Chief Delphi

Chief Delphi (http://www.chiefdelphi.com/forums/index.php)
-   General Forum (http://www.chiefdelphi.com/forums/forumdisplay.php?f=16)
-   -   Week 2 Analysis (http://www.chiefdelphi.com/forums/showthread.php?t=127665)

Karthik 12-03-2014 13:03

Re: Week 2 Analysis
 
Quote:

Originally Posted by Travis Hoffman (Post 1358101)
I've not seen much if any discussion on this (maybe I'm not looking in the right place). Does anyone have any firsthand experience with such a scenario happening to them? Is this actually a thing?

There's been some of this discussion in the Hot Goal Timing Issues thread: http://www.chiefdelphi.com/forums/sh...5&postcount=16

Travis Hoffman 12-03-2014 13:25

Re: Week 2 Analysis
 
Quote:

Originally Posted by Karthik (Post 1358110)
There's been some of this discussion in the Hot Goal Timing Issues thread: http://www.chiefdelphi.com/forums/sh...5&postcount=16

Hmm one Crossroads mentor I just spoke with said he observed the same type of behaviour in some matches at CR.

Wonder if that happened to our post-match rolling outside the field boundary tech foul alliance partner dudes. My guess is they were going to turn right to align themselves with the ball for a late pickup, but power went kaputt early and they coasted over the wall before they could turn.

Reasonable doubt, wooo!!! I learned that by watching Law and Order. :rolleyes:

RallyJeff 12-03-2014 13:33

Re: Week 2 Analysis
 
Quote:

Originally Posted by Travis Hoffman (Post 1357378)
So the pedestal lights upon an end cycle confirmation, and it turns off following the initial possession entry for the next cycle. Can ANYONE give me any justification whatsoever for that blasted light to not be controlled locally by a "pedestal ref" holding a toggle switch hooked to a power supply, lol. Please don't tell me the light inside is donated and has to be controlled by Ethernet I/O or some such nonsense.

Scoring a goal and ending the cycle doesn't just light the pedestal. It also clears the posessions from the tablets of the refs tracking that alliance and the "assist" indicator LEDs (both the ones on the goals and the ones in the driver stations). For all this to happen seamlessly, it all has to be linked through the FMS.

Quote:

You can get rid of all delays related to the pedestal and the field by severing the lighting completely from the cycle process. Pedestal ref turns it on when the ball is through the goal (presume lights on the opposite scoring goal turn off once the end cycle is confirmed - PRETTY OBVIOUS SIGNAL TO TURN THE PEDESTAL ON). Turn it off once the HP inbounds and the first possession/assist light turns on. Note any early pickup infractions and communicate to nearest foul-calling ref. It's not like they'll be busy doing anything else for a bit once the HP picks the ball up....

You will have pedestal lighting times of sub 1-second if you simply have humans do it and make it the only thing they have to track during the match. [More gameplay. Less blar.] <-- Spotlight worthy, no? Make it happen, people. :-P
It's not as simple as you make it sound. There are many situations where a ball through the goal doesn't - and shouldn't - cause a lit pedestal: an illegal goal, for instance, or clearing a dead ball that got freed.

And there are cases where lighting of the pedestal should be legitimately delayed. Balls don't count as scored if they're touching an alliance bot, so if the scoring robot is still reaching into the goal and contacting the ball, the ref can't end the cycle yet even if the ball is completely through the goal. Also, one of the requirements to score is that the ball remains in the goal, so the ref can't count it as scored until it's clear that the ball won't bounce out.

Even if a ref was only worried about monitoring scoring, they'd still need to be aware of what's going on over the entire field, and there could still be perceived delays in illuminating the pedestal.

Overall, I think the system we have takes some practice to use quickly, but it's pretty good. I understand that Week 1 events reported a system lag, but the updates since then seemed to have dealt with that issue. At GTR East, when I could, I'd try to check how quickly the pedestal would illuminate. Every time, it would already be lit within the fraction of a second it took to raise my head and look after hitting "end cycle".

mathking 12-03-2014 13:48

Re: Week 2 Analysis
 
At Crossroads there were some very noticeable delays in the pedestal lighting up. Q24 that Travis posted a link to was just one of several I noticed. Most of the time when it happened I saw a referee working on their tablet. They have to toggle between screens and sometimes that delays things. Particularly if there is a foul just before or after a goal is scored. The delays are a problem, but the inconsistency is an even bigger issue. In Q32 at Crossroads we had a 7-8 second delay once, but the other alliance had one over 15 seconds. That necessitated a replay. Which is why I still think the best solution is spreading the responsibility around more. If there were one person watching each robot for possessions and scores I believe we would see fewer errors and delays. The other referees can concentrate on fouls. It is not just as simple as assigning more people, you would have to change the FMS to allow this.

Travis Hoffman 12-03-2014 13:59

Re: Week 2 Analysis
 
Quote:

Originally Posted by RallyJeff (Post 1358129)
Scoring a goal and ending the cycle doesn't just light the pedestal. It also clears the posessions from the tablets of the refs tracking that alliance and the "assist" indicator LEDs (both the ones on the goals and the ones in the driver stations). For all this to happen seamlessly, it all has to be linked through the FMS.

I'm not suggesting any of those functions be severed from the FMS; just the pedestal light.

Quote:


It's not as simple as you make it sound. There are many situations where a ball through the goal doesn't - and shouldn't - cause a lit pedestal: an illegal goal, for instance, or clearing a dead ball that got freed.

And there are cases where lighting of the pedestal should be legitimately delayed. Balls don't count as scored if they're touching an alliance bot, so if the scoring robot is still reaching into the goal and contacting the ball, the ref can't end the cycle yet even if the ball is completely through the goal. Also, one of the requirements to score is that the ball remains in the goal, so the ref can't count it as scored until it's clear that the ball won't bounce out.

Even if a ref was only worried about monitoring scoring, they'd still need to be aware of what's going on over the entire field, and there could still be perceived delays in illuminating the pedestal.

I like the low tech flag method mentioned earlier, combined with a manual pedestal light. Referee signals somehow that end cycle is complete (heck who needs a flag - they could just put their hands up in the ay-ahhh like they just don't cay-ahhhh...whatever). Pedestal attendant/ref/drone/whatever then turns the dang light on. Due to the overt and flamboyant (jazz hands???) action of the scoring ref, the pedestal attendant/ref/drone/whatever doesn't have to guess what scoring ref is thinking re: any of the situations you mentioned above.

Quote:

Overall, I think the system we have takes some practice to use quickly, but it's pretty good. I understand that Week 1 events reported a system lag, but the updates since then seemed to have dealt with that issue. At GTR East, when I could, I'd try to check how quickly the pedestal would illuminate. Every time, it would already be lit within the fraction of a second it took to raise my head and look after hitting "end cycle".
Canada must be a magical fairyland of FMS functionality and wonderment! :)

tavar321 13-03-2014 20:03

Re: Week 2 Analysis
 
https://docs.google.com/document/d/1...c_ekZaS7U/edit

^ This document compiles content from many people’s posts on key subjects from http://www.chiefdelphi.com/forums/sh...d.php?t=127313 and http://www.chiefdelphi.com/forums/sh...=Week+Analysis
(Week 1 Analysis up till the beginning of page 28, and Week 2 Analysis up till pg 5). Team 1389 made this compilation to learn from others' opinions, observations and questions about tournament play so far. We are happy to share this document with other FIRST teams in the spirit of gracious professionalism.

Bill_B 13-03-2014 21:08

Re: Week 2 Analysis
 
Tavar, 2 points.
great idea to consolidate this mountain of commentary.
use message numbers instead of page numbers - not everyone shows the same number of messages per page.

tavar321 13-03-2014 22:09

Re: Week 2 Analysis
 
Thank you, edited those in.


All times are GMT -5. The time now is 21:28.

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