Robot in 3 Days : Feedback

[Steven Donow]

Originally Posted by colin340

I'm with Lavery on this one, hold off until week 2 so crazy fringe ideas and non-conventional strategies have time to grow legs. it's hard for super unpolished ideas to be debated against polished, proven, videotaped ideas.

RI3D I appreciate the passion you have for pushing first to a higher level I'm just concerned that too much of this in week one will lead to incredibly uniform robots and strategies.

The "dillema" I see with pushing Ri3D back from Kickoff weekend is then, in some ways, it's not truly a "Robot in 3 Days" from a design standpoint. I don't see any way that, if the Ri3D team is exposed to the game prior to starting the 3 days, that they'd NOT be thinking about what to do.


Some could argue that this would be good, giving the team more time to develop what they're defining as the MCC(if upper-level MCC is what Ri3D intends to make), but on the other hand, it defeats the intent of Ri3D(if I'm interpreting their intent properly) of being, "build season in 3 days".

Also a possible downside to not exposing your team to Ri3D until later in the season is, what if, upon seeing that what Ri3D does is drastically different than how your team interpreted the game, your team feels as though they've been doing the game "wrong", thus possibly derailling the design process even further.

[Lil' Lavery]

They could still do the project over the first three days of build season, but simply not post the results online until later on (midway through week 1 or week 2). That would also give them time to prepare plenty of supporting documentation and fully explain their processes and iterations.

My gripe doesn't only necessarily come from my team, though I used it as an example. How many non-wheeled frisbee shooters did you see last year? I only know of 1024 and 1503. Everyone else was using the wheeled launcher design. I don't know of a single Nerf-inspired frisbee "flicker" or rotating arm flinger. I know a handful of teams took the time to prototype other concepts and didn't like their results, but the bulk of teams went immediately into the wheeled shooter concept demonstrated by RI3D.

Don't take my comments to mean I don't like RI3D, but I think delaying the release some could lead to more variety across FRC. The right-after-kickoff nature of the project leads to a tremendous amount of influence over FRC teams, perhaps an unhealthy amount so. I recognize the potential to allow it to help raise the bar of many mid-level teams, but I also don't want it to be used as a crutch.

[techhelpbb]

Originally Posted by Lil' Lavery

Don't take my comments to mean I don't like RI3D, but I think delaying the release some could lead to more variety across FRC. The right-after-kickoff nature of the project leads to a tremendous amount of influence over FRC teams, perhaps an unhealthy amount so. I recognize the potential to allow it to help raise the bar of many mid-level teams, but I also don't want it to be used as a crutch.

Has anyone considered the decade of FIRST before RI3D?
What is the data to support the idea that the RI3D was even the influence?
How do we know it's not just that the use of the Frisbee are the cause?

Not being argumentative, just curious to see the supporting evidence.

[James1902]

I think the argument could be made that we would have seen a majority of wheeled shooters this past season even if Ri3D didn't happen. It was the most efficient way most teams could shoot discs and there were already a number of videos showing the concept posted from non-FIRST related projects before kickoff.

[Oblarg]

Originally Posted by Lil' Lavery

They could still do the project over the first three days of build season, but simply not post the results online until later on (midway through week 1 or week 2). That would also give them time to prepare plenty of supporting documentation and fully explain their processes and iterations.

My gripe doesn't only necessarily come from my team, though I used it as an example. How many non-wheeled frisbee shooters did you see last year? I only know of 1024 and 1503. Everyone else was using the wheeled launcher design. I don't know of a single Nerf-inspired frisbee "flicker" or rotating arm flinger. I know a handful of teams took the time to prototype other concepts and didn't like their results, but the bulk of teams went immediately into the wheeled shooter concept demonstrated by RI3D.

I do not see this as a bad thing, necessarily. Cribbing designs that work is an essential part of good engineering.

Delaying the release of Ri3D information would remove a very, very important early resource for inexperienced teams. I do not think the marginal benefit of a possible increase in robot variety (I'm not entirely convinced of the magnitude of this effect on well-established teams, at any rate - I know 449 prototyped a very wide variety of designs) compares to the utility of having that information out there for teams which it benefits greatly.

Anecdotally, I know for a fact that 4464 would not have had the success we did last year without Ri3D. We had a working shooter prototype by the second weekend - it was the only prototype mechanism we were able to build, due to budgetary constraints. Without Ri3D, we would not have had a tested, feasible design to attempt, and as a team we simply could not afford to try things which we did not know would work.

[Spatel7793]

If you guys want to have a completed robot, I suggest just focusing on building the actual robot and less about the videos. Honestly, if you want to share, vine vids are a quick way to do so, coupled with tweets. Quick, simple, and you're not wasting manpower by sitting in front of a computer, dealing with setup for your stream and whatnot.

Honestly, I think Ri3D is a good concept, and while I don't endorse copying other peoples' designs, it is definitely a good place to start. There were definitely a good amount of rookies who seemed to know what they were doing whereas previous years, a "turtle bot" was the way to go to show that you could at least drive. Definitely made the game more interesting.

Once again, not saying anyone copied directly (because wheel based shooters hands down were probably the best way to go with this game).
There were plenty of changes that would have been made to Ri3D's design.

Didn't read through all the comments, but I saw that someone mentioned to release the footage sometime at the end of week 1 or week 2. I agree with this. It gives teams time to think about how they want to approach the problem themselves, and Ri3D would just be either another way of looking at it, or a way of executing it.

In summation: I think Ri3D is a great event. Building the robot would be more efficient with tweets and vines during the project and releasing a final documentary style video later.

[Michael Corsetto]

Originally Posted by Lil' Lavery

My gripe doesn't only necessarily come from my team, though I used it as an example. How many non-wheeled frisbee shooters did you see last year? I only know of 1024 and 1503. Everyone else was using the wheeled launcher design. I don't know of a single Nerf-inspired frisbee "flicker" or rotating arm flinger. I know a handful of teams took the time to prototype other concepts and didn't like their results, but the bulk of teams went immediately into the wheeled shooter concept demonstrated by RI3D.

I see your point about how RI3D could limit variety in design. However, how many non-wheeled shooters did we see in 2012, the pre-RI3D era? I can think of 16 and was it 1523? Also a team from Israel used a tennis racket. That's all I can think of, maybe I'm missing a ton. Can we really say RI3D was to blame for lack in shooter variety?

A lot of teams used 2006 and 2009 as their base for designing their 2012 bot. I know the frisbee was a new game element, but like James said, there were already videos of wheeled frisbee shooters online, and more videos from other FRC teams by the end of week 1. It seems like whatever is out there, teams will use. If not RI3D, something else.

In your team's design process, are you against looking at online videos/testing/data, and instead rely mostly on in-house designs? Or is that your goal at least? Sounds like RI3D was a point of contention/frustration in your team's design process last year. Just curious.

Originally Posted by Lil' Lavery

Don't take my comments to mean I don't like RI3D, but I think delaying the release some could lead to more variety across FRC. The right-after-kickoff nature of the project leads to a tremendous amount of influence over FRC teams, perhaps an unhealthy amount so. I recognize the potential to allow it to help raise the bar of many mid-level teams, but I also don't want it to be used as a crutch.

You use a certain set of skills by ignoring outside data and brute forcing through a problem yourself. You use another set of skills by blindly copying what is available on the internet. You use another set of skills by taking outside data, synthesizing it, and utilizing the data to improve your designs.

Some would call using RI3D a crutch. Some would call it smart. I fall with those in the second camp. But then again, I'm a terrible engineer and copy everyone else all the time

-Mike

[Joe Ross]

Right now, Robot in 3 Days is a powerful marketing tool to showcase the rapid prototyping products made by iR3 Creative Engineering and the other sponsors. With the robot done in 3 days, teams can see all the cool products used, and still have time to purchase them and use them to prototype their robot this year. If a delayed release was done, teams wouldn't have chance to design those products into the current year's robot.

[Mr_K]

As a 15-year participant in FIRST I am impressed with "Robot in 3 Days" ability to product a robot as they do.

I am sadden to see all the FIRST students viewing their site wanting to copy this concept instead of having to think to develop their on their own robot concept. I thought the purpose of FIRST was to show students HOW TO apply STEM in their thought process.

Giving student the answer up front does not really help them to think outside the box.

I do agree that many new teams and teams with limited technical support and resources find what is presented as a "God sent". I will agree it is.

I would like to see the "Robot in 3 Days" roll some STEM into their presentation so students can see how and why the robot was built the way is was.

[Lil' Lavery]

Originally Posted by Michael Corsetto

I see your point about how RI3D could limit variety in design. However, how many non-wheeled shooters did we see in 2012, the pre-RI3D era? I can think of 16 and was it 1523? Also a team from Israel used a tennis racket. That's all I can think of, maybe I'm missing a ton. Can we really say RI3D was to blame for lack in shooter variety?

A lot of teams used 2006 and 2009 as their base for designing their 2012 bot. I know the frisbee was a new game element, but like James said, there were already videos of wheeled frisbee shooters online, and more videos from other FRC teams by the end of week 1. It seems like whatever is out there, teams will use. If not RI3D, something else.

I remember at least a couple catapult or other weird launcher designs at each event I went to in 2006 and 2009. One of my alliance partners at Peachtree in 2006, 1139, used a pneumatic piston to launch balls (they also made it to Einstein that year). 384 and 1086 both built catapults in 2009, and faced each other in the finals at VCU.

2012 saw less, that's for sure. 16 and 2474 are the most notable I can think of in terms of shooting from the key (and 2474 had a wheeled shooter for barrier shots). But we still did see plenty of "dumping" and "dunking" mechanisms for 2-pt shots that didn't rely on wheeled shooting. I was hoping to see more mechanisms specially crafted to score on, say, the pyramid goal (like 1024). Wheeled shooters were likely the optimum design for the "cycling" and full court shooting robots, but there were other strategies that would have suited other shooting mechanisms that were left mostly unexplored.

Originally Posted by Michael Corsetto

In your team's design process, are you against looking at online videos/testing/data, and instead rely mostly on in-house designs? Or is that your goal at least? Sounds like RI3D was a point of contention/frustration in your team's design process last year. Just curious.

No, not in the least. But, like most good teams, we want to arrive at a design driven by our approach to game strategy and consider all opportunities. Outside influence is very important in our design process, but it's not the initial step. Thus the immediate release of RI3D tainted our brainstorming process some, with a large portion of it almost immediately following along those lines (aside of a couple wacky climber ideas that almost certainly wouldn't have worked). It wasn't at all a point of contention or frustration for the team as a whole, but it didn't allow for pure design brainstorming as much as I would have liked. Research has a lot of value in engineering design, but it can also stifle creativity if timed poorly.

Originally Posted by Michael Corsetto

You use a certain set of skills by ignoring outside data and brute forcing through a problem yourself. You use another set of skills by blindly copying what is available on the internet. You use another set of skills by taking outside data, synthesizing it, and utilizing the data to improve your designs.

Some would call using RI3D a crutch. Some would call it smart. I fall with those in the second camp. But then again, I'm a terrible engineer and copy everyone else all the time

I didn't mean to say that RI3D was a crutch to every team, but it is certainly a crutch to many. If used responsibly it can be beneficial with minimal side effects, but that is not always the case.

[AdamHeard]

I think the potential distraction to established teams is a minor side effect that is well worth the huge potential gain for newer teams and teams without mentors.

We can all idealize and say teams should follow a perfect design process that is posted in presentations, but many teams have no mentors, no experience, etc... If they get a functional scoring machine because they copied Ri3D then that's AWESOME. That certainly puts them in a better place, both in terms of experience and morale, to grow for the future than a robot that just drives.

In summary, release it as early as possible.

[Ryan Dognaux]

Originally Posted by AdamHeard

We can all idealize and say teams should follow a perfect design process that is posted in presentations, but many teams have no mentors, no experience, etc... If they get a functional scoring machine because they copied Ri3D then that's AWESOME. That certainly puts them in a better place, both in terms of experience and morale, to grow for the future than a robot that just drives.

This x1000. I think we forget just how many teams would show up to regionals with maybe a running drive base a few years ago. I saw much less of this last year and I think Ri3D was the primary factor in this. Release everything as you did last year and if teams don't want to be influenced by it then they should just not view it.

[techtiger1]

Life was better without all the COTS stuff out there. This forced teams to actually design and think as well as budget for a robot. Now I feel like teams do certain things because they already know it works or have it on hand. The idea of bringing up each teams level of play is just putting it nicely. We all know who the great teams are and at the end of the day, they just want more capable robots to choose from in alliance selection in the second round. Robot in three days is awesome, they should keep it up, and can/ will release the robot whenever they feel like it.

[Ryan Dognaux]

Originally Posted by techtiger1

Life was better without all the COTS stuff out there.

You and I have very different definitions of better. I like knowing that I can integrate an existing gearbox into our mechanism and have confidence that it will just work. It lets us spend more time designing and iterating our mechanisms instead of designing a gearbox to properly power it. This is only one example, but if we can use COTS parts, we probably will. Time is the most valuable resource during build season.

[AdamHeard]

Originally Posted by techtiger1

Life was better without all the COTS stuff out there. This forced teams to actually design and think as well as budget for a robot. Now I feel like teams do certain things because they already know it works or have it on hand. The idea of bringing up each teams level of play is just putting it nicely. We all know who the great teams are and at the end of the day, they just want more capable robots to choose from in alliance selection in the second round. Robot in three days is awesome, they should keep it up, and can/ will release the robot whenever they feel like it.

I'm not sure what point you're really making. That COTS items are bad? That good teams only want other teams to be good so they can serve themselves by picking better teams?