I’m sure many of you are preparing for the rapidly approaching build season, and so are we here at 'Snow Problem! As part of our preparations, we are looking to get some amount of feedback about what kind of content (white papers, videos, CAD files, code) you would prefer us to release-- our goal, as it has always been, is to provide resources and “mass mentorship” to the community, and to provide you the best possible resources, we are hoping to focus our content creation on those that teams find most useful.
For reference, last year’s white papers can be found here on CD-Media, and our videos may be found on the Robot in 3 Days https://www.youtube.com/watch?v=
We’re looking forward to the 2018 FIRST Power Up season!
Hype robot reveal or robot walkthrough?
Are white papers useful? Are there any papers you would like to see?
What do you find interesting/useful in daily recap videos?
I personally find the robot walkthroughs and white papers integral to Ri3D’s usefulness simply because of the exact details provided, especially those pertaining to controls and motor/gearbox mounting configurations. Flashing lights and hype music makes for a good reveal video, but it isn’t really helpful in the context of build season.
I think there’s merit to both. The hype videos are crucial to get my design team’s attitude shifted from “how in the heck are we going to do this” to “this is totally doable”. The more thorough technical details help us actually design though.
If I had to make one request, it would be to show videos of the robot in action without cuts, from several angles and with a wide shot. We can learn the most from seeing how the robot actually interacts with a field and gamepieces in real time.
Another thing that would be useful is to highlight the robot’s flaws and where it messes up with actual game pieces. Then design teams can know what to avoid and what works and what doesn’t.
I’m not necessarily your target audience, but I have always found prototyping videos of mechanisms more helpful than any other aspect of 3iRD (including video of the finished robot.) If you could include videos of all of your prototypes in action (the good, the bad, and the ugly) with what you changed between each trial, that would be amazing.
Do some soul searching and ask yourselves what your goals for participating in RI3D are first. That’s my number one suggestion.
I think it would be great to see the RI3D teams spend more than a few minutes actually thinking about the strategic importance of the machine they are building and the features it has (fuel shooters, I’m looking at you…).
For the love of jibbers crabst - please work on centralizing all of the RI3D content and streams… I know you are a bunch of disparate groups but if no one can easily find it all in one spot then it becomes more detrimental than helpful.
+1 to this, I want to know what doesn’t work! Although that might cause us to dismiss ideas out of hand, so on second thought that may be a tossup.
I personally liked all the RI3D regardless of what they did, and I don’t really look to them for strategy. This year the collective failure (relative to the competitive reality) of ball shooting wasn’t great, but honestly I don’t think I would have done it anyway. RI3D consistently shows what the above-average team is capable of building in an entire build season anyway.
Wasn’t there a single youtube channel with all the RI3D reveals and things last year?
As someone who has been in and around the Ri3D machine for the last couple years: you’re not wrong. I think it took most people more than those first eight or ten hours to realize a Really Good gear system would be more valuable than a big passive gear pocket and a fuel shooter. (Credit where it’s due, WCP’s MCC got this balance closer to right in their video released early Week 2.)
For the love of jibbers crabst - please work on centralizing all of the RI3D content and streams… I know you are a bunch of disparate groups but if no one can easily find it all in one spot then it becomes more detrimental than helpful.
This is, admittedly, a thing I’ve noticed too. (And we didn’t exactly help in 2016 when the Digit board code stayed on the Garnet Squadron GitHub.)
What would be the most useful thing for a live stream, assuming that the stream will be something to set and forget?
So I’ve been involved/led an Ri3D team for 3 years now and I can say you never get enough time to strategize. It’s really hard to set aside a large portion of your time for strategy when there’s so much building to do. We’ve gotten to the point where we basically familiarize ourselves with the game and immediately begin prototyping mechanisms to perform each task. Then we take all our best prototypes and scramble to fit them all within the frame perimeter on the last day. Our approach leads to a robot that has effective/semieffective mechanisms for each element of the game, but not necessarily a robot that will be competitive from a strategic standpoint.
Our logic is that teams will have way more time to dig into strategy than we do, and that the best way we can benefit them is by getting as many prototypes on camera as possible. With that said, we could really do a better job on our daily update videos…
As for the centralization of Ri3D content, I think Dan and the original Ri3D team do their best to keep everything in one place. There is actually a registration process for Ri3D now (as there was last year) that helps to ensure that Ri3D teams are getting content to the Ri3D Channel on YouTube amongst other organizational things. The issue in my opinion is that Ri3D has blown up in recent years and there are more teams to follow. That inevitably makes it harder to keep things organized. I think it’s great for FRC teams to have a lot of content to look through, but I worry that if Ri3D keeps growing that quality content will start to get lost in the noise. Hopefully we don’t reach that point this year, but I like what you said in your first paragraph Marshall.
Ri3D’s purpose is to provide mass mentorship to FRC teams, not so a bunch of college kids can keep building FIRST robots. I think the teams that have participated so far have done a good job, and hopefully that continues as more and more Ri3D teams form.
Wow, thanks so much for the fantastic responses so far (and seriously, keep the feedback coming!). Part of the reason I made this thread is we’re doing a bit of soul searching for how we’ll be doing our build and documentation this year, so especially the critical feedback is **very **appreciated. A lot of this feedback is going directly into our ('Snow Problem) planning Slack, since most of our members don’t really keep up with Chief Delphi.
These in particular are suggestions that I’m definitely going to bring up with the 'Snow Problem group. I think you guys (and all of the comments along those lines from others) are hitting the nail on the head in terms of the focus we’ve been missing.
Definitely. I can speak for 'Snow Problem in that we really have a few guiding principles that we emphasize when we’re doing our strategy.
0: It doesn’t matter what we build if we don’t document and create excellent content to go along with it. That would be missing the entire point.
1: Build a robot such that the design would be reasonably competitive at a Minnesota regional.
2: Create a robot that is **inspiring **and cool-- be able to say “we built this cool thing in three days, what awesome stuff can you build in six weeks”-- by pushing the limits of what is possible to do in three days.
3: Build a robot that we can use for demonstrations for the rest of the year, especially to satisfy our primary sponsor, CSE Expo, which is an event where middle school students drive our robot.
I don’t think we’ve adequately explained these goals and how they shaped our strategy in the past-- in particular, 2 is basically the sole reason we’ve been pushing towards powder coating and CNC machining in 2017 (and likely will continue to do so in 2018). 3 is why, although we identified gears as being the 100% most important thing in the game, we built our robot around fuel. Frankly, it’s a cooler demo, and the kids have absolutely loved it, even if it was a poor competitive decision. I realize this isn’t necessarily the ideal way to be running, but like an FRC team, we have to keep our sponsors happy if we want to continue building robots.
Completely agree. Last year 'Snow Problem posted an aggregated spreadsheet of social media affiliated with each team, but it was kind of buried in one of our threads. I probably should actually send an email or two about that…
We’ll be keeping this in mind when we do our strategy process this year. We try to schedule our process approximately like we would an FRC build season-- from noon to midnight, we do strategy until approximately dinner time (6:00 ish), then split into prototyping and drive train build. We try to have a robot driving by the end of “week two”-- noon on Sunday, and have an iteration of the manipulators done by week four. We haven’t been entirely successful with this approach in the past, but hopefully we can better communicate what we’re doing to you guys so it makes sense what we’re doing and why we’re doing it.
It might be good for the Ri3D teams to point to where good information on how to strategize is, so that teams that may not know how can learn. Just like a list of good resources on the subject.
Along the lines of prototype videos/discussion, one of the areas I think could be expanded would be discussion of critical dimensions and features. It could be as explicit as detailed engineering drawings released on game piece mechanisms, to as a vague as “we found that we really want a lot of compression on [game piece] when using [X wheel].” Given that some of these Ri3D teams have access to more game pieces and field elements than teams do on the first weekend of build, finding out the characteristics of the game pieces and field elements and how they interact with the robots we’re building can be incredibly useful information to have. Going into detail about the iterations you made with your prototypes and what dimension/feature changes resulted in performance differences can be very useful for setting teams off in the right direction.
I really like the hype reveals, on an emotional level. The pitfall is that I can expect my students to react on emotion as well, leading to pigeonholing designs based on what was shown off as working or failing.
The hype shouldn’t be more than 30 seconds.
One interesting thing… When watching Ri3D with my team, sometimes I notice my students give off a “wow they built this awesome thing and we can’t even make ours work” vibe, likely linked to having “growth mindsets” and other educational concepts that I haven’t finished wrapping my own head around and tweaking our program to take into account & teach our students. That’s a pretty team-specific concern.
It would be legit to follow up a day or two later with “The Making Of A Hype Reveal Video” that shows off how a team can get the same (:ahh:) in their videos with a cell phone and some editing software and maybe one nice light. IIRC at least one of the Ri3D teams did something similar to this, with a bloopers/behind-the-scenes reel after the hype train rolled through.