Chief Delphi

Chief Delphi (http://www.chiefdelphi.com/forums/index.php)
-   General Forum (http://www.chiefdelphi.com/forums/forumdisplay.php?f=16)
-   -   Week 1 Observations (http://www.chiefdelphi.com/forums/showthread.php?t=135220)

excel2474 02-03-2015 16:20

Re: Week 1 Observations
 
Quote:

Originally Posted by waialua359 (Post 1452309)
I disagree.
We were 30th after our 1st 3 matches and by the end of the day, ranked 3rd.
We held onto the #1 spot and lost it in the final match by 6 total points.

Good example. That's probably because you weren't the only ones that had a couple bad rounds. It all works out.

Breadbocks 02-03-2015 17:33

Re: Week 1 Observations
 
I disagree with the "Robots control their own fate" thing. This year, more than ever, you're vulnerable to bad matchmaking.

This time, you're running the risk of being shafted by the opposing alliance not doing coop. Coop points are half of the baseline 80 points that's going to be expected of every single team every qualification match in coming weeks, and you 100% cannot do the entire coop stack yourself. If you get matched against potatoes, your ranking is going to drop, and drop precipitously. Our team dropped from top 4 to what eventually became 9th because the opposing alliance never put a 4th tote on the coop stack in one match, which was brutal on our average.

Boltman 02-03-2015 17:50

Re: Week 1 Observations
 
Quote:

Originally Posted by Breadbocks (Post 1452357)
I disagree with the "Robots control their own fate" thing. This year, more than ever, you're vulnerable to bad matchmaking.

This time, you're running the risk of being shafted by the opposing alliance not doing coop. Coop points are half of the baseline 80 points that's going to be expected of every single team every qualification match in coming weeks, and you 100% cannot do the entire coop stack yourself. If you get matched against potatoes, your ranking is going to drop, and drop precipitously. Our team dropped from top 4 to what eventually became 9th because the opposing alliance never put a 4th tote on the coop stack in one match, which was brutal on our average.

And in Dallas ONE bot put up 95. So relying on Co-op is not a good strategy sure it helps and is arguably the easiest 40 points but it does rely on others and that is never a good thing , scouting could ferret out those that do typically and those that don't there is video on prior matches...if you were scouting.

You team "shafted itself" by not being prepared for unsuccessful co-op pts. Teams that scored higher may have had the same issues yet scored higher...why? You create your own points based on how well your alliance performs as a unit.


Take IE Regional (Listed by QA in Quals):

Now strictly Cooperation

80 <---------------#1 1572 Hammer Heads (2 matches with co-op)
120
200
80 <--------------- #4 973 Greybots (2 matches with co-op)
160
200

....

240 <-------------------------#13 Golden Gears (6 matches with co-op....Best Co-Op% and 17th pick in draft)

So Co-op #'s are not indicative of Finals placement.


Hammer Heads NEVER did one single co-op attempt nor did Grey Bots and both were in Playoffs and NOW Hammer Heads in World Championships as non co-op bot.

PayneTrain 02-03-2015 18:02

Re: Week 1 Observations
 
Quote:

Originally Posted by Breadbocks (Post 1452357)
I disagree with the "Robots control their own fate" thing. This year, more than ever, you're vulnerable to bad matchmaking.

This time, you're running the risk of being shafted by the opposing alliance not doing coop. Coop points are half of the baseline 80 points that's going to be expected of every single team every qualification match in coming weeks, and you 100% cannot do the entire coop stack yourself. If you get matched against potatoes, your ranking is going to drop, and drop precipitously. Our team dropped from top 4 to what eventually became 9th because the opposing alliance never put a 4th tote on the coop stack in one match, which was brutal on our average.

1) Wait, you're telling me that you think each team should be averaging 80 on their QA in the coming weeks? Currently available data for mean QA for week 1 didn't even break HALF of that. I would be shocked if mean QA ever hit 80 before CMP.

2) While this is a true thing that does suck, 148, 3824, and other high ranking teams ranked high with <50% co-op stack conversion rate. I know our rank dropped precipitously because by the time we fixed our robot we hit a stretch of matches without co-op capability on the opposite side of the step. It's an issue we're looking to solve in two ways for our next event.

AdamHeard 02-03-2015 18:05

Re: Week 1 Observations
 
Quote:

Originally Posted by Boltman (Post 1452370)
Take IE Regional (Listed by QA in Quals):

Coopertition

80 <---------------#1 1572 Hammer Heads
120
200
80 <--------------- #4 973 Orange Crush
160
200

Not really a fan of our new name.

Boltman 02-03-2015 18:12

Re: Week 1 Observations
 
Quote:

Originally Posted by AdamHeard (Post 1452384)
Not really a fan of our new name.

Ha fixed

AdamHeard 02-03-2015 18:14

Re: Week 1 Observations
 
Quote:

Originally Posted by Boltman (Post 1452391)
Ha fixed

Ehhhhhhhhhhh. Try again.

Code Orange is 3476.

Mark Sheridan 02-03-2015 18:16

Re: Week 1 Observations
 
Quote:

Originally Posted by Boltman (Post 1452370)
80 <--------------- #4 973 Code Orange (2 matches with co-op)

When did we trade names with the greybots?

Boltman 02-03-2015 18:16

Re: Week 1 Observations
 
Quote:

Originally Posted by AdamHeard (Post 1452392)
Ehhhhhhhhhhh. Try again.

Code Orange is 3476.

Got it thx anyhow both still got far without needing co-op

dave1027 02-03-2015 18:20

Re: Week 1 Observations
 
Quote:

Originally Posted by DareDad (Post 1451941)
But, a marginal team with a kit-bot could manage to do autonomous mobility and a 10 point climb. This year, a basic kit-bot can be part of a robot set, but only if the other two members of the alliance don't blow it off as unneeded points.

Beyond that, if they try to push single totes onto the scoring platform, they're likely to get yelled at by their partners for "cluttering" the field and getting in the way. There's certainly a place for a less advanced team with a good driver to pick up litter, but that would require engineering and manufacturing an effective noodle grabber which would mean that they aren't a less advanced team.

Last year, it was about all 3 robots on an alliance working together, this year leaving the third alliance robot off the field in eliminations has been an effective strategy.

This year is a great year for teams with less engineering and manufacturing power. With a kit bit chassis, power drill, some basic hand tools, and a trip to home depot, a competative robot could be built capable of a four stack, plus topping with an RC. I know because this is what we did for a prototype. We took our prototype a few steps further, but we could have competed with the prototype. No special expertise required. I'm not saying that it would be a power house/rock star, but it would be competitive, it would have contributed.

Nuttyman54 02-03-2015 18:20

Re: Week 1 Observations
 
Quote:

Originally Posted by Mark Sheridan (Post 1452394)
When did we trade names with the greybots?

Why are these so-called "Greybots" orange? Some questions can never be answered...

Breadbocks 02-03-2015 19:54

Re: Week 1 Observations
 
Quote:

Originally Posted by PayneTrain (Post 1452380)
1) Wait, you're telling me that you think each team should be averaging 80 on their QA in the coming weeks? Currently available data for mean QA for week 1 didn't even break HALF of that. I would be shocked if mean QA ever hit 80 before CMP.

2) While this is a true thing that does suck, 148, 3824, and other high ranking teams ranked high with <50% co-op stack conversion rate. I know our rank dropped precipitously because by the time we fixed our robot we hit a stretch of matches without co-op capability on the opposite side of the step. It's an issue we're looking to solve in two ways for our next event.

1. I'm projecting that the average will be around 80, for good teams at least, in coming qualifiers, unless there's major changes to the way scoring works. Just how large the potential points from noodles are as well as coop didn't really seem to hit until a day or two in, but now that word's out teams will know to train throwing noodles and plan around them and coop.

2. Part of why I expect matchmaking to matter a lot more is because of 1. With coop becoming a thing everyone knows to do, or at least attempt, there's more pressure on everyone else to be able to do it as well. If everybody does coop, it doesn't matter. If nobody does coop, it doesn't matter. If everybody except a few teams do coop, it's brutal on the few teams because they can't and it impacts teams that get matched against them because then they get dragged down by a significant margin, unless in the time it takes your team to grab the yellows and dump them in the middle is equal to the time it takes you to get 40 points from stacks.

notmattlythgoe 02-03-2015 21:46

Re: Week 1 Observations
 
Quote:

Originally Posted by Breadbocks (Post 1452357)
I disagree with the "Robots control their own fate" thing. This year, more than ever, you're vulnerable to bad matchmaking.

This time, you're running the risk of being shafted by the opposing alliance not doing coop. Coop points are half of the baseline 80 points that's going to be expected of every single team every qualification match in coming weeks, and you 100% cannot do the entire coop stack yourself. If you get matched against potatoes, your ranking is going to drop, and drop precipitously. Our team dropped from top 4 to what eventually became 9th because the opposing alliance never put a 4th tote on the coop stack in one match, which was brutal on our average.

I actually have the exact opposite opinion you do on matchmaking. Since the randomness of your opponents ability to stop you from scoring is taken out of the equation you are only worried about the randomness of your partners.

Previously you could get paired with 2 good robots and end up against an alliance of 3 great robots. That's a loss usually. This year if the opponent is great it doesn't matter, you can still put up the points and the fact that they did more doesn't effect you.

EricH 02-03-2015 23:05

Re: Week 1 Observations
 
Quote:

Originally Posted by Nuttyman54 (Post 1452402)
Why are these so-called "Greybots" orange? Some questions can never be answered...

I've never been able to figure that out, though they were wearing BLACK yesterday (I was looking for orange...).


For Tom B.: I think Glenn missed an opportunity to plug their third partner, who kept very busy running smaller stacks into position. VERY busy. One medium-tall stacker (359, #2 seed and #1 pick), one RC specialist that could handle short tote stacks (1572, #1 AC), and one short-stacker (Livewire--forget their number), one set of dominating matches.

Sperkowsky 03-03-2015 00:11

Quote:

Originally Posted by PayneTrain (Post 1451964)
As much as I would love this, it makes the rarely used but still existing "processing litter" mechanic hard to execute. I'd prefer if they go ahead and "recycle" all of the noodles tomorrow morning, though.

With regards to the numbering on robots, I can't tell you how surprised I wasn't how the numbering turned out. Bonus points to robots at Palmetto with vertical and split numbering, as well as teams just throwing on the number plate they get in the pits and rolling out to the field with them (this actually happened).

With regards to the game being hard to understand, I really don't know how the GDC expected it to be easy to understand. I know hindsight is 20/20 but to improve the understandability of the game, I would have done this:

1) Get rid of Litter
2) Divide all remaining point values by 2 (not really necessary, but it helps later)
3) Instead of gratuitously stamping the FIRST logo on every game piece even though the banners, scoring overlay, and logos on the field walls apparently aren't enough, and replace every logo with the point value stamped on to the object. Every tote has a big white "1" printed on each side and each can has each FIRST logo replaced by a big white "x3" stamped around the cylinders.
3-a) Change auto point values so we don't see the most useless auto period since 2010. Tote set is worth 6, robot set is worth 4, and stack set is 12, but the containers are x3 to your auto score for both consistency but also to encourage better auto play.
4) Replace the gold totes on the field at the start of the match with gray totes, or a different color. Keep gold totes off the field. Allow the new totes in the staging zone to be scored on the bumps. Stamp the +1 on those
5) Replace the vision targets no one uses with the "grabsomefeet" co-op logo on the cans. On each side of the step have a display that shows the current point value of the co-op stack. In eliminations, have the display show the seeds playing in the matches.
6) Maybe have displays on top of each backstop showing how many points currently sit on each backstop.
7) (Totally optional and a little crazy) Somehow get the bumps to illuminate where a scored stack is placed

If I had no clue what this game was when I saw it, I would wonder why teams are throwing scoring objects to the other alliance, why teams are delivering yellow totes to a center object in qualification matches for a lot of points but don't ever touch them in eliminations (when points matter more, right!), why a stack of 6 totes tall is not worth any more than 6 totes laying alone on the bumps, and understand exactly how much cans help.

I'm pretty sure as much as 11 years ago FIRST had scoring object values printed on the objects (Frenzy, right?). This way I could look at the field and figure out exactly what was happening. You count "1, 1, 1, 1, x3... that's 12 points!" You see the co-op display and know how much that is worth.

Post-match overlay could be really good if some thought was put into it. One idea I have at 2:30 AM is this:

1) Take a screenshot of the field before the field gets the green lights.
2) Have the first overlay take up the lower third, like the small one we see in the playoffs, while you see the field screenshot.
3) After that screen, move to a screen that shows the final score of the match as well as a sample of the standings. Have 16 slots to show at once. You get the top 10 + the 6 robots in the last match on the screen showing their rank and change arrow, QA and change arrow, Co-Op score, and Next Match. The teams not in the match are not highlighted while the ones from the match are highlighted in their alliance colors. If one of the teams from the last match are in the top 10, you show the 11th place team, 2 means you show up to 12th, and so on.

One concerning thing I have gotten out of this game so far has nothing to do with FRC the program. FRC is a product. A game like this is hard to sell as a product to potential sponsors, volunteers, media partners, schools, etc.

I agree with most of what you said but keep those vision targets we use them.


All times are GMT -5. The time now is 17:16.

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