Chief Delphi - Team Update 14 (2016)

Page 12 of 13

« First

Show 100 post(s) from this thread on one page

Chief Delphi (http://www.chiefdelphi.com/forums/index.php)

- General Forum (http://www.chiefdelphi.com/forums/forumdisplay.php?f=16)

- - Team Update 14 (2016) (http://www.chiefdelphi.com/forums/showthread.php?t=145043)

marshall

03-03-2016 09:30

Re: Team Update 14 (2016)

Quote:

Originally Posted by GeeTwo (Post 1550764)

Bag and Tag actually simulates real world situations fairly well.

Absolutely right! The struggle is real:

http://www.cnn.com/2012/08/10/us/mar...ss_igoogle_cnn

Quote:

Originally Posted by Article

It landed Monday and will spend the next four days installing operational software that will give it full movement and analytic capabilities, scientists said at a news conference at NASA's Jet Propulsion Laboratory in Pasadena, California.

And to think, all those NASA dudes had was a rover in a bag that they couldn't test with back here on Earth... Good thing they had Matt Damon to science the crap out of the firmware for them.

KevinG

03-03-2016 09:38

Re: Team Update 14 (2016)

Quote:

Originally Posted by GeeTwo (Post 1550764)

Bag and Tag actually simulates real world situations fairly well. You've got to get the main piece on a barge or slow ship across the ocean a couple of weeks before installation/use on the other side of the world, but the techs and engineers can bring a limited amount of stuff in excess baggage. COTS stuff can be procured at the install site, or shipped directly there from the manufacturer. This very closely resembles the situation of my department's data collection branch.

As a simple way to prevent further proliferation of things which look like a robot but aren't, and to avoid incredibly long, complex rules, how about these:

The bagged robot (and spare parts) must weigh no more than 160 pounds (or pick another number) total for all bags. This applies both for initial bag-and-tag and rebagging after any demo, open-bag period, or event. Bags will be weighed by inspectors before teams may un-bag. Any overage is deducted from the weight allowed under the withholding allowance.
At check-in to an event, each team is issued a "robot button" (e.g. 2 inches in diameter backed with velcro loops) bearing the team number. The robot must have this button affixed in order to enter the practice field or the match queue.

This would allow duplicates of assemblies, but in order to have duplicate robots, they would have to be considerably underweight or have a lot of quickly removable COTS parts.

Mandating that inspectors weigh bags first puts a very large and unnecessary burden on them, particularly in the early hours when we're already having to deal with all of the teams that forgot their lockup form, have a torn bag, or (my favorite) bagged their lockup form inside with the robot. In a district model where our goal is to get robots inspected ASAP it's even less viable. That said using a weight limit is a viable approach.

I would make the following rule changes:

1. The maximum weight of the bag may be no more than 130 pounds. Bumpers do not count as part of the bag weight limit.

2. Bags will be weighed at the discretion of the LRI. It is recommended that teams weigh their bags and mark the amount on the sheet. In the event that a bag is found to be overweight teams will be instructed to remove items until the weight is met. Those items will be quarantined until the end of the event.

3. Robots that have not passed inspection may not be powered outside of the pit area. This includes practice fields.

4. COTS items assembled in accordance with the manufacturer's specifications are not considered prefabricated items. The assembly may only consist of components purchased as part of a single COTS item.

These changes would allow teams to ship a robot (even with bumpers on) with 10 pounds of leeway, and not place an undue burden on the inspectors unless the LRI sees something that deserves attention. The requirement that robots pass inspection before being powered on outside of their pits eliminates the utility of having two robots and also is a logical safety requirement. The COTS change provides a bit more flexibility to teams in terms of their spares, and also eliminates what I consider to be a fairly silly rule.

D_Price

03-03-2016 09:53

Re: Team Update 14 (2016)

it could for sure be of some use!

Quote:

Originally Posted by Doug G (Post 1550607)

This will fit nicely in my pit :yikes:

Daniel_LaFleur

03-03-2016 11:13

Re: Team Update 14 (2016)

Quote:

Originally Posted by EricH (Post 1550525)

Here's my suggestion to improve bag-and-tag:

Cut the withholding allowance to 20 lb SPARE parts. (Or remove it altogether.) COTS items remain unlimited.

I'd vote for remove it altogether.

Quote:

Originally Posted by EricH (Post 1550525)

The loose translation would be that if it doesn't match what's on the robot already*, it's gotta be in a bag on bag day.

[insert "grumpy mentor" "back-in-my-day-we-boxed-up-the-robot-and-all-its-spares" section here]

*The definition of "match" is intentionally left a little bit fluid, because no two parts will be 100% identical--maybe something got an extra hole somewhere or something like that. I hate to use the "reasonably astute observer" standard but that may need to be what is used.

Define match as "functionally equivalent"
Also, limit the weight of the bag to 200 LBS.

AdamHeard

03-03-2016 11:58

Re: Team Update 14 (2016)

Quote:

Originally Posted by Chris is me (Post 1550766)

I mean, doesn't that kind of prove his point though? The robot is on its side in the picture, when did 254 ever tip over in 2014? 973's wooden wonders were good robots and all, but I don't think anyone would say they were nearly as good as 254's machine, 973 themselves included.

As an aside: I don't think copying in FRC will ever be as rampant as it is in Vex. In Vex, you are mostly using a set of COTS parts that can be put together in a relatively finite number of different ways to achieve the game objective. There is less customizability and design flexibility than there is in FRC. The nature of Vex itself makes it much easier to copy and compete, and removing the bag won't make that suddenly happen in FRC to the same extent it happens in Vex.

With the argument that a powerhouse team would just build several different robots and give them out to alliance partners or whatever, when has that ever happened in Vex? That should be really easy in Vex, right?

It's not really a fair entry into this argument either.

It was a robot designed in ~ 2 hours and made in a day out of 2x4s that was top heavy as heck.

The_ShamWOW88

03-03-2016 12:17

Re: Team Update 14 (2016)

Quote:

Originally Posted by Mr V (Post 1550372)

To those that think if there was no bag day that there wouldn't be teams that choose a week 5 or 6 event with the idea that they will finish their robot after watching a week 1 or 2 event are being quite naive. There would certainly be some that do that with the goal of copying the top performing robot from those early weeks. I'm not saying that they will all necessarily build a robot that performs as well as the original but it is likely that some will come close and maybe a couple will build one that does even better. I also think that there would be a few teams that don't decide which route to follow until seeing week 1. In the context of this year's game I could see a team building both a low bar and a non low bar robot and then deciding which one to finish perfecting to take to their event.

Even if the desire is not to copy a top performing robot you can not deny that there would be teams that pick later events to give them more time to perfect their robot, driving and code. It could make it quite hard to fill up those week 1 and 2 events and I believe it would be even more of a problem with areas in the district system.

This.

Believe it or not, teams will attempt this.

But the argument is, you either prefer bag-and-tag and it prevents extreme reiteration or you don't prefer it and put up with the inevitable cons it brings, such as this.

Chris is me

03-03-2016 12:21

Re: Team Update 14 (2016)

Quote:

Originally Posted by The_ShamWOW88 (Post 1550857)

Who's going to perform better though - the team that copied a robot off of a few photos in 5 weeks, or the team that built their own robot in 12 weeks, with several iterations and weeks of drive practice and tuning?

Being good at FRC consists of so, so much more than the actual design of the robot. Too much emphasis is placed on the high level robot concept being the absolute key to success, when it is ALL in the implementation and the details.

Michael Corsetto

03-03-2016 12:22

Re: Team Update 14 (2016)

Quote:

Originally Posted by The_ShamWOW88 (Post 1550857)

But the argument is, you either prefer bag-and-tag and it prevents extreme reiteration or you don't prefer it and put up with the inevitable cons it brings, such as this.

Your point assumes that "extreme reiteration", or just outright copy-ing, is a "con".

I don't think it is. Copy away!

For anyone looking to copy our 2016 robot, good luck ;)

-Mike

The_ShamWOW88

03-03-2016 12:22

Re: Team Update 14 (2016)

Quote:

Originally Posted by Chris is me (Post 1550858)

Oh I totally agree with you. My point was simply that teams will do it.

GeeTwo

03-03-2016 12:23

Re: Team Update 14 (2016)

Quote:

Originally Posted by KevinG (Post 1550788)

Robots that have not passed inspection may not be powered outside of the pit area. This includes practice fields.

Currently teams are allowed to play on the practice field or regulation field during their designated practice matches without passing inspection. At the risk of some additional paperwork, what if inspection were split into two parts?

Pre-inspection would only include major safety items, such as sharp or otherwise unsafe parts, insulated connections, no electrical connection to frame, many if not all of the pneumatics rules, and probably a few others.
Full inspection would begin with or build on on pre-inspection and also include wiring details, final weigh-in, bumper rules, frame perimeter measurement

A robot must pass pre-inspection to power up on the practice fields or for scheduled practice matches, and both pre-inspection and full inspection for the filler line or competition matches. Only one pre-inspected or fully inspected robot allowed per team.

The_ShamWOW88

03-03-2016 12:27

Re: Team Update 14 (2016)

Quote:

Originally Posted by Michael Corsetto (Post 1550860)

Your point assumes that "extreme reiteration", or just outright copy-ing, is a "con".

I don't think it is. Copy away!

For anyone looking to copy our 2016 robot, good luck ;)

-Mike

Not exactly. Some may view it as a con, others may view it as a pro. Either way we're both right.

I guess the point I was trying to make was either you get rid of bag and tag and some will attempt to copy/reiterate an existing design or you keep it and people continue to worry about other teams having enough resources to build practice robots....

marshall

03-03-2016 12:27

Re: Team Update 14 (2016)

Quote:

Originally Posted by GeeTwo (Post 1550863)

Pre-inspection would only include major safety items, such as sharp or otherwise unsafe parts, insulated connections, no electrical connection to frame, many if not all of the pneumatics rules, and probably a few others.
Full inspection would begin with or build on on pre-inspection and also include wiring details, final weigh-in, bumper rules, frame perimeter measurement

This is actually what I thought the update was going to be based on some brief conversations with Jon and Frank in Palmetto... I'm sad that it didn't go this way but it is what it is.

The_ShamWOW88

03-03-2016 12:28

Re: Team Update 14 (2016)

Quote:

Originally Posted by GeeTwo (Post 1550863)

Pre-inspection would only include major safety items, such as sharp or otherwise unsafe parts, insulated connections, no electrical connection to frame, many if not all of the pneumatics rules, and probably a few others.
Full inspection would begin with or build on on pre-inspection and also include wiring details, final weigh-in, bumper rules, frame perimeter measurement

It seems to be hard enough to get teams to inspect the first time. Asking them to do it twice. Especially at District events where we don't have a full practice day to get it done.

Richard Wallace

03-03-2016 12:44

Re: Team Update 14 (2016)

From the Manual:

Quote:

Originally Posted by FRC 2016 Game Manual, Sectio 5.5.2

ROBOTS are permitted to participate in scheduled Practice MATCHES prior to passing Inspection.
However, the FIRST Technical Advisor (FTA), LRI or Head REFEREE may determine at any time that the ROBOT is unsafe, per Section 3 (3.4.1 Safety Rules), and may prohibit further participation in Practice MATCHES until the condition is corrected and the ROBOT passes Inspection.

Many years ago as a new LRI, I tried to implement what Gus suggested. Turned out it was too confusing for teams and volunteers alike. The Manual (quoted above) has it right, IMHO.

BrennanB

03-03-2016 12:59

Re: Team Update 14 (2016)

Quote:

Originally Posted by AdamHeard (Post 1550850)

It's not really a fair entry into this argument either.

It was a robot designed in ~ 2 hours and made in a day out of 2x4s that was top heavy as heck.

It is however a decent indication that if 973 can build a semi competitive clone in a day, a much more average team should be able to do it in a few weeks?

1477 chimed in saying that they tried to do it and weren't overly successful, so maybe less teams than I thought would be able to pull the copy off.

2826 did a copy of 2013 254, which was (lets be honest here) a ridiculously complex robot. I dunno what timeline that was, but probably a bunch of time.

Of course all these examples show that the copies aren't as good as the originals, but that was the premise from the start anyways, so not sure why we are still commentating on it. The point is that it could very well up their competitiveness.

So then what don't I like about it then? Seems like a win win?

In my experience in other programs (albeit easier to modify and lower budget) teams have this need to be all secretive of their designs and compete late. Someone brought up VEX competitions filling early september and october. Basically those aren't any of the good teams. And usually whenever a design gets revealed/leaked before world championships it means it was done against the team's will/it was an old design and they already have a much better robot. A good example of this is the New Zealand vex teams which are incredibly strong and consistently put out championship winning worthy robots. They don't stream besides like their national championship very late in the season. Robots and teams there feel very distant to me at least. It is worth saying that there are some regions with webcasts and such do stream, Ontario being one of the main ones. I think the culture is slightly shifting into more of a sharing one too, but i'm not too involved in the VEX community anymore to be a great judge of that.

Plenty of teams already don't reveal anything until they absolutely have to keep their competitive edge. eg 1114, 2056, 2826?, 1678? I'm sure there are others, and for 2826 and 1678, I don't actually know those are mostly guesses in general they tease/reveal later in the season. Not that I don't understand the reasoning behind it and see how it's advantageous, it's just that I really enjoy seeing the robots in a polished way like 118 and so many other teams that put out videos earlier. Removing bag day I think would cause more teams to do this.

Copying is smart. Why re-invent the wheel? I just don't love the secretive nature culture that results from it, and I would prefer FRC not to be like this.

All times are GMT -5. The time now is 05:37.

Page 12 of 13

« First

Show 100 post(s) from this thread on one page