Do you think a practice robot is needed to compete?

[techhelpbb]

Originally Posted by Wayne TenBrink

The early results of this survey demonstrate the gap between the "Chief Delphi" crowd and everybody else. What percentage of total teams build a practice bot? Its nowhere close to 80%.

The only way we can insure valid metric is with FIRST doing it.

I used to go around as CSA and, in addition to my duties, ask about which programming language and speed controls (also CAN or PWM) teams use. In that capacity I could get real data on what people are really doing and, as the CSA, I could see what's working and not working. This also made it possible to be ahead of an issue - if everyone with a particular part is seeing a problem I know where to go immediately if it becomes apparent.

So this is, at best, a barometer of the sentiment of the people who bother to contribute and should not be used in place of a proper study. However the fact it's not all inclusive doesn't remove the fact that there is a clear majority from those that do contribute at this time.

Also at this time only 8.63% see no value in a practice bot at all.
So 91.37% can find a good reason to do it or would consider it if they had the resources.

[AdamHeard]

It'd be nice to redo this poll w/ noting who is in districts and who isn't.

A lot of mid to top level CA teams build a somewhat decent practice bot, and I think they'd stop doing so under bag and tag.

[Ryan Dognaux]

We aren't in the district model and built a practice robot for the first time last year. We didn't manage our time as well as we should have and ran into rushing to complete the competition bot near the end of the 6 weeks. We also had quirks pop up on the competition robot that we didn't have on the practice bot.

The biggest thing it gave us was more drive team practice and the ability to iterate between events. Without it we wouldn't have been able to add can grabbers or practice with our ramp, both which were added mid-season.

We have already decided we will be building 2 again this year. Once we go to the district model we may not because of the un-bag windows.

[John Weissman]

Since we have been building practice bots for the past 4? Years, our results are much improved. Since we have used swerve since I and my sons joined 1640, using the practice bot has greatly improved our drivers ability to do the things necessary for each game. It also allows us to make any adjustments needed on doue ( the practice bot ) and install them on prime ( our competition bot ) during the season.

[GreyingJay]

For those that don't build a practice bot, do you have other "experiment bots" that you can use as a drive platform for testing mechanisms, drive trains, autonomous modes?

My team hopes to build a practice bot if for nothing else than to have something for the programming and controls teams to have something to experiment with in parallel with other stuff being developed. And then for next year to have something driveable during the pre-season.

[Amanda Morrison]

Originally Posted by Wayne TenBrink

The early results of this survey demonstrate the gap between the "Chief Delphi" crowd and everybody else. What percentage of total teams build a practice bot? Its nowhere close to 80%.

Let's be careful to separate speculation from fact, here. I don't know if it is close to 80%, but I doubt anyone does at this juncture. I was surprised by Collin's post - even though IRI has many upper-tier teams, it is not solely comprised of them. I would imagine that if FIRST expected to take any further action on this "extend the build season" proposal that they would ask the teams directly for their fact-gathering instead of relying solely on CD information. If FIRST did take that route, the results could vary as follows:

- regionally, in areas where there are fewer events vs. a greater number of competitive opportunities
- by team age, showing a greater number of rookies or 2nd year teams not (/not yet?) utilizing this strategy
- by enforced competition model, district areas vs. open competition areas
- by declared team sponsorship level/yearly budget amount

...and potentially by other socioeconomic or program-specific factors I'm not considering. We're also being pretty broad at what a "second robot" constitutes. Are we talking about teams that knowingly mill or punch a second or third set of pieces during the build season and assemble within the current 6 week window? Are we counting those who use scrap wood in the shop, but only after the robot has been bagged? What if the second robot doesn't have electronics, but they've mocked up a new arm on an immobile frame? Or what if the team continues building new appendages without a base?

My point is, we've got enough potential variables here and zero data aside from CD's poll as far as "second robots" are concerned, and even that has many different definitions. I've got a gut feel and an opinion myself, but I am sure everyone else does, too. I think the real question for FIRST is: how many teams put down the tools completely until their event? I'm not talking just CAD, I mean teams that have any physical work going on whatsoever.

If that number proves to be the overwhelming majority, what is lost by giving them the opportunity (but again, NOT the obligation) to continue building until their first competition and on throughout the FIRST season?

And with that question, we've once again returned to this thread.

[D.Allred]

Originally Posted by techhelpbb

Pretty clearly we should add some options to this poll I slammed together.
I don't see a way to change this on this forum. Is there a way?

I was more curious to see if there was a significant majority leaning towards wanting or needing a practice bot and so far there is good reason to look into this further.

The real answer is simple. Drive time makes you more competitive. Practice robots are one method.

Improving your design and fixing problems make you more competitive. Practice robots are one method.

There are several threads swirling around this same issue and the polls do not reflect simple available options. I feel a six week build plus an out of bag system for regionals would be the best fix for now. I would probably drop the practice bot and make incremental improvements during those windows.

Of course, I would not need a practice bot if the 6 week build limitation was gone. Our last regional is week 2. Assuming that we qualify for championship, I would take those practice robot resources and build a new machine.** Is that what we want as a community?

Personally I'm a fan of incremental improvement.

David

Disclaimer... this is my opinion and does not reflect my team's views.

** I understand trying to copy components of another design is not simple, nor will it automatically be successful.

[Peyton Yeung]

On 461, the past few years we have built a majority of a full practice bot but after this year with districts we will likely be building just one bot and spend more time on iterating and testing sooner. The unbag windows were so large and the time between events was so short that we hardly used the practice bot since we could just unbag the main bot and practice/improve.

[GeeTwo]

Originally Posted by techhelpbb

The only way we can insure valid metric is with FIRST doing it.

Not necessarily. First of all, let us recognize that even if FIRST compiled these metrics, they would have to be collected based on what teams say about themselves. So, all we need to do is come up with a way to ask every competing team the same set of questions. If we came up with a standardized set of questions, and sent someone "pit scouting" each event, we could compile a valid data set. I'm fairly confident that there are no regionals, and few district events which are not represented by someone on CD who could either do this pit scouting, or recruit someone to do so.

I've created this thread to try to generate such a canvas/questionnaire.

[AllenGregoryIV]

My favorite reason to build a practice bot is because we get to hang out with our friends on other teams more. During the competition season I normally attend three-four regionals (2 competing with 3847 and 1-2 as a volunteer). The other 3-5 weeks we are regularly practicing with our friends in Houston. Often it's at 118's facility but last year we went to 624's home base and 5414's shop on different weekends. We have also been up to visit Texas Torque as well and practice with them. Those days are honestly some of my favorite days of the entire year. 2-6 teams just driving around having fun and working to improve during the season. You can't have that experience with out a practice bot (or at least most teams don't).

[MrForbes]

Allen, that sounds like a great reason to build a practice bot.

[Jared Russell]

Originally Posted by GreyingJay

For those that don't build a practice bot, do you have other "experiment bots" that you can use as a drive platform for testing mechanisms, drive trains, autonomous modes?

My team hopes to build a practice bot if for nothing else than to have something for the programming and controls teams to have something to experiment with in parallel with other stuff being developed. And then for next year to have something driveable during the pre-season.

We build a practice bot, but we also find lots of value in building an "experiment bot" out of an old chassis. For example....

(There was also a 2015 version of Skystalker, but it was far less comical to watch).

If you are lucky/guess right/use a similar chassis and drivetrain every year, there is nothing to prevent you from having a rolling base ready to go at kickoff. This (along with the ability to have design, fabrication, and driving experience carry over from season to season) is a huge reason why many successful teams do not tend to deviate a lot in drivetrain design from year to year. A "free" facsimile on day one!

[Jared Russell]

Originally Posted by MrForbes

Allen, that sounds like a great reason to build a practice bot.

One of my favorite practice bot anecdotes was in 2014, a week or so before World Championships. A bunch of California teams brought their practice bots to our field to practice. We invited some other great California teams (who unfortunately had not qualified for the Championship) to bring their now-retired robots to help fill out the field. Our drive teams all scrimmaged together for hours. They went over every possible scenario, practiced beating defense, inbounding, perfecting positioning - you name it. (It must have paid off, since there wound up being 3 Einstein teams and one Division Finalist in the building that afternoon).

But even disregarding the future on-field success, I thought it was really cool. It was a unique bonding experience for the teams in the region. Hopefully we will do the same thing in 2016 (2015's game didn't have quite the same dynamic, so most teams could practice alone).

[Mark Sheridan]

Originally Posted by AllenGregoryIV

My favorite reason to build a practice bot is because we get to hang out with our friends on other teams more. During the competition season I normally attend three-four regionals (2 competing with 3847 and 1-2 as a volunteer). The other 3-5 weeks we are regularly practicing with our friends in Houston. Often it's at 118's facility but last year we went to 624's home base and 5414's shop on different weekends. We have also been up to visit Texas Torque as well and practice with them. Those days are honestly some of my favorite days of the entire year. 2-6 teams just driving around having fun and working to improve during the season. You can't have that experience with out a practice bot (or at least most teams don't).

Yep, great point. it might the best thing about practice bots with a good game. If we really like the game, why not play it more? Hopefully a few more teams in our area will join us this year.

[Gdeaver]

Besides practice and development with the practice bot, we use the practice bot as a source of spare parts. It's stripped and then put back together for each competition.