Championship 2008 - Atlanta - Your Thoughts Please

[Covey41]

Originally Posted by artdutra04

They only two years they had the black cloth on the GA Dome roof was in 2005 and 2008; there was just the regular roof in 2004, 2006, and 2007.

The main reason behind the cloth in 2005 was for the CMUcam, which was not targeted at an active light that year, but rather a passive green target tetra. And for this year, it seems that they were impressed with how WPI has used climatic lighting at BattleCry@WPI for the past few years, and they wished to imitate it during the Einstein Finals.

According to Steve Sargent, (Sargent productions) the black cloth covering the roof was left over from the cosmetics show from the week before the championships, same as 2005. It had nothing to do with the CMU cameras.

[dangerousdave]

Originally Posted by Hachiban VIII

A shadow hung over the Georgia Dome this year... Literally...

It was really dark in there with the black roof. Very different atmosphere than years prior... I believe the roof was covered to cover holes caused by the Tornado.

I would assume that the black fabric hanging below the outer white colored roof of the GA Dome, was done to allow First to control the lighting and make for a more intimate setting, if that can be done in a 71,000 seat stadium. You could not see upper deck rows above about row 10. I don't believe that the tarp had anything at all to do with any storm damage as the dome suffered very little roof damage.

This year was not the first time the black tarp was used at the GA Dome for the Championship's. It was also used in 2005.

Built in 1992, the Georgia Dome is still the largest cable-supported fabric roof in the world. Stretching more than 395,000 square feet, the Teflon-coated Fiberglas fabric roof is quite an engineering marvel and the roof fabric weighs just 68 pounds. For those interested in the technical aspects of the roof go here.

Dave

[GaryVoshol]

Originally Posted by Nica F.

Good:
-I think that we all owe a lot of thanks to those who worked on the Georgia Dome + pits area + all over atlanta, because despite all of the damage, were still incredibly hospitable and continued to allow us to hold the event.

I agree - so I told them at http://www.gwcc.com/feedback.html

The staff was especially friendly this year.

[Doug G]

I've been thinking a lot about the size of the Championships and the inevitable growth of teams. I agree with many that 8 FRC fields, probably won't work (I wish it could). Limiting the number of teams seems like a bummer since attending the championships is a huge motivation for many FIRSTers.

I guess that leaves one very viable solution and that is to design a game that extends the alliances to 4v4 or 3v3v3 . Doing so would most likely mean a larger field (40' x 80' - very doeable for the 4 fields that exist now) - can smaller regionals accomodate this? I think most can. Having 8 or 9 teams on the field at once would be quite a chaotic sight!! The new control system is clearly capable of this. In a 4v4 setup, each division could go to about 100 teams and you still get 100 matches per division and each team plays 8 matches.

So can anyone tell me why a change in the number of teams on the field for each match won't help solve some of these issues being discussed?

[chessking132]

I thought championships went great this year. I had allot of fun and i believe that team 75 also had a blast.

I only have one suggestions for next year, a sundial system that FIRST can use to help que teams and send messages to teams to warn them if there is a back up in queuing like Friday morning. This would help teams make it on to the field on time and eliminate teams being in queuing for a long time. other then that one small detail i thought it was a well planed, safe, professional event.

Matthew Simpson
Team 75 driver

[Bongle]

Originally Posted by Doug G

So can anyone tell me why a change in the number of teams on the field for each match won't help solve some of these issues being discussed?

My ideas as to why those won't solve the matches problem:
Because with more teams per match comes more randomness, and a consequent decline in the quality of the rankings. By that, I mean you'll have more teams at the top of the rankings who don't really have the scoring power to have earned that spot, and who were just lucky with alliance pairings.

Plus it isn't a linear response in terms of matches. With more robots per match comes more teams carrying more robots through the same choke points on the side of the field. It'd take longer to set up the robots, for the field reps to ensure they're all turned on and aligned, and for teams to set up their OIs. So your match turnaround time will probably increase, potentially offsetting your ability to run more matches per team.

I love the idea posted above of having other fields within the GWCC. Even if you can't have B fields for all 4 divisions, maybe have 2 divisions share one field, and the other two share another. Even if only one extra field could be installed, you're still increasing capacity by 25%.

I like the idea of restricting champs to only those that have earned it that year. If a team wants to compete more in a year, they can raise money and go to a second regional. If a team wants to compete at champs, they have to both raise money AND perform. Championships should require as much, if not more effort to attend than attending a regional.

[Jay Lundy]

Originally Posted by Doug G

I guess that leaves one very viable solution and that is to design a game that extends the alliances to 4v4 or 3v3v3 .

FIRST abandoned 1v1v1 because the 2 weaker teams would gang up on the strong team. I don't see 3v3v3 being much different, unless they somehow came up with a game that discouraged that.

I don't like 4v4 because it further reduces the contribution of one robot to the overall alliance. You may get more matches but the outcome of each match depends even more on your alliance partners, so it may make the rankings an even poorer judge of skill. (Edit: Same as what Bongle said)

[TKM.368]

Originally Posted by Doug G

I guess that leaves one very viable solution and that is to design a game that extends the alliances to 4v4 or 3v3v3 . Doing so would most likely mean a larger field (40' x 80' - very doeable for the 4 fields that exist now) - can smaller regionals accomodate this? I think most can. Having 8 or 9 teams on the field at once would be quite a chaotic sight!!

Smaller emerging regionals would have a hard time not necessarily in accommodating the size, but there would be 1/4 queuing, 1/4 playing, 1/4 leaving the field, and only 1/4 left with time in the pits. As soon as you get to the pits, you'd have to turn around and get back in line.

I guess this comes down to accomodating regionals or accommodating the Championships.

This has probably been suggested someplace else before, but perhaps the winners from 5-10 regional competitions should qualify for a larger Super Regional in April. Winners of these would then earn their way to the World Championships in the early summer.

Something has to give as it seems the current model is beginning to be maxed out.

[dlavery]

Originally Posted by dangerousdave

I would assume that the black fabric hanging below the outer white colored roof of the GA Dome, was done to allow First to control the lighting and make for a more intimate setting, if that can be done in a 71,000 seat stadium. You could not see upper deck rows above about row 10. I don't believe that the tarp had anything at all to do with any storm damage as the dome suffered very little roof damage.

This year was not the first time the black tarp was used at the GA Dome for the Championship's. It was also used in 2005.

The black tarp is installed in the Georgia Dome every spring. It stays up throughout the summer, and is usually removed just before the start of football season. The primary reason for the tarp is to keep sunlight filtering through the translucent fabric of the roof from completely overwhelming the air conditioning system and heating the Dome to the point of being uninhabitable. The tarp is usually installed a few weeks after the FIRST Championship, but it may be installed earlier if other Dome clients request it. This year the Arbonne Cosmetics show, which preceded the FIRST Championship in the Dome, paid a hefty sum (you don't want to know how much) to have the tarp installed early to create the right "lighting mood" for their show.

From above, the tarp looks like this (taken while I was crawling around the catwalks above the tarp). You can also see the top rows of seats that are currently being replaced. They were scheduled for replacement prior to the tornado, and are not being replaced due to tornado damage.

-dave



.

[David Brinza]

I think that ~85 teams/div with four divisions each sending an alliance to Einstein will be the formula for the Championship for some time to come. Eight division fields and adding Einstein quarterfinal matches are just too much to accommodate from a space-time perspective (sorry for the physics pun) .

Conducting a set of four "super-regional" events prior to the World Championship is not practical since teams will require time to arrange travel and funding to attend.

As FIRST grows, the Championship event will have ever fewer "open registration" slots in order to accommodate the qualifiers (CA, EI, RA, and winners) of more regionals. The quantity of teams at the Championship will remain about the same, the quality will continue to improve. Eventually, maybe only regional qualifiers, hall-of-fame teams and prior year champions will participate in the Championship event. So, what will happen when FIRST has 100 regionals??

[EricH]

Originally Posted by David Brinza

As FIRST grows, the Championship event will have ever fewer "open registration" slots in order to accommodate the qualifiers (CA, EI, RA, and winners) of more regionals. The quantity of teams at the Championship will remain about the same, the quality will continue to improve. Eventually, maybe only regional qualifiers, hall-of-fame teams and prior year champions will participate in the Championship event. So, what will happen when FIRST has 100 regionals??

With 44 regionals next year and ~ 20 teams with a slot already, here's the math:
6 teams/regional*44 regionals +20=284 teams currently qualify through past performance/regional performance. (Next year)

So how long would it take to max out? Add 1 team per year (HoF addition) and run some numbers. I'm going to assume steady growth of 3 regionals per year. For the 2010-2012 seasons:
303
322
341
and the next year, we're over the current team limit. Or, how many more regionals will it take? At 3 a year, it takes 3 years, so 9-10 more regionals (after the 2009 season).

Not all teams will come. But, there still won't be a whole lot of room for non-qualifying teams.

So for the next 3-4 Championship Events, the current system would work. But after that, either a different qualification method (e.g. super-regionals that you get to through regionals) or more (or bigger) divisions would be necessary.

[BornaE]

Originally Posted by Doug G

I've been thinking a lot about the size of the Championships and the inevitable growth of teams. I agree with many that 8 FRC fields, probably won't work (I wish it could). Limiting the number of teams seems like a bummer since attending the championships is a huge motivation for many FIRSTers.

I guess that leaves one very viable solution and that is to design a game that extends the alliances to 4v4 or 3v3v3 . Doing so would most likely mean a larger field (40' x 80' - very doeable for the 4 fields that exist now) - can smaller regionals accomodate this? I think most can. Having 8 or 9 teams on the field at once would be quite a chaotic sight!! The new control system is clearly capable of this. In a 4v4 setup, each division could go to about 100 teams and you still get 100 matches per division and each team plays 8 matches.

So can anyone tell me why a change in the number of teams on the field for each match won't help solve some of these issues being discussed?

The problem is the randomness of the game and adding more teams to each alliance would make it more random. I rather a 2v2 or even each robot competing independently rather than alliances. the competition is strongly skewed toward the stronger teams. there are a few strong teams and lots of average teams and no matter how good you are, you need to get really lucky to win, and that is a major problem.

[JaneYoung]

Originally Posted by dlavery

This year the Arbonne Cosmetics show, which preceded the FIRST Championship in the Dome, paid a hefty sum (you don't want to know how much) to have the tarp installed early to create the right "lighting mood" for their show.

I studied the tarp work from my seat in Curie. It was excellent. I figured the installation would come at a very hefty price. Really amazing work. Thank you for this photo, Dave.

[Kevin Sevcik]

I haven't seen it mentioned, but I think the placement of the screens in the pits was pretty inconvenient. Having screens positioned such that they can't be seen from 90% of the pits seems pretty poor. I think positioning them back to back over the central display area would make it a bit easier for teams to see matches without having to worry about deserting the pits and missing the judges. Better still, place two screens at either end of the pits. I suppose this might lead to a little more loitering in the aisles of the pits, but honestly, I don't see how it could get much worse than it usually is anyways.

Also, I kind of like the idea of 6 divisions. I think 6 independent divisions is a better idea than 4 divisions with 8 fields. Mostly because of the scheduling and coordination nightmares inherent in two fields. I think it's a pretty safe assumption that an extra field only gets you somewhere between 50%-75% more matches, because the fields absolutely have to stay in sync at all times. You can't just dump teams on the first available field, because they need to know which field to show up on. And with the turnaround times for teams you'd be talking about, I don't think you could let the fields get more than 2 matches out of sync. Maybe just 1. So on the whole, I'd prefer 8 divisions to 2 fields per division.

Anyways, back to running Elims for 6 divisions. You do realize that Major League Baseball solved this problem for us about a decade ago, right? That's right, you could have two wild card teams. Determine them as you wish, but I would suggest a short round robin between two 3 alliance groups of finalists of the different divisions. I mean how often have we fretted about how strong Galileo is and how much of a shame it is that only one winning alliance could come out of it? How awesome would it be for a wild card team to rematch against the alliance they faced in their division. There are other solutions to the problem, of course. You could do the 3 alliance round robin of division champions and take the top two teams from each group. Heck, this would work well for an 8 division championship. Two 4 alliance round robins take 6 matches to finish, barring a tiebreaker match, and they could be played on parallel fields flanking Einstein. Which would be a lot faster than the minimum 8 matches you'd have to play for a single elimination playoff.

[Bongle]

Originally Posted by Kevin Sevcik

Things about double field that made me remember GTR used to run double-field

When GTR ran double-fields, each of your matches would indicate which field it was to be on (Pontiac field or Chevy field). The fields would alternate Pontiac-Chevy-Pontiac-Chevy so that while one reset, the other ran a match.

As for how much you increase playing-time:
When GTR ran double-field in 2006, they managed 124 matches in 435* minutes of venue time, for a length of 3.5 minutes of venue time per match. This includes introductions and 2m15s of play time, so that's pretty impressive.

At the Lone Star Regional in 2006 (biggest regional that was the same week that year), they managed 81 qualifying matches in 485 minutes* of venue time, for a length of 6.0 minutes per match.
*All end times rounded to next 15 minutes.

Edit: Looking again at the scheduling, my times are based off of the explicit times that teams were expected to show at the field, not when the matches actually happened. This means that this calculation only shows how much FIRST expects double-fields to increase play, not how much they actually do. GTR may have actually gone long that year (I don't remember) and LSR may have been ahead of schedule, which would throw off these calculations.

So based on that, we can say that FIRST expects double-fielding to increase match capacity by ~70%. If we say that they actually increase it by 50%, then that is still resulting in a very-good 11-12 matches per team at championships compared to the 7-8 they've gotten in recent years. Or maybe it ends with the competition ending on time. Would still require additional refs though, especially if the fields are split between GWCC and the dome.