Chief Delphi

Chief Delphi (http://www.chiefdelphi.com/forums/index.php)
-   General Forum (http://www.chiefdelphi.com/forums/forumdisplay.php?f=16)
-   -   Week 1 Impressions of Rack 'N Roll (http://www.chiefdelphi.com/forums/showthread.php?t=55183)

Render 04-03-2007 15:42

Re: Week 1 Impressions of Rack 'N Roll
 
1885 played 1895 in every qualifying match as well...

Eldarion 04-03-2007 15:45

Re: Week 1 Impressions of Rack 'N Roll
 
Quote:

Originally Posted by PandaMan (Post 590112)
I was talking about teams in general. :yikes:

So was I! :) Judging from what I saw (one or two robots had autonomous out of the entire regional and lots of fast, effective arms), I think a lot of teams may have had a similar discussion.
Quote:

Originally Posted by PandaMan (Post 590112)
Even so, Having an autonomous that works 75% of the time would have made a huge difference in many of those 0-0, 2-2, 2-4 type matched. In such cases, similar to last year, low scoring matches can be greatly affected by autonomous.

I will concede that point.

henryBsick 04-03-2007 15:57

Re: Week 1 Impressions of Rack 'N Roll
 
Of the many thoughts that I have had, two stick out.
1)New 'randomization' algorithm for qualifiers sucks. Plain and simple.

2)I was very dissapointed at the lack of offensive-defense that I saw at NJ. Think Connect 4 people!!!

Predictions:Teams with a consitant arm and RELIABLE dual 12" ramp systems will enjoy victory. (Teams like 67 and 179 pop into my mind) They are rare, but there is one spot out of 3 on an alliance for one of this type of team. Couple that robot with two amazing tubers and you have yourself a win.

Madison 04-03-2007 16:28

Re: Week 1 Impressions of Rack 'N Roll
 
I guess I'll start with my gripes so that I might end with some positive comments.

The bad:
  • Someone, somewhere needs to do something that'll stop these first regionals from being guinea pigs for everyone else's benefit.
  • It seemed like enforcement of the rules was phased in throughout the day Friday and through Saturday. On Friday, our lifts were never measured once deployed and we were awarded a 30 pt. bonus. On Saturday, during the elimination rounds, our lifts were measured every time they deployed and we found that there were circumstances that resulted in our partners dipping below 12". That's an important bit of information to know before starting the elimination rounds. Enforce all of the rules all of the time.
  • We were penalized once for exceeding a 72" x 72" box outside our home zone when the corner of our one, deployed lift passed over the tape. I didn't see anyone actually measure anything, though. I can't eyeball 6' within one or two inches, so I'm happy that FIRST found some volunteers that could.
  • The match generation algorithm is simply terrible and needs to be changed before the next set of events. It's not fair to anyone.
  • We were disallowed from testing an autonomous mode that used about 3/4 of the practice field because it interfered with a side of that field that folks were using to practice throwing tubes. I get that FIRST's not about robots, but can we all agree that the robots are just a bit more valuable to the message than the human players?
  • After alliance selection, the practice field was closed and packed away. With so many different sorts of ramp/lifting robots around and a match generation algorithm that ensures you won't play with a majority of the teams at the event, we felt that making sure our partners (that had never played with or against our machine) knew how to get atop our lifts was important and the practice field makes doing that a lot easier. Let's keep the practice field around until the elimination rounds start next time.

The good:
  • The game is great fun to strategize and play. I can't speak to how things appear from the stands, but the field gets pretty crowded at some times and coordinating between three teams to stay out of everyone's way and still play effectively was challenging.
  • The 'random' matches meant that we played against team 492 (another group from near Seattle) in every match. As a result, they got consistently better at defending us each time we played and we had a great time trying to find better, more effective ways of playing defense while trying to avoid being trapped on the other side of the field.
  • We're exceptionally happy with our robot and strategy. As the weeks pass and teams get better, it will become increasingly difficult for lifters to dominate completely, but I think they'll be an integral part of winning alliances.
  • Blair has called our robot the "White Wonder". We like it.
  • The random placement of the Rack isn't nearly as giant a variable as it seemed at first. Some teams were able to consistently score during autonomous mode.
  • 30 pts. of penalties for sitting in an opponent's end zone was a large enough disincentive to stop most teams from opting to try such a strategy. Those teams that did lost as a result every time I'm aware it happened.
  • Congratulations to Eric Stokely for winning the Regional Woodie Flowers Award
  • Congratulations to team 1540 -- The Flaming Chickens -- for winning the Regional Chairman's Award. Our kids would love to know more about your presentation so they know how to better prepare themselves next year.
  • Having a functional robot means we got to do more to help other teams.
  • The referees were great about addressing questions from our students, even if we weren't always happy with the answers.

My gripes, excepting the match list and consistent use of rules, are all pretty inconsequential. The event ran well, we had a great time, the volunteers all did a great job to keep things running smoothly, and we like the game and think that it's dynamic, challenging and fun to play. We're really looking forward to heading down to San Jose, CA in two weeks to see how things have changed and we hope our machine is as effective and valuable then as it was this weekend.

Athleticgirl389 04-03-2007 16:33

Re: Week 1 Impressions of Rack 'N Roll
 
Quote:

Originally Posted by Alex Cormier (Post 589228)
The game is lame.

I am in total agreement with you Alex. There is no excitment in watchin this game WHAT SO EVER. There is some in playing it, but it is dull. Unlike last year, you did not know who won until the very end; this year you know when that fianl buzzer goes off. Suspense is good and it is deffinately missing this year.

wilsonmw04 04-03-2007 17:30

Re: Week 1 Impressions of Rack 'N Roll
 
Quote:

Originally Posted by T. Hoffman (Post 590016)
I wholeheartedly second this request. I'm surprised it wasn't a natural part of the original scoring software design.

There was a count down during each match. It was posted in the middle team's control area for each alliance. It counted down the two different periods of each match. Now if you are only talking about the big screen, then you are correct.

Noah Kleinberg 04-03-2007 18:22

Re: Week 1 Impressions of Rack 'N Roll
 
To the people who think autonomous mode isn't very important this year, don't think of it as two points, scoring is exponential. especially if more than one robot on your alliance scores during autonomous, you're at a pretty big advantage for the human control part of the match.
I also noticed several times in autonomous that teams would position their robot perfectly to score, but not release the keeper; it's not too hard to use a timer and just make sure that at the end of autonomous mode you drop your keeper no matter what.

Also, while the game can be boring during qualification matches, I thought it got pretty exciting during eliminations. One tube can change the score by a lot, compared to for example one ball last year.

Duke_of_Hazard 04-03-2007 19:14

Re: Week 1 Impressions of Rack 'N Roll
 
Well, from my point of view, the outcome of matches is not going to be decided from robot design as much as from driver skill. This game demands coordinated drive teams and precise movements. If you watch, the really dominant teams not only had a good robot, but

1- the base driver was in a class above, out-maneuvering the opposition with ease (and also didnt fall off of ramp-bots as much)
2- the operator had a good sense of timing to release the tube and score
3- both drivers new what each other was about to do, and were able to change tactics on the fly

Last year it wasnt as obvious, with the balls *literally* flying, but in nearly all the games from FIRST, if you look in the background, and not the pretty robots(though its hard not to!), its the more skilled and coordinated alliances that win.

Sgraff_SRHS06 04-03-2007 19:28

Re: Week 1 Impressions of Rack 'N Roll
 
Not only looking at the final scores, but also looking at how the bots played, I'm a bit disappointed, especially because of the way the game is being played so slow.

It's like watching a match between 2 very poor alliances virtually every single match in Triple Play.


Sure, there were lots of teams that compromised. But something has to be done about match pairing. What should happen is that everyone should have either played with or against everyone else. And based on what happened, it did not necessarily happen.

But I think the real fault is the failure of the game to have any kind of loading zone. It would at least save half of the confusion of this game (with projectiles being thrown everywhere).

A bot that can be dominant in this game is...truthfully speaking...going to have to go through match play like there's no tomorrow and score as many as possible. Speed more than brute force is going to win this game.

wilsonmw04 04-03-2007 20:09

Re: Week 1 Impressions of Rack 'N Roll
 
Quote:

Originally Posted by Sgraff_SRHS06 (Post 590347)
A bot that can be dominant in this game is...truthfully speaking...going to have to go through match play like there's no tomorrow and score as many as possible. Speed more than brute force is going to win this game.

I can't disagree more. Big bots who play D will be able to control the rack. Defense is going to win games, not offense. The best alliance will be 1 ramp bot (with 2x12" lifting), one scoring bot ( needs to score 2-3 ringers with heavy D) and a good defensive Bot. Winning matches will be in the 68-76 point range.

Pavan Dave 04-03-2007 20:11

Re: Week 1 Impressions of Rack 'N Roll
 
I think I'lI think I'll answer my own questions now.

After my initial reaction to the game I have had a COMPLETE change in mind toward the whole thing thanks to a couple of people, (you know who you are...I hope).

From the regionals I have seen I thought that it is slow. The game seems to last forever if there are not as competitive teams on the field, but if there is a tube on a robot constantly, and the drivers are great, it makes for a great match up. This will tremendously change though because of the technologies that are present that teams can utilize (See this topic!) I expect to see two full turns scored around the racks (2X8) by two teams by the end of championships.

I also think that the minor disappointments I have with the use of certain game pieces (spoilers) will dramatically change throughout the course of the season due to teams learning from others' mistakes. Although they will not be used as much in qualifying rounds I think that by Lone Star and Las Vegas, that there will be at least six spoilers placed per regional at the end of the days and definitely at championships there will be six to ten placed a day.

Besides the slow start I believe that this game holds a giant opportunity for teams to work together in a different sense. Instead of trying to win or lose I believe this year more friendships will form due to the nature of this game and the different interactions throughout game play. My only rant for this game is the "initiating end sequence" thing which really got annoying for me after a couple of matches, but I guess I can live with it!

Peace,
Pavan.

sovierr 04-03-2007 20:12

Re: FIRST Impression
 
Quote:

Originally Posted by Hieb (Post 589153)
I'll respectfully disagree with this, at least the boring part. Maybe it's because I'm a soccer fan and can enjoy 1-1 tie more than a 100 point win, but I saw some very exciting matches today, including one we lost 13-15. Just because teams aren't consistently scoring 100+ points doesn't mean it's a boring game.

I guess there is a difference between low scoring due to good defence or low scoring due to poor offense. The former can be quite exciting, while the latter is sure to be a snooze fest!

sovierr 04-03-2007 21:36

Re: Week 1 Impressions of Rack 'N Roll
 
Quote:

Originally Posted by Eldarion (Post 590093)

Also, a little note as to the rack. There are several times when the rack was moved such that both lights were out of the field of view of the camera. There probably should have been lights above all 8 columns, not just the four columns.

There are 4 lights on the rack, not just 2. There can't be an orientation that prohibits a bot from seeing all 4. Also, if there were 8 lights there would be times that the robots saw 3 at the same time - and possibly even 4. This would increase the programming difficulty quite a bit.

ldeffenb 04-03-2007 21:42

Re: Week 1 Impressions of Rack 'N Roll
 
Quote:

Originally Posted by Eldarion (Post 590093)
As a programmer, I am very disappointed by the lack of emphasis FIRST placed on autonomous. 2 points is something that can be gotten far easier by making the arm a little faster, playing a little better defense, etc.

I'm also the team programmer and we didn't bother to even THINK about the emphasis on autonomous. We used it to drive the team to implement a controllable, reproducible, reliable tube handling mechanism. That enabled autonomous scoring but truly benefited the operator-initiated robot operation even more.

Quote:

Originally Posted by Eldarion (Post 590093)
Also, a little note as to the rack. There are several times when the rack was moved such that both lights were out of the field of view of the camera. There probably should have been lights above all 8 columns, not just the four columns. This would also have allowed semi-automatic scoring during manual mode, and may once again have increased the number of ringers and spoilers placed on the rack.

Our experience is that you will ALWAYS see at least ONE light if you point your robot at the center of the playfield. If you look straight down the side, you might not see any, but if you angle the initial placement towards the center of the field, no matter how they translated or rotated the rack, at least one light would be visible. But then, we also changed the lens on our camera to a 90 degree wider angle.

I disagree with 8 lights making it easier. That would have just made more cases with multiple lights in view which (at least in our implementation) is a much more difficult autonomous situation to solve.

As for semi-automatic scoring in manual mode, I'm going to go public with a hint from Team 386 for the upcoming Central Florida Regional. Just a single word, EasySCORE (tm). Watch for us at UCF!

Lynn (D) - Team Voltage 386

Gene F 04-03-2007 21:45

Re: Week 1 Impressions of Rack 'N Roll
 
I like this years game! I agree that some of the low scoring was not what I expected but this is the first weekend of competition. The quarter-final matches at VCU between #2 and #7 are what would expect to see more of as the weeks progress.

Scores were 48 to 68
92 to 32
and 100 to 62.

Each alliance had a good defensive ramp, each had two robots that could score. Both scoring robots could also get onto the ramps reliably. What's left... scoring on the rack makes the difference. Which means offence, defence and counter defence all come into play. This makes for some very exciting and fun to watch matches as the members of the alliance switch roles throughout the match. What decided the matches was the rack. Both alliances knew that winning the rack was paramont to winning the match to the extent that they risked (and at times lost) not having enough time to get on the ramp reliably.

As teams become more confident in their ability to score offensively, defence will fade a bit but will remain a major part of the game.


All times are GMT -5. The time now is 22:49.

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