot with the bonus light. And if you have a 2ball auto and got the bonus, woohoo! The randomness forces the coding sub team to design autos that compensate for it. It's not unfair, just more challenging.
Well, it is unfair since even if you have a 2 cube auto to do the opposite side switch/scale, your opponent can do their 2 cube auto faster and potentially a 3 cube auto. While you can coordinate with your partner to split the task, it now requires one side to have two robots that have a good auto vs one robot. It probably won't be a big deal at that level anyways but at its core it is kind of unfair.
__________________
2012 Dean's List Winner
2014-? Mentor

Reply With Quote
  #20   Spotlight this post!  
Unread 01-09-2018, 02:09 PM
JeffB JeffB is offline
Registered User
FRC #5052 (RoboLobos)
Team Role: Mentor
 
Join Date: Apr 2015
Rookie Year: 2014
Location: Austin
Posts: 296
JeffB has a reputation beyond reputeJeffB has a reputation beyond reputeJeffB has a reputation beyond reputeJeffB has a reputation beyond reputeJeffB has a reputation beyond reputeJeffB has a reputation beyond reputeJeffB has a reputation beyond reputeJeffB has a reputation beyond reputeJeffB has a reputation beyond reputeJeffB has a reputation beyond reputeJeffB has a reputation beyond repute
Re: Unfairness in plate assignment

Quote:
Originally Posted by brandn03 View Post
I agree. Red also has the advantage of knowing that their scale and switch will always be on opposite sides of the field. Whereas, blue has to plan for a scenario where they are on the same side or opposite sides.
Your next post hits on this point. But, that's a bad assumption. As it can be randomized to do the exact opposite, both alliances must plan for both cases.

The problem the OP is looking at takes place when only one robot on an alliance can handle autonomous. If each side has a single bot that can do this, one has a slight advantage 1/8 of the time. It's an advantage that very few teams worldwide can exploit. The actual liability isn't all that much of a concern.
__________________


Any thoughts I share are meant to be my interpretation of rules/events/intents. They shouldn't be viewed as anything even approaching an official viewpoint.
Reply With Quote
  #21   Spotlight this post!  
Unread 01-09-2018, 02:15 PM
Bennett548's Avatar
Bennett548 Bennett548 is offline
Engineering Mentor
AKA: Steve Bennett
FRC #0548 (Robostangs)
Team Role: Mentor
 
Join Date: Feb 2013
Rookie Year: 2004
Location: MI
Posts: 107
Bennett548 has a brilliant futureBennett548 has a brilliant futureBennett548 has a brilliant futureBennett548 has a brilliant futureBennett548 has a brilliant futureBennett548 has a brilliant futureBennett548 has a brilliant futureBennett548 has a brilliant futureBennett548 has a brilliant futureBennett548 has a brilliant futureBennett548 has a brilliant future
Re: Unfairness in plate assignment

Quote:
Originally Posted by Mr. Tatorscout View Post
Have you forgotten Aerial Assist auto? You had a 50/50 shot of being the robot with the bonus light. And if you have a 2ball auto and got the bonus, woohoo! The randomness forces the coding sub team to design autos that compensate for it. It's not unfair, just more challenging.
It was not fair to each team, but it was fair for the alliances. They both had a hot side and a not hot side. Once you drive to the hot goal you can shoot both balls there.

It's like calling a coin flip is fair.

If Aerial Assist were like power up, there is a chance that red alliance would score both balls on the same side and get two bonuses, and the blue alliance must score in one side for the bonus then drive and score in the other side to get the second bonus.
Reply With Quote
  #22   Spotlight this post!  
Unread 01-09-2018, 02:33 PM
Brian Selle's Avatar
Brian Selle Brian Selle is offline
Software/Mechanical Mentor
FRC #3310 (Black Hawk Robotics)
Team Role: Engineer
 
Join Date: Jan 2012
Rookie Year: 2012
Location: Texas
Posts: 224
Brian Selle has a reputation beyond reputeBrian Selle has a reputation beyond reputeBrian Selle has a reputation beyond reputeBrian Selle has a reputation beyond reputeBrian Selle has a reputation beyond reputeBrian Selle has a reputation beyond reputeBrian Selle has a reputation beyond reputeBrian Selle has a reputation beyond reputeBrian Selle has a reputation beyond reputeBrian Selle has a reputation beyond reputeBrian Selle has a reputation beyond repute
Re: Unfairness in plate assignment

Top level teams will be able to handle all scenarios, however, some paths will physically take longer than others for all teams and likely have different point value outcomes. In a close elimination round, your team could be handed one or two unfavorable cases and lose due to the FMS. I'm good taking the L if we were outwitted or outplayed on the same field, but playing against an alliance with an easier layout seems unnecessary and unfair. I love the random concept but think it should be symmetrical for red/blue.
__________________
http://www.team3310.com/
Reply With Quote
  #23   Spotlight this post!  
Unread 01-09-2018, 02:56 PM
brandn03's Avatar
brandn03 brandn03 is offline
Registered User
AKA: Brandon
FRC #6802 (The Mean Caimans)
Team Role: Mentor
 
Join Date: Oct 2015
Rookie Year: 2011
Location: Va
Posts: 86
brandn03 is a splendid one to beholdbrandn03 is a splendid one to beholdbrandn03 is a splendid one to beholdbrandn03 is a splendid one to beholdbrandn03 is a splendid one to beholdbrandn03 is a splendid one to beholdbrandn03 is a splendid one to behold
Re: Unfairness in plate assignment

Quote:
Originally Posted by JeffB View Post
Your next post hits on this point. But, that's a bad assumption. As it can be randomized to do the exact opposite, both alliances must plan for both cases.

The problem the OP is looking at takes place when only one robot on an alliance can handle autonomous. If each side has a single bot that can do this, one has a slight advantage 1/8 of the time. It's an advantage that very few teams worldwide can exploit. The actual liability isn't all that much of a concern.
Yep, I was completely missing the point of the post. I thought the OP was saying the two examples given were the only possible configurations for the randomization of the plates. I now realize I'm an idiot.
Reply With Quote
  #24   Spotlight this post!  
Unread 01-09-2018, 02:57 PM
EricH's Avatar
EricH EricH is offline
New year, new team
FRC #1197 (Torbots)
Team Role: Engineer
 
Join Date: Jan 2005
Rookie Year: 2003
Location: SoCal
Posts: 21,887
EricH has a reputation beyond reputeEricH has a reputation beyond reputeEricH has a reputation beyond reputeEricH has a reputation beyond reputeEricH has a reputation beyond reputeEricH has a reputation beyond reputeEricH has a reputation beyond reputeEricH has a reputation beyond reputeEricH has a reputation beyond reputeEricH has a reputation beyond reputeEricH has a reputation beyond repute
Re: Unfairness in plate assignment

Quote:
Originally Posted by Bennett548 View Post

It's like calling a coin flip is fair.
And you think a coin flip is fair?

I once did some statistical analysis on an everyday quarter... It's been a while but if I recall correctly there was about a 5% bias towards one side. Think about that one for a minute.
__________________
Past teams:
2003-2007: FRC0330 BeachBots; 2008: FRC1135 Shmoebotics; 2012: FRC4046 Schroedinger's Dragons

"Rockets are tricky..."--Elon Musk



Reply With Quote
  #25   Spotlight this post!  
Unread 01-09-2018, 03:04 PM
Hjelstrom's Avatar
Hjelstrom Hjelstrom is offline
Mentor
FRC #0987 (High Rollers)
Team Role: Mentor
 
Join Date: Mar 2008
Rookie Year: 2005
Location: Las Vegas
Posts: 232
Hjelstrom has a reputation beyond reputeHjelstrom has a reputation beyond reputeHjelstrom has a reputation beyond reputeHjelstrom has a reputation beyond reputeHjelstrom has a reputation beyond reputeHjelstrom has a reputation beyond reputeHjelstrom has a reputation beyond reputeHjelstrom has a reputation beyond reputeHjelstrom has a reputation beyond reputeHjelstrom has a reputation beyond reputeHjelstrom has a reputation beyond repute
Re: Unfairness in plate assignment

Quote:
Originally Posted by Brian Selle View Post
Top level teams will be able to handle all scenarios, however, some paths will physically take longer than others for all teams and likely have different point value outcomes. In a close elimination round, your team could be handed one or two unfavorable cases and lose due to the FMS. I'm good taking the L if we were outwitted or outplayed on the same field, but playing against an alliance with an easier layout seems unnecessary and unfair. I love the random concept but think it should be symmetrical for red/blue.
I agree, it seems really bad to have a match where one team has a longer distance to travel than the other. I hope they fix this so the match is random but symmetrical.
Reply With Quote
  #26   Spotlight this post!  
Unread 01-09-2018, 03:07 PM
Rivet Man's Avatar
Rivet Man Rivet Man is offline
Registered User
AKA: 97525A425
no team (Better than Bolts)
 
Join Date: Mar 2016
Rookie Year: 2014
Location: Jacksonville
Posts: 72
Rivet Man has a reputation beyond reputeRivet Man has a reputation beyond reputeRivet Man has a reputation beyond reputeRivet Man has a reputation beyond reputeRivet Man has a reputation beyond reputeRivet Man has a reputation beyond reputeRivet Man has a reputation beyond reputeRivet Man has a reputation beyond reputeRivet Man has a reputation beyond reputeRivet Man has a reputation beyond reputeRivet Man has a reputation beyond repute
Re: Unfairness in plate assignment

Hi EricH,

I thought about that, it doesn't make any sense!

Best Regards,
RM
Reply With Quote
  #27   Spotlight this post!  
Unread 01-09-2018, 03:11 PM
EricH's Avatar
EricH EricH is offline
New year, new team
FRC #1197 (Torbots)
Team Role: Engineer
 
Join Date: Jan 2005
Rookie Year: 2003
Location: SoCal
Posts: 21,887
EricH has a reputation beyond reputeEricH has a reputation beyond reputeEricH has a reputation beyond reputeEricH has a reputation beyond reputeEricH has a reputation beyond reputeEricH has a reputation beyond reputeEricH has a reputation beyond reputeEricH has a reputation beyond reputeEricH has a reputation beyond reputeEricH has a reputation beyond reputeEricH has a reputation beyond repute
Re: Unfairness in plate assignment

Quote:
Originally Posted by Rivet Man View Post
Hi EricH,

I thought about that, it doesn't make any sense!

Best Regards,
RM
Nope, it doesn't. But that is how it panned out.

Of course, that knowledge is useless unless you know 1) how the coin flip was done, 2) how many times it was done, and 3) which side the difference was towards--and 4) a second (different) coin had the opposite bias. None of which I've stated...
__________________
Past teams:
2003-2007: FRC0330 BeachBots; 2008: FRC1135 Shmoebotics; 2012: FRC4046 Schroedinger's Dragons

"Rockets are tricky..."--Elon Musk



Reply With Quote
  #28   Spotlight this post!  
Unread 01-09-2018, 03:12 PM
barn34's Avatar
barn34 barn34 is offline
Isn't this how Skynet got started?
AKA: William Barnickel
FRC #2481 (Roboteers)
Team Role: Mentor
 
Join Date: Feb 2012
Rookie Year: 2009
Location: Metamora, IL
Posts: 263
barn34 has a reputation beyond reputebarn34 has a reputation beyond reputebarn34 has a reputation beyond reputebarn34 has a reputation beyond reputebarn34 has a reputation beyond reputebarn34 has a reputation beyond reputebarn34 has a reputation beyond reputebarn34 has a reputation beyond reputebarn34 has a reputation beyond reputebarn34 has a reputation beyond reputebarn34 has a reputation beyond repute
Re: Unfairness in plate assignment

Quote:
Originally Posted by Brian Selle View Post
Top level teams will be able to handle all scenarios, however, some paths will physically take longer than others for all teams and likely have different point value outcomes. In a close elimination round, your team could be handed one or two unfavorable cases and lose due to the FMS. I'm good taking the L if we were outwitted or outplayed on the same field, but playing against an alliance with an easier layout seems unnecessary and unfair. I love the random concept but think it should be symmetrical for red/blue.
Pretty much all of this. The mere fact that a RNG can put either alliance at a potential disadvantage (perceived or not) is a pretty big faux pax for the GDC here. Here's hoping it's just a minor oversight and they get the software fix in so the RNG just cycles between the 4 symmetrical configurations once competition season kicks off.
__________________



2018 St. Louis Regional Winner, Central Illinois Regional Winner, Wisconsin Regional Winner
2017 Central Illinois Regional Winner
2016 Einstein Field WORLD CHAMPIONS, Carver Sub-Division Winner, Central Illinois Regional Winner, Smoky Mountains Regional Winner
2015 Rock City Regional Winner, Central Illinois Regional Winner
2014 Einstein Field Semi-Finalist, Galileo Division Winner, Wisconsin Regional Winner


--- Industrial Design Award (2010, 2015) ------- Innovation in Control Award (2013, 2014, 2016) ---
--- Quality Award (2015) ---- Creativity Award (2014) --- Excellence in Engineering Award (2018) ---
--- Regional Chairman's Award (2018) --- Judges Award (2017) --- Visualization Award (2009) ---
Reply With Quote
  #29   Spotlight this post!  
Unread 01-09-2018, 03:19 PM
cglrcng cglrcng is offline
Registered User
FRC #0060
 
Join Date: Jan 2013
Location: Kingman, AZ
Posts: 518
cglrcng has a reputation beyond reputecglrcng has a reputation beyond reputecglrcng has a reputation beyond reputecglrcng has a reputation beyond reputecglrcng has a reputation beyond reputecglrcng has a reputation beyond reputecglrcng has a reputation beyond reputecglrcng has a reputation beyond reputecglrcng has a reputation beyond reputecglrcng has a reputation beyond reputecglrcng has a reputation beyond repute
Re: Unfairness in plate assignment

Quote:
Originally Posted by RoboChair View Post
Code for all scenarios you need to handle and pray to RNGesus.
This is the nightmare. Due to A04 you will never program to head for the 3rd letter of any of the 8 possible random combinations as crossing the far null lines leads to a penalty to red card. So, the opponents switch plates are out of bounds.

NOW, you are down to 4 targets (scale left, scale right, switch left, switch right), 3 robot positions left, center, right and a ton of positional variations in each spot but all at least touching the alliance wall), and now the most important of all with randomization..... 3 moves (which robot moves first, second, and third to avoid each other for each chosen path taken to get to which of the target(s) chosen).

That increases the amount of code you will need avail. as choices. The necessary delays will take a lot of work to both work out between alliance partners. And most of it will be completely untested and theoretical when tested on the field.

Having tested from each robot starting position, for those taking the long path(s) to say scale on either side, and running stop watch tests, auto bell start to a point rounding the first corner near each switch woukd be highly useful to plan delays.

If you are an auto scale bot and I were an auto scale bot (both depositing in different sides of the scal plate or same), or I am an auto switch bot, I would allow you to move first, if I knew you would round that corner at 2.6 seconds, I could then choose my delay of auto start timing easier to avoid messing up both of our auto routines by collision. I could then much safer follow ypu and we both stand a much higher rate of success.

This year, testing the 4 target paths for success from 3 robot positions is not enough. TEST THEM, Document them fully, Time them and record many points for clearance along the way so you can fully strategize, plan, and collaborate early with alliance partners.

DO NOT WAIT until in queing lines to plan, leave that for last second adjustments for best laid plans of mice and men issues that arise. And know exactly how long it will take your robot to clear a certain spot, so that others will be able to plan perfectly the delays needed to avoid robot collisions.

And most important of all do not expect to be able to just adjust a slider on ds to enter necessary delays if you are not the first move robot. Have all paths 4 X 3 X numerous delays as default choices that can be code chosen in hands off drivers behind the line position. ASSUME that other alliance partner robots will be crossing your path every single match unless you are the first move robot as it will happen in more matches than a few no matter your robot starting position..

Avoiding collisions like your alliance partners robots have the plague is the proper approach, so 12 auto paths, (4 from each of 3 robot positions) X 3 or 4 or more delay of start choices would be about fair IMHO as a fair starting point.

Then fine tune that with very early alliance partner collaboration. Testing together on a practice field will rarely be possible so do not count on it since practice field time will be a premium this year.

3 things this year will test Woodie's favorite GP to the extreme limit this year in my opinion....1. Who gets to move first each match each side, collisions amongst both alliance partners robots and opposition alliance robots in null zones and beyond, and cubes dropped in wrong color plates at any time (Those points lost will be impossible to recover, be very careful in assessing any blame outloud).

Further, chock up the many possible mistakes that will be made to 2 things....50 percent to the randomization rule and shared null zones in auto, and no physical barriers as pertains to the A04 rule, and 50 percent to to not early enough collaboration with alliance partners if you do not heed the warning.

How many teams have ever written in any year.....and developed and fully tested and perfected (then documented for alliance partners) 36 total auto routine choices for their first and succesive matches of the season? I'm just curious.

And, that only covers 4 possible paths X 3 robot positions (12) X 3 possible preprogrammed delays (36) auto choices. Robot auto start delay(s) 1 second, 3 seconds 6 seconds as an example.

Now, the best will have 2 scale cube deposit positions or more so they can adjust to opposing alliance to avoid contact, and 3 or more switch positions also.....ooops those choices just skyrocketted.

Now, how many auto routine choices can be pre-entered into the ds to respond to the FMS data?

Last edited by cglrcng : 01-09-2018 at 03:24 PM. Reason: Correct spelling error
Reply With Quote
  #30   Spotlight this post!  
Unread 01-09-2018, 03:22 PM
John Bottenberg John Bottenberg is offline
Unregistered User
AKA: JABot67
FRC #2930 (Sonic Squirrels)
Team Role: Engineer
 
Join Date: Feb 2009
Rookie Year: 2007
Location: Redmond, WA
Posts: 559
John Bottenberg has a reputation beyond reputeJohn Bottenberg has a reputation beyond reputeJohn Bottenberg has a reputation beyond reputeJohn Bottenberg has a reputation beyond reputeJohn Bottenberg has a reputation beyond reputeJohn Bottenberg has a reputation beyond reputeJohn Bottenberg has a reputation beyond reputeJohn Bottenberg has a reputation beyond reputeJohn Bottenberg has a reputation beyond reputeJohn Bottenberg has a reputation beyond reputeJohn Bottenberg has a reputation beyond repute
Re: Unfairness in plate assignment

Quote:
Originally Posted by barn34 View Post
Pretty much all of this. The mere fact that a RNG can put either alliance at a potential disadvantage (perceived or not) is a pretty big faux pax for the GDC here. Here's hoping it's just a minor oversight and they get the software fix in so the RNG just cycles between the 4 symmetrical configurations once competition season kicks off.
Is there a way that teams can design for their robot to be able to take advantage of the asymmetrical situations, e.g., by mitigating the negative effects of the "bad" asymmetrical situations or taking further advantage of the "good" asymmetrical situations? If so, there may be some blowback from teams who designed for these scenarios if it turns out the scenarios never happen.

Also, is it possible that the asymmetric situations simply require different strategies to be employed, and that at the end of the season it will be hard to use match result data to back up the assertion that these situations create a meaningful bias toward one alliance or the other?

I really have no idea.
__________________
FLL Team "Dark Matter": 2003-2005
Robofest Team "Dark Matter": 2005-2008
Team 67 Programmer: 2007-2010
Team 3322 Programming Mentor: 2012-2014
Team 2930 Engineering Mentor: 2015-????
Reply With Quote
Reply


Thread Tools
Display Modes Rate This Thread
Rate This Thread:

Posting Rules
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

vB code is On
Smilies are On
[IMG] code is On
HTML code is Off
Forum Jump


All times are GMT -5. The time now is 10:39 PM.

The Chief Delphi Forums are sponsored by Innovation First International, Inc.


Powered by vBulletin® Version 3.6.4
Copyright ©2000 - 2018, Jelsoft Enterprises Ltd.
Copyright © Chief Delphi